Get the most recent info and news about Mini Mania on HackerNoon, where 10k+ technologists publish stories for 4M+ monthly readers.

Parts and Accessories for Classic Mini, BMW MINI Cooper, Austin Healey Sprite, MG Midget & Morris Minor Automobiles
1-800-946-2642

Mini Mania

meesho

Mini PCs are a great option for folks working with limited desktop space. Not only do they offer many of the same features found on their larger siblings, but they’re usually highly affordable ...

Tons Of Mini PCs Are On Sale During Prime Big Deal Days

As the name implies, JLab’s JBuds Mini true wireless earphones are smaller and more discreet than most competing models. And at just $39.99, their enjoyable sound quality, adjustable EQ ...

JLab JBuds Mini

Fujifilm Instax Mini deals can bring the prices of these popular instant cameras way down, and some discounts also include additional film packs to boot. The latest release, the Instax 11 ...

The best Instax Mini prices and deals for October 2023

Some exciting rumblings suggest Apple could release the next generation of the iPad mini, the iPad mini 7, very very soon. Apple's diminutive tablet is due for an upgrade, having last had a refresh in 2021. However, don't expect any major design updates.

iPad mini 7 rumors: Everything you need to know

We tried to find more codes today! In the age of multiversal mashups, Anime Mania is one of the unique anime experiences on Roblox that we have played. Inspired by All Star Tower Defense, Anime Mania shifts away from the top-down tower defense genre toward ...

Anime Mania Codes (October 2023)

Its boss fights and mini-games are disappointingly pedestrian, and very little in this game feels as glorious as Sonic Mania, but compared to the usual output from Sonic Team even a mild success feels like a revelation. In Short: It’s no Sonic Mania ...

Sonic Superstars review - celebrity makeover

Cheats are taking advantage of cricket fans' desperation for tickets to the India-Pakistan World Cup match on October 14. The difficulty fans have faced in purchasing tickets from official sites ...

India-Pakistan World Cup: Fear of fakes puts match mania on sticky wicket

We may earn a commission if you make a purchase from the links on this page. The mini still has its devout fans, silently praying that Apple might bring it back in some form of 3-year refresh cycle. For the time being, if you want one, the iPhone 13 mini ...

iPhone 13 mini review: the small phone that's actually good

Here’s how you can check your PNB account’s mini statement via a simple missed call: Step 1: Open the Phone app on your mobile and dial 18001802223 from your registered mobile number. Step 3: Check the notification panel or open the message app to ...

How to check PNB mini statement by missed call, SMS, net banking, app and more

The best mini PC combines a lot of power in a small package - giving you the flexibility to work how you want and where you want. If you’re looking for an alternative to a large desktop ...

Best Mini PC in 2023

After the world’s richest person Elon Musk acquired the micro-blogging platform Twitter for 44 billion dollars, he started his first day in the office by firing Twitter’s top executives, including CEO Parag Agrawal, Chief Financial Officer (CFO) Ned Segal, and legal affairs and policy chief Vijaya Gadde. Just a few days after that, he reportedly fired a whole team of Data Engineers. Within a week after, Chief Twit announced layoffs for half of all Twitter employees, which are about 3,700 employees out of 7,500. Unfortunately, Twitter is not the only giant to lay off its employees. The long list of companies sacking their employees includes Facebook’s parent company Meta, Google’s parent company Alphabet, Netflix, Byju’s, Microsoft, and many more. Meta has announced layoffs of more than 11,000 employees, which is 13 per cent of the company’s workforce. “Today I’m sharing some of the most difficult changes we’ve made in Meta’s history. I’ve decided to reduce the size of our team by about 13 per cent and let more than 11,000 of our talented employees go”, Meta co-founder Mark Zuckerberg, said in a message to the staff. Situations are similar in India. Around 15,708 people have lost their jobs. The Indian companies included unicorns like BYJU’S, Ola, Meesho, and MPL. The previous year, the Indian education platform Unacademy is reported to have laid off more than 1,000 employees. This year, US companies had around 45,000 employees laid off as of October 2022. Tencent also froze hiring as a measure of cost-cutting. The world's biggest e-commerce company Amazon Inc., has also announced the layoff of more than 10,000 employees. The unicorn startup Byjus announced it will lay off 2,500 employees, that is 5 percent of its workforce, over the next six months. The company is sacking around 2,500 of its 50,000 employees. What Is the Reason for Sacking Employees? Tech firms around the globe are fighting economic slowdowns. The slow consumer spending, higher interest rates by central banks, and strong dollars overseas are hinting toward a possible recession. As a result, tech firms including Alphabet, Meta, and Microsoft are laying off employees or freezing hiring. Recently, Indian IT companies including Wipro, Infosys, and TCS have fired hundreds of employees for moonlighting (simultaneously working for two or more companies). IT firms are worried that employees taking up secondary jobs after regular work hours will affect productivity, lead to conflicts of interest, and possibly data breaches. The most common reason for laying off employees is the pressure from investors and cost-reducing measures. Investors invest in companies in which they see growth potential. So, these companies are removing their excess employees to cut off their excess expenses, as these startups hired workers even when there was no immediate need for them. What’s Next for Those Who Have Lost Their Jobs? There is good news too, Twitter has reached out to its employees who were fired before to return to their positions. Some of those who are being asked to return were laid off by mistake, according to two people familiar with the moves. Others were let go before management realized that their work and experience may be required. It is not easy to get a new job during this time of economic crisis. Those who have lost jobs should invest their time in personal development and skills improvement and should regularly practice. For example, if you design UI, you must keep prpractisingour skills so that when you get time to apply for a new job, you perform better than others. You can also learn something new, if you are good at frontend development, try learning backend development. Remember to keep looking for new job offers on LinkedIn. Tech giants’ employees are likely to be hired by small startups as they have lots of experience. Also published . here

Tech Giants are Laying off Thousands of People. Why?

tl;dr Meeshkan now covers 100% of the Keras Public API! When moved from using a DSL to using Keras for Machine Learning ingress, the Meeshkan core covered only 25% of the . To keep the business running without interruption, we used a small script to determine if a Keras model could be deployed on the Meeshkan backend and, if not, fell back to using either or . Meeshkan Keras Public API Theano Tensorflow What this meant for Meeshkan was higher costs and potentially slower processing for about 75% of our jobs, as these jobs could not take advantage of our distributed infrastructure. We decided not to pass these costs along to our customers but rather keep one low rate for all Machine Learning jobs. Needless to say, we’ve been working tirelessly since the to finish our custom backend so that every Keras job that doesn’t explicitly call can run on our low-cost, pretty-darn-fast network. launch of the Meeshkan Public Beta K.some_proprietary_method I’m thrilled to report that we have finally reached our goal! We will be rolling out a new version of Meeshkan to production over the next few weeks that covers . We’ll start with A/B testing and then using for all customers doing Machine Learning on our network. 100% of the Keras Public API Ugh…this is boring…now that Meeshkan covers 100% of the Keras Public API and all of the tests are passing, there is nothing left to fix. How am I gonna spend my Sundays from now on? With my family? Quick, I better write in a couple bugs! What I’d like to write about today is how we achieved this milestone, what we learned, and why I think this step is particularly important in our company’s evolution. Getting to 100% Writing a Keras backend is a pretty daunting task — aside from the 3000ish lines of Python code needed to glue everything together in Keras, the underlying library on which the backend relies needs to be rock solid, as Keras will not test your library’s code and will only report numerical inconsistencies between your computed results and those of other (assumedly correct) backends. Furthermore, while Keras does a lot of helpful things, it is ultimately up to the backend writer to implement RNN, Convolution, Max Pooling and other thorny tensor manipulations from ML’s greatest hits. Over the course of three months, it took us four rewrites of the Meeshkan core and three rewrites of the backend to get to a version that was production-ready. All the while, CI servers spinning on Amazon made sure that tests were passing and that code coverage was impecable in the most critical places. Lessons Learned For the future Keras backend writer, here are some lessons I learned (the hard way!) about how to put together a backend for Keras. As not all of this information is available in the Keras documentation, I hope it saves people time and provides useful information. 1. Use this gist to organize your work There is no one central place in the Keras code base that tells you what functions you need to implement to get your backend to both pass the tests AND cover the Keras Public API. The closest they provide is at , which lists certain functions that only work for some backends and a few functions that are not part of the Public API (meaning not used in the Keras source outside of the backend code). The dictionary above maps keys to functions in the Meeshkan backend, but you can just grab the keys and grep them in the various Keras backends (i.e. ) to see how they are put together. I’m sure this list will change over time, but that’s what it is on Sunday, January 21st at 5:23PM Helsinki time. https://keras.io/backend grep “def normalize_batch_in_training” 2. Reverse engineer with reckless abandon For 50ish percent of the Keras backend API surface, you can write a mini- backend in less than a week by reverse engineering the Tensorflow backend and comparing your version to the real one. By doing this exercise, you’ll grock how the internals of Keras work and have a pretty failsafe baseline as you begin to implement a backend for your library of choice. 3. **_keras_shape, _keras_shape, _keras_shape** Keras has a more-or-less undocumented interface called that can basically be called anywhere, anytime to get the shape of a tensor. _keras_shape $ git grep -c _keras_shapebackend/cntk_backend.py:7backend/meeshkan_backend.py:1backend/tensorflow_backend.py:8backend/theano_backend.py:79engine/topology.py:23models.py:4 So attach a to, well, everything. We use a sneaky decorator to tack it on . _keras_shape a posteriori 4. Use other Keras backends as a template Keras has three open-sourced fully-featured backends: CNTK, Tensorflow and Theano. Between these three backends, there is almost always one from which you can gain inspiration as you are writing yours. Especially for idiosyncratic Keras conventions like that have no real equivalent in other ML projects, scaffolding your implementation from another backend’s code will save you lots of time. It will also allow you to make logical breakpoints in your code to debug at the same place across different backends. batch_dot 5. Go on, treat yourself to a fully-connected dense neural network Writing a Keras backend, you will invariably get bogged down in the minutia of implementation details for core tensor operations and will not see the forest for the trees. When you’re feeling blue, never forget that you are a dot-product, a ReLU and a L2 Normalization away from a fully connected neural network. Of course, don’t neglect the low hanging fruit! I’d start by just implementing a function that returns your backend’s name. def __backend(self):return 'meeshkan' But once you’ve done this, don’t hesitate to choose your favorite backend functions so that your library can build cool things with Keras right away. It’s insanely motivating! 6. When in doubt, WWND (what would numpy do) A normal human brain cannot possibly anticipate every tensor operation necessary to implement a Keras backend and must read through the code to know what’s going on. In doing so, you will invariably find a few operations that are not present in your base library and that are inefficient to compose from primitives or are just not possible with primitives. So, you’ll be forced to write new functionality into your core library. When doing so, I found that if you aim for API parity with , you can’t go wrong. There are just a couple functions, like , where the numpy implementation doesn’t help much for tensors. Of course, there’s a whole class of functions (like those in ) that can’t by numpy-fied. But the majority of odds and ends you’ll have to implement to glue everything together can be thought of in a numpy state of mind. So keep your API calls similar to numpy — you’ll be thankful you did when others read your backend code understand what’s going on out of the box without having to peruse your library’s documentation. numpy np.gradient tf.nn Deploy with Confidence to Meeshkan Meeshkan has recently achieved a lot of cool milestones — we had some great meetings with , kicked off the program, and passed 6000 student mark on our . But the geek in me thinks that this Keras milestone is the most important of all. Here’s why. partners in San Francisco Accelerace Udemy course Keras is, in my opinion, the of AI. Like the English language, it is imperfect and idiosyncratic, but like the English language, so many enthusiasts are inspired by what it offers that they converge around it to communicate. Meeshkan is no different — we are enamored with the Keras project and feel that lots of great ideas in AI can be articulated with ease using their concise API. de facto lingua franca More and more companies and hobbyists are using Keras every day to make their AI models. And as AI gets easier, people want to do more of it but are blocked by the limits of their own machines. In order for Meeshkan to have a compelling business argument, it needs to give people a substantial advantage over their PC while at the same time providing the same high-quality AI that other Keras backends do. Because Meeshkan is tested against all other Keras backends and passes 100% of the tests, you can deploy to the Meeshkan network with confidence knowing that your AI jobs will yield the same results that one would obtain from other backends like Tensorflow and Theano. At the same time, you get the benefit of the blazingly fast Meeshkan backend, which is the best time-to-money bargain on the market and will only get better once the new backend is fully deployed. I hope that you’ll start building some ! We can’t wait until our new backend is fully rolled out, at which point we will report back with some neat benchmarks. We hope to earn your trust as we strive to offer the industry’s best mix of quality, speed and affordability. incredible ML projects on Meeshkan

