NuBinary is a fractional-CTO service and software development consulting firm working with startups in various stages, MVP development, scaling, team building, and post-series A/B growth

NuBinary

nubinary

Let's learn about via these 71 free stories. They are ordered by most time reading created on HackerNoon. Visit the to find the most read stories about any technology. The Markup /Learn Repo 1. We Ought to Get Better at Recognizing Dark Patterns Companies advertising online have developed a series of unethical practices that some describe as Dark Patterns, which can mislead consumers on the internet. 2. Facebook Crypto Scammers are Imitating Zuckerberg, Musk, and Bezos The ads, until recently available for view in Facebook’s public ad library, were frauds that slipped through Facebook’s content moderation process... 3. Does Amazon Follow the Law? Amazon lists its own brands and exclusives above competing products with better ratings. 4. What Happens When you Try to Opt Out of Personalized Ads The Digital Advertising Alliance has a website that you can use to opt out of personalised ads. The problem is, it doesn't really work like it is supposed to. 5. A Letter from TheMarkup.org President Nabiha Syed My dad brought home a computer when I was 3 years old. He hoped that it would teach me things — and it did. I learned to spell in English by playing Reader Rabbit and the Fabulous Word Factory. 6. Who Can Spy on My Kids Online? What the United States’ children’s privacy law does and doesn’t do 7. Hey Google, Show me Non-Discriminatory Job Ads Google's advertising system as allowed employers and landlords to discriminate against nonbinary and some transgender people by excluding them from ads. 8. What Is “Suckers List” And Why Texas Drivers Sued Its Creator A joint investigation by The Markup and Consumer Reports was cited in a lawsuit over discriminatory car insurance rates 9. The Markup Investigation: Congressman Says that Google engineered a “Walled Garden” Comment comes as part of big tech CEO grilling by antitrust committee 10. Is There a Valid Cause for Using Facial Recognition Tools in Schools The deployment of facial recognition technology in a school in the US marks a disturbing trend with the protection of identity and the safety of students. 11. Are Time Clock Machines With Fingerprints Legit? You don’t always have to give your boss the finger 12. Are Students Being Entrapped by Fake Test Answer Sites? There are fake test answer websites that are being used to entrap students looking for answers during online tests. 13. What is Causing the Worrisome Increase in Ransomware Ransomware has become a much more common crime online in recent days. But why is this the case and what can people even do about this massive problem? 14. Does Anonymization of Data Guarantees Privacy? Anonymization of data isn’t enough to guarantee privacy 15. Facebook Peeked at Your Info When You Applied for Student Aid Online For millions of prospective college students, applying online for federal financial aid has also meant sharing personal data with Facebook, unbeknownst to them. 16. Can I Avoid Facial Recognition At The Airport? Technically, yes, but it is not always easy 17. A Letter from TheMarkup.org Editor Julia Angwin Dear Readers, 18. Life360 to Stop Selling Precise Location Data Life360, a safety app, claims it will stop selling precise location data. 19. Many Minneapolis Uber and Lyft Drivers Were Victims of Reported Carjackings Last Fall A lot of ride-share drivers were victims of carjacking in Minneapolis last fall. Here is what spokespeople, the police, and the drivers have to say about it. 20. Online Unemployment Systems Are Showing Signals of Strain Newly laid-off workers face crashes, long load times, and messages offering phone callbacks as states struggle to adjust 21. How Unemployment Benefits Systems Are Failing During The Pandemic Long wait times, passwords sent by mail, and shutdowns thwart applicants 22. How Are Gig Workers Coping During the Crisis Workers fight for hazard pay, protective equipment, and better sick leave 23. Over 50 U.S. Gig Workers Murdered on the Job in the Past Five Years A new study says more than 50 gig workers have been murdered in the U.S. over the past five years. 24. Could There Eventually Be a Ban on Behavioral Advertising? Could behavioral advertising be on its way out? 25. Why One State's Testing Procedures Vary From a Neighbor's? We want to know the ways one state’s testing procedures may vary from a neighbor’s, and the reasons why. 26. What Do U.S. Politicians Have to Say About Mortgage Discrimination Against Minorities? American politicians react to The Markup's findings that minorities are getting discriminated against when they apply for a mortgage. 27. The Concerning Results of Investigation Into Google’s YouTube Ad Placement Blocklist An investigation by The Markup has shown that YouTube has been surpressing Black Lives Matter-related ads while doing nothing to prevent white supremacist ads. 28. Are You Paying 400 Times More for Your Internet Than Your Neighbor? You Might Be AT&T, Verizon, EarthLink, and CenturyLink disproportionately offered lower-income and least-White neighborhoods slow internet service for the same price... 