Rizwan Virk, the Founder of Play Labs @ MIT and author of The Simulation Hypothesis explores how the current crisis and VR tech might affect long term social trends. There’s a memorable scene in the 1993 movie , where Sylvester Stallone’s character is asked by Sandra Bullock’s character if he wants to have sex.  Since he has just been awakened into an unfamilair future after being in suspended animation for decades, he eagerly says yes.  She then hands him a headset and says let’s start, leaving him more than a little confused.  She explains that that sex is now done without any physical contact and without “exchanging any bodily fluids” because that was what led to the downfall of western civilization! Demolition Man Figure 1: In the movie, Demolition Man, Sandra Bullock and Sylvestor Stallone had a virtual encounter While they are unlikely to be as extreme as depicted in science fiction, it’s very possible that the recent outbreak of the corona virus (COVID-19), will cause behavior shifts in our society that will end up affecting us much longer than the duration of the outbreak itself.  And these long term trends will be impacted by the capabilities of our tech. Fear of the virus is already affecting businesses in the US and around the world.  Anyone who had planned to attend a conference in person in the next month has probably seen their conference cancelled, including SXSW (South by Southwest), GDC (the Game Developers Conference), Google and Facebook developers conference, and so many others. Some of these conferences are scrambling to put some content online.  UPDATE: the NBA has just cancelled the rest of the season too. But travel to conferences isn’t the only thing affected. I was at a company recently where they told me they were not sitting next to each other during meetings in their conference room, but leaving an empty seat. Figure 2: Behavioral changes are starting ... where will they lead? I suppose it’s all part of the “social distancing” that is leading to changes in social behavior. While the handshake has been around for thousands of years, it hadn’t really become a default greeting until the nineteenth century and even then only in the west. Recently, handshakes are being replaced by fist bumps, elbow or even foot bumps.  Some are suggesting that Spock’s Vulcan greeting, holding up fingers in a V (and saying “Live Long and Prosper”) will or something similar that requires no contact will become the new normal before too long. Stanford, Columbia, Harvard and MIT are a few (among many) Universities that have cancelled in person classes, replacing them with online lectures.  More than that, even though young people are less likely to be affected by the virus, MIT is sending students home – essentially kicking them out of their dormitories! Figure 3:  Classes move online for many universities, including Stanford Life here in Silicon Valley is already being affected.  Large tech companies like Google and Facebook have already told their employees to work from home.  Coffeeshops, the favored place to work for many budding entrepreneurs and freelancers, are markedly less crowded and keeping shorter hours, though this is being offset by the number of people who aren’t going into the office and working at coffeeshops instead. Recently, I wanted to go see an old movie at the Stanford Theater in the heart of Silicon Valley. It is a1930s era classic theater that shows old movies, complete with an organ that is played in between films.   It turns out they had cancelled all their films out of an “abundance of caution”, not because there were any specific cases. Will other movies theaters and public places follow suit?  Of course we are already experiencing a serious disruption in the near term.  The question I wan to explore is how will the changes brought upon by this virus (and perhaps even more deadly viruses in the future) affect social interaction using technology (particularly VR) in the long term? In the Starz network TV show , we were shown a parallel version of Earth which had been ravaged by a virus. When individuals from our “world” went into this parallel world, they were surprised at how empty the public spaces were, even years after the virus had disappeared.  The survivors of the virus, it seemed, had changed their social habits, and these habits stuck.  This was in stark contrast to “our world” where restaurants and other public spaces were full of people.  Could we see a similar aftermath here? Counterpart In the short term, conferencing technology companies are enabling alternatives.  Companies like Zoom and Webex and other teleconferencing and remote working solutions, not to mention distance learning solutions, will reap the benefits as more and more people cancel in person meetings, classes and conference and opt to do things remotely. Welcome to the Virtual World So what’s virtual reality (VR) got to do with all this? When Facebook bought Oculus in 2012 (for $2 Billion), it was a shocker and many people suddenly bought into the idea that VR was going to become the “next big thing” for gaming, education, social networks, business travel.  HTC soon came out with the Vive headset, Samsung with its Smartphone based VR, called Gear VR (also based on Oculus technology) , and Sony with its Playstation VR. Suddenly, VR content companies were being funded and seemed to be flourishing, making everything from VR games to VR movies to VR exercise and VR social networks and VR business meetings, etc. Only they weren’t really flourishing.   To quote a friend of mine who worked at Facebook in those years – Facebook’s own internal studies found that people only wanted to use VR very rarely – maybe once a week. If you compare to how often people used their mobile phones, this was a bit of a disappointment and signaled the coming decline in VR startups. Suddenly,consumer facing VR startups, many of whom had raised quite a bit of funding from 2013-2106, found that it was hard to raise for VR content and they started to fold. With the failure of the few VR social networking apps like Altspace VR, etc., the dream of VR, which was shown in the 2018 fim by Steven Spielberg, , died an untimely death. Ready Player One In , the physical world is depressing (as a result of some tragedy that isn't fully explained).  As a result, the favored way of doing is to put on your lightweight VR headset and enter a VR MMORPG. Ready Player One anything the Oasis, Instead of going to schools, students would attend virtual classrooms.  Instead of playing sports in real life, everyone exercised in VR, and they could be anywhere from the comfort of their homes.  Relationships were virtual, using an idealized version of ourselves as avatars. And although they didn’t get into it in the PG-13 rated movie, you could extrapolate that like in Demolition Man, an all pervasive virtual world was likely to have virtual sexual encounters! Figure 4: In Ready Player One, the Oasis VR was used for most social interactions It wasn’t the first time a vision of people escaping to the virtual world seemed like it might happen but didn’t.  When Second Life was introduced, there were numerous articles on how people lived, played, and worked in the virtual world, had relationships and even kids.  But that petered out as well, though Linden Labs, the company behind the virtual world released a VR version of the world in 2017, which didn’t really take off either. Game Over, right? The Resurgence of VR in these times? Surprisingly, even before the corona virus hit in early 2020, VR was on a comeback track. With the successful release of the cheap and standalone headsets like Oculus Quest in 2019, and the rise of corporations using VR applications for training (it is much cheaper to train employees in VR),in Silicon Valley and elsewhere prospects for VR were optimistic for the first time in years. In fact, Facebook had manufactured enough headsets for a modest demand; in stead, the Quest was sold out for months (and remains so to this day at many retailers). Amidst this resurgence, Facebook was also already working on its own version of the Oasis, called Facebook Horizon, which was set to go into closed beta in early 2020.  As of the writing of this article, no specific date for the release of horizon has been set, but you can be sure that they are looking closely at the need to hold conferences in different ways. One interesting tidbit with all of the cancellations was that HTC, makers of the Vive headset, announced that instead of having a physical developer’s conference, would hold their conference in – virtual reality!     They will use VR Education’s ENGAGE platform.  And it seems like the death of AltspaceVR, which was written about extensively, was exaggerated, and they are not only around, but getting requests to host events in VR! And it makes sense.  One of the biggest reasons that people go to conferences in any industry is not just to watch the presenters, but to meet other people in their industry who they don’t see in person very often. Facebook hasn’t announced any VR component of its cancelled F8 developer conference yet, though they are promising lots of online content. While online conferences are great just for watching content, they don’t provide the visceral ability to interact with other people, and VR may just be the solution for this.  With the release of Horizon, expected this year, you can bet that Facebook will pushing for virtual world conference and interactions in the near future. Figure 5: Will Facebook Horizon be the first VR world on the road to the Oasis? As more and more schools around the country cancel classes for a few weeks, it’s very possible that contingency plans will need to be made for longer term closures, and with the release of Horizon and other VR-meeting platforms.  When I was running Play Labs @ MIT, one of our companies was creating science lab VR sims, which could be used to allow students to run experiments that would be too expensive for the schools to have in physical labs.  Could the same concept be expanded to all labs? As the fear of going to public spaces increases, will people start to avoid the gym? Just like in the Oasis, will VR be used to exercise at home.  In fact, one VR company that provides a VR set for biking, VRZoom (a company that i'm an investor in), has told me the most convincing way for them to show their product to investors is to do the presentation inside virtual reality! Coming back to my desire to go to a movie theater to see an old movie, one of the more unexpected popular activities in Second Life was to have avatars sit in a movie theater and watch a movie on a virtual screen! At one point, the Netflix VR app, which provided a similar experience of sitting in a virtual theater watching any of the Netflix programming, was the most popular app in the Samsung Gear VR (and presumably, the Oculus) app stores! Many companies have cancelled all non-essential travel. Most large tech companies, including Google, Twitter, and Facebook, have told many employees that the can work from home.  Using zoom or skype or slack is great for messaging, but it just doesn’t have the same impact when you need to have a meeting with colleagues that you might know, and those that you don’t know. While the debate thus far has been on whether the media and the government is over-reacting or under-reacting to the Corona virus, perhaps what we should be watching is the change in consumer and social habits that will go with the aftermath of this (and potential future) pandemic. As people spend more time at home, forgoing gatherings with lots of other people, in addition to the teleconferencing providers, delivery services like Amazon, will no doubt benefit in the short term. It’s also very possible that this may be the behavior change will prompt people to put on their VR headsets rather than venturing out to the phsyical world. The virtual world may end up being the best way to interact, especially with idealized versions of ourselves as our avatars. And that was the promise of VR all along: that it’ll be more fun (and much safer) to use VR for everything from going to school, attending a conference, and perhaps even virtual relationships! Welcome to the future.

Amazon

Facebook

Funding

Google

Netflix

PlayStation

Samsung

Slack

Sony

Twitter

2022 - HackerNoon Contributor of the Year - Augmented Reality

Read free chapters of my books!

Nominated for 2022 - HackerNoon Contributor of the Year - Augmented Reality

Will the Corona Virus Aftermath Lead to a Resurgence in Virtual Reality?

Cosmic JS just released their new Extension functionally, which enables anyone to embed their own custom app inside the Cosmic CMS! In this tutorial we’ll be going through some of the steps required to make our Amazon Product Search Extension. TL;DR View the full codebase on GitHub Install the Extension in your Bucket in Your Bucket > Extensions > Browse Extensions for more information on building an Extension View the Extension documentation The extension we’re building We’re building a simple Extension that allows you to search the Amazon Product Catalogue and add Objects to your bucket with products’ name, description, image, and an affiliate link. This is quite a simple app, built using react, and we won’t be going into too many details on the parts that don’t directly relate to the new Cosmic JS Extensions API. Getting access to the Cosmic API in your Extension Cosmic JS embeds extensions as an inside its CMS interface, the extension is given a url with a querystring providing all the information needed for your app to access the relevant bucket like so: <iframe> https://43d32000-5ce7-11e7-8fc4-c1f6eec4f920.cosmicext.com/?bucket_slug=my-bucket-slug&read_key=foo&write_key=bar Extracting these variables, is very easy, and even easier using a library like qs: qs ; { bucket_slug, read_key, write_key, } = qs.parse( .location.search.slice( , )); .log({
   bucket_slug,
   read_key,
   write_key,
}); import from "qs" const window 1 Infinity console These keys can then be used to make API calls on the bucket your currently using the extension in: fetch( ,
   opts,
); `https://api.cosmicjs.com/v1/ /object/amazon-credentials ` ${bucket_slug} ${
      read_key
         ? + read_key
         : } "?read_key=" "" In our app we use these keys to add new object for Amazon Products that we find. extension.json One file that must be included in your extension is the file. This tells Cosmic important metadata about your app, like it's name, icon, display image, and the default Objects and Object Types that should be created when the Extension is first installed. The for our app is as follows: extension.json extension.json { : , : , : , : [
    { : , : , : , : [
        { : , : , : , : },
        { : , : , : , : },
        { : , : , : , : }
      ]
    }
  ], : [
    { : , : , : , :[
        { : , : , : , : , : , : },
        { : , : , : , : , : , : }
      ]
    }
  ]
} "title" "Amazon Product Search" "font_awesome_class" "fa-shopping-basket" "image_url" "http://fla.fg-a.com/shopping-cart/shopping-cart-black-3.png" "objects" "title" "Amazon Credentials" "slug" "amazon-credentials" "type" "amazon-credentials" "metafields" "key" "amz-key" "title" "Key" "type" "text" "value" "" "key" "amz-secret" "title" "Secret" "type" "text" "value" "" "key" "amz-tag" "title" "Tag" "type" "text" "value" "" "object_types" "title" "Amazon Products" "slug" "amazon-items" "singular" "Amazon Product" "metafields" "key" "image_url" "type" "text" "value" "" "title" "image_url" "parent" false "children" null "key" "affiliate_link" "type" "text" "value" "" "title" "affiliate_link" "parent" false "children" null The field is the name that your extension will display inside Cosmic The field must be a valid icon name the field is a display image that will show with your extension. title font_awesome_class Font Awesome image_url You also have space to specify a list of objects that will be created when your extension is installed, here we’ve detailed an object that will store our Amazon Product Search credentials. This means that Cosmic will handle saving this information for us! Deployment and Continuous Integration Uploading your extension to Cosmic is very easy, you just need to take a folder that contains you file, and an file (and any other assets you need), zip it up, and upload it. extension.json index.html Our build folder looks like this: build
├── extension ├── index └── static
 └── js
 └── main .json .html .a651cd8a .js You can either upload your zip file through the Cosmic JS CMS, or by using the rest API. We use Bitbucket to host our code, which has a built in CI service called Pipelines. You can configure pipelines to run on every commit to the branch of your git repo, and to upload the newest version of your app to Cosmic's servers. master We’ll configure our Bitbucket Pipeline like so: image: codogo/pipelines-universal:latest pipelines: branches: master: - step: script: - yarn install - yarn run lint - yarn run build - ./uploadNewExtension.sh Where contains: uploadNewExtension.sh EXTENSION_NAME= PREVIOUS_EXTENSION_ID= ) null There is no extension on cosmic to delete Found id : deleting old extension... https://api.cosmicjs.com/v1/ /extensions/ {\ -H > /dev/null curl --progress-bar --url --header --form --form > /dev/null #!/bin/bash # $BUCKET_SLUG, $READ_KEY, and $WRITE_KEY are environment variables " " ${EXTENSION_NAME:-Amazon Product Search} echo "geting previous version extension id..." " ) | .[0].id" $( curl --progress-bar "https://api.cosmicjs.com/v1/extensions-2?hide_metafields=true&read_key=foo" | jq -r ".bucket.extensions | map(select(.title == \"$EXTENSION_NAME\" ) "
# jq is a neat little program for extracting data from json, it is available here: https://jqplay.org/
if [ " $PREVIOUS_EXTENSION_ID " == " " ] ; then
echo " '$EXTENSION_NAME' "
else
echo " for '$EXTENSION_NAME' $PREVIOUS_EXTENSION_ID "
echo " "
curl --progress-bar -X DELETE " $BUCKET_SLUG $PREVIOUS_EXTENSION_ID " -d " "write_key\":\" \"}" $WRITE_KEY "Content-type: application/json" fi echo "upload new version of extension..." "https://api.cosmicjs.com/v1/ /extensions" $BUCKET_SLUG "content-type: multipart/form-data" "write_key= " $WRITE_KEY "zip=@./build.zip" This script will remove any previously uploaded versions of our extension, then take our zipped Extension and upload it to our bucket. Summary In this tutorial we’ve had an introduction to the way Cosmic JS hosts and integrates with your Extension, and an explanation of how to upload your Extension using Bitbucket’s Continuous integration. If you’ve learnt something, please ! share this article If you’re making an extension, or anything else, with Cosmic JS or , we’d love to see what you’re making. get in touch on our Slack Twitter This post was written by , an award-winning digital agency with a passion for creating amazing digital experiences. Codogo This article was originally published on . the Cosmic JS Blog