The Meeshkan Keras Backend Covers 100% of the Keras Public API: Why This Matters

Happy New Year, Hacker Noon readers! We’re proud to announce that we’ve finished our first full-length course called . The course is available on and you can follow it free of charge. Meeshkan: Machine Learning the GitHub API Udemy _Learn how to plan, deploy and run a Machine Learning problem on AWS and Meeshkan - Free Course_www.udemy.com Meeshkan: Machine Learning the GitHub API | Udemy The basic idea of the tutorial is the following. Let’s say that you are a venture capitalist or talent hunter or you offer a service to developers and you would like to discover exciting new projects on to invest in them, or hire the authors, or strike up a partnership. What type of projects will really take off? Maybe Machine Learning can be used to help us hone our intuition about projects on GitHub. GitHub Meeshkan has a crush on the Octocat. To cut to the chase, by the end of the tutorial, you’ll see that This is even after introducing satisficing criteria to account for the skewed distribution of some targets (i.e. a threshold of will only have a handful of projects in the category while the rest of us mortals hang out with the s). The way we figure this out is by feeding Meeshkan lots of different that serve from the , uploading various to , and passing the results to a in order to visualize how our models are performing. [drumroll] dense, fully connected neural networks learn faster about stars than forks, and models perform surprisingly well when measuring mean-squared-loss between actual and predicted star count in addition to categorical cross entropy for star counts above a certain threshold . stars > 80000 ? 1 : 0 1 0 webhooks data collected GitHub API Keras models Meeshkan small express server Results displayed on our data visualize, which you can find at Our model learns pretty darn fast but then plateaus. Epochs on the X axis, loss on the Y axis. I ran a first batch on a dataset with around 200 000 datum, which was obviously overkill. The lowest line is predictions for a higher star-count (0 for under 1500, 1 for over), whereas the upper two are for 300 stars. https://github.com/meeshkan/data-visualizer. The course takes a day or so to complete and by the end you’ll not only have done some Machine Learning, but you’ll have deployed an entire environment to Amazon Web Services that will automatically ingest publicly-available API data for Machine Learning. If you want to modify the tutorial to do learning about one of the , go for it! I’m excited to see what you learn :-) hundreds of open APIs on the internet In this article, I’d like to talk about the making of the tutorial and, specifically, reveal a few neat things I learned about using Meeshkan to analyze results of large data-collection projects on public APIs. 308 550 That is the number of EC2 server instances that were spawned on Amazon to analyze and inspect the GitHub API. The batch collected data on 995 977 repos and a whopping 14 329 073 commits. The average lifespan of a server was 203.45 seconds, and we bid 0.0040 USD per hour per server. So, when you do the math, 308550 * 203.45 * 0.0040 / 3600 = $69.75 to collect all that data. The main reason I let the job run so long is because parts of the setup (EC2, GitHub, MySQL, the tutorial’s code) can sometimes flake out unexpectedly, which means that you won’t get the number of commits you need to do machine learning. For example, only 125 082 of the repos in the database have over 50 commits. If I run: t2.micro SELECT full_name, stargazers_count, COUNT(*) AS commit_count FROM repos JOIN commits on repos.id = commits.repo_id GROUP BY repos.id ORDER BY commit_count ASC, stargazers_count DESC LIMIT 5; I get: +--------------------+------------------+--------------+| full_name          | stargazers_count | commit_count |+--------------------+------------------+--------------+| impress/impress.js |            32868 |            1 || Automattic/kue     |             6852 |            1 || cdnjs/cdnjs        |             5801 |            1 || square/cube        |             3871 |            1 || enyojs/enyo        |             1941 |            1 |+--------------------+------------------+--------------+5 rows in set (7.51 sec) In other words, false positives, as none of those repos have only 1 commit. But even when we only study repos with more than 20 commits, 100 000 data points is more than enough to get us started. Our database is ingesting a blazing 40 repos per second and around 1000 commits per second! You’ll learn how to deploy this setup in a few clicks on the tutorial. Udemy Also, there are some fun diamonds in the rough. The command: SELECT full_name, size, stargazers_count FROM repos WHERE size = 2 ORDER BY size ASC, stargazers_count DESC LIMIT 5; yields: +-------------------------------+------+------------------+| full_name                     | size | stargazers_count |+-------------------------------+------+------------------+| atg/chocolat-public           |    2 |              197 || edankwan/Jesus.js             |    2 |              114 || bancek/django-smtp-ssl        |    2 |               84 || boucher/stripe-webhook-mailer |    2 |               81 || tlatsas/bash-spinner          |    2 |               50 |+-------------------------------+------+------------------+5 rows in set (0.45 sec) Yes, is really a repo, yes, it has almost nothing in it, and yes, it has managed to rack up 114 stars. Actually, now 115 with one from me :-) https://github.com/edankwan/Jesus.js If you’re not following the tutorial on Udemy and have at least a Yellow Belt in JavaScript and AWS infrastructure, you can deploy this all yourself with just a few clicks from . Pull requests are welcome! https://github.com/meeshkan/github-tutorial-stack 73 This is the number of webhooks that I passed to Meeshkan to get the results for this tutorial. Because the webhook generates the data dynamically based on the path, it is really easy to keep the model the same but change the webhook. A screencast as I was putting together the tutorial— launching a (batch) job in Meeshkan takes less than 30 seconds! For example, the same model can be used to analyze how many distinct authors there are over a 3 or 10 commit window just by changing a parameter in the webhook. The model doesn’t change at all. For each webhook, I uploaded between 3 and 5 models, all of which run in parallel, so we see the download icon for the the first results in minutes and the whole thing finishes up within a few hours for the longest jobs at the Meeshkan network’s very reasonable price of 0.08 USD per hour. 4 This is the number of things I can think of off the top of my head to do after you finish following the course. I would choose to focus only on stars with a threshold of around 1 000 stars for a project, running a few deeper neural nets on just this range. I would run this against my validation set, which is really easy. Instead of passing the webhook , I use and voilà, we are on our validation set. This way, we’ll know if we need to regularize our model. Because you are not prematurely regularizing, right? RIGHT??!? tutorials.meeshkan.io/github/80_10_10_/train/... tutorials.meeshkan.io/github/80_10_10_/validate/... I’d get more data to look at a slightly longer commit history and write a new Keras model that treats commits as a sequence of events. I’d get a version of the model into a production environment ASAP and start using it on new GitHub repos. That way, if we decide that an early version of the model is good enough to start helping us make business decisions, we have a small stash of predictions already waiting for us. No, wait, there’s a fifth thing…you should crack open a nice bottle of Penfolds Grange Hermitage 1951 because you’re going to be making bank with your awesome AI ninja skills after you take our course :-) Udemy Your very own API-crawler on Meeshkan The is itching to do your Machine Learning! We are a small company with a big heart that is taking on the likes of Google and Amazon by offering a low-cost Machine Learning sandbox where anyone can explore new ideas. Here are some great things to do once you sign into Meeshkan: Meeshkan Public Beta . Following a tutorial is great, and building upon it is even better. There are endless hyper-parameters to tune. So once you’ve finished the tutorial, tune away! And remember: if you don’t want to follow the full tutorial, you can just use the webhook to get started, where is explained in video 3 of the tutorial, using a model from video 5 of the tutorial. Deploy the Udemy tutorial materials https://tutorials.meeshkan.io/github/... ... . Machine Learning is for everyone, not just big companies with lots of money, but sometimes it can be daunting to get started. After you watch our tutorial, we hope you’ll have ideas for cool projects you can do on Meeshkan. We’d love to help you get up and running! Sign up for a free consultation through your or directly on . Book a time to chat Meeshkan Dashboard calendly . You can join the group through the Meeshkan console or by signing up using . Join our Slack group this form Thank you very much for checking out our Machine Learning service. We think you’ll like it a lot and we are working every day to make it faster, cheaper and easier to use. Did I mention you get 100 free hours of Machine Learning and a free fifteen-minute consultation to get your ML job up and running? ! See you on Meeshkan