29. Why Are Prices On Amazon So Insanely Volatile? Dynamic pricing algorithms on Amazon aren’t just about supply and demand 30. Official Retailers Dragged Into Battle In April 2019, fans of the makeup influencer Jaclyn Hill noticed her new eyeshadow palette, normally $38, on Walmart.com for the unbelievably low price of $16.39. Many rushed to buy; others were skeptical: Hill had announced that the palette would be available only through certain retailers, and Walmart wasn’t one of them. 31. What To Do if Denied Housing or Apartment Due to Inaccurate Background Report Getty Images and Ali Wijaya 32. Access to Coronovirus Testing Will Depend On Your Location Five scenarios that show the differences in states’ testing algorithms 33. Pitfalls Of Court Hearings On Zoom Court hearings are going virtual in response to COVID-19. Studies show they can lead to harsher outcomes for defendants 34. What To Do if The Bank Mistakenly Reported You as Deceased Investigate, complain, and maybe sue 35. Already Struggling with School? The Pandemic Made it Worse. The Pandemic Didn’t Start Educational Disparities—but It Has Made Them Worse. Vulnerable students get hit the hardest via online learning. 36. Algorithmic Auditing: Can We Actually Eliminate Algorithm Bias Algorithmic auditing got press recently when HireVue, a popular hiring software company used by companies like Walmart and Goldman Sachs faced criticism. 37. Can You Tell What Products On This List Are Prohibited On Amazon? Amazon is the largest retailer in the world, but a majority of the sales on the site come from third-party sellers. Along with toasters and mops, the site’s a target for people trying to sell dangerous and illicit products. 38. This Online Abortion Pill Provider Used Tracking Tools That Gave Powerful Companies Your Data The trackers notified Google, Facebook’s parent company Meta, payments processor Stripe, and four analytics firms when users visited its site. 39. New Report Finds That Facebook Still Runs Discriminatory Ads We found discriminatory ads can still appear, despite Facebook's efforts 40. Are Your Photos Exposing You? You may be accidentally sharing personal information in your photos 41. Big Tech Vs. Big Government: The Choice is Pretty Bleak A group of bipartisan lawmakers hopes to pass antitrust legislation that takes aim at tech giants buying out the competition and hoarding data. 42. Can You Speak The Google Language, Please? The Markup obtained internal documents that coach new employees to avoid creating “very real legal risks” in using words like “market” and “networkeffects” 43. Can Robots Manage Health Benefits? In an attempt to manage soaring health care costs, some government officials and health care companies are turning to algorithms to determine how to allocate limited benefits, who to provide care to first, or whether a person should receive care at all. 44. How Tenant Screening Companies Can Use Your Speeding Ticket Against You Estimated 9/10 landlords, under pressure to ensure their properties are safe, use companies like Rentgrow to perform background checks on potential tenants. 45. The Truth About Less Biased Data-Informed Predictive Policing Critics say it merely techwashes injustice 46. How Geofence Warrants Are Affecting The Legal System Geofencing itself simply means drawing a virtual border around a predefined geographical area. Data can then be gathered on users who enter that area. 47. How To Make Millions With Unemployment Sites That Don't Work In at least one case, that same company is now getting pandemic-relatedcontracts 48. Will Google's and Apple's Contact Tracing Preserve Privacy? The proposed system is anonymous but vulnerable to trolls and spoofing 49. Reasons Why Your Internet Is Slow What’s the problem here? It might be you (sorry). It might be the infrastructure you’re working with. It might be your internet provider. 50. How Popular is Right-Wing Content on Facebook? Facebook has been criticized in the past because many believe they are allowing right-wing content to run rampant on the site. 51. How Facebook Serves Targeted Ads by Big Pharma We found drug ads targeted at users interested in everything from bourbon to therapy on Facebook, with significant implications for privacy for users. 52. Introducing a Browser Extension to Detect Amazon Brands Amazon isn't completely upfront with what brands they own or don't. That's why The Markup created a browser extension to help with this problem. 53. What Was Different About The 2020 Census And Its Challenges The count is going mostly digital for the first time. Oh, and there’s a pandemic 54. Facebook Ads Market Potentially Dangerous “Abortion Reversal” Procedure The Markup has found that Facebook is serving up ads and posts for the so-called “abortion pill reversal” procedure... 