Amazon Product Search

Fetch

cosmic js

Read My Stories

Building and Publishing a Cosmic JS Extension Using Bitbucket Pipelines

Product engineer is a specialisation within the career of software engineering where the engineer has put focus and effort into building out their capability of delivering products. I often get asked how this differs from a “normal” software engineer, why it’s important and “how do I become a (better) product engineer?”. This post covers the why, what and how of the product specialisation for engineers. Why product engineers matter If you’re developing products, then as a rule software engineering or some type of engineering is needed at some point in your production path to create your products for sale to your customers. If you’re building tech products then there is a good chance engineering is a significant part of the time, cost and effort involved in shipping product. This means that it makes sense to have people that are both knowledgeable and passionate about working in this type of environment and towards the outcome you’re looking for (successful products). They can bring their knowledge to the translation between the technical and engineering world through to the product world, which ultimately translates into your customers world and the financial success of your organisation. In this way, it’s often good to have a team with 1-2 engineers with a product specialisation paired with engineers that may be more technically specialised. You can have business, sales, product management and marketing people with all the product capability in the world but if the engineers are too technically focused you’ll find challenges having the software you develop embody your ideas within your commercial constraints. Sometimes what seem like small technical nuances can turn into serious product issues. A product minded engineer will be able to balance the technical with the product considerations. Ideally yes but in reality people have different strengths and interests. In an organisation of any reasonable size you’re better off playing to people’s strengths (it’s kind of what our entire economy is built on – specialisation). Time spent researching product related practices for engineering is time not spent researching the latest in self-training machine learning algorithms. It’s too hard for someone to cover everything and you usually need a high performing team not a set of rockstars. So, why isn’t every engineer a product engineer then? Shouldn’t we make every engineer product minded? Side note: Look at most sporting teams that win consistently, they usually have stars but everyone acts as equals. If you’re an engineer, why specialise as a product engineer? To the engineers reading this, treat product as a specialisation just like any other. Product, team lead, AI and ReactJS front end are all specialisations that you can pursue. At the very least, getting a base level understanding of product will set you up to function better as a product development team member. You will come to understand why decisions are being made, whether the team is following appropriate practices and probably gain some empathy with your customer and other stakeholders. At the very best, you’ll be an indispensable member of a product development organisation. You’ll be able to see shortcuts that make people say “wow, they delivered what the customer needed without the customer knowing they needed it”. You’ll become relied upon by your non-technical colleagues to find solutions that span technology, product, customer and timeline. You’ll likely understand the customer more and gain a different appreciation for your job. You’ll probably become a welcome mediator between your more technical colleagues and your less-technical colleagues. You might even be in a good spot to transition into a more product focused role (like product ops, product management or product analytics). What is a product engineer? Just like someone might specialise as a or they may also specialise as an engineer that excels at building products, a . back end  .NET Engineer mobile Android engineer product engineer The product engineer specialisation is no different to the more technical specialisations. As a product engineer you also focus on learning particular patterns, techniques and technologies, just in this case where they are relevant to product development specifically. I have seen product engineers pitted as an either/or to being a software engineer but a product engineer is just a sub-specialisation. You can specialise in both being a .NET engineer and a product engineer. The capabilities and knowledge areas you want to focus for a (or as a) product engineer are guided by the : Principles of a Product-led Approach Work from the outside-in Understand how to understand the customer ( ) Ultra quick start: Jobs-to-be-Done Understand and apply product analytics Understand personas Understand markets (e.g. total addressable market) Rapid, early validation Get involved in developing hypotheses Build prototypes, MVPs, work in ‘slices’ ( ). engineers find it easy to do technical prototypes but often find customer/feature focused ones ‘not meaningful’ enough Know when to use throw away code and build for scale Maturing through iteration Pragmatically apply agile Be confident and comfortable with an evolving approach to technical challenges (e.g. performance/scaling) Be able to design software in product iterations ( ) sometimes what you first create will need to evolve towards the ideal architecture. Also, refactor later as it may not be tech debt . Disciplined prioritisation Apply frameworks to how you prioritise Invest time in (re)evaluating your priorities How to develop or become a product engineer The best way to increase your product engineering capabilities within yourself or your team is a two pronged approach: learn and apply, then mature through iteration. Learn To kick-off your learning get started with some of the links above as well as research into the topics called out. You might want to read online, get books or listen to podcasts. Apply Start bringing each concept into your daily work as you learn it.

Different

2022 - HackerNoon Contributor of the Year - Agile

2022 - HackerNoon Contributor of the Year - Agile Software Development

Don't talk to McKinsey, talk to Terem

Nominated for 2022 - HackerNoon Contributor of the Year - Agile

Nominated for 2022 - HackerNoon Contributor of the Year - Agile Software Development

Who is Product Engineer

If you are new to Agility and Retrospective, I’ll offer in this post a novel introduction to it. We will explore how retrospectives and your team organization can take inspiration from A/B Testing. A/B Testing In 2008, Eric Ries in his now world-famous book “Lean Startup” described a mindset and a set of principles intended to be used by startups. This was the beginning of the lean startup movement. The core concept of lean startup includes learning, failing fast, and a build-measure-learn approach. Startups are fragile system aiming at validating a business model, or trying to find their ” ”. Startups are looking at creating exponential growth while reducing costs. product/market fit That approach has proven to be extremely useful and is now embraced by the whole ecosystem. The ” ” terms kind of vanished of time, as it is now accepted that a startup has to be lean. lean startup One of the techniques used by startups and enterprises during their phase of experimentation is called . A/B testing The idea behind A/B testing is to do an experiment where you would test two versions of something and see which ones perform the best. As an example, we could build two versions of a webpage, drives 50% of our traffic to version A, drives the remaining 50% to version B, and see which one has the best , or more explicitly, which ones sell the most. conversion rate Just as the lean startup approach, A/B testing became very popular, and is now present in most e-commerces websites, social networks and email campaigns. How does that apply to my team? The Lean startups approach and specifically A/B testing is about making a hypothesis, trying it in the wild, measure the results and finally decide what to do about it. This is exactly the kind of culture that you are trying to create with retrospective meetings. Of course, you are not here trying to optimize the click rate of a button, but a team-related metric. If you think about it, the goal of the retrospective is to decide collectively about some actions, based on the belief that applying that actions will move the team forward toward a defined goal: That is the hypothesis. Of course, during the next iteration, that action has to be applied, and its results measured and discussed. I proposed in the article to take some time during each retrospective to collectively decide what to do with the past actions. Feed the feedback loop So, if you are new to agility and are more confident with technical optimizations, one way to define retrospective and the approach behind it is : Retrospective is A/B testing applied to your team process But please, continue reading ;-) For the People, by the People , applying the A/B testing concept to your team . Before trying to do so, you need to understand some key differences. Done wrong can be extremely dangerous The first thing to understand before going further is that when you could by yourself decide to change the text of a button on a webpage, is that . during a retrospective YOU don’t decide anything The team is responsible to reach a collective decision on actions that should be integrated into the next iteration. Not the manager, the product owner nor the retrospective facilitator: THE TEAM. Actually, that’s the whole goal of the meeting: Create a shared vision and collectively decide what we can do to improve the day to day work. Because humans matters, are not machines and have FEELINGS the velocity and impact of changes should carefully be taken into consideration. Secondly, you also want to reconsider what is the that you are trying to optimize. When A/B testing is about focusing on one core metric at a time, here the metrics will be more diffused. metric Sometimes you can’t measure things scientifically, and should rely on simples questions such as “Are you happy with that change ?” or “Do you think we should go back to our previous process ?“. But for sure, you don’t want to measure solely your team velocity or happiness. One important point of divergence with A/B testing, and linked to the somewhat less scientific approach, is that with a team, you can’t really create two parallels universes wherein one on them, the team will do A, in the second one, people will do B, wait for one iteration, and see which one performed the best. You are part of one team, that can only have one reality at a time. On top of that, the team exists in a much broader system with no control or so on it. The team should adapt over time to this moving environment. Moreover, the team also evolves. The team can grow, welcome new members, say goodbye to others. People’s priorities and vision can also change over time. That means that because of that, you will have to adapt constantly. The only way to nail it is to adopt this collaborative learning spirit. Welcome to the retrospective revolution. Still here? Go further and join Retrolution Previously published at https://www.retrolution.co/A-B-Testeams/