Machine Learning the GitHub API with Meeshkan

tl;dr The Meeshkan Machine Learning Public Beta is officially live. Go learn something on it! Cloudera In 2010, Cloudera played one of the Internet’s great April Fools jokes. Called , they claimed that Apache Hadoop was being run on 1,000,000 Nexus devices to help Google process images. Little did they know that, as mobile computing got more powerful, this is exactly what companies would do all the time — using the machines that connect to their network as an extension of their servers. Pushing the Limits of Distributed Processing Putting aside the Orwellian overtones, this brings up a great question — what if we actually worked together to make distributed systems that could help people crunch data while paying device owners for usage time? Of course, most people in 2010 did not have access to massive amounts of data, let alone the knowledge to analyze it. But it’s 2017. Andrew Ng has published his second on Coursera. Companies are hiring dedicated ML engineers and using (MLaaS) in their workflows. Artists are using Machine Learning to create . Individuals’ personal devices, , are getting more and more powerful. People to mine virtual currency. The game is different… Machine Learning course Machine Learning as a Service crazy cool work even wearables fly GPUs across the globe A World of SETIs The project describes itself as a network of “Internet-connected computers in the Search for Extraterrestrial Intelligence (SETI). You can participate by running a free program that downloads and analyzes radio telescope data.” Your computer helps SETI sift through an obscenely large, ever-increasing dataset to find aliens. How cool is that?!? SETI@home In 2017, the business world is starting to look a lot more like a bunch of mini-SETIs. With techniques like data augmentation as well as awesome projects like the and , people can run their Machine Learning models on giant open datasets and data-streams. So while data is becoming accessible, the problem is that most companies and individuals cannot afford experimenting with Machine Learning in a sandbox environment to gain intuition and understanding about their models. IOTA Data Market OpenMined Meeshkan is a network where anyone can train and test Machine Learning models in a massively parallel way on participating host devices. In return, the hosts make a few bucks. This brings Adam Smith’s invisible hand into GPUs and CPUs and allows people to work with big data without the big price tag. Of course, a Raspberry Pi can never be as fast as an NVIDIA 1080Ti. However, thousands of these little guys will, like the , divide and conquer to get things done in a reasonable time. As device owners make money, businesses can test out ideas cheaply. Enter Meeshkan . African Wild Dog Meeshkan: What’s in a Name? Meeshkan is short for the Hebrew word , which roughly translates to “your dwelling places” and is found in Numbers 24:5. Balaam, unexpectedly moved by the Israelites’ makeshift encampment in Sinai, proclaims “How beautiful are your tents, Jacob, your dwelling places, Israel!” ( ). meeshkenotecha Ma tovu ohalecha Ya’akov, meeshkenotekha Yisrael Biblical distributed systems — Numbers 24:5. Meeshkan takes this small anecdote as a starting point for the entire service. The company’s mission is to make something beautiful out of our scattered digital dwelling places working . together Let’s think about who wins and who loses… Device owners, called “sellers” in Meeshkan, because they make money just by connecting to the network. win Machine Learning practitioners, called “buyers” in Meeshkan, because they can run massively parallel jobs at a fraction of the cost of services with their own machines. This is because a market of thousands of devices dictates the price. win Meeshkan because our company takes a small fraction of each transaction from buyer to seller, not unlike other P2P and B2B services like eBay, Uber or AirBnB. wins Companies that overcharge you for Machine Learning . No, just kidding, everyone wins, so they too because they can lower their prices. In fact, in our private alpha, a Machine Learning consulting company automatically drew extra capacity from the Meeshkan network when demand was high. lose win Go Learn Something! Mosey on over to , then click on one of the “Try Now” links on the homepage to get to the . Meeshkan web app A few random musings: We give you of ML on Meeshkan. Use it, break stuff, report back to us, and help us make Meeshkan better. 100 free hours Once you’re in Meeshkan, click on , scroll to the bottom, and sign up for our Slack #public-beta channel. Settings Subscribe to @ on Medium and @ on Twitter to watch our first end-to-end tutorial, called “Machine Learning the Github API,” which will be available next week. meeshkan MeeshkanML We are open-sourcing our codebase as fast as possible. Check out our projects on . https://github.com/meeshkan Do you like coffee? Do you want more money to buy some? ! Or download our and we’ll transform your droid into a Deep Learning workhorse and pay you for usage time with . Put your device on Meeshkan Android App Iota Do you like coffee? Do you want more money to buy a lot of it? really jobs@meeshkan.com What are you waiting for? on Meeshkan! GO LEARN SOMETHING Your partner in machine (and human) learning,Mike SolomonC.E.O., Meeshkan Machine Learning P.S. Did anyone else pull an all-nighter last night to finish before the deadline? I haven’t been this excited for new installments of a series since the final Harry Potter books. BOOYAH! Convolutional Neural Networks

Meeshkan Machine Learning

Crack Long-Form Blogs: Complete Guide & Experts' Insights 

An exact clone of youtube with all functionalities from View count to Subscribe to everything (Without Youtube Api) Using Firebase, FFmpeg, And React First of all, I would like to say sorry for my writing inconsistency. My channel has been dead for a year now but the fact that I still had been getting views until I checked today made me realize. So, yeah Here I am back again :) with some really cool projects that I made during the last 1 year. I hope To get support back again Now let’s come for what you guys have come for :)) It’s my summer vacation and I devoted a much longer time completing this project with all the functionalities which I planned to add. Not to be joking, This Project contains tons of features. ? See it on Your own Don’t believe me https://youtu.be/5PrkNg5yIu0 As seen from the video, You Can probably tell that UI is not too bad too, In the project I tried to focus on every part (as it was my 1st coding project after a long time). I made the app fully responsive, Animations can be seen, And Here are the additional features I forgor to tell in the video: Animation And Ui ( ) For eg You can pass a property in your Component something like this on any page you want: And Good Authentication System Forgot password, Reset Password, Email Verification System emailVerified ProtectedRoute . 3 View Count System Based On your Ip Views are counted per 1 IP. So, that doesn’t result in an extra view when you refresh Your page. You can use any library/api to get Ip but I am using the following Api: (You can Search Videos on the header based on a video’s title (Not tags) ) Search Functionality (The Related Side videos on the watch screen and on the home screen are fetched based on the tags you upload when you upload a video…) Get Related Videos Based On Tags (Make Your video Public or Private (You can edit it later too) ) Publicity Functionality (quite self-explanatory bruh) Subscribe Your Favorite Creator and Like/Dislike and Comment Functionality (Only a video owner can edit his video and change the visibility or other things of videos) A Video Owner Can edit his Video (After Uploading Video, The app automatically generated thumbnails for your video, and the thumbnails and videos are stored in your server like this: ) Generate Thumbnails For Your video Using FFmpeg (I pretty much wasted a whole day on making good firebase rules but at last it was worthy) Strong and some advanced Firebases Rules I hope this project is exciting for you as much as it is exciting me rn. And as always it’s gonna be a project in parts and I promise to post an article per week (On Saturday/Sunday). But besides that I have various projects aligned to show you all from an SOS react native App to your Own Game to much more. And not gonna Lie, these projects gonna be fun and full of learning So, I think to let’s continue from the next article setting up our project and go on till the end of making this project. For those ones who are curious to know the project and file structure, here are these :) File Structure Bruh Lmao these are too many files not even coming on 1 screen Firestore Videos Collection Structure {
"data": {
"GVSA4od75f5BOq2CfmBK": {
"Categories": ["Ballcat", "Cats", "#meme"],
"Userpfp": "https://lh3.googleusercontent.com/a-/AOh14Gg6y6DrYdtYNYl-p_6eV3OyUAARBD0J4H4bTgML=s96-c",
"Usersname": "harshvardhan jain",
"comments": [
{
"Comment": "Good video",
"VideoId": "GVSA4od75f5BOq2CfmBK",
"name": "harshvardhan jain",
"timestamp": {
"__datatype__": "timestamp",
"value": "2022-06-13T09:58:16.376Z"
},
"userId": "EK9nslWIRFgwjnhPuFS9GnPgMQN2",
"userPfp": "https://lh3.googleusercontent.com/a-/AOh14Gg6y6DrYdtYNYl-p_6eV3OyUAARBD0J4H4bTgML=s96-c"
}
],
"descriptions": "This is how you make slime....I am gonna right a long description for testing, so you have to cope.......Me trying my best to make it long but still short aaaaaaaa GOD help me.....Should i just copy paste some dummy text? hmmmm SUS aaaaaaaaa",
"dislikes": [],
"duration": 14.84,
"filePath": "Uploads\\file-1655114264218-cute_cat_relax_on_outdoor_ground_6892533.mp4",
"likes": ["EK9nslWIRFgwjnhPuFS9GnPgMQN2"],
"ownerId": "EK9nslWIRFgwjnhPuFS9GnPgMQN2",
"publicity": "Public",
"thumbnail": "http://localhost:5000/Uploads/thumbnails/thumbnail-file-1655114264218-cute_cat_relax_on_outdoor_ground_6892533.png",
"timestamp": {
"__datatype__": "timestamp",
"value": "2022-06-13T09:57:45.800Z"
},
"title": "This Is How Cat Relax Uwu",
"views": ["122.173.27.25"],
"__collections__": {}
},
"UAge5Pgww7Z1vWsYro4l": {
"Categories": ["Tankster.io", "Online games", "#gaming"],
"Userpfp": "https://lh3.googleusercontent.com/a-/AOh14Gg6y6DrYdtYNYl-p_6eV3OyUAARBD0J4H4bTgML=s96-c",
"Usersname": "harshvardhan jain",
"comments": [
{
"Comment": "A comment from new user",
"VideoId": "UAge5Pgww7Z1vWsYro4l",
"name": "A NEW USER",
"timestamp": {
"__datatype__": "timestamp",
"value": "2022-06-13T03:47:49.003Z"
},
"userId": "MmQFFNPBNgW2BNiHH1oJYxas1F82",
"userPfp": "https://cdn-icons-png.flaticon.com/512/149/149071.png"
}
],
"descriptions": "Tankster.io A pretty nice game that I found online !! It's discord is good place Yea changed description",
"dislikes": [],
"duration": 34.835,
"filePath": "Uploads\\file-1655053918168-Tankster.io - Google Chrome 2022-05-15 21-59-05 (online-video-cutter.com).mp4",
"likes": ["EK9nslWIRFgwjnhPuFS9GnPgMQN2", "MmQFFNPBNgW2BNiHH1oJYxas1F82"],
"ownerId": "EK9nslWIRFgwjnhPuFS9GnPgMQN2",
"publicity": "Public",
"thumbnail": "http://localhost:5000/Uploads\\thumbnails\\file-1655114375973-Screenshot 2022-03-02 165903.png",
"timestamp": {
"__datatype__": "timestamp",
"value": "2022-06-12T17:12:00.247Z"
},
"title": "Tankster.io An Online Game pretty nice Game",
"views": ["122.173.31.158", "122.173.27.25"],
"__collections__": {}
}
}
} With That, I think we come to the end of our article. But the next article is not too far ;) Hope to See you guys excited and we’ll meet you in the next part Till then stay safe, stay healthy Thank you Also Published here