55. How Pandemic Testing Protocols Vary Across The USA To determine how testing protocols for COVID-19 vary across the United States, we sent requests under public records laws to all 50 states, New York City, and Washington, D.C. The requests were sent to health departments the week of March 16 and were identical. The database below contains responses we have received, as well as publicly available guidance from some jurisdictions. 56. Amazon's 'Essential Item' Policy Upsets Everyone Currently that appears to include a 10-pack of rubber chickens 57. Why are so Many Opposing the Right-to-Repair Laws The right-to-repair movement has drawn an unusual coalition of stakeholders, including environmentalists, libertarians, engineers, gamers, and hobbyists. 58. Testing the Perfornace and Privacy Settings of COVID-19 Vaccine Websites The government websites responsible for helping citizens get their COVID-19 Vaccines have been found to have less-than-ideal accessibility and privacy features. 59. If You're Ready To Protest Your Phone Should Be Ready Too Simple steps to take before hitting the streets 60. Allstate's Car Insurance Algorithm: How Insurance Algorithm Squeezes Big Spenders Seven years ago, Allstate Corporation told Maryland regulators it was time to update its auto insurance rates. The insurer said its new, sophisticated risk analysis showed it was charging nearly all of its 93,000 Maryland customers outdated premiums. Some of the old rates were off by miles. One 36-year-old man from Prince George’s County, Md., who Allstate said in public records should have been paying $3,750 every six months, was instead being charged twice that, more than $7,500. Other customers were paying hundreds or thousands of dollars less than they should have been, based on Allstate’s new calculation of the risk that they would file a claim. 61. How Much Exactly are Podcasts Able to Track us Traditionally, the podcast ecosystem has been tracking-resistant, in part because podcasters release their shows through RSS, free technology dating back to '99 62. Tell Me Your Secrets: How Many Workers Review Posts on Amazon? We found banned items for sale on Amazon.com 63. How Automated Background Checks Can Freeze Out the Wrong Renters Computer algorithms that scan everything from terror watch lists to eviction records spit out flawed tenant screening reports. And almost nobody is watching 64. How Many Americans Don't Have Internet Access? The COVID-19 crisis highlights the costs of the U.S. digital divide 65. Is the Data We Share with Pay Apps Private at all Using a payment app of any kind, no matter how privacy-preserving it is, always introduces a middleman: an agent that uses data to execute a demand. 66. Tech Titans Hammered By Congress At Antitrust Hearing Lawmakers allege abuse of monopoly power and political bias 67. Identifying Fake Reviews On Amazon Spotting the fakes isn’t always possible, but here are some tips 68. How We Pay For Free Websites With Our Privacy An array of free website-building tools, many offered by ad-tech and ad-funded companies, has led to a dizzying number of trackers loading on users’ browsers. 69. Allstate's Car Insurance Algorithm: How Insurance Algorithm Was Analyzed State regulators and consumer advocacy groups have scrutinized Allstate Corporation’s use of big data and personalized pricing in the way it calculates how much the company charges its private auto insurance customers. 70. There's Something Wrong With Google Search Results The search engine dedicated almost half of the first page of results in our test to its own products, which dominated the coveted top of the page 71. What Analyzing Google Search Results Revealed About Google We designed an experiment to measure the quantity and placement of
these Google-created and self-referential search results and how they compare to others. Thank you for checking out the 71 most read stories about The Markup on HackerNoon. to find the most read stories about any technology. Visit the /Learn Repo

71 Stories To Learn About The Markup

376 Stories To Learn About Google

xCollective is an up-and-coming female and nonbinary project that aims to help women mint NFTs and find jobs in the crypto web3 industry. They have an emphasis on achieving fair pay for women using a decentralized recruiting network. The project should also increase women's and nonbinary individuals' role in crypto adoption and development of web3.  Women and nonbinary individuals have a unique view of web3 helping the community grow through efficient marketing as well as attracting a diverse population. A study conducted by Deloitte found that will start accepting crypto as a form of payment in the next 2 years. (Deloitte) Women's involvement in crypto adoption is detrimental as it is projected to grow approximately four times the rate of initial internet adoption. Currently, crypto is a male-dominated industry but, hopefully, that gap may eventually start to close. 