Velocity

Retrospective is A/B Testing for Teams

At PEAK6 Capital Management we operate a variety of different systems in support of our trading teams. As we improve and evolve these systems, we sometimes run into hurdles along the way that are not all that easy to diagnose. This is the story of one of these hurdles our systems and core team ran into that came about while updating our pricing system. So, relax, stay a while, and listen. A Head-Scratcher As you might expect, market data and pricing systems are central to our trading platform and they process a LOT of data (around 7MM messages/second or about 4TB/day) just for pricing (in aggregate, our systems can pump more than 30MM messages/second at 10Gbps). All this data is needed so these systems can give us visibility into market volatility, position risk, and liquidity. Additionally, even though we are not a high-frequency trading shop, our ability to create, monitor, and execute trading strategies relies on up to date market data and derived calculations like NBBO, the greeks, and others. So, imagine our surprise when a routine server hardware upgrade resulted in our high-performance pricing system using new processors like this in our test environments: That’s right. Every. Other. Processor. Pegged… 🤔. Thanks to our systems team,  we put a pause on the rollout and started to dig in. Our goal being to ensure our pricing system was operating at least as good as it had been on the existing hardware and was taking advantage of better processors, memory, and bus speeds. So, the first order of business was to take a look at our existing rollout as we thought there might be something strange with this machine or that we might need to dig into thread-pooling and scheduling within the app. And that’s when we scratched our heads again. Our production environment had the same profile, every other processor pegged… HyperHerring One possible explanation for the above behavior would be HyperThreading. HyperThreading is Intel’s implementation of simultaneous multi-threading in which each physical CPU has two threads available for execution in order to improve overall usage and throughput. When HyperThreading is enabled, each physical CPU will act as if there are two logical CPUs available and the logical CPUs will appear in tools like or . In the image below, you can see how HyperThreading is advertised and shown to impact execution, essentially reducing CPU idle time by packing in the two threads worth of work. htop lscpu We’re running on the new hardware with 20 physical cores supporting 40 threads and on the existing  hardware with 14 physical cores supporting 28 threads. Given both processors come with HyperThreading enabled and we were seeing similar behavior, we decided to do an easy test by disabling HyperThreading and reviewing the behavior. Intel Xeon Gold 6230s Intel Xeon E5–2690s SSH, reconfigure, reboot, and… unfortunately we found we now have 20 CPUs with every other processor pegged. It was the same behavior with HyperThreading disabled. At this point our engineers working on the pricing system started to question their sanity and began trying to reproduce in our testing environments. Easy enough, one variable down, let’s keep asking why and see if we can get to root cause. Works on My Machine After the shenanigans above, we went back to the development environment for our pricing system. Our development environments are used for testing functionality more so than performance or system utilization, but are great sandboxes. As we brought the system online, we saw the following profile (this is on a machine running two 6230s): This shows the pricing system running across all 80 logical CPUs with fairly even distribution and is exactly what we’d expect the behavior to look like. There were a few sighs of relief seeing the expected behavior, but more than a few furrowed brows as we now wanted to know what was different between environments. Pricing In A Box I should mention that most of Peak6’s applications rely on Kubernetes and are containerized. Moving our standalone apps to containers and using an orchestrator like Kubernetes means that we have a common interface and application runtime across all of our systems. We have around 200 nodes across four geographically isolated data centers running around 1,500 applications at any given time. Additionally, as we have systems like pricing and use some low-level hardware features for things like shared memory and multicast, we do end up with some highly tailored Kubernetes configurations. These allow us to continue using the same tooling, monitoring, etc that Kubernetes gives us while fully taking advantage of all the hardware we have available. Now, in our development environments, we typically launch applications outside of docker as we want to do local testing, debugging, etc, without building images. The down side is, we may run into differences in application behavior depending on whether the application runs locally or within docker. When we tested the pricing system in the development environment as shown above, we tested both natively and using a containerized version of the application. In both cases, the pricing system had a fairly even load distribution across CPUs. So this was another variable we could put to rest, … or so we thought. 7 Out of 5 Whys Getting stumped while triaging software behavior isn’t a great feeling. If anything, computers are predictable as they only do as instructed. Although, there are many factors that make behavior non-deterministic, emergent, and difficult to understand. Kerninghan’s law, “debugging is twice as hard as writing code in the first place,” carries weight up and down the stack. At times like these, the best thing you can do is go full detective and review your case notes. Let’s recap what we’ve seen so far: UAT pricing systems with new hardware show an every-other CPU load distribution Production pricing systems with existing hardware show an every-other CPU load distribution In UAT, the behavior is invariant to HyperThreading In Development pricing systems show even CPU load distribution natively and containerized It was in reviewing the above, that we realized that in each case we saw every other CPU being utilized, it was always every odd or every even CPU. Why not all CPUs for specific processors? Why not just one processor? This led us to look more at the processor stats where we saw the following: Does it jump out at you? Every other CPU either odd or even lines up with NUMA nodes! NUMA and You The first thing you need to know about NUMA is that it’s really neat. It’s a way for processors to use local memory as they multitask executing threads across processors on a system. As long as the same workloads end up being scheduled on the same cores, the memory will be locally available and execution time reduced. As an example, in a four-core system running with two threads per core, it might look like the following: In cases where the scheduler moves a thread across cores and memory is no longer available locally, there is still fairly efficient access to the remote memory within the same NUMA node. In the worst-case scenario, a thread gets scheduled on a processor that has to cross from one NUMA node to another via an “interconnect”, which has the highest latency. On a Linux system, the completely fair scheduler (CFS) will attempt to allocate threads using a local node first. In the case a thread cannot be scheduled locally, the scheduler will migrate the thread, but will not necessarily migrate the memory, as seen below where core 2 reads across the bus remotely for memory that’s available locally for core 3. Diving deeper, we used the tool to get more info into how NUMA was performing and the trail got hot as we saw the following: numastat Here the and tell us that our pricing system is running almost exclusively on node0. It also explains why threads running non-locally are not executing as efficiently and the workload is unbalanced. If we look at our dev environment, we see a much better profile: numa_miss numa_foreign That’s no or and almost 1:1 memory use. This gets us to the root of the behavior being inefficient thread scheduling across NUMA-bound processors. However, it doesn’t explain why this was happening. numa_miss numa_foreign local_node As a note, I’m simplifying here a bit as there’s more complexity with scheduling algorithms, balancing latency targets, time slicing, etc that can all be tuned and could all be their own posts. Great Scott! It’s Kubernetes! Me: Knock, Knock You: Who's there? Me: Kubelet You: Kubelet who? Me: Kubelet gonna schedule all your containers' threads randomly Let's recap the facts again: UAT pricing systems with new hardware show an every-other CPU load distribution Production pricing systems with existing hardware show an every-other CPU load distribution In UAT, the behavior is invariant to HyperThreading In Development pricing systems show even CPU load distribution natively and containerized UAT and Production systems show NUMA misses and CPU imbalances Development systems show no NUMA misses and no CPU imbalances The last and final clue to our case came as we compared Kubernetes pod configurations between environments looking for any additional hints for why we’d see a pricing system in development work as expected while a Kubernetes scheduled pricing system in UAT or Production continue to show a CPU imbalance. As it turns out, our pod configurations for the pricing system did not include any CPU requests in UAT or Production. This seemed benign, but it resulted in Kubelet (Kubernetes worker node orchestrator) assigning only the minimum cpu shares to our pricing system containers. You can see this in source in the conversion to milli cpu to shares returning MinShares. here In production, we confirmed this by reviewing cpu.shares in the kubepods cgroup: More Equal Than Others Alright, you’re thinking. I get it, there were only two cpu shares defined for these containers, but what does that have to do with NUMA and CPU balancing? Here’s where we go all the way down the rabbit hole. Kubelet is configured to use the default CPU manager (essentially passing CPU assignment down to the OS(Linux)) Kubelet configures the containers for the pricing system with no CPU limits and two CPU shares Kubelet also configures these containers for burstable quality of service, meaning no guaranteed reservations Linux, seeing the above, divides out available CPU across all CPUs based on any affinities that are set as well as memory stickiness, etc (but the pricing system doesn’t have any) Linux, then divides out all remaining CPU cycles to processes based on CPU shares (but the pricing system comes in at a much lower priority because of the CPU shares it’s been assigned). Threads for the pricing system end up running across CPUs on the system and include performance hits for both migrating threads as well as migrating memory pages across NUMA nodes. randomly Further evidence of this behavior can be seen by checking out which shows us virtual memory statistics including and . These show that we’re updating memory to be moved within the numa nodes around 19 Billion times and 3,495 for huge pages! /proc/vmstat numa_pte_updates numa_huge_pte_updates This random allocation also shows up in NUMA memory usage across nodes. Instead of a more equal distribution between each of the NUMA nodes, we end up with one node ~90% allocated while the other is closer to ~60%. A One-line Solution Simply adding a : “70” to the pod configuration changes everything. It allows kubelet to properly hint that the pricing system’s threads will need compute, which results in more contiguous processor allocation and better memory locality in NUMA. This one-line change completely changes the pricing system behavior giving us better system saturation across all available resources. resources.requests.cpu First, we see CPU shares appropriately set at around 72k instead of 2. Reviewing our /proc/vmstat shows a much better profile, with no huge page table updates and ~475MM moves (compared with 19B before!). Additionally, the memory allocation is better balanced: Finally, we see a much more even distribution across processors: We’re rather lucky here as we see the Linux scheduler react nicely to the CPU request. Kubernetes versions continue to add better support for NUMA and CPU topology and assignment. If we were still running into issues after narrowing down CPU shares as the root cause, we could use the finer grained numactl controls or cpuset and memset configurations to continue to tweak the allocations. Thankfully, we don’t have to yet. Outcomes After applying our one-line fix for a CPU request, we watched our pricing system performance over the next few days in our UAT environment. The change not only corrected the CPU performance imbalance, but also resulted in about a 3x improvement in calculation speed along with a 75% reduction in overall calculation queuing during market spikes. Heck Yes! Where to Go From Here Hopefully you’ve gotten a taste of some of the complexity that comes with maintaining a high-performing pricing system. As we talked a little about CPU features like HyperThreading and affinity, NUMA and memory allocation, and Kubernetes, we also shared how our engineers worked together and triaged this issue from reproducing it to evaluating a fix. If you’re interested in diving into these kinds of problems or working on systems that really scale, check out our . We’re always looking for passionate, curious engineers who like working on hard problems! open positions References Helpful links and blogs if you want to dive deeper. https://www.kernel.org/doc/Documentation/cgroup-v1/cpusets.txt https://kubernetes.io/blog/2018/07/24/feature-highlight-cpu-manager/ https://techdecoded.intel.io/resources/measuring-the-impact-of-numa-migrations-on-performance/#gs.x6s95v http://man7.org/linux/man-pages/man2/migrate_pages.2.html https://jbd.dev/numa/

Intel

Routine

Make It Awesome

A Parable of Pricing, Processors, and Memory

Many of us using Docker to make our lives easier have ran into the issue of conflicting ports due to running more than 1 docker container. It’s easy to get bogged down into changing config files and squander an entire afternoon. I intend to write this article as a reminder to my future self whenever I need to set up multiple instances of docker container on a single machine. My development environment consists of — A Linux machine and two separate Laravel projects to run in distinct containers. The files to modify There are essentially 2 files which require modification: .env docker-compose.yml Both of these files are located at the root directory of the Laravel projects. The format of PORTs in Docker config Before we proceed let’s refresh the format of the ports as defined in the docker-compose.yml. The format is: <External>:<Internal> This is super important to remember. So, if you see the MySQL port defined as you immediately know that 33060 is the external port that the container exposes to the outside world. Whereas 3306 is the internal port used inside the container itself. 33060:3306, Now, let’s get into business! Modifying the docker-compose.yml file Initially both of my projects came with a file containing the ports as follows: docker-compose.yml mysql: 
        ports: appid:
        ports: redis:
        ports: '3306:3306' '8000:80' '6379: 6379' Notice that both the projects will attempt to run on the same ports since both of them have the exact same ports exposed (defined as external). Now, in order to solve this issue, you need to change one of the project’s config file and update the exposed ports to different ones. But another issue pops-up. It is likely you will be collaborating with multiple developers on the same project. A config change pushed by you might not be on par with their development environment. You are left in a conundrum! The idea to tackle this issue, lies in creating a separate config file that ONLY serves your dev environment and put it into .gitignore The steps goes like this: Create a docker-compose-dev.yml Add it in .gitignore Update it’s contents with PORTs of your choice and not conflicting with other docker containers running My looked like this: docker-compose-dev.yml mysql: 
        ports: appid:
        ports: redis:
        ports: '33060:3306' '8001:80' '6378: 6379' Notice that, the internal ports on the right are still the same. Cause other containers on the machine don’t care about any internal ports running in any other container. Thus it is safe to change the external ports only! Modifying the .env file The key thing to remember here is: the .env file should always contain the external ports as defined in the docker-compose.yml Keeping that in mind we update our .env file and add the external ports to it. The modified file should have the following ports defined: =http://localhost: = = APP_URL 8001 DB_PORT 33060 REDIS_PORT 6378 Considering the case where a separate application is running on the 8000 port of localhost. Wrapping it up To wrap up everything, we need to follow just 1 rule while running our docker-compose cmds. > docker-compose -f docker-compose-dev.yml build/ run /<other docker cmds> The key thing to remember here is, appending this line: docker-compose -f <your-updated-docker-compose-filename> giving cmds. before any This should do the job! Previously published at https://medium.com/consol/running-multiple-docker-containers-with-custom-config-files-777a0f57fe89

Super

Diary of an aspiring software engineer

Running Multiple Docker Containers With Custom Config Files

The Rust programming language is an ambitious project of the Mozilla Foundation – a language that claims to be the next step in evolution of C and C++. Over the years of existence of these languages some of their basic flaws still haven’t been fixed, like segmentation errors, manual memory control, risks of memory leaks and unpredictable compiler behavior. Rust was created to solve these problems while improving security and performance along the way. Evrone has found use of Rust in a lot of projects, and our engineers have amassed substantial experience with the language. In this article we’ll tell you about the main features of Rust. Feature rundown Strong static typization Lack of garbage collection and an ability to manually control where data is stored through pointers A powerful built-in static code analyzer that helps to avoid memory control and multi-threading issues C-like syntax with brief keywords History of Rust Work on Rust was started in 2008 by enthusiast Graydon Hore. In 2009 Mozilla expressed interest in the project, and just a year later the project was announced publicly. 2012 saw the release of the first alpha version of Rust. A year later, developers of the Servo web engine that is based on Rust announced that they had received support from Samsung. Thanks to that the engine was ported to the ARM architecture. Rust 1.0 was released in May 2015. The same year, the language was placed third in Stack Overflow’s poll of favorite developer tools. Starting from 2016 and to this day, Rust tops these rankings. Reasons to use Rust Pros A uniform compiler from Rust developers with a built-in package builder and manager, test system and documentation generator Safe memory management that helps to avoid segmentation errors An ability to use abstractions, which makes manual memory control easier Fix suggestions for most common compile errors, plus clear and concise pattern errors Pointers could only be used in unsafe code – safe code only includes links to objects guaranteed to exist Great compatibility with Mac and Unix-like systems Cons Lack of classes and succession, which makes writing object-oriented code harder Very strict compiler that sometimes polices memory addresses too much Use cases of Rust The language supports the main programming paradigms: object-oriented, parallel, functional and procedural. It provides enough control over memory and is secure enough to become a popular tool for OS and key application development. Its main drawback, however, is insufficient support from hardware manufacturers who prefer using more ubiquitous C and C++. Here are some of the successful software projects written in Rust: Redox, a Unix-like OS based on a micro-kernel, and most of the software for this OS is also written on Rust Servo, multithread-optimized web engine Firecracker, a micro-virtualization system created primarily for serverless environments Blockchain in Rust Distributed ledger-based systems have to be able to quickly process requests within the network with minimal computing load for the device. C++ tools are very well-suited for this task, so developing blockchain architecture using Rust will prove even more effective. Those are the notable examples: Parity.io – an alternative client for Ethereum and Bitcoin Polkadot.network – heterogeneous blockchain networks Exonum, a framework for blockchain-based projects MaidSafe, a distributed data processing and storage system Solana, a platform for creating blockchain-based applicationsRust in web development Rust can be used for creating web projects: the language’s SDK can be utilized for both the frontend and the backend sides of the application. For example, the client side is done via Yew – a framework inspired by React and Angular. Web servers can be easily created using Actix-web – a very performance-oriented framework with support for WebSockets, TLS and HTTP/2.0. Other tools enabling the use of Rust for the web are also available: rocket, conduit, gotham, pencil. Some of the successful Rust projects: Dropbox, a cloud storage service OpenDNS, a web service for creating public DNS servers Coursera, an education portal created by IT professors at Stanford University Machine learning in Rust Rust-based neural networks look like a very prospective venture. The Rust API can become a popular tool for this purpose thanks to its performance and a low-level memory control using high-level abstractions. But at the present moment, machine learning applications using Rust are only at an experimental stage. The Rust ecosystem lacks proven and reliable libraries to develop neural networks similar in features to Python-based ones. Future of Rust C++ has been dominating the programming language landscape for almost 40 years, becoming an industry standard and rightly holding onto that title to this day. Rust is being actively developed and refined trying to solve the most critical flaws of C++ and other programming languages. Among Evrone projects built with we can point out at a large ERP system for a restaurant chain. Within it, Rust handles micro services serving the users data about the nearest restaurant. It is also the base of self-service information kiosks. Rust Another notable use of Rust is Vexor – our continuous integration service. There, Rust is responsible for managing and scheduling tasks, the automatic computing power purchaser and the logging system, as well as serves as the basis of an agent handling isolated executions of tasks on workstations.