Build a Fully Functional Youtube Clone Using Firebase, FFmpeg, And React. Part 1

Why we use unmock for our full stack Let’s face it, mocking sucks. Every programming language and platform has different tools to help you mock API calls, but all of them share some common fragilities: No clean way to verify if the mocks are correct representations of the API. No way to avoid reverse engineering the API you’re mocking. No way to quickly mock internal APIs. No way to monitor in production if the mocks from your tests resemble the data that you’re getting back in the wild. As a result, teams lose tons of time and energy writing mocks for their unit and integration tests, which leads to buggy tests, less test coverage, and ultimately a less reliable production environment. unmock Enter . Unmock is great because it basically takes care of all of this stuff for you. There is a good YouTube overview of the service (currently in closed beta), but I thought I’d walk you through how unmock speeds our team up at Meeshkan. unmock

Never mock an API again

tl;dr yarn add --dev faceplant You’ve developed your killer app. It’s passing unit tests, integration tests, end-to-end tests... You’re so content that you print the code out and pin it to your fridge. You deploy to production, but in a few hours, you get complaints that the service is more or less completely useless. You dig through debug logs, but to no avail — there’s nothing obviously wrong, and tests are passing, so your users must just be exaggerating. All of them. express Or maybe you forgot to set an environment variable in production. From my experience, this is the #1 source of service failure in well-tested production apps, and this has reared it’s ugly head at Meeshkan as well. To fix this, we made , a TSLint plugin for production workloads. The goal is to replace start commands like with things like so that a linting step is done on the production environment and the service will fail fast if the lint check doesn’t pass. [faceplant](https://github.com/Meeshkan/faceplant) ts-node dist/server.ts tslint --project . && ts-node dist/server.ts The first rule in this package, , allows you to check to make sure that your env variables are set in both dev and production environments. There are options for reading from an file or using the actual environment. You can even have it check for variables not present in your code. All this adds up to safer production code and a more pleasant development experience. no-unset-env-variables .env If you’ve never developed a TSLint plugin before, it’s dead easy. We used this to get started and figured out how to make it production-ready by following the conventions in . great boilerplate this repo Otherwise, when in doubt, use . [http://i-forgot-to-set-my-env-variable.com](http://i-forgot-to-set-my-env-variable.com) axios.get(process.env.API_URL || "http:// i-forgot-to-set-my-env-variable.com") welcome! Pull requests

Forgetting to set environment variables in nodejs apps

TL;DR — Apollo Server & Typescript ⇒ graphqlgen At Meeshkan, we were writing our own typescript interfaces for JS object representations of our GraphQL schema, which proved to be error-prone and laborious. The new package by makes that crazy easy, but if you are not comfortable with GraphQL schemas yet, it can feel like another layer of incomprehension in an already incomprehensible stack. I wanted to take a moment to share a few key learnings from working with this new tool in case you are struggling to get started: [graphqlgen](https://github.com/prisma/graphqlgen) Prisma How resolvers in Apollo Server actually work. Why is awesome. graphqlgen Why you should use prisma types for your server’s schema’s types. never Bonus: How to unit your test resolvers. How resolvers work in Apollo Server Before you can work with effectively, you need to understand the basics of how resolvers work in Apollo Server. graphqlgen The main thing that helped me in grocking resolvers was to learn how they cascade, not unlike reducers. The from Apollo Server is a helpful start, but I couldn’t quite figure out how everything tied together I concocted a bad example showing the problem they’re trying to solve. redux documentation Let’s assume that your schema is: type User {id: ID! @uniqueposts: [Post!]!} type Post {id: ID! @uniqueauthor: User!} type Query {user(id): User} One (bad) way you could write your resolvers is like this: export const Query: MyQueryResolverType = {user: async (_, { id }, __) => {const user = await getUserFromDatabase(id); // fetch userconst posts = await getUserPostsFromDatabase(id); // fetch postsreturn {...user,posts,}}} export const User: MyUserResolverType {id: ({ id }, _, __) => id,posts: ({ posts }, _, __) => posts,} export const Post: MyPostResolverType {id: ( { id }, _, __) => id,author: ({ author }, _, __) => author,} There are numerous problems with the code above: we have had to write our own types for resolvers like , and . will take care of these for us, which we’ll see in a bit, but for now we’re stuck with that fragility. MyQueryResolverType MyUserResolverType MyPostResolverType graphqlgen The main problem for this section is that the query is doing both too little and too much work. We have no clue if the incoming graphql query is going to even ask for the posts, so why would we fetch them here? Furthermore, if the person does something like: user {user(id) {posts {author {posts {author {id}}}}}} …this will break, as it won’t recurse down the tree. So let’s fix it with proper resolvers. export const Query: MyQueryResolverType = {user: (_, { id }, __) => getUserFromDatabase(id); // a promise} export const User: MyUserResolverType {id: ({ id }, _, __) => id,posts: ({ id }, _, __) => getPostsFromUser(id), // a promise} export const Post: MyPostResolverType {id: ({ id }, _, __) => id,author: ({ id }, _, __) => getUserFromPost(id); // a promise} And no more breakage. ApolloServer traverses down the tree, calling the correct resolver as needed. This is because the first argument passed in to the resolver function is the representation of the object returned from the previous resolver, which we use to get important information about the object like the . In the ApolloServer documentation, this object is called , which I think is a poor choice for a name. I would opt for or or . id parent partial_obj obj_hint obj_with_stuff_missing Why is awesome graphqlgen The “good” example above has (at least) two pesky anti-patterns: the Resolver type needs to be written by hand, and there are functions like that are basically identity functions. fixes this. It will automatically generate Typescript interfaces for resolver types default resolvers for resolvers. Goodbye boilerplate! Post.id graphqlgen and To use , simply include a file in your root directory, and then somewhere in your build to generate the goods. The file has, at the time of this writing, six keys that we use, all of which are pretty self-explanatory: graphqlgen graphqlgen.yml yarn add --dev graphqlgen npx graphqlgen graphqlgen.yml # The target programming language for the generated codelanguage: typescript # The file path pointing to your GraphQL schemaschema: ./src/schema.graphql # Type definition for the resolver context object# The context object is passed into ApolloServer.# It usually contains stuff like a database connection# or user id or whatever.context: ./src/types.ts:IContext # Find TS models that correspond to the graphql schema and map them# NOTE: this part is tricky and is the subject of the next section,# so read on to find out more.models:files:- path: path/to/my/resolver/types.tsdefaultName: 'I${typeName}' # my convention- path: path/to/my/generated/prisma/typesdefaultName: '${typeName}Node' # prisma convention # Where all of this boilerplate code is writtenoutput: ./src/generated/graphqlgen.ts # In case you need inspiration for your resolvers, this generates# stubs that you can fill inresolver-scaffolding:output: ./src/generated/tmp-resolvers/layout: file-per-type # one file per type in the Now, if we revisit our previous example, things get a whole lot easier to follow: export const Query: QueryResolvers.Type = {user: (_, { id }, __) => getUserFromDatabase(id); // a promise} export const User: UserResolvers.Type {...UserResolvers.defaultResolvers,posts: ({ id }, _, __) => getPostsFromUser(id), // a promise} export const Post: PostResolvers.Type {...PostResolvers.defaultResolvers,author: ({ id }, _, __) => getUserFromPost(id); // a promise} All of the resolver types are hanging out in , and we no longer need to roll our own. Furthermore, default resolvers are generated for every type, so we no longer need to write all that boilerplate. Sweet! src/generated/graphqlgen Why you should use prisma types for your server’s schema’s types never Prisma is fantastically efficient. So efficient that you will just want to ping your database from your client using GraphQL and dispense of Apollo Server entirely. This is tantamount to reading and writing to a database directly from a web page which is, in short, a very bad idea. Now that is around, it is even more expedient to create an Apollo Server in record pace, and in doing so, you may be tempted to use the interfaces generated by Prisma as the interfaces representing types in your schema. This is also a bad idea. For example, let’s imagine that your prisma contains: graphqlgen datamodel.prisma type UserNode {id: ID! @uniqueemail: String!internalFieldUsedByDevs: String!anotherInternalField: String!} whereas our contains: schema.graphql type IUserNode {id: ID! @uniqueemail: String} If you do not want and to be reachable by your client, you probably do not want the same Typescript type as prisma generates. The more sensible alternative would be to write your own type and use that. No problem, except how does know how to use yours and not prisma’s? Let’s go back to . internalFieldUsedByDevs anotherInternalField graphqlgen graphqlgen.yml models:files:- path: path/to/my/resolver/types.tsdefaultName: 'I${typeName}' # my convention- path: path/to/my/generated/prisma/typesdefaultName: '${typeName}Node' # prisma convention Here, . will first use the types in and then try to get whatever it does not find from the generated prisma types. That way, will correctly be documented and type-checked as the that we wrote and not generated by prisma. order matters graphqlgen types.ts Query.me IUserNode UserNode While this seems trivial at first, as your server schema and prisma schema start to diverge, you will wind up in situations where your server schema is not a strict subset of your prisma schema. Thus, separating the two will become essential, and it’s better to get in the habit of enforcing that early. Bonus: How to unit test your resolvers Doesn’t really have anything to do with , but hey, since we’re talking about writing resolvers, let’s not forget to test them! graphqlgen import * as request from "supertest";import app from "../../path/to/my/express/app"; const ENDPOINT = "/graphql"; test("test hello", async () => {const { text } = await request(app).post(ENDPOINT).send({query: "{ hello }"});expect(JSON.parse(text).data.hello).toBe("Hello!");}); Underwhelming, I know, but the cleanest things often are. As I couldn’t really find anything online that stated this this succinctly, I thought I’d put it here. This assumes that you have set up an express app in a separate file from your server so that you don’t have to start and stop the server for each test, which can lead to all sorts of wacky port-related errors. We’re using Jest at Meeshkan, but mocha or really anything is fine. At some point, you’ll need to mock your database communication object (in our case, the instance generated by ), which is pretty straightforward on any of these platforms. prisma prisma generate So there you have it! Almost-painless type checking, default resolver scaffolding and unit testing. w00t!