75% of US retailers The lack of women and nonbinary individuals in the cryptocurrency space is often related to being intimidated when trying to find employment in a male-dominated field. This has contributed to a lack of diversity over time. Many of these women have skills for the cryptocurrency industry but are left out. xCollective has made it its mission to assist women and nonbinary individuals become involved in NFTs and assist them in finding jobs in the crypto web3 industry. Rachel Price, the founder of xCollective, describes her company as: "a professional growth community, with a mission to open virtual doors and help people get paid jobs in web3 by providing career coaching, networking and wage negotiation tips. We are focused on empowering women & nonbinary individuals to make hiring in web3 more inclusive, but our community is open to all races, genders, religions, etc. Our long term vision is to turn our community into a decentralized, crowdsourced recruiting platform - where we can support people across the globe when it comes to enhancing their professional opportunities and careers, specifically in web3 roles." xCollective reaches out to crypto projects founders and co-founders to discuss diversity issues we currently face and also the pay gap between women and men. They also survey women's pay scale for the crypto industry which can then be used in negotiating a paid job. A conducted by McKinsey & Company found that companies practicing gender diversity outperformed peers by over 25% (McKinsey). Hopefully, over time the benefits of increased diversity will motivate the crypto industry to hire more women and nonbinary individuals with equitable wages. Increased diversity may be hard to achieve if women are intimidated to apply for work in the industry. Rachel of xCollective’s advice when applying for a web3 position is: study "There are so many transferable skills you can take with you from web2 to web3. Don't discount your previous experience. You do not have to start from scratch if you want to pivot into a web3 role. A huge part of what we discuss within the xCollective community is how to recognize your skills and strengths and translate that into the job you are applying for in the web3 space. Don't underestimate the importance of practicing storytelling and crafting your narrative." Another way X-Collectives assists in finding jobs is through the refer-to-earn program which rewards people for referrals that lead to employment. Rachel at xCollective summarizes this program stating: "There is a lot of money in recruiting-it is a lucrative business. I truly believe if we can help our community earn on referrals and redirect recruiting dollars, we can create better incentives in the system, and ultimately help people get access to networks they wouldn't have otherwise had access to. This is why I am so driven to turn xCollective into a decentralized recruiting network and create a refer to earn platform to drive it." Women often face discrimination and bias during the interview process or when trying to obtain a web3 position. Hiring should not be based on looks, gender, or how you speak. Women and nonbinary individuals can become frustrated while learning about crypto and the web3 environment in a male-dominated industry. Sometimes this leads these individuals to give up on their aspirations to enter the industry. Rachel of xCollective has expressed many of the challenges women and nonbinary individuals face when starting a career and have a mission to support them. In this supportive role, Rachel states: "I am excited to be exploring the ways in which web3 can reduce discrimination and bias within the interview process and make hiring in the space more inclusive and diverse."Often, women and nonbinary individuals underestimate their skills and how much they should be paid for working in the web 3 field. Entering a world transitioning from web 2 to 3 is difficult and even more trying as a female but with Rachel Price and xCollective women and nonbinary can find support and guidance while achieving their goals.” xCollective will also be assisting with the minting of NFTs. They will be launching their genesis NFT project on the Ethereum blockchain and anyone interested in gaining access to the whitelist can contact Rachel via xCollective’s twitter which is xCollectiv. Another way to get whitelisted and get to know the community is by joining and participating in their discord. Although xCollective is early on in their mission, other people are starting to take notice of its project. HedgeHUG is an advisory program that helps pre-mint NFT projects. xCollective was recently chosen for the hedgeHUG accelerator program and hedgeHUG will be providing 90 days of free mentorship to xCollective’s genesis Ethereum-based NFT launch which started in July. They will help open doors for partnerships in the HUG ecosystem. This also includes helping promote the project utilizing their social channels with an impressive audience of over three hundred thousand people. xCollective’s future is to mold its community into a decentralized recruiting platform. Although their focus is on more underrepresented groups, they are open to all people. Over time with xCollective’s dedication and fiery motivation, an inclusive and diverse web3 environment will evolve for women and nonbinary individuals.