Solana

Polkadot

Chain

Dropbox

Mozilla

Winner of Noonies 2020 for the award Hacker Noon Contributor of the Year - INTERVIEW.

2020 - Hacker Noon Contributor of the Year in INTERVIEW

Evrone.com

Why Rust is Meant to Replace C

This is a tutorial on how to setup an Electron app on , so that new versions are deployed to with a simple pull request. Travis CI Github Releases Boilerplate I created a o that has all the necessary configuration to deploy a minimalistic app to Github releases. If you get lost during the tutorial you can look to that as an example. Also, if you don't have an electron app yet and just want to start with it as an example feel free. boilerplate rep Electron Builder We need a package that will handle packing the app into an executable and deploying to Github releases. is a fantastic npm package that handles building, signing, notarizing, and deploying an electron app on all three operating systems. Add it using yarn ( ): Electron Builder recommended by Electron Builder electron- --dev yarn add builder Electron Builder uses your app's file for most configuration. package.json { : , : , : , : , : { : , : , : }, : , : [ , , , , ], : , : { : , : , : { : [
        { : , : },
        { : , : , : , : }
      ]
    }, : { : }, : { : , : }
  }, : { : , : }
} "name" "{APP_NAME}" "version" "{VERSION_NUMBER}" "description" "A minimal Electron application that deploys on CI servers" "main" "main.js" "scripts" "start" "electron ." "release" "electron-builder" "test" "echo success" "repository" "https://github.com/{USER_NAME}/{REPO_NAME}" "keywords" "electron" "ci" "travis" "tutorial" "demo" "author" "{USER_NAME}" "build" "appId" "{APP_ID}" "publish" "github" "dmg" "contents" "x" 110 "y" 150 "x" 240 "y" 150 "type" "link" "path" "/Applications" "appImage" "license" "LICENSE" "nsis" "createDesktopShortcut" "always" "license" "LICENSE" "devDependencies" "electron" "^4.0.1" "electron-builder" "^21.2.0" Replace all the configuration variables with your own values. The configuration variables are all caps in {BRACKETS}. You should have a license file named LICENSE at the root of your directory, as well as a copy called in your build folder ( ). Electron builder uses those licenses as the license agreement for the installers. license_en.txt build/license_en.txt Good practice for an appId is a reverse domain name. For example, ours is . io.qvault.app You can setup your test script to actually run tests if you want, the above just prints "success" to the screen. We will configure Travis to run those tests on the CI server. At this point you should be able to run yarn release --publish never which will build and package your app locally into the directory. However, it will only build the package for your local operating system, which is expected. dist Travis CI Navigate to and sign up using your Github account. Once signed in you should be able to select which repository you want to connect to Travis. https://travis-ci.org/ Copy this code into at the root of your repository: .travis.yml : node_js language : : : : : : : : : : : : : : : : : : : : : : : : : : : : node_js - '11.6.0' # Always run two parallel builds: one on mac and one on linux # the linux build will use wine to be able to build windows and # linux apps matrix include - os: osx osx_image xcode10.2 language node_js node_js "11.6.0" env - ELECTRON_CACHE=$HOME/.cache/electron - ELECTRON_BUILDER_CACHE=$HOME/.cache/electron-builder - os: linux dist trusty sudo required services docker language generic notifications email false # cache some files for faster builds cache yarn true directories - node_modules - $HOME/.cache/electron - $HOME/.cache/electron-builder # add git lfs for large file support before_install - | if [ "$TRAVIS_OS_NAME" == "osx" ]; then mkdir -p /tmp/git-lfs && curl -L https://github.com/github/git-lfs/releases/download/v2.3.1/git-lfs-$([ "$TRAVIS_OS_NAME" == "linux" ] && echo "linux" || echo "darwin")-amd64-2.3.1.tar.gz | tar -xz -C /tmp/git-lfs --strip-components 1 export PATH="/tmp/git-lfs:$PATH" fi before_script - git lfs pull # on PRs and merges to master and prod run tests and build the app script - | if [ "$TRAVIS_OS_NAME" == "linux" ]; then docker run --rm \
        -v ${PWD}:/project \
        -v ~/.cache/electron:/root/.cache/electron \
        -v ~/.cache/electron-builder:/root/.cache/electron-builder \
        electronuserland/builder:wine \
        /bin/bash -c "yarn --link-duplicates --pure-lockfile && yarn test" else yarn test fi # only deploy to github on a merge to the prod branch deploy provider script script bash deploy.travis.sh skip_cleanup true on branch prod before_cache - rm -rf $HOME/.cache/electron-builder/wine # only run this script on pull requests and merges into # the 'master' and 'prod' branches branches only - master - prod The comments in the above file should explain what each step does, but the basic idea is to on each pull request to verify that a pull request doesn't break the app. Then, once code is merged into the prod branch, we trigger the following deploy script to build and push our code to Github Releases: yarn test Copy this file to deploy.travis.sh [ == osx ]; yarn release docker run --rm -e GH_TOKEN -v :/project -v ~/.cache/electron:/root/.cache/electron -v ~/.cache/electron-builder:/root/.cache/electron-builder electronuserland/builder:wine /bin/bash -c #! /bin/bash if " " $TRAVIS_OS_NAME then # deploy on mac else # deploy on windows and linux " " ${PWD} "yarn --link-duplicates --pure-lockfile && yarn release --linux AppImage --win" fi In order for your script to have permission to upload code to Github Releases, then you will need to set an environment variable that contains an API token. .travis.yml In Github Navigate to your personal settings / Developer Settings / Generate New Token. Then go to your repository settings in Travis and you can add an environment variable. The variable name is and the token is the one you created on Github. Make sure to keep the variable private (the default) on Travis so that it won't print the token in the logs. GH_TOKEN Done! Now all pull requests to master and prod should run tests, and all code merged into the prod branch should trigger a new release. The released assets and downloadable installers will be published to your Github repository under the . releases tab The release will be a draft so after each deployment you need to go in manually and convert from a draft to a published release, which is just the click of a button. Confused? If you get lost feel free to keep looking back at the and also make sure to look at the logged errors in Travis. If you have questions I am available on our discord server at the following: working example repo https://discord.gg/EEkFwbv Also, if you are looking for more advanced options take a look at the . It uses this method along with support for code signing, auto updates, and notarization. Qvault code By Lane Wagner on twitter @wagslane Previously published at https://qvault.io/2019/08/08/automatic-cross-platform-deployments-with-electron-on-a-ci-server-travis/

A minimal Electron application that deploys on CI servers

{APP_NAME}

2022 - HackerNoon Contributor of the Year - Algorithms

2022 - HackerNoon Contributor of the Year - Cryptography

2022 - HackerNoon Contributor of the Year - Github

2022 - HackerNoon Contributor of the Year - Job Hunting

2022 - HackerNoon Contributor of the Year - Mathematics

2022 - HackerNoon Contributor of the Year - Online Education

Coding Guru of 2022

Super Career Strategist of 2022

Learn to code on Boot.dev

Nominated for 2022 - HackerNoon Contributor of the Year - Algorithms

Nominated for 2022 - HackerNoon Contributor of the Year - Mathematics

Nominated for 2022 - HackerNoon Contributor of the Year - Cryptography

Nominated for Coding Guru of 2022

Nominated for 2022 - HackerNoon Contributor of the Year - Github

Nominated for 2022 - HackerNoon Contributor of the Year - Online Education

Nominated for 2022 - HackerNoon Contributor of the Year - Job Hunting

Nominated for Super Career Strategist of 2022

Automatic Cross-Platform Deployments with Electron on a Travis CI Server

Component-based Website Development with React and the Cosmic JS CMS

Using Cosmic JS’s Webhooks, we can sync with to provide fast and powerful search functionality to our projects. Algolia Cosmic JS TL;DR Demo Source Code Intro If you’ve ever tried to implement search yourself, then you know it can be difficult. Thankfully, Algolia has created a product that makes it easy. In this article we’re going to go over how you can connect Algolia to your Cosmic JS backend and add it to your Apps. The demo App was built with React.js and an Express.js server. If you’re using something different, you can still follow this pattern. Just modify it for your use-case. There are four main steps: Create an Algolia App Create Cosmic JS Webhooks Create API endpoints to receive the Cosmic JS Webhook POST requests Add search widgets to our App Part 1 — Create an Algolia App Navigate to . Log in or sign up for a (free) account. https://www.algolia.com Once you’ve logged in, navigate to the Dashboard. In the Dashboard, click ‘NEW APPLICATION’. Give your App a name and click ‘CREATE’. The next page will ask you to choose a region. Pick the closest one. Then click ‘Continue’. You should now see the Dashboard for your newly created App. Click ‘API Keys’ on the left side of the screen. This is a list of API Keys that you will need for later steps. Part 2 — Create Cosmic JS Webhooks Log in to your Cosmic JS account and go to your Bucket. Select ‘Settings’, then ‘Webhooks’ in the Dashboard menu. You should connect three different Webhooks. Replace the endpoints with whatever you’d like them to be for your project Event: Object created and published; Endpoint — https://<__YOUR_DOMAIN__>/api/algolia/create Object edited and published; Endpoint — https://<__YOUR_DOMAIN__>/api/algolia/edit Object deleted; Endpoint — https://<__YOUR_DOMAIN__>/api/algolia/delete Click ‘Save Webhooks’. Now your server will receive POST requests whenever an Object is created, edited, or deleted. The next step is setting up your server to receive those requests and use them to keep Algolia synced. Part 3 — Create API endpoints to receive the Cosmic JS Webhook POST requests The demo App uses a Node.js Express server. Here’s the code: ( ).config({ : }); express = ( ); bodyParser = ( ); corser = ( ); next = ( ); routes = ( ); algoliasearch = ( ); convertDataToAlgoliaObject = ( ); fetch = ( ); port = (process.env.PORT, ) || ; dev = process.env.NODE_ENV !== ; app = next({ dev }); handler = routes.getRequestHandler(app); client = algoliasearch(
  process.env.ALGOLIA_APPLICATION_ID,
  process.env.ALGOLIA_ADMIN_API_KEY,
); itemsIndex = client.initIndex( );

app.prepare()
  .then( { server = express();

    server.use(corser.create());
    server.use(bodyParser.json()); server.post( , (req, res) => { { data } = req.body; (data.type_slug === ) { algoliaObject = convertDataToAlgoliaObject(data);
        itemsIndex.addObject(algoliaObject)
          .catch( .error(err));
      }
      res.status( ).send();
    });

    server.post( , (req, res) => { { data } = req.body; (data.type_slug === ) { algoliaObject = convertDataToAlgoliaObject(data);
        itemsIndex.addObject(algoliaObject)
          .catch( .error(err));
      }
      res.status( ).send();
    });

    server.post( , (req, res) => { { data } = req.body; itemsIndex.deleteObjects(data)
        .catch( .error(err));
      res.status( ).send();
    });