Using graphqlgen for our cloud stack

Many programmers are moving towards data science and machine learning hoping for better pay and career opportunities — and there is a reason for it. The has been ranked the for last a couple of years and the average salary of a data scientist is over in the United States according to Indeed. Data scientist number one job on Glassdoor $120,000 Data science is not only a rewarding career in terms of money but it also . IMHO, that’s the main motivation many good programmers are moving towards , , and artificial intelligence. provides the opportunity for you to solve some of the world’s most interesting problems data science machine learning If you are in the same boat and thinking about becoming a data scientist in 2018, then you have come to the right place. In this article, I am going to share some of the by learning about data science, machine learning, and artificial intelligence. best online courses you can take to become a data scientist in 2018 When it comes to programming languages and libraries, and are clearly two leading programming languages in the areas of machine learning, data analytics, and artificial intelligence. Many exciting developments are happening in those languages and they already have some useful libraries to start with. Python R If you’ve got some programming or scripting experience, these courses will teach you the techniques used by real data scientists and machine learning practitioners in the tech industry and prepare you for a move into this hot and lucrative career path. 5 Courses to Learn Data Science in 2018 Here is my list of the top five online courses you can take to learn data science and machine learning and become a data scientist. These courses will provide you with valuable insights, advice, tips, and recommendations to take your data science career to the next level. It rightly contains the courses on both and , two of the most popular programming languages of data scientists. Python R : 1. Python for Data Science and Machine Learning Bootcamp Many of my friends and colleagues who have moved towards data science have chosen Python as their choice of language. One reason is that Python is probably the most popular language and offers everything. They were already using for scripting and and it’s easy to learn data science if you already know the language rather than learning a new language like R. Python object-oriented programming Python for Data Science and Machine Learning Bootcamp This is one of the best courses to master data science and machine learning python libraries. You will learn how to use most of the popular Python machine learning and deep learning libraries like , , Seaborn, Matplotlib, Plotly, , , , , and more! NumPy Pandas Scikit-learn Meeshkan TensorFlow Keras This course is designed for both beginners with some programming experience and experienced developers looking to make the jump to data science! It is comparable to other data science boot camps that usually cost thousands of dollars, but now, you can learn all that information at a fraction of the cost — with just if you’re lucky to get it during Udemy’s flash sale. $9.99 : 2. Data Science, Deep Learning, and Machine Learning With Python This is one of the rare hands-on courses to learn the neural networks, artificial intelligence, and machine learning techniques that employers are seeking. This course is designed by analyzing the real requirements in data scientist job listings from the biggest tech employers, which means it covers the machine learning and data mining techniques real employers are looking for. : 3. Careers in Data Science A-Z If you have an interest in data analytics and machine learning and are wondering about how to become a top-level data scientist, then this is the right course for you. In this course, you will not only learn the basic steps on how to become a data scientist but also real-world tips and tricks to take your data science career to the next level. Careers in Data Science A-Z™ | Udemy 4. Tableau 10 A-Z: Hands-On Tableau Training for Data Science! Tableau is one of the popular tools among data scientists and that’s because there is a great demand for data scientists who know Tableau. This course will teach you Tableau 10 for data science step-by-step. . It contains real-life data analytics exercises and quizzes to give you first-hand experience with Tableau You’ll learn all of the features in Tableau that allow you to explore, experiment with, fix, prepare, and present data easily, quickly, and beautifully. 5. R Programming A-Z: R For Data Science With Real Exercises! R is another popular programming language for data science and this course provides a good overview of R from a data science perspective. There are lots of and lectures out there. R courses However, and students often get overwhelmed. R has a very steep learning curve This course is truly step-by-step. In every new tutorial, we build on what had already learned and move one extra step forward. That’s all about . As I said in the first paragraph, data science is a very lucrative career with lots of exciting opportunities to solve some of the most interesting problems in the world. some of the best courses to become a data scientist in 2018 If you have a passion for data analytics and problem-solving and want to make a difference both in your career and the world, this is the right time to make your move. I’ll write more about data science and machine learning for both and developers as my knowledge grows in this areas. Python Java Many of my readers have asked about libraries, frameworks, and books in and that’s on my list of things to do this year. These online courses are a good point to starts with. Java and Python for machine learning BTW, here are some other programming and development courses you may like to explore in 2018: 5 Big Data and Apache Spark Courses for Java Developers 5 Spring Microservice Courses for Java Developers 5 Courses That Can Help You Become a Scrum Master in 2018 5 Courses to Learn Python in 2018 5 React Native Courses for JavaScript Developers 5 Free Courses to Learn Spring Boot and Spring MVC in 2018 10 Free Docker Courses for Java Developers to Learn DevOps 5 Free Jenkins and Maven Courses for Java Developers Closing Notes Thanks, You made it to the end of the article … Good luck with your Data Science and Machine learning journey! It’s certainly not going to be easy, but by following this roadmap and guide, you are one step closer to becoming a . Data Scientist If you like these data science and machine learning courses, then please share with your friends and colleagues, and don’t forget to follow on Twitter! javinpaul P.S. — If you need some FREE resources, you can check out to start your journey. Deep Learning Prerequisites: The Numpy Stack in Python

Top 5 Data Science and Machine Learning Course for Programmers

Friday morning at … Meeshkan Machine Learning pylaver (0.10.3)================ >>> send('toosoonsmallcodedoubleplusungood') send('wtf?') <<< >>> send('c733b44a686ec4d7af75e14e0160d375fdae39fe') send('it must be friday...') <<< >>> from pantheon import knuth>>> send(knuth('Being overly eager in the rewriting of code ''so as to maximize efficiency and speed in the retrieval, storage, ''and execution of its compiled vicissitude is an act whose ''repeated doing metaphorically anchors every conceivable ''manifestation of wrongness, badness and immorality to the ground ''and from which these dark forces derive their energy.')) send('ah, yes. true that.') <<<

What Life Is Like Sometimes

Uncle Bob’s seminal post has been an inspiration for many software designers and has become increasingly relevant as JavaScript allows business logic code to be reused. There never was an excuse for spreading business logic out all over a code base, but there is even of one now that we can and then wherever we need it. Clean Architecture less npm publish core-logic npm install --save core-logic This article takes as its point of departure the as defined by Uncle Bob: Use Case The software in this layer contains business rules. It encapsulates and implements all of the use cases of the system. These use cases orchestrate the flow of data to and from the entities, and direct those entities to use their business rules to achieve the goals of the use case. application specific enterprise wide Use cases are often described as: Post an image of ; or my cute cat Fetch a list of movies made by J.R. Bookwalter However, software is messy, abstractions are leaky and often times, if you’re like me, you find yourself dreaming up leviathan Use Cases that spin out of control or trivial Use Cases whose logic you’ve already coded into the UI and that you won’t be removing anytime soon. What I’d like to present below is some low-hanging fruit for Uncle-Bobbing your Redux app with Middleware. is a simple JavaScript state container that has become a pattern unto itself that is reproduced in libraries for and . I strongly recommend you use it — introducing it into your app is trivial and after ten minutes of working with a redux state you’ll fall in love and never turn back. Redux Kotlin Swift Motivating example — A Log In Screen As a motivating example, I’ll use the Log In screen that actually applies this pattern. As a side note, the whole app applies this pattern, so feel free to download it, poke around and reach out to us if you want to know how certain things are done. Meeshkan Let’s start with a simplified version of this log in screen: // login.jsimport React from 'react';import { reduxForm, Field } from 'redux-form';import { connect } from 'react-redux';import { Button, MyTextComponent } from 'ui-lib';import { Text, Alert } from 'react-native';import Actions from 'react-native-router-flux';import analytics from 'analytics-lib';import { login } from 'kludgy-implementation'; const onSuccess = () => {analytics.report('successfully logged in');Actions.main(); // this is our navigation action} const onFailure = () => {analytics.report('log in error');Alert.alert("Sorry...", "Something went wrong");} const submitter = ({email, password}) =>login(email, password, onSuccess, onFailure); const LogIn = ({handleSubmit}) => (<View><Field component={MyTextComponent} name="email" /><Field component={MyTextComponent} name="password" /><Button onPress={handleSubmit(submitter)}><Text>Log In</Text></Button></View>); export default connect()(reduxForm({form: 'LogIn'})(LogIn)); There are several problems with this implementation: Our function must bubble up to the UI to accept success and failure callbacks. What if we want to change how many steps are in the log in process or add more options than success and failure? Refactoring doom… login Ditto for our analytics function, and even worse, we now have to test that every component calls it correctly. What a mess! We have to remember the order of success and failure and why/how it matters, sprinkling logic all over the UI without any spec that tells us why this is the case. The day that we want to change navigation, analytics or alert libraries we are in for a week of refactoring every UI component. Hello subtle code breakage. Middleware to the rescue! Let’s rewrite the above example using middleware. To start, I’ll show our new and improved component straightaway: // login.jsimport React from 'react';import { reduxForm, Field } from 'redux-form';import { connect } from 'react-redux';import { Button, MyTextComponent } from 'ui-lib';import { Text} from 'react-native';import { loginAction } from 'better-implementation';import Ize, { navSuccessIze, alertIze, alertFailureIze } from 'ize'; const login = Ize(loginAction,navSuccessIze('main'),analyticsIze(),alertFailureIze("Sorry...", Something went wrong")); const LogIn = ({handleSubmit, login}) => (<View><Field component={MyTextComponent} name="email" /><Field component={MyTextComponent} name="password" /><Button onPress={handleSubmit(login)}><Text>Log In</Text></Button></View>); export default connect(null, {login})(reduxForm({form: 'LogIn'})(LogIn)); Some reasons to celebrate: Gone are the explicit calls to libraries like navigation or analytics! Gone is spreading logic all over and callbacks that we can’t unit test without triggering all sorts of side effects! onSuccess onFailure This is way shorter! This is way easier to read! This is way easier to test! Ok, but how do we get there through middleware? 1. Using Action Creator Creators via the Ize pattern We use to implement the Action Creator Creator pattern. Basically, all of that stuff takes an action and adds a bunch of useful metadata for middleware. [redux-ize](https://www.npmjs.com/package/redux-ize) Ize 2. Use a library like redux saga to handle async calls Check out to see how one may handle our async log in call and dispatch success or failure events. As an example: [redux-saga](https://redux-saga.js.org/) import { call, put } from 'redux-saga';import { loginSuccessAction, loginFailureAction } from 'actions';import Ize, { navIze, alertIze } from 'ize'; function* logInSideEffect({payload: {email, password},meta: {navSuccess, alertFailure}}) {try {call(login, email, password);put(Ize(loginSuccessAction, navIze(navSuccess)));} catch(e) {put(Ize(loginFailureAction, alertIze(alertFailure)));}} Note how we use the pattern again here to move and information to a normal and track. This will make sure it is picked up by the middleware when the success or failure action is dispatched. ize navSuccess alertFailure nav alert 3. Create some middleware to handle analytics, navigation and alerts Easy like Sunday morning… // analytics.jsimport analytics from 'my-awesome-analytics-provider'; export default store => next => action => {action.meta && action.meta.analytics && analytics(action.type);next(action);} // nav.jsimport Actions from 'react-native-router-flux'; export default store => next => action => {action.meta && action.meta.nav && Actions[action.meta.nav]();next(action);} // alert.jsimport Alert from 'react-native'; export default store => next => action => {action.meta &&action.meta.alert &&Alert.alert(...action.meta.alert);next(action);} We have even more reasons to celebrate! Want to change your analytics provider? No prob — two lines of code. Want to test your middleware? No prob — use a stub in one place. sinon Want to kill off certain navigations or alerts depending on the app’s state? You have access to the store now! Want to test the order of all this stuff? No prob — just write one test to verify your redux store configuration. Summing it all up — Redux is More than a State Container Redux markets itself as “a predictable state container for JavaScript apps.” This is true, but one thing you’ll notice is that I haven’t mentioned a single reducer that propagates these actions to a state. Of course, these actions be tied to a state, and in practice they usually are, but I wanted to keep this simple so that you see how powerful middleware is. could It’s easy to start using Redux Middleware (or any middleware, like composed Rx object transformations) to handle things like navigation, analytics and alerts. Your app more predictable, less work to code, easier to test, more human-readable and, most importantly, squeaky Clean. Who doesn’t want that? Thanks Uncle Bob!

Keep Your UI Clean With Redux Middleware

Update: A reader asked me to publish this as an npm package. Done and done! https://npmjs.com/package/redux-ize . Enjoy! In redux apps, we have actions, we have action creators, but at , we use Redux Action Creator Creators. Sound like overkill? Maybe. But let me show how we use it and then you be the judge! Meeshkan So let’s say that we have an action creator that creates an action and, thanks to , automatically dispatches it through our middleware when we invoke it. The middleware can add all sorts of good stuff (timestamps, device info…), log events, run asynchronous tasks and a whole lot more. Let’s zoom in on our action creator: react-redux export default clapForPost(nClaps) => ({type: "CLAP_FOR_POST",payload: nClaps}); Let’s zoom in on our analytics middleware: import analytics from 'my-analytics-tool' export default store => next => action => {if (action.meta && action.meta.analyticsData) {analytics(action.type, action.meta.analyticsData)}next(action);} Hm… in our analytics data, we’d really like to report what page the claps are coming from. So we go back to our file and write a new function: action.js export default clapForPostWithAnalytics(nClaps, analytics) => ({type: "CLAP_FOR_POST",payload: nClaps,meta: { analytics },}); Do you smell the code smell yet? What if we also want to introduce navigation data based on where we are in the app? If we have five possible things to add, hello 2⁵ different functions. What would be nice is if we could add a bit of analytics data to one action, a bit of navigation data to another… So let’s do it! const ize = n => r => (...argsOuter) => f => (...argsInner) => {const primo = f(...argsInner.slice(n));return {...primo,meta: {...(primo.meta ? primo.meta : {}),...r(...argsInner.slice(0, n), ...argsOuter),},}} export const analyticsIze = ize(0)(a => ({analytics: a || {},})); export const promiseIze = ize(2)((resolve, reject) => ({form: {resolve,reject,}})); export default (...args) => args.reduce((a,b) => b(a)); And then in our file with the action creator… import Ize, { promiseIze, analyticsIze } from './ize';import { clapForPost } from './actions/self-aggrandizing'; class MyComponent {render() {const { clapForPost } = this.props;const clapper = () =>new Promise((res, rej) => clapForPost(res, rej, 3));return (<button onClick={clapper}>Hello world</button>);}} connect(null,{clapForPost: Ize(clapForPost,promiseIze(),analytics({placeInApp: 'MyComponent'})),})(MyComponent); And the action that will be sent is: {type: "CLAP_FOR_POST",nClaps: 3,meta: {analytics: {placeInApp: 'MyComponent',},form: {resolve: [Function resolve],reject: [Function reject],},},} Voilà our action creator creator! Interestingly, this strategy grew from our adaptation of a in the world. Action creator creators are a generalization of this strategy. much-used hack redux-form At , the rule of thumb has become: Meeshkan When an action contains information that cannot possibly be linked to the UI, use middleware. When an action contains only information that is obviously pertinent to the action (i.e. e-mail for login), use an action creator. When you need to dispatch UI-related information or random top-level hacks without putting them in the action’s payload, the is your friend. action creator creator pattern Happy hacking!

Redux Action Creator Creators

At , we are heavy users of in our web and mobile apps. After first using and middleware, we are now happily using as our way to deal with asynchronous operations. If you are not using yet in your React apps, you should drop what you’re doing and read the tutorial! Meeshkan [redux](http://redux.js.org/) [redux-promise](https://www.npmjs.com/package/redux-promise) [redux-thunk](https://github.com/gaearon/redux-thunk) [redux-saga](https://github.com/redux-saga/redux-saga) [redux-saga](https://github.com/redux-saga/redux-saga) The is a great way to describe side effects through pure functions. There are many different ways to understand the pattern, but in my own mind, I like to think of it as a script. Take, for example, the saga: Saga Pattern import { call, put } from 'redux-saga/effects';import fetchKittenPictures from './kittens'; function* fetchKittenPicturesSaga(action) {try {yield put({ type: 'FETCH_KITTEN_EXECUTING' });const critters = yield call(fetchKittenPictures);yield put({ type: 'FETCH_KITTEN_SUCCESS', payload: critters });} catch(e) {yield put({ type: 'FETCH_KItTEN_FAILURE' });}} Giving this script to an actor that will declaim my saga, she will triumphantly pronounce: KITTEN FETCH (a play in one act) Teller: I am executing a kitten fetch!Teller: I am making a call to my kitten fetch api![she goes offstage, comes back]Teller: I have my kittens.[in the event that she breaks her leg, she yells "NO KITTENS!"] But what if we gave our saga to the late great Robin Williams. Would he follow our script? No! It may look something like this: KITTEN FETCH (a play in one act) Robin: I am executing a kitten fetch!Robin: Wait, kittens don't fetch, dogs fetch. And why are we executing kittens? That's cruel.[pauses for laughter]Robin: Ah, we're talking about an API.  I could sure use an IPA![pauses for laughter]Robin: I am making a call to my kitten fetch api![he goes offstage, comes back]Robin: I have my kittens.[in the event that he breaks his leg, he yells "NO KITTENS!"] A different teller will tell the saga differently. In the world of React development, where sagas contain core business logic, sometimes sagas need to change slightly depending on the platform on which they’re running. For example, we may need two “tellers” for the same saga — one for web and one for mobile. They need to be able to share most of their “script” while making slight adjustments based on their unique way of telling the saga. An anti-pattern would be to hardcode into a saga all of the different ways it could be told. The cleaner solution is to give the saga to a teller and have her tell it. So, we have put together a hack called that does exactly this. The API is dead simple: saga-teller function sagaTeller(saga, rules); The first argument, , is the saga that the teller will tell, and is a blueprint for how the teller will modify the saga. saga rules The rules object itself is in the form: // match// note that put and call should be mutually exclusive {when: <'before' or 'after'>,put: <an action if we are responding to PUT>,call: <a function if we are responding to CALL>,effect: <any valid redux-saga effect like all, put, take...>} // rules{first: <call or put>,matches: <Array of match>,last: <call or put>} Using our kitten saga, we may do this: const myVeryOwnFetchKittenSaga(fetchKittenSaga, {first: call(lookMomIKnowHowToUseSagas),matches: [{when: 'after',put: 'FETCH_KITTEN_EXECUTING'effect: put({type:'MOM_YOU_SHOULD_BE_MORE_PROUD_OF_ME'}),},],}); And there you have it. Now, when I tell the saga, the first will do is call a function telling my mom that I know how to use sagas. Then, the chugs along, and after it emits a action with , it sends out a action. This way, the original saga stays intact and my personal saga can be used wherever I want. This is also really easy to test—you just need to assert that the rules object contains the correct logic. react-saga fetchKittenSaga put FETCH_KITTEN_EXECUTING MOM_YOU_SHOULD_BE_PROUD_OF_ME In a more real-world example, this allows you to write one saga and then write a few rules that tweak its behavior based on if you are web or mobile, Android or iOS, etc. Note that the pattern is a strategy. You should never drive your app down the field with it. Think of it as a sort of patch for your saga — it should represent the minimum amount of work needed to modify a core saga, but anytime the teller’s version dwarfs the saga itself, it becomes an unwieldy kludge. In this case, just write (and test) two different sagas. saga-teller goal line Below is a gist that shows a rudimentary implementation of a . The function at the bottom just allows you to chain together multiple tellers, so you can do . Think of this like the telephone game, where each teller hears the version of the previous one and modifies it with their special addition. saga-teller tell tell(mySaga).by(teller0, teller1, teller2) I will definitely not be making anything other than a gist for this because it is based on a internal API and is really fragile in case of updates. But it is short enough that you can copy it, paste it, and modify it to your heart’s content. We’ve been using a version of it at with lots of success. Enjoy telling your sagas! redux-saga Meeshkan