    server
      .listen(port, (err) => { (err) err; .log( );
      });
  }); /* eslint-disable no-console */ require 'dotenv' path './.env.production' const require 'express' const require 'body-parser' const require 'corser' const require 'next' const require './routes' const require 'algoliasearch' const require './utils/convertDataToAlgoliaObject' const require 'isomorphic-fetch' const parseInt 10 3000 const 'production' const const const const 'items' => () const // ---------- API endpoints for synchronizing Algolia ----------- '/api/algolia/create' const if 'items' const => err console 200 '/api/algolia/edit' const if 'items' const => err console 200 '/api/algolia/delete' const // data is an Array of one or more Object _ids => err console 200 if throw console `> Ready on http://localhost: ` ${port} First, an client is created. It uses the Algolia Application ID and Admin API Key (which can be found in your Algolia App Dashboard). algoliasearch Then endpoints are created for each Webhook that we’d like to receive from Cosmic JS. The Webhook POST requests include the Object their it’s body. A custom ‘convertDataToAlgoliaObject’ utility function is used to convert that Cosmic JS Object into an object formatted for Algolia. Here’s the code for ‘convertDataToAlgoliaObject’: convertDataToAlgoliaObject = { asin; brand; categories; featured; price; thumbnail;

  data.metafields.forEach( { (metafield.key) { :
        asin = metafield.value; ; :
        brand = metafield.value; ; :
        categories = metafield.value && metafield.value.split( ); ; :
        featured = metafield.value.includes( ); ; :
        price = metafield.value; ; :
        thumbnail = metafield.value; ; : }
  }); {
    asin,
    brand, : categories || [], : data.content, : data.created_at,
    featured, : data.modified_at, : data._id, price, : data.published_at, : data.slug,
    thumbnail, : data.title,
  };
}; .exports = convertDataToAlgoliaObject; const ( ) => data let let let let let let ( ) => metafield switch case 'asin' break case 'brand' break case 'categories' ',' break case 'options' 'Featured' break case 'price' break case 'thumbnail' break default // Do nothing return categories content createdAt modifiedAt objectID // eslint-disable-line no-underscore-dangle publishedAt slug title module This function will be different for each project. You basically just want to extract the information that will be ‘searchable’. Now, Algolia’s algoliaseasrch library can be used to upload, edit, or delete the corresponding entry in Algolia. The backend setup is complete! You should note, however, that no syncing will take place until you deploy your App. The custom API endpoints you created don’t exist yet, so the Algolia Webhook POST requests won’t be intercepted. Part 4 — Add search widgets to our App Not only does Algolia provide awesome search functionality, they also provide many libraries that can be used to display your data. I used their for the demo project. It provides React components that are essentially plug-and-play. You can customize their styling with your own CSS. The is very well-written, so there’s no need to repeat it here. React Instant Search library documentation Conclusion Hopefully, you found this article helpful. If something isn’t clear, check out the or documentation. They’re both great! Cosmic JS Algolia This article originally appears on . the Cosmic JS website

Connecting Algolia to Cosmic JS for Awesome Search Functionality

Big Data's value, popularity, and scale of usage in business today come from a few of the indisputable benefits it has to offer: The data is coming from various resources that contribute to its accuracy; The constant influx of data guarantees its relevance; Instead of manual analysis, it is processed automatically with the help of algorithms, which reduce the error probability and allows increasing the data volume; Centralized data access has become possible with the help of Big Data technologies. The world statistics show that since 2015. the number of companies using Big Data has grown from 17% to 50% The increase rate is impressive, and the truth is, it’s no longer just a fancy word a cool tech guy would use when discussing what’s behind supercomputer VIKI from . I, Robot In fact, many large industries have already adopted Big Data technologies and are gaining profit. In this article, I'll look at three industries in which big data is making a dent in 2020. Big Data in Medicine and Healthcare When discussing Big Data in healthcare, one should distinguish the three aspects of its application. Firstly, it’s the so-called . clinical aspect that involves disease diagnosis, treatment, and prevention Then comes aimed at better management of healthcare institutions by improving their operational and economic efficiency, on both regional and governmental levels. the management aspect Finally, is the one where Big Data can play a truly revolutionary role. Disease studies and drug development are the core areas here. the scientific aspect How exactly can Big Data technologies be useful to a particular patient and a specific doctor? The fundamental concept lies within the development of electronic health records (EHR). is a system that related to an individual’s health. EHR contains the , all the hospitalization records and doctor visits, laboratory tests, vaccinations, examinations, allergies, or any other related facts. Apart from serving as a storage of accurate medical data, EHR can control the patient undergo regular screenings and treatments by sending corresponding reminders and notifications. EHR collects and stores medical and other personal data patient’s disease and ailments’ history Many experts claim Blockchain to be a good technology choice for EHR: it fulfills some of its major requirements, such as completeness of data, its protection, and verification. makes it possible to use millions of EHR’s for as social, territorial, age-related, demographic, genomic, etc., which in its turn, allows providing a patient with exceptionally individual recommendations. The synergy of these three rising technologies allows creating a diagnostic system of a brand new quality level. Combining Blockchain with Big Data and AI detecting hidden dependencies between a disease along with other side-factors is a prototype example of such a system. Its database as well as 2 000 000 text pages from medical journals and clinic tests. By processing this data, the program . Watson for Oncology by IBM includes more than 600 000 diagnosis records can produce the most optimal treatment plan A retrospective analysis of treatment plans prescribed says that in regards to lung cancer, . In comparison, the rate of around . Watson selected efficient treatment tactics in 90% of the cases US hospitals averaged 50% Big Data also represents the core technology for forecasting medical institutions’ spending. It relies on a multifactor analysis of such statistical data as the number of repeat visits, ailments recorded against specific experts and departments, the prevalence of certain pathologies, number of chronic patients, epidemiologic indicators, etc. serves as a good example here: they . Repeated hospitalization, missed doctor appointments, drug supply are the major factors included. British National Health Service use Big Data to analyze the efficiency of surgery sub-departments McKinsey’s recent research says that . McKinsey reports that it can be achieved through Big Data capability to unite the data stored in separate sources wholly isolated from one another. For instance: Big Data helps to not only reduce healthcare costs but also contributes to the quality of life enhancement Consolidating these types of data from the sources above opens up unique possibilities for the medicare. Clinic data (medical history, laboratory tests, etc.); Patients’ behavior insights based on purchase preferences - the data collected from portable and home medical devices; Scientific research and tests; Medical establishments’ records of the services provided: pharmacies’ reports on the drugs sold, price statistics from the healthcare market. Optimizing the operational activities is essential for bringing the established treatment practices to a whole new level. Personalized and preventive approaches based on remote monitoring can become the new reality thanks to vast volumes being analyzed. Big Data in Retail Every minute millions of people are purchasing by carrying out millions of transactions, so it’s no surprise that retail is one of the top industries leveraging Big Data. The significant tasks retail is trying to solve from both are the following: technological and entrepreneur standpoints Any retail business is looking for very explicit recommendations which steps one should take in a particular situation. Nowadays, Big Data analytics can help with that. One of the aspects of a company’s steady income depends on — is the . Price correction is one of the principal retail management mechanisms. The studies show that, on average, managers spend half of their work time on goods reevaluation, which still isn’t accurate enough due to human factors. In addition, one cannot guarantee that the process operates such crucial variables as seasonality, stock availability, supply and demand structure, competitors’ activities, etc. That is why large . smart pricing policy retail chains switch to automatized price reevaluation using Big Data technologies Process huge amounts of data;Get updates on the sales performed and the stock dynamics in the entire retail chain; React quickly to the changing situations and produce forecasts by addressing all the contributing factors (weather conditions, seasonality, holidays, distribution points, competitors, etc.); Analyze promo activities. , has implemented a similar solution, which has . What’s also important here, the new price wasn’t just somebody’s judgment call but was carefully deduced, relying on the variables mentioned above. Macy’s, which operates 800 stores with 73 trade items spared the need for pricing correction by 22 times Using the same method, . Amazon is adjusting its prices every two minutes Apart from the optimizations the technology brings to the existing businesses, it plays a big role in making strategic decisions when developing a new entity. By analyzing competitors’ locations nearby, transport access, potential customers’ income rate, their habits, and preferences, it gives a comprehensive idea of the territory’s commercial potential for future business. Big Data in Law Enforcement Historically, law-enforcement facilities have been working with huge volumes of data. There are few directions where data-based insights can be of particular value to the law-enforcement establishments. First and foremost, it’s cybersecurity. As unfortunate as it may seem, but fraudulent activities and malware have become a very often type of crime for today’s police. is said to be . Big Data algorithms can detect suspicious activities before they actually occur and alert consumers or companies about a potentially fraudulent action. It’s becoming a must for many businesses to take proper care of enterprise data security. Big Data analytical function today’s most efficient instrument for fraud identification and prevention Big Data analytics is also behind modern policing strategies. It allows the police to take a more proactive role in combating criminal activities. By using various sources of data such as the number of arrests, crime rates, fines issued in regards to a particular neighborhood, the police can monitor the areas prone to criminal activities more closely. Predictive policing — is the term used to describe this method in criminology. Lastly, on various levels. It gives the authorities unprecedented access to very different kinds of precise information from all over the world to help understand the global picture of a situation and make well-informed decisions. Economy, health, climate, social issues are just a few fields the law enforcement agencies need data science analytics for. advanced analytics Big Data offers is an efficient tool for governing establishments Summary “Information is the oil of the 21st century, and analytics is the combustion engine” — Peter Sondergaard at Gartner. It’s safe to say that data truly is the new oil, today. But a great opportunity can be a big challenge at the same time. So if you are having a hard time with structuring and organizing your business data and you feel like there is a better way to deal with it, reach out to me!

Every Two Minutes

Plan A

3 Industries Harnessing the Power of Big Data: Healthcare, Law, and Retail

A code should be organized. That’s the fact and I think is a very good one for the opening of the text about nesting. So... CSS is used for describing the presentation of a document written in a markup language. There are few more texts to check on CSS, e.g. , , . Every one of us is familiar with that and knows that you first need a to detect an element and after that comes . positions in CSS text underline relative units selector the styling For the best way to select the desired element, we should use every trick there is - . However, , and every element must be selected separately. In another word, if there is a child element it should be defined completely separate from the parent. selectors CSS doesn’t allow nesting Less and Sass Less and Sass are which extend CSS language in valuable ways. Just one of many improvements they offer is an easier and more natural way of defining nested elements and its relative position among themselves, nesting! CSS pre-processors It is an option which has same interpretation in less and sass so the next explanation is universal for both of them. Nesting makes the code easy to read, extend, and maintain. Nesting An example will tell everything (and then I will explain it, just in case): Less/Sass: .grandparent{ .parent1{ .child1{} .child2{} .child3{} } .parent2{ .child1{} .child2{} } } CSS: .grandparent .parent1 .child1{} .grandparent .parent1 .child2{} .grandparent .parent1 .child3{} .grandparent .parent2 .child1{} .grandparent .parent2 .child2{} So, instead of the defining child elements with space in the same line (CSS syntax), pre-processors enable you to use the same structure as in e.g. HTML. So if the child3 element is a child of the parent1 and parent1 is a child of the grandparent, all that is defined just like that. Ampersand As a developer, I'm full of best practices and pieces of advice and here comes one on this. Nesting ~~should~~ must be used for pseudo-classes because it is a must. To define a specific state of an element, it is necessary to add pseudo-class next to the selector and with nesting, it's ultra-intuitive - just put ampersand next to pseudo-class. Besides this use, it can (but it's unnecessarily) be used next to the child selector but with . The `&` always refers to the parent selector when nesting. `_` .child{ &:active{} } If the ampersand is located right next to the child element in nesting, it is compiled into a double class selector. Less/Sass .child{ &.grandchild{} } CSS .child.grandchild{} Conclusion Deep nesting, which is actually recommended, - it’s hard to reuse, override and make CSS file bigger, but developers who use it are aware of it all and it’s not a problem (believe me). has its downsides Use everything available to make your code writing experience as fast and as easy as you can. Nesting is naturally one of those things, so use it maximally!