Telling a (redux) Saga

tl;dr: Algerians and Ivoirians are kicking some major Machine Learning ass! Earlier this month, I shared the Meeshkan Machine Learning service with a friend from Côte d’Ivoire and his immediate reaction was “you have to get the word out in Africa.” I’m a big believer in enthusiasm, and he was genuinely enthusiastic, so we gave it a shot. We initially ran a small advertising campaign in four countries — Côte d’Ivoire, Senegal, Algeria, and Morocco. The organic growth from these campaigns has been much more promising than anything we’ve seen in Europe or North America. Why? Below are some musings. Meeshkan — A short introduction Meeshkan is an IoT service where people can convert almost any device into a client that executes distributed Machine Learning tasks. People are paid for the time their device is used. Meeshkan is currently in its public beta. You can download the app and checkout for more information. here meeshkan.com Mobile culture in Africa The first users of the Meeshkan Public Beta came from Africa and represent over 50% of our current user base. Algeria accounts for 33% alone! As with any small company, it is difficult to determine if there is a statistically significant relationship between trends in one’s own data and more global phenomena. That being said, there are several reasons that Meeshkan’s early adoption in Africa is not surprising. Note that these reasons come from my own personal research and discussions with friends and colleagues, but no one on our team has significant knowledge about African economics. I am interested to hear what others think! As , there are major incentives for people in Africa to own cell phones. For many Africans, they serve as access to FM radio, as flashlights, as lifelines to health care and as banks on a continent without robust physical infrastructure. CNN reports The banking culture is particularly interesting, and is perhaps the one that is most linked to Meeshkan’s early adoption. For example, a out of Harvard’s Kennedy Center by Jay Rosengard shows a jump in Kenyan’s with access to a financial account from 45% to 75% in 2014. A report from claims that ordering a cab in Nairobi is easier via mobile than in New York. of the Huffington Post calls this phenomenon “leapfrogging” — jumping over intermediary technologies to go straight to the state of the art. 2016 paper The Economist Diana Brazzell An advertisement for mfukoni, a popular banking app in Kenya. A continent that has leapfrogged into mobile banking should, in theory, be ripe for leaping into mobile markets connecting buyers and sellers, which is what Meeshkan is. However, popular solutions like PayPal and Stripe only provide payout services for a limited amount of countries, almost all of which are in the top third of countries in terms of GDP per capita. So the mobile-first culture is, in my opinion, not enough to explain the early adoption rate. One major factor may be Meeshkan’s use of cryptocurrency. Cryptocurrency African monetary units are notoriously unstable, and it makes sense that cryptocurrency would become a viable alternative for individuals looking to secure their assets. This is, however, a guess — I cannot corroborate this with any research. Most of my poking around in this area comes from the blog. However, startups are already starting to hedge their bets on this idea. For example, KodePay is a Kenyan based startup bringing blockchain currencies into their local market. They have reported that they have already created partnerships with Kenyan banks. cryptoafrica.com Meeshkan currently uses Iota as its cryptocurrency of choice. Meeshkan, as an IoT app, trades in Iota, a Distributed Ledger cryptocurrency that allows for fee-less nanotransactions. This is done by attaching Proof of Work to the transaction itself, where each transaction must approve two others in order to be approved. Meeshkan has recently approved its first transactions in Iota with African IoT sellers, although we have approved less in Africa than for other continents. So, while Africa seems ostensibly ripe for a cryptocurrency revolution, we have a lot of work to do to evangelize Iota and cryptocurrency with our African user base. We’re working hard on that and hopefully we’ll have good news about that in a future blog post. Curiosity The most interesting part of all this is that, to date, the only emails and tweets we’ve received showing curiosity about Machine Learning and/or the service are from Africa. We’ve been asked “what’s Machine Learning”, “will machines take over the world”, and “how much money to people make with Machine Learning” to name a few questions. We’ve pointed these people to various online resources and have responded with our own modest answers to these big questions. The first tweet mentioning Meeshkan ever came from Africa! Effusive joy and minor debauchery ensued after seeing this. All this has got me thinking — can Meeshkan convert of CPU/GPU on the network to of CPU/GPU? Meaning, can the service incite people to become practitioners of Machine Learning? This would be super cool for Meeshkan because we’re greedy and want more buyers in the system. But it would be even for humanity because it would break the first-world hegemony over Machine Learning and make it accessible to all. sellers buyers cooler So, in the next release of the Meeshkan Android app, we’ll be providing a tab where people can explore Machine Learning. The tab leads to two resources: an RSS for trending news in Machine Learning and a link to the seminal Andrew Ng course where people can learn. We’ll be collecting analytics data to see how popular these options are and, if we find that they are widely used, we’ll explore ways that the service can convert Meeshkan IoT’ers to Meeshkan (non-)Machine Learners.

Machine Learning in Africa

The First Rule of Program Optimization: Don’t do it. The Second Rule of Program Optimization (for experts only!): Don’t do it yet. — Michael Jackson In our era of modern, speedy machines with oodles of memory, performance is something that few coders ever need to think about; but we think about it anyway. We think about performance even when we don’t need to, and it’s much to our detriment. It complicates our lives and the lives of other coders in our codebases. Thus the first rule of : . optimization Don’t do it In this post, I’ll illustrate the reasoning behind the first and second rules of optimization, and I’ll provide some tips for the times when you really truly do need to undertake performance improvements. Note: When we talk about performance we’re often referring to the speed a program is run, but the same rules apply to optimizing memory usage, battery usage, or whatever other resource might important to your program. Part I: The Bottleneck My first important real-world performance problem happened during a summer of research in college. A program for helping biologists identify genes was too slow. My advisor suggested that there was a part of the algorithm that took a long time, and I should start with optimizing that. graph coloring So I did. I had a great time squeezing CPU cycles out of C++. It’s a fun form of puzzle solving. I improved the performance of the graph coloring algorithm by 70%. After a week of intricate optimization, I ran the whole program on the full dataset. It was only 15% faster — hardly noticeable. What was going on? I dug around a little and discovered that when a process finished, it wrote long results to a file. The program caused many processes to write at once. As the operating system’s scheduler switched among the processes, the disk had to do tons of seeks to write each file. Disk seeks are expensive. I added a few lines of code that created a file system lock so that only one process could write to disk at once. Sequential writes are much faster than seeks, so with that change the program was about 60% faster. With a few lines of code, I’d more than cut the runtime in half. My week of C++ optimization had not. My advisor had pointed me in the C++ direction because he’d spent the most time on it. He’d probably only spent a couple minutes writing the code that writes to files. I can’t tell you how many times in my years of performance work across many platforms I’ve seen people waste time on optimizations that do almost nothing. The most common failed performance effort is optimizing something that’s not the bottleneck. You often think you’re optimizing the bottleneck but you’re not looking at the whole picture. Ironically it’s often the most experienced engineers that make this mistake first . It’s easy to get caught up in our own systems and forget there’s a whole other world out there. I once saw a group of senior backend engineers spend an entire hack week rewriting a complex web endpoint to use Go instead of Python. At the end, they found out that the bottleneck for the page was on the browser side of the app (and not the server side) making their performance improvement completely irrelevant. These were experienced engineers who knew extremely well what made the server fast and slow, but their detailed knowledge of one part of the system prevented them from looking at the bigger picture. Always ask yourself, what performance does the end-user care about? How does your code impact that? Part II: Complexity Performance improvements that don’t get at the bottleneck can actively harm your codebase. Almost all performance improvements increase the complexity of the code. Unneeded complexity creates a lot of problems that are not performance-related by making code harder to understand. Let’s take a look at another example (slightly simplified) from a friend who used to use Python for graphics: for i in xrange(1000):my_mesh.draw() Imagine that before pushing this, the author noticed that draw() is being called 1000 times. Since the Python interpreter has to look up methods on each invocation, time could be saved by caching the reference to the draw method in a local variable: draw_method = my_mesh.drawfor i in xrange(1000):draw_method() This seems simple enough but it’s liable to cause a lot of problems down the line. While right now it might not seem like it will cause bugs, imagine that it’s in a large code base. Over time, many new lines of code might be added to the loop by many different engineers. Eventually draw_method might be far away from , and the engineer will have to skip around and lose context to figure out what that means as they read through the code. draw_method = my_mesh.draw It will definitely slow people down and it could cause bugs. For instance, somebody might add live updating: draw_method = my_mesh.draw for i in xrange(1000):…if my_mesh.outdated():my_mesh = updated_mesh…draw_method() Now there is a bug. The draw method will call draw on the first mesh, not the updated one, because the wrapper to my_mesh.draw wasn’t updated. It’s quite common to see small performance “improvements” thrown in as part of larger changes, when those improvements haven’t been evaluated for how much they help. Even what seem like simple performance improvements add a cost. The cost is complexity. Complexity means bugs and more time reading, testing, and maintaining code. Particularly in large or long-lived code bases, complexity is the biggest impediment to progress. The change above is consistently faster on my by about .05ms on average. Is .05ms enough of an improvement to warrant the complexity? Most of the time, no. machine You are simply not in a position to evaluate whether the complexity cost is worth the improvement in performance unless you have a big picture view of the performance of the system. Maybe you can optimize, but you certainly shouldn’t do it yet. Part III: How to approach performance problems But let’s say you think your system is really too slow, and you really do need you to improve its performance for empirical reasons. Here are few rules to get you started in a disciplined way. Get the right metric A lot of the problems I covered above come from measuring the wrong things. In my project in college, I measured the time for the graph algorithm to finish and not the time for the results to be written out. The hack week team was measuring server response time and not the time for the web page to appear to a user. Measuring the wrong thing leads to solving the wrong problem. Making sure you’re measuring the right number is the single most important thing to do when you’re tackling a performance problem. Without that, you have no idea how much you’re helping. Similarly, don’t necessarily trust those more experienced than you to tell you the bottleneck. Make sure you evaluate it yourself. How important are marginal gains in this system? Once you’re measuring the right thing, it will be a lot easier to tell which improvements are worthwhile. Nonetheless, it’s worth it to spend some time thinking about what your goals are. You need to know not only how fast (or memory-intensive, etc) the system is, but also how much marginal gain you’ll get from improvements. Do you save your company money? Do you save your users time? If it’s a script that runs once a week that nobody is dependent on, even savings of an entire minute (basically forever in computer time) might not be worth adding complexity. But if it’s a function run a million times per second across a fleet of thousands of servers, savings of microseconds could save a lot of money. If you understand what your performance goals are before beginning your work, you can make the right call on performance/complexity tradeoffs later on. If you’re being honest with yourself, you’ll often see that you should scrap marginal gains and focus on major wins. Measure right Lots of things can affect your performance measurement. You might be a mobile developer with a slick test device that has only one program running on it. If your goal is to produce a response in < 10ms on average, that’ll be a lot easier on your test device than on your user’s four-year-old phone with low battery and a hundred apps running in the background. It could also cause you to misunderstand your program’s characteristics. You might be I/O bound on a test device, when on most devices you’re actually CPU bound. All sorts of external factors can affect performance including: Memory usage Server load Network bandwidth and latency Battery level CPU usage Disk usage Various caches at all layers When you’re measuring, try to make sure as many factors as possible are held constant so you can accurately compare different approaches. But also make an effort to understand what these factors usually look like, so you can make sure your appraisal of bottlenecks is realistic. One simple strategy for cutting out the natural ebbs and flows of resources on a machine is to run your code many times. Python’s easy-to-use library defaults to running your sample a million times. This can help average out some fluctuations in system resource availability. timeit Try to integrate performance tests into your build Some software projects fail their builds if a commit causes a performance regression. If performance is important for your project, consider adding performance as part of your continuous integration. It can let you prevent performance regressions before they get shipped. The sooner you are aware of regressions, the easier they are to fix. In large codebases, small regressions can build up over time unless they’re aggressively kept in check. Tests can keep them out of your system. Finally, the optimization Armed with the right philosophy and information about your system, you’re ready to begin performance optimization. Don’t do it until you’ve profiled and analyzed and figured out a coherent strategy — then go forth and code! Once you understand the basics of how to tackle general performance problems, you can start to delve deeper into your system. This is where things can get really fun and exciting. Small systems are elegant logic puzzles. Large systems are universes of slowness whose complex interlocking parts can take months to unravel. Improving performance in both environments can be beautiful. It’s just not often the right thing to do. For my guidance on web performance see So your website is slow? Let’s fix that . is how hackers start their afternoons. We’re a part of the family. We are now and happy to opportunities. Hacker Noon @AMI accepting submissions discuss advertising &sponsorship To learn more, , , or simply, read our about page like/message us on Facebook tweet/DM @HackerNoon. If you enjoyed this story, we recommend reading our and . Until next time, don’t take the realities of the world for granted! latest tech stories trending tech stories

The Rules of Optimization: Why So Many Performance Efforts Fail

I had to look up isomorphic in a dictionary the first time I came across this word in web development. reports that Wikipedia Isomorphic JavaScript , also known as Universal JavaScript , describes JavaScript applications which run both on the client and the server . Building a package that works out of the box in both the server (ie Node) and the client (ie a browser) is hard. As soon as you start mixing in dependencies, if even one of them makes a call to a node-only package (think ) or browser-only object (think ) , it will tank your whole build. Worse yet, you may not even be able to reasonably infer this problem exists, as they may be buried in sub-sub-sub-dependencies that have no bearing on the code you are writing. [fs](https://nodejs.org/api/fs.html) [window](https://www.w3schools.com/jsref/obj_window.asp) This document describes a couple easy conventions for developing isomorphic node packages and ends with one opinionated missive about how they should be packaged. Ignore requires in webpack Even if you use a package like to create conditional node vs browser imports, does not care about conditionals because it has no clue how the runtime will resolve them. So, in the following scenario: [detect-node](https://www.npmjs.com/package/detect-node) webpack const foo = isNode ?r equire("fs") :     require("fs-web"); it will happily try to package all of the code you require, meaning that in this case, it will try to package . It will then fail with the following error message. fs Module not found: Error: Can't resolve 'fs' in [insert file here] In response to my , Alex Rokabilis writes: question on how to get around this conundrum [check out] the not very well known function. This is a webpack specific function that will instruct the parser to avoid bundling this module that is being requested and assume that a global function is available. __non_webpack_require__ require While is a documented part of the API, it is buried deep within the documentation, thus its obscurity. However, it works like a charm with one minor tweak. [__non_webpack_require__](https://webpack.js.org/api/module-variables/) If you try to write: const foo = isNode ? [__non_webpack_require__](https://webpack.js.org/api/module-variables/)("fs") :     require("fs-web"); Node will barf with the following error message: __non_webpack_require__ is not defined Typescript will also not recognize . __non_webpack_require__ This can be overcome in the following ways: For typescript users, require . Do not use for this, it will not work. [@types/webpack-env](https://www.npmjs.com/package/@types/webpack-env) [@types/webpack](https://www.npmjs.com/package/@types/webpack) In your code, somewhere before you use , write the following hack: __non_webpack_require__ if (isNode) {(global as any).__non_webpack_require__ = require;} There are a few different ways to accomplish this, most of which will not raise an error in an IDE and will compile with and but will fail in your node environment. So use this one — it’s been extensively tested by our team and works. babel typescript Use dependency injection and TS interfaces is a popular pattern that is famously evangelized by Uncle Bob in that I’ve mentioned several time on this blog. The idea is that dependencies should always import inwards, meaning that “core” code should never know about dependencies, but should make a contract with the calling code via interfaces, and the calling code implements classes that make good on this contract by, amongst other things, pulling in relevant dependencies. Dependency injection this article In , we use dependency injection for the logging and persistence mechanisms, both of which are defined in an options object that is injected into the main function at runtime. If no option is given, sensible defaults are chosen based on the package, which is the most popular and reliable way to detect if an environment is node or not. Let’s see how that works. unmock-js unmock detect-node In our dependency injection, parameter is defined to contain, amongst other things, the following interfaces: options export interface IUnmockOptions {// ... some stuff, then ...logger?: ILogger;persistence?: IPersistence;// ... more stuff} and themselves contain various methods, such as and , that call functions specific to node or the browser. For example, here is the difference between node and jsdom code for one method in the interface: ILogger IPersistence ILogger::log IPersistence::saveHeaders IPersistence // fs-persistence.ts export default class FSPersistence implements IPersistence {public saveHeaders(hash: string, headers: {[key: string]: string}){fs.writeFileSync(`${this.outdir(hash)}/response-header.json`, JSON.stringify(headers, null, 2));}} // local-storage-persistence.tsexport default class LocalStoragePersistence implements IPersistence {public saveHeaders(hash: string, headers: {[key: string]: string}){window.localStorage[`${this.outdir(hash)}/response-header.json`] = JSON.stringify(headers, null, 2);}} Don’t use separate packages I’ve flipped back and forth on this issue quite a lot. On one hand, using separate packages for different environment solves the “1980s Christmas lights” phenomenon where an error in one environment tanks all of the others. This can, of course, be toxic for package and project management. On the other hand, separate packages can lead to a usability problem if the developer is trying to develop an isomorphic package and now needs to include and manage multiple packages. This is currently the case, for example, with versus . If you are developing a package, for example, and are trying to reuse server and client code that relies on , this can lead to a huge mess of clauses. I’m sure that Sentry will fix this, but the general problem should be avoided if possible. [@sentry/node](https://www.npmjs.com/package/@sentry/node) [@sentry/browser](https://www.npmjs.com/package/@sentry/browser) Next.js Sentry if / then As a result of this philosophy, is a single package that works flawlessly across Node and browser environments, even in complicated scenarios that mix code from the two. [unmock-js](https://www.npmjs.com/package/unmock) The goal is to ease developer experience so that they have a batteries-included way to start working with mock data as they build out their web app. tl;dr When making isomorphic JS packages: Use . __non_webpack_require__ Use dependency injection. Thanks for reading! Who doesn’t like Morph?!?

Building isomorphic JavaScript packages

Ah, another beautiful spring morning in Helsinki. The birds are singing, the drunks are leaving the bars, and I have given myself 30 minutes to write an appallingly ugly hack to get fenced code tabs working in mkdocs material. I made this hack because there seem to be several people that and yet no one has stepped up to do anything about it. The hack is provisional — I’m sure there are 100s of cleaner ways to do it, but this gets the job done! really need the feature First, I grabbed , which more or less works in several mkdocs themes but fails in mkdocs material. [markdown-fenced-code-tabs](https://github.com/yacir/markdown-fenced-code-tabs) Scouring the internet for how to fix this, I found a promising lead on . So the CSS styling works, but when one clicks on the tabs, nothing happens. Digging into the HTML, I realized that all of the stuff doesn’t work for reasons beyond my understanding. this GitHub issue aria Nothing that a quick JavaScript hack can’t fix… First, in your , add: mkdocs.yml extra_javascript: javascript/tabhack.js Then, swallow your pride and include: And there you go! Fun times, fun times…

Stories, News & Mentions about MEESHO

Meesho’s Stories on HackerNoon

Meesho's News & Mentions