Nesting in Less and Sass

or how to spread the word and win contributors Hello folks! Congrats to those who made it to the 4th part of the Open Source journey! For those of you who haven’t read the previous parts or are wondering what I planned for the next parts: Table of content: Intro Starting a project Documentation Publicity Issues and PRs (WIP) Automation (WIP) Versions management (WIP) Recap We’ve already discussed what it means to start a project, how to do it optimally and how to write good documentation for it. Today we’re moving forward and we’ll be discussing how to attract the publics attention to ones project and optimizing it to both attract and correctly manage contributions. The baseline for this part is: You already have an open source project, it is available on GitHub, well documented and can be consumed easily via one of the package registries. Spreading the word Let's address the elephant in the room: as your project grows you simply wont be able to handle everything by yourself. Hence, you need more people to work on the project if you want it to live long and prosper. To get more people involved in your project you need more people to know about it and above all else to believe in it. From my experience the best way to expose your open source to the right audience is to and to write a blog post about your project. The resource might be purely dev oriented (like or not (like Medium), or tech focused (like ). use one of the well known resources dev.to ) here on Hacker Noon One commonality between all these resources is that they have an established audience and it is the relevant audience. You can also your article between various online resources so that you’d cover an even larger audience but be aware! Cross posting has a few downsides: cross-post Each of these platforms may have a different mark up language and you’ll have to redo all the formatting Maintenance — if something changes (and things change) you’ll need to update your blog post across all of the resources. will If you go for Medium I’d highly recommend you submit your article to one of the . It will require additional effort on your end, since you’ll have to align your article with the publication requirements. However, it will also ensure that your article is exposed to a large and, which is even more important, audience. large publications relevant You can also decide going behind a : "Stories that are part of the paywall are also eligible for distribution to Medium readers through topics, which power recommendations on Medium on our home page, on our topic pages, in our Daily Digest and in our apps..." metered paywall I can’t tell you which one is better, but my personal preference is publications because it ensures your article is exposed, instead of the vague “eligible for distribution” term. If your blog post goes viral then it can create a cascading effect and add even more leads to your open source. For example, if after publishing the article your Github project received a few dozens of stars in one day it can get to the Github page which is in itself another source of exposure. Trending A few points to make your blog post great (again): Start with a problem statement. It might even be the title of the blog post. Usually people are looking for a solution to a specific problem and before they decide investing the time to read your post they should have an idea whether it’s what they’re looking for. An example: https://medium.com/angular-in-depth/customizing-angular-cli-build-an-alternative-to-ng-eject-v2-c655768b48cc As you can see it doesn’t say “custom-webpack builder — a new fancy project”, it clearly states the problem that it solves in the title. If you google “Customizing Angular build” this is the first result you’ll get and right from the search page you can see which problem it addresses. Describe why and how exactly your project solves this problem Provide a detailed step-by-step guide, starting with installation and finishing with a working example. Create an example project that uses your open source and link to the sources in the blog post. There are lots of developers who prefer a working example to any blog post. Get some feedback before publishing it. Make your friends go over it without telling them what it’s about and see if they can figure it out themselves. If they can’t then probably it’s not that clear and you need to elaborate. After you’ve published your blog post, make sure you share it on social media, with your friends, your family and strangers on the street. This will increase your project’s exposure but you also have to make people to contribute to your project. want Making your project attractive to contributors The best thing is to start an open source with other people. This way from the very beginning you have a team with which you can split the burden. However, that’s not always the case. If you started alone you have to attract contributors and in my experience there are two types: Someone who wants to make an impact and looks for a project to contribute to (these are rare but still exist). Someone that uses your package and found either a bug or lack of certain functionality. In both cases just having your source code shared on Github and a single blog post on how to use it are not enough. Here are a coupe of things that can make people want to contribute: Have a pending implementation list. It might contain known bugs, planned features or something else. This list will make it simpler for contributors of type #1 to pick the right item and issue a PR. It can be a standalone list or you can (and probably should) use issues and labels on GitHub. Have a . contributors guide A very basic contributors guide should explain the repository structure and have a step-by-step guide for building and running your project and tests. The expanded guide can contain architecture, design decisions, code of conduct and more. A good example is Atom’s contributors guide . Don’t underestimate it’s value! It’s something that takes a decent amount of time to make when the project has grown, and I wish I created it at the very beginning and updated it gradually as the project evolved. Unfortunately, I didn’t have someone to point out its importance, and today my project has no contributors guide. It’s always on my TODO list but there is always something more urgent than that. Recognize your contributors Listing your contributors on the main page of the project would give them additional incentive to make a contribution. Just adding user names can be enough, but I’d recommend you to use All Contributors . Apart of creating a fancy section with profile images and badges for your all your contributors, it automates new contributors addition by creating PRs that add contributors to this section. Wrapping it up In this part we’ve discussed several things that will increase your project’s exposure and provide people with an initial incentive to open a PR or an issue. But this won’t keep them as contributors neither will it make sure they will finish the work they started. In the next chapter we’ll discuss how to manage as well as how it affects the peoples’ perception of your project. Issues & PRs If you enjoyed my writing, learned something new or insightful or if you just don’t want to miss the next part, make sure you’re following me on or here. Cheers! Twitter

Follow my blog

Open Source Series: Publicity

The main indicators of car competitiveness are price, quality, technical specifications, design, operating costs, and safety. In order to increase the competitiveness of automobiles, it is necessary to improve product development methods, introduce new design technologies, use computer capabilities, and mathematical modeling in all aspects of experimental design work. But the central aspect is to follow the latest . Important tools, such as AR and VR, are already widely used in the automotive industry. technology trends VR & AR in Auto Production VR development services are widely used by large companies such as Peugeot, Citroen, Land-Rover, Jaguar, Boeing, etc. The use of virtual reality-based solutions can further improve the efficiency of product development. It provides an opportunity to instantly evaluate the results of designers work when the real product is presented in the form of a digital model. Let’s consider the main advantages of using VR/AR in the automotive industry: Virtual reality systems allow you to use a virtual prototype instead of a full-scale model. The design of the car is developed in parts. If there are flaws in the model assembly, then using virtual reality technology, you can quickly fix design flaws. VR allows replacing real tests with computer ones. By removing crash tests, companies will reduce the time and cost of creating a new model. The use of services of AR/VR development companies can dramatically improve the design quality, significantly reduce development time, and the cost of the design process. The use of Augmented reality technology at the stage of car maintenance and repair will allow service stations to work more efficiently. To take advantage of additional reality, you need a tablet or VR glasses. By pointing the device’s camera at the car, it is possible to integrate the necessary additional information into the existing image. This approach will significantly reduce the preparation time for work and simplify the maintenance process. BMW’s Mixed Reality Laboratory BMW has been working with VR technology for a long time. The company has created a mixed reality lab combining Unreal Engine with real physical equipment. It delivers experience with a much greater degree of immersion compared to virtual reality alone. consists of a simulated car interior with a steering wheel, pedals, and adjustable seats, as well as a VR system running on . The Mixed Reality test bench Unreal Engine 4 With the possibilities that virtual reality developments provide, BMW engineers can improve the aesthetic characteristics of the car’s interior, as well as focus on the design of the interface, usability and other tasks related to the human factor. The solutions to these problems would have been impossible within the framework of traditional two-dimensional design on the monitor screen. For example, the designer wants to check how convenient it is to control the car’s infotainment system when the steering wheel is completely locked or, for example, when the seat is folded back. A poorly designed user interface can significantly affect such important parameters as safety, accessibility, and comfort. In the BMW Mixed Reality Laboratory, tasks like this, as well as many others, can be modeled both visually and physically. High-Tech Motorcycle Helmet F-35 fighter pilots have been using helmets with augmented reality for a long time, which allows them to follow the readings of instruments and weapons systems without looking at the dashboard. As a result of the collaboration of BMW and DigiLens, this technology is also available for motorcyclists. The company believes that expanding the AR development to motorcycles is quite natural. This technology can enhance both security and ease of management. Using alternative technologies, the driver needs to take his eyes off the carriageway and lower it onto a small screen where various information is displayed. The two dead zones in the review are the cause of most accidents. But, with augmented reality technology and a number of sensors installed on the motorcycle, you can display warnings about the proximity of other road users to the helmet screen, thereby preventing the number of accidents. A display for motorcycle helmets allows drivers to monitor the level of fuel and speed, without taking their eyes off the road. The helmet is also equipped with cameras, which allows you to use it as a recorder. Using magnetic fasteners, the display is fixed inside the helmet in front of the driver’s eyes. Repairing Glasses , playing the role of an interactive assistant. Looking at any BMW car model with such glasses, the mechanic sees a structure with the main components and assemblies without a body shell. Prompts appear on the screen which tool to use and how to get to the problem unit. BMW has equipped its technicians with augmented reality glasses BMW augmented reality technology is available not only to the mechanic but also to the driver. So, during normal road traffic, glasses can tell you how to behave in various traffic situations, go around a traffic jam, or warn about an accident ahead. The developers also intend to integrate the glasses with the infotainment system of the car and with a smartphone. And when parking, due to information exchange with a rear-view video camera, the driver wearing these glasses will literally feel that he has eyes on the back of his head.

Assembly

Boeing

Instantly

PEUGEOT

Zones

Automotive Industry Revolution: How BMW Leverages AR & VR in Car Production

In RSpec, there are two different ways to write DRY tests, by using or . Their purpose is to create variables that are common across tests. In this post, we will explore differences between before and let and explain . before let why let is preferred by the Ruby community let Let creates lazily-evaluated local variables. This means that let() is not evaluated until the method that it defines is run for the first time. It DRYs up the spec and makes it more readable. $count = describe let( ) { $count += }
  it expect(count).to eq( )
    expect(count).to eq( ) it expect(count).to eq( ) 0 "let" do :count 1 "stores the value" do 1 1 end "is not cached across examples" do 2 end end Let should not be used for local variables, which have to be saved to the database, as they will not be saved to the database unless they have already been referenced. In this case, you should use or . let! before blocks The the tests we are creating are done using . RSpec and Capybara Also, have a let block inside of a before the block, this is what let! is made for! never let! Unlike let, you can . It means that, even if you didn't invoke the helper method inside the example, it will be invoked before your example runs. use let! to force the method's invocation before each example $count = describe invocation_order = []
  let!( ) invocation_order << $count += it invocation_order << expect(invocation_order).to eq([ , ])
    expect(count).to eq( ) 0 "let!" do :count do :let! 1 end "calls the helper method in a before hook" do :example :let! :example 1 end end As with let blocks, if multiple let! blocks are defined with the same name, the most recent one will be executed. The core difference is that will be executed multiple times if used like this, whereas the will only execute the last time. let! blocks let block before(:each) Before(:each) block will run before each example, even if the example doesn't use any of the instance variables defined in the block. This can noticeably slow down the setup of the instance variables. @tests = [] describe User before( ) @user = User.new describe it @user.should have( ).tests it @user.tests << Object.new it @user.should have( ).tests class User def tests || end end do :each do end "initialized in before(:each)" do "has 0 tests" do 0 end "can accept new tests" do end "does not share state across examples" do 0 end end end Depending on your personal preference you could use before blocks when: In nearly every situation, it is better to use let over before blocks. There is a reasonable amount of variables. There are variables that don't need to be referenced directly but are required. There are many commands to be executed because its syntax is more clear when many commands are involved. Creating mocks/stubs. Give it a try and create your own model or ! controller tests before(:all) This block is executed only once, before all of the examples in a group. There are certain situations this can cut down on execution and effort. @tests = [] describe User before( ) @user = User.new describe it @user.should have( ).tests it @user.tests << Object.new it @user.should have( ).tests class User def tests || end end do :all do end "initialized in before(:all)" do "has 0 tests" do 0 end "can get accept new tests" do end "shares state across examples" do 1 end end end It runs outside of transactions, so the data created here will bleed into other specs. Using before(:all) in RSpec will cause you lots of trouble unless you know what you are doing! Conclusion It all depends on what and how you need to make the work or consider using for creating data. Let blocks bring more to the table than before blocks. RSpec tests FactoryGirl Besides being slower, one of the major problems with before blocks is that spelling errors can lead to bugs and false positives, allowing certain types of tests to pass when they shouldn't. before( ) @user = User.find( )
    @user.logout it expect(@usr).to be_nil :each do username: "kolosek" end "should log the user out" do end Since @usr wasn't previously defined, the test will pass, @usr is nil by default. The same test using let would raise NameError because @usr isn't defined. Hope this will help you to better understand the differences between let and before blocks. Thank you for reading! Previously published at https://kolosek.com/rspec-let-vs-before/

Initialized

RSpec Testing: Let Vs. Before

2020 is off to a tumultuous start, with an ongoing global pandemic . The Dow already experienced its since the 2008 crisis, and even cryptocurrency markets are blood red, with Bitcoin suffering its in over a year. likely to trigger a global recession greatest decline biggest price crash This has occured alongside ongoing , which were thought to be largely over, until just recently when two Americans and one British soldier died in a rocket attack on an Iraqi base. US-Iran tensions Further, global oil demand will or even drop for the first time since the Great Recession. remain stagnant Finally, those behind a well-known , PlusToken (which stole around $2 billion worth of cryptocurrency from investors), have moved around $100 million worth of Bitcoin to "mixers" in order to disguise their traceability, and ultimately make a getaway to fiat through an exchange. crypto Ponzi scheme What do these all have in common? These seemingly disparate events all have one thing in common: They've had undoubtedly large impacts on markets, both traditional and alternative (like the Bitcoin market). The seemingly impossible question to ask is " " how much impact did each event have? Some causal chains are pretty obvious. The spread of the coronavirus means that fewer people are travelling, which is the main reason airline stocks are tanking. Others, not so much. Is Bitcoin crashing because it's not the safe-haven we thought it was, and investors are just as afraid in the crypto market as traditional markets? Is it because people are freaking out over PlusToken? Is it none of the above? If you're expecting a simple answer like "Coronavirus is 60% of the cause, US-Iran is 5%, PlusToken is 20%, and Oil is 15%," then I'm sorry to dissapoint you, because those numbers are completely made up, regardless of who makes a go at it. Instead, I'll propose this: Focus on blockchain innovation, not on price movements. By its very nature, price is an infinite, undying source for the news. When there's nothing better to report on, there's always the option to make headlines that read "Bitcoin Up/Down X%." We can do better. Fortunately, there are many better things to report on. For starters, you can now contribute to clean energy and fight climate change on the blockchain, by investing in a (disclaimer: I'm a consultant at the group behind it, Invictus Capital). tokenized solar fund In spite of all the doom-and-gloom, real-world solutions are making their way into the markets, and it's just up for them to be widely adopted. Companies like Odem are in the fight against coronavirus. using blockchain Of course, the efficacy of these solutions, whether it's blockchain-related or mass-market, like Trump , are all yet to be seen. suspending travel from Europe We're in the midst of some of humanity's greater challenges, so let's shine a light on potential solutions, not the price of Bitcoin.

Alongside

Let's connect!

Coronavirus, Oil Shocks, US-Iran, or PlusToken—What's Behind Bitcoin's Crash?

I, Yuri Filipchuk, partner at CYFRD investment company, discuss the nature of the biotech industry, its peculiarities and prospects. In 2011, Marc Andreessen, the investor, and founder of the a16z fund (Andreessen Horowitz) : stated “Software is eating the world. My own theory is that we are in the middle of a dramatic and broad technological and economic shift in which software companies are poised to take over large swathes of the economy. ” And it’s hard to disagree with him: in every industry, you can find a leader who relied on software. Amazon was the largest seller of books, Spotify and iTunes were the biggest music companies, while game developers dominated entertainment. Eight years have passed, and the situation has not changed much at first glance. The very same Amazon, Apple, and Spotify only strengthened their positions, and in many industries, new stars were born. But Andreessen Horowitz Foundation : biology comes to the fore. It is no longer an empirical science, but an engineering discipline. Previously, in this area, we applied tools created by ourselves, but then we learned to apply nature’s mechanisms. One can literally program the body and create, for example, bacteria or cells to fight diseases. is positive Biotech features I should probably highlight the essentials. Biotechnology involves the use of living organisms’ processes for creating commercial products. It’s essential to divide the concepts. Mankind has been developing in the field of biology for a long time, but the invention of beer, for example, refers to long-standing biotechnologies, whilst DNA manipulation is a modern biotechnology industry. Today, biotech companies and startups are mainly focused on four industries. Red biotech refers to healthcare and medicine: diagnostic, treating illness and pharmaceuticals. Companies find genes and proteins associated with specific diseases and create drugs for them. This approach allows for drug development in a wide variety of cases, including cancer, neurological and infectious diseases. Medicines for the diseases of the endocrine and cardiovascular systems are underway. Oncology therapy is the most promising area in red biotech Source: 2019 Emerging Therapeutic Company Trend Report Green biotech pertains to agriculture. It implies a slightly different approach. Mankind has long been manipulating crops and livestock to increase the yield and quality of the final product. Bioengineering has proposed a similar but more effective method. Now, genetically modified foods have become the norm both from crops and animals. Let’s take Beyond Meat startup as an example. The company produces vegan meat using pea protein, rice and mung bean. Beyond Meat can imitate the taste of chicken, beef and pork without the use of animal meat. Conversely, in China scientists Crispr technology to breed giant pigs that grow much faster than ordinary species and tolerate winter better. use There are white and blue biotech: the industrial sector and the marine environment industry. Agriculture and industry are now considered the most promising. In the next five years, they are expected to be worth $36.4 billion and $274 billion respectively. According to cell biologist Olga Maslova, uneven progress in the industry is the main feature of biotechnology. At the same time, she emphasizes high risks for investments, and R&D requires costly premises, equipment, specialists and time for testing. However, biotechnology is extremely promising as it covers the most important areas of human life, including health. Market development and its current position The modern biotech industry was born in 1973. Back then, American scientists Herbert Boyer and Stanley Cohen first changed the DNA of bacteria by combining it with another sample’s genes. In 1976, Boyer, along with venture capitalist Robert Svenson, founded Gentech, a private biotechnology company, which became the first IPO company in the industry four years later. Biotech began to commercialize. Since then, the domain has shown sustained progression. So much so that in 1991, 35 biotech companies went public, and by 2000 there were 63 of them in the USA alone, which was an absolute record at that time. The average amount of funds raised has also increased from $20-30 million to $85 million. Investment in biotech grew from $5.2 billion in 2013 to $ 17.5 billion in 2018. Basically, it’s a more than three-fold increase during five-year period. Therefore, the market is very hot. Over the past five years, an average of about 70 companies launched IPOs annually. At the same time, 2018 was a record year for the total amount of investments through this procedure — $ 8.1 billion. We’ve seen oncology projects closing impressive rounds for clinical trials within the last couple of years. To be more specific, series A in this market varies between $30-50 million for the first phase of clinical trials. After that, a pre-IPO takes place, projects raise $150-200 million to complete the tests and get FDA approval. Investors expect that with a proven cure for cancer, the startup will nail IPO with the highest possible assessment. Red biotech remains the largest industry segment $314 billion. Biotechnology companies use DNA data to quickly determine the patient’s disease, for instance, malignant tumors or STDs. worth In August 2019, binx permission from the U.S. Food and Drug Administration (FDA) to start selling a portable device for diagnosing chlamydia and gonorrhea in women, the most common STD. In binx, they promise a result within 30 minutes; this allows you to determine the diagnosis and get an appointment with a doctor. received Gene therapy is another area of ​​red biotech. Companies are developing drugs that modify human DNA and fight the disease. Today, 17 drugs are for cell and gene therapy in the US market. One of them, Luxturna, treats a rare retina condition which gradually leads to blindness. Another medicine, Provenge, is for the treatment of prostate cancer. approved Johnson&Jonson, Roche, Novartis are giants with hundreds of thousands of employees and billion-dollar sales. Using biotechnology, they medicines not necessarily for gene therapy, but which treat a wide variety of diseases: from heart problems to nervous system disorders. Now there are about 7,000 such drugs at various development stages. create Unlike other verticals, in biotechnology, a ready-made product that is introduced into the market raises a very small amount of funds. Investors are betting on companies whose products can undergo clinical research and immediately become leaders in the newly created market. Source: 2019 Emerging Therapeutic Company Trend Report White biotech or the industrial sector is the second-largest biotech business worth $248 billion in Q4 2017. Such companies use microorganisms and enzymes to create industrial materials, from energy and biofuels to plastic. Avantium, for example, creates renewable packaging materials for Coca-Cola and Danone. The German AMSilk produces protein from spider silk, which can be used in various fields. The company has already agreed with Adidas on the creation of biodegradable sneakers as it did with Airbus on the development of materials for aircraft construction. Agriculture (green biotech) and aquaculture (blue biotech) are two relatively small areas of biotechnology. The first is known for the development of genetically modified products, this market is worth $32.9 billion. Here, companies modify the DNA of crops to increase yield, improve their quality, and help adapt to environmental conditions. Monsanto, Syngenta and Bayer are among the largest players in the segment. I should point out, that the field of aquabiotechnology is worth $ 3.5 billion. Marine organisms are deemed a source of more diverse biomolecules. They are used in food additives, medicines and cosmetics. Norwegian Aker BioMarine, for instance, creates foods rich in healthy omega-3 fat from krill, which are small shrimp-like crustaceans. Marinova, another market leader, produces fucoidan brown algae extract, which is then used in food, pharmaceuticals and skincare products. Tech improvement itself also contributed to the growth of investment in the field: DNA sequencing, for example, became ten times cheaper, neural networks and machine learning have enabled numerous process automation. In addition to the United States, startups from Asia raise the most funding in red biotech. Source: 2019 Emerging Therapeutic Company Trend Report Industry Challenges and Prospects First and foremost, discoveries take quite some time: biotech companies spend years researching, developing and testing. I would call it the main shtick of the industry. According to estimates, the process usually takes between 8-10 years. The result is not necessarily predictable: a product can be both a historical breakthrough and a resounding failure. Biostars According to , in 2017, pharmaceutical companies alone spent more than $71 billion on R&D. Not every dollar invested results in something specific. According to IQVIA, only 11.4% of biotech products successfully the path from the first phase of testing to regulatory approval in 2018. PhRMA covered Entrepreneur Daria Loseva, co-founder of the MyWayDNA biotech startup, industry projects share common complexities in the early stages: a need to minimize the risks associated with the product, finding understanding investors and partners, and proving to them the technology works and will be highly relevant in the future. Despite all the difficulties, analysts predict rapid growth of biotech. According to , by 2025 the market will be worth $729 billion, which is almost a twofold increase compared to 2018. At the same time, the average annual growth rate (CARG) will reach 8.3%. Global Market Insights The skyrocketing areas will be agricultural and industrial sectors with 10% GAGR (thanks to a wide range of application areas and production cost reduction). Marine biotechnologies and the healthcare sector will stay a little behind — 9.6% and 6% respectively. Andreessen Horowitz believes that the biotech industry now resembles the 50-year-old IT industry. Well, I am positive that pretty soon the industry will make a rapid breakthrough and become a part of every aspect of our lives, from the treating of illnesses to the materials we use and food we eat.

Andreessen Horowitz

Apple

Coca-Cola

Animal Cloning and New Drugs: How Biotech Works

Blockchains are notoriously challenging network structures to scale, both in performance - such as latencies, throughput, and number of validators - as well as in feature sets - such as customizability of the bytecode that the blockchain executes. In the blockchain implementations we have seen to date, there is a problematic relationship between the number of nodes on a network, and the . speed at which the network can process transactions Furthermore, almost all blockchain implementations feature a monolithic architecture, which means that they implement a single functionality set, thus making them highly dependent on the work of a small number of core developers. To combat the performance problems, layer 2 solutions are aiming to create lightweight data structures to help mitigate the inherent computationally-intensive requirements of existing layer 1 protocols. Sharding, zk-proofs, side-chains, and other advances are helpful in lowering or offsetting computational load, but lack the fundamental protocol corrections needed to compete with centralized solutions like the card networks or the various government sponsored digital payments rails. Furthermore, they introduce their own sets of problems. On the other hand, to combat the monolithic design problem, some blockchain projects have proposed interoperability solutions, but to date these solutions remain improperly designed, such as the shared security model, leading to high overhead and development complexity. In 2018, a new family of protocols that addresses the issues in layer one blockchain technology was published The Snow family of protocols, including the Avalanche consensus protocol, can support millions of nodes with latencies down to less than a second and throughput of thousands of TPS with high security guarantees. by Team Rocket. This is accomplished by using a base proof of stake security model, and a subsampling technique that captures sufficient state information across the network without requiring full node sync for finality. In essence, when a state change (e.g. transaction) occurs, a node broadcasts it to a subsample of nodes to detect if the sample agrees that the state change is valid. This sampling technique is repeated for all sampled nodes until agreement is reached. Once a state is verified, a hashed signature proves the validation to the network to create global consensus. Since 2019, we have been building a platform using the Snow family of consensus protocols. The platform is designed to take advantage of the Avalanche consensus protocol, as well as numerous layer 2 optimizations such as sharding, pruning, and a DAG data storage model. Just as importantly, AVA implements a different network model, which, like the internet, it is built to host and interoperate many different blockchains, spanning the whole spectrum from permissioned to permissionless. Blockchains run in their own subnetworks, which include custom validator sets that may have custom rules for membership, including rules about whether the blockchain is secured by proof of stake, proof of work, or other spam or membership protections. Furthermore, they run custom virtual machines, which dictate the functionality of the subnetwork by encoding any arbitrary logic, and which extends further than simple smart contract platforms, since AVA allows programmability at the network layer, not just the application layer. These are clearly bold claims in the competitive and often hype-fueled crypto space. In an effort toward transparency, we have released the codebase for the AVA platform. This codebase has been publicly released as an open source project seeking community involvement for developing and testing. Developers will have access to the AVA toolset, which includes a wallet, faucet, and libraries for building applications. Designed to promote involvement from the global development community, we have also launched the first stage of our Developer Accelerator Program (DAP) - a bug bounty! More details on this program can be found here: . avalabs.org/dap We invite the Hackernoon community to join AVA and test our code! | | Developer Accelerator Program Documentation GitHub Repo By Kevin Sekniqi, Co-Founder and COO of AVA Labs

Avalanche

Open Codebase and DAP Launch to Incentivize Network Participation [AVA Labs Update]

How to become a Python Programmer in 2023? The best resources you can take to start your journey as Python developer If you have just started learning to code as a Computer Science graduate and thinking to then you have come to the right place. In this article, I am going to share some of the . learn Python in 2023 best online courses to learn Python in 2023 Python is one of the most popular programming languages and it’s used in many domains e.g. Web development, Automation, Data Science, Machine learning etc. In recent years, Python has also become a default language for Projects and that’s another reason why many experienced programmers are learning Python in 2023. Data Science and Machine learning It’s also one of the top languages on my list of . But, for all its power and flexibility, Python is still a very simple language to learn and often touted as a beginner’s programming language. programming languages to learn in 2023 If you are a Java developer or a Web developer using to create web applications, it still pays off to learn Python because becoming a Polyglot programmer i.e. learning more than one programming language not only improves your knowledge and experience but also make you more attractive for companies who value . JavaScript Polyglot programmers If you are thinking to learn a new programming language then also Python is a good choice, particularly if you are looking to move towards a lucrative career path in Data Science and Machine learning which has lots of opportunities. In past, I have blogged about to start with programming but things have come a long way since them. In recent years, Python has grown a lot which resulted in more and more opportunities for Python developers. why a beginner should choose Java over Python Particular in developed countries like the USA and UK, the rise of Python is aphenomenon, and part of that can be attributed to academics where Python has become a go-to language to learn Programming as opposed to , , and . Java C C++ Top 5 Python Courses for Beginners and Intermediate Programmers Anyway, here is my list of from and , two of the best place to learn a new technology online. top 5 online courses to learn Python in 2023 Udemy Pluralsight 1.Complete Python Bootcamp: Go from zero to hero in Python 3 This is one of the most popular courses to learn Python on Udemy and more than 250,000 students have enrolled in it. That speaks volumes about the quality of the course. This is a comprehensive but straight-forward course to learn programming language on Udemy! and useful for all levels of programmers. Python In this course, you will learn in a practical manner. You will start by downloading and setting up Python on your machine and then slowly move on to different topics. Python 3 Complete Python Bootcamp: Go from zero to hero in Python 3 It’s also a where an instructor will show you live coding and explain what he does. practical course The course also comes with quizzes, notes, and homework assignments as well as 3 major projects to create a Python project portfolio! which complements your learning. In short, and it’s also cheap and I bought it for just $10.99 or $9.99 on . one of the best courses to learn Python in 2023 Udemy flash sal e 2. The Complete Python Developer in 2022: Zero to Mastery This is one of the best online courses to learn Python 3 and become a successful Python developer in 2023. This course is created by Andrei Negaoie, one of my favorite instructors who got his fame from Udemy and then moved to open its own Academy known as ZTM Academy (Zero to Mastery). I really love Andrei's teaching style and how it presents concepts that make it easy for beginners to learn things they have never heard before. For example, this Python course will not only teach you Python syntax and programming rules but also a lot of things about programming like variables, loops, functions, modules, and much more. Here are things you will learn in this awesome beginner level Python course Python basics and Programming Fundamentals Object-Oriented Programming with Python Machine Learning with Python Data Science with Python Web Development with Python How to test, debug and run Python programs and applications You will also build more than 12 real-world Python projects to gain practical experience like a Twitter bot to automatically like and retweet your favorite content, a web scrapper with Python and Beautifulsoup to scrap websites, and a professional portfolio website to learn web development with Python. Overall, T is a massive, engaging, and up-to-date course to learn Python in a fun, and engaging way. he Complete Python Developer in 2023: Zero to Mastery Their learn-by- doing method adopted by this course really makes learning easy and you will retain most of the things you learn. In short, one of the best Python courses for beginners. Here is the link to join this course - The Complete Python Developer in 2022: Zero to Mastery 3. [Best Python Course on Coursera] Python for Everybody Specialization Many learning platforms have recognized the power and the future of this beautiful programming language. They have created courses on how to use it and be a professional programmer; of these platforms is Coursera. is a specialization offered by one of the best universities called, Michigan university, and the instructor is one of the best teachers on this platform, showing you how to start from the beginning and be an advanced python programmer through many projects such as web crawler and using SQL databases with Python and . And, if you find Coursera courses useful, you should consider joining ,  a subscription plan from Coursera which gives you unlimited access to their most popular courses, specialization, professional certificate, and guided projects. It costs around $399/year, but it's completely worth your money as you get unlimited certificates. Here is the link to join this course - Python  for Everybody data visualization Coursera Plus Python  for Everybody 4 . Python Fundamentals This is another fun course to learn Python on Pluralsight and great if you have a Pluarlsight membership. I have taken a couple of courses on Python on but this one stick with me. Pluralsight The authors Austin and Rober have done a wonderful job to keep learning fun but at the same time meaningful. Like other courses, this one also starts from ground zero and in first 30 minutes, you will learn how to download and install on your workstation. Don’t worry about your OS, the course covers installing Python on , Windows as well as Mac, so you have covered. Linux The course covers all major Python topics like Strings and Collections, Modularity, Objects, Collections, Handling errors, Iterables, Class, Files and Resource management as well deployment of Python applications. In short, one of the best course to learn Python in , but you need a membership. If you don’t have yet then you can signup for get access to this course for almost FREE? Pluralsight 10-day free trial and 5 . Complete Python Masterclass This is another fantastic course to learn Python on Udemy. I am a big fan of Tim Buchalka and have attended a couple of his courses. His passion and experience really show in his courses and that’s why I am recommending this Python course to you if you are learning Python first time. In terms of content, this is not very different from the other two courses but the style is what matters most. If you can connect with @Tim Buchalka, which is very much possible then you will learn Python very easily and in a fun way. The extensive experience of instructors, both Tim Buchalka and @Jean-Paul on Software development and teaching, which is more than 60+ years together will certainly help you to learn Python in the right way. This course also covers Python 3.0 which is what most of the organization is using now. It’s not the latest but still the major latest version to learn. . The Python Bible | Everything You Need to Program in Python Another awesome course to learn Python on Udemy. This is a project-based course and you will build 11 Projects int this Python Course. If you enjoy hands-on learning while working on the project rather than learning individual concept then this course is for you. This is comprehensive, in-depth and meticulously prepared course and teaches you everything you need to know to program in Python. It delivers what is promised in the title, A-Z, it’s all here! The author, Ziyad is a prize-winning University lecturer of foundation level computer science and delivering project-based learning which is truly reflected in this course. In short, one of the best course to learn Python, if you like project-based learning instead of going through individual concepts one by one. 7 . 30 Days of Python | Unlock your Python Potential In early 2016, Python passed Java as the #1 beginners language in the world. Why? It’s because it’s simple enough for beginners yet advanced enough for the pros. You can not only write simple scripts to automate stuff but also create a complex program to handle trades. You can even use Python for , , , , , and more IoT Web Development Big Data Data Science Machine learning This is a very practical course and useful not just for beginners but also for programmers who know other programming languages like , and want to learn . Java C++ Python In 30 days this course will teach you to write complex Python applications to scrape Data from nearly any website and Build your own Python applications for all types of automation. It’s perfect for busy developers who learn by doing serious stuff. That’s all about some of the . Python is a very powerful language and allows you to create simple scripts to automate stuff from complex object-oriented applications but it is still a simple language to learn. best courses to learn Python Programming in 2023 I highly recommend Python to all beginners who are starting with programming and Coding. Along with , experience will help you in your career for a long time. Java Python programming Other you may like Programming Resources 5 Courses to Learn Web Development in 2023 5 Courses to Learn Spring Microservices with Spring Boot and Cloud 5 Online Courses to learn Core Java for Free 5 Courses to Learn Big Data and Apache Spark 5 Courses to Learn React Native in 2023 Top 5 Courses to Learn TensorFlow and Machine Learning Top 5 ReactNative Courses for Mobiel Application Developers 10 Free Docker Courses for Senior Developers 10 Free Data Science Courses for Beginners 10 Free Machine Learning Courses for Beginners My favorite books to learn Python Programing 5 Best Computer Vision Courses for Beginners 10 free Database and SQL Courses for Beginners Thanks for reading this article so far. If you like this article then please share it with your friends and colleagues. If you have any questions or feedback then please drop a note. — All these are paid courses, if you are looking for some free stuff, then you should check out this list of . P.S. FREE Python Courses

Apache

Keep

Paid

Winner of Noonies 2020 for the award Hacker Noon Contributor of the Year - CLOUD.

2020 - Hacker Noon Contributor of the Year in CLOUD

2022 - HackerNoon Contributor of the Year - Django

2022 - HackerNoon Contributor of the Year - Google Cloud Platform

2022 - HackerNoon Contributor of the Year - Neural Networks

Read my blog

Nominated for 2022 - HackerNoon Contributor of the Year - Django

Nominated for 2022 - HackerNoon Contributor of the Year - Google Cloud Platform

Nominated for 2022 - HackerNoon Contributor of the Year - Neural Networks

Top 5 Online Courses to Learn Python in 2023— Best of Lot

It’s an open secret among devs, almost a joke, that Javascript datepickers suck, hard. But why? As I’ve implemented story scheduling here at Hacker Noon, so we can schedule a story to be published at a specific date/time, I’ve come to understand why they suck so badly (at one point, even trying to write my own). Some datepickers suck worse than others, as I quickly discovered early on in the dev process. Hell, half of the work involved here was picking the right datepicker. First of all, I should mention we’re using React on app.hackernoon.com, so that limits our options right off the bat. See, some datepickers have an event and a prop, mirroring React inputs. Some take that a step further, allowing you to control the state changes yourself. But wait, there’s more! onChange value Setting the state of a datepicker isn’t that bad, whether it’s handled for you or you handle it yourself. The capital-H Hard Part™️ is configuration. Should the user be able to pick a time in the past? Should they be able to pick a date one year in the future? Should they be able to pick any time, or should it go in 5 minute increments? 15 minutes? Do the date picker and time picker have controls? Can these controls be modified? What about the date and time displays? The displays in particular were a pain in the ass. The thing is, we had to customize the displays such that we could see the number of stories scheduled for each day, and ideally, each time slot as well (in 15 minute increments). It wouldn’t do just to click on the day/time and see the number -- you want to be able to see the big picture at a glance and select a day based on where stories are needed. The first method described would be too cumbersome. So what does this have to do with the price of butter? Many of the libraries I looked at offered either time interval configuration, or interface customization, but not both. They offered date validators, but not time validators. They would offer some subset of the features we needed, but it seemed that nothing quite fit our use case. So I decided, with some nudging from , to write my own date/time picker. In retrospect, this was probably not the best route, but we figured that in keeping the interface and rules simple, we could vastly simplify the datepicker for our use case. Dane It turned out that there were a ton of edge cases related to time (surprise, not surprised). I mean, I fixed all of the edge cases that came up, but I was certain there would be cases I missed (like odd time zone changes/consolidations/splits in other countries, for example). At suggestion, I ended up trying , one library I had tried before, but found the time restraint config lacking. The solution ended up being to check for times in the past, and prompt the editor to either publish immediately or choose another time -- it’s worked well so far. Storm’s react-datepicker The only hitch with react-datepicker is that you can’t really customize the date or time displays beyond adding classes. This ultimately means that we can color-code the divs, but we can’t add a number or anything like that unless we use some sort of or selector...we’re still iterating on this, of course. ::before ::after So how does it look for Hacker Noon editors? I added a button near the end of the editor flyout menu: Hacker Noon clock icon And then once the clock is selected, the editor chooses a time (pro-tip: ). noon is a good time to publish And then it publishes at the scheduled time, going off into our story distribution machine: Speaking of : we will be improving workflows, leveling up communication around publication times, making a better editor view to scale our publishing themes , and eventually getting the story scheduler in the hands of our writers. iterating at Hacker Noon In the end, no library is ever perfect, but it seems date/time pickers are a special case where all the behavior you need is there except one key feature, or split between two tools, or what have you. I’m glad we found a solution that works for us (and I’m glad I was able to make the editors’ lives easier), but I would like to tackle the datepicker problem at some point. I have just enough hubris left from my early 20s to think: maybe I could do it right? Eh, we’ll see. For now, back to the internet!

Latest writings on my blog

Story Scheduling at Hacker Noon, or, Why I Hate Javascript Datepickers

My journey of becoming an iOS developer begins in the summer of 2019. Back these days, I did not know anything about programming at all, so it was hard for me even to understand the most straightforward code. As a big fan of Apple, I decided to learn something related to iOS; therefore, I chose to learn Swift. Since there were a lot of tutorials on YouTube, I decided to start learning there. After a week of intense studying, I gained fundamental skills, which allowed me to enroll in Udacity Nanodegree program. With every new project, the difficulty of the assignments steadily increased. Luckily, the motivation that I had was strong enough to keep me on this challenging path to success. As a result, I successfully graduated from this course in less than three months. With all the experience that I got from this program, I decided to build my first complex app. Since I’ve always wanted to build my own messenger, I decided to create it as my first portfolio app. Due to the flexibility that Firebase Database provides, I was able to create a complex friend network for messenger users. Besides that, users receive data almost instantly, which makes a messaging process more enjoyable. Also, mChat uses a Mapbox API, which allows users to choose custom-designed maps and share their location with friends(if the anonymous mode is disabled). Overall, this messenger presents a modern design with lots of animations and custom chat color designs. I look forward to hearing your thoughts about my project! :) Github Repository: https://github.com/realpaliy/mChat

YouTube

From a “Hello World” Project to a Complex Messenger

HackerNoon Tech Stories Archive March 12th, 2020

@rizstanfordRizwan Virk

@tonyspiroTony Spiro

@scottmiddletonScott Middleton

@maxiredMaxence

@dcollins_peak6Dan

@abrarabrar

@iurii-gurzhiiIurii Gurzhii

@wagslaneLane Wagner

@tonyspiroTony Spiro

@tonyspiroTony Spiro

@tanyatopaltanyatopal

@kolosekNesha

@jebJeB

@sophie-zoriaSophie Zoria

@kolosekNesha

@FrederikBusslerFrederik Bussler

@yuriiYurii Filipchuk

@CosmosCosmos

@javinpaulJavin Paul

@austinAustin Pocus

@PaliyVitaliy

@rizstanford
Rizwan Virk

@tonyspiro
Tony Spiro

@scottmiddleton
Scott Middleton

@maxired
Maxence

@dcollins_peak6
Dan

@abrar
abrar

@iurii-gurzhii
Iurii Gurzhii

@wagslane
Lane Wagner

@tonyspiro
Tony Spiro

@tonyspiro
Tony Spiro

@tanyatopal
tanyatopal

@kolosek
Nesha

@jeb
JeB

@sophie-zoria
Sophie Zoria

@kolosek
Nesha

@FrederikBussler
Frederik Bussler

@yurii
Yurii Filipchuk

@Cosmos
Cosmos

@javinpaul
Javin Paul

@austin
Austin Pocus

@Paliy
Vitaliy