xCollective: The Future Of Diversity

In my first interview with Courier, when my now-boss was telling me about the company, he stopped and said, “I see your display name is not the same as the name on your application. Are there any other pronouns that I should be using?” It was a small gesture, but it made an impact on me. The moment hinted at a company culture built around treating people with respect and appreciating authenticity. Courier is still a very small company building the foundation of a diverse team, but even as we grow, I’m confident that this will remain a priority for us in the future. Encouraging employees to be their true selves in big and little ways is important because it’s the right thing to do—it also happens to be the best way to recruit a diverse team of employees. And increasing diversity improves many facets of an organization: Productivity, profits, employee retention, and even the ability to navigate conflict. In honor of Pride month, let’s talk about the power of diversity and how tech companies should improve inclusion in their organizations. Diverse Teams Make More Money It doesn’t matter whether you’re talking about racial diversity, gender diversity, or other forms of diversity—people with different experiences often come up with different ideas from each other. Those unique insights drive innovation and better decision-making, which, in turn, increases profits. When looking at your demographics, it’s important to consider power and hierarchies. If most of your diversity is at the entry-level while management and leadership remain primarily white, cis, able-bodied, and male, you aren’t going to see the same results as companies with diversity at every level. than companies with below-average diversity in management. The same study found that teams with above-average diversity in management also had (Earnings Before Interest and Taxes) margins that were 9% higher than organizations with below-average diversity in management. Organizations with above-average diversity in management see 19% higher innovation revenue EBIT Studies that zero in on the results of increased diversity when it comes to specific demographics see interesting results too. A study that compared two genders (men and women) found that companies where , outperform their peers. 30% or more of their executives were women Unfortunately, the study didn’t acknowledge or account for other genders. Available research indicates these results apply to LGBTQ+ diversity as well. A study of the LGBTQ+ inclusivity at 657 publicly-traded U.S. companies found strong evidence that “ and higher stock market valuations.” Diversity makes sound commercial sense. more LGBT-friendly firms have higher profitability Differences Improve Collaboration and Innovation Diversity also affects how teams collaborate and innovate in complex ways. Overall, the more diverse a team is, the more creative and innovative they’ll be. Mixed-gender teams can compared to homogenous teams and can better maximize creativity among team members. better manage group conflict But some situations can thwart the benefits diversity brings. For instance, . In turn, this can discourage employees who belong to marginalized groups from speaking up. Because these insights are no longer being shared, the benefits of diversity are lost, while employees feel detached and alienated. stress makes people in groups start leaning back into homogenous thinking It’s also important to remember that employees who experience oppression are more likely to burn out. Glassdoor reviewed 209,112 workplace reviews and found that than their coworkers. A 2021 McKinsey report shared that LGBTQ+ women, disabled women, and Black women than their peers and noted that, “Women who regularly experience microaggressions are…almost three times as likely to say that in the past few months they have struggled to concentrate at work as a result of stress.” LGBTQ+ employees are 51% more likely to report burnout experienced more microaggressions In other words, if you want to experience the benefits of diversity, you must create a place where people can genuinely let their guard down and be themselves. That means increasing diversity in your team so employees don’t feel tokenized. This also means making your workplace a safe place to be a marginalized person. How do you do that? By making it an unsafe place for discrimination and harassment. For instance, leadership and management need to step in and say something when they see a microaggression happening. They need to make it an uncomfortable place to foster exclusion and hate. Harvard Business Review and Glassdoor partnered up for a series of and published an article on their findings in 2021. The author recommends “encouraging diverse cultural ideas while fostering agreement among employees about the importance of a common set of organizational norms and beliefs.” studies on diversity in the workplace Inclusivity is a Major Factor in Employee Retention Organizations looking to reduce employee turnover should consider that the level of inclusion employees feel at work greatly influences their career decisions. Whether or not your employees decide to take another job could come down to how comfortable they feel at work. A survey of 1,789 full-time U.S. employees and 2,000 live participant experiments found “high belonging was linked to a , a 50% drop in turnover risk, and a 75% reduction in sick days. For a 10,000-person company, this would result in annual savings of more than $52M.” whopping 56% increase in job performance A 2020 Catalyst survey of 2,100 employees at large corporations found that and 20% of an employee’s decision to stay or leave their current organization is determined by inclusivity. 35% of an employee’s emotional engagement in their job That’s not surprising, given that feeling included at work is likely to make people happier, and happier employees are more likely to stay. What’s more interesting is that the retention benefits go beyond the demographics most affected. Workplaces with more gender diversity in women and men. (Unfortunately, the research on inclusion didn’t include nonbinary employees.) may see lower turnover 4 Tips for Increasing Employee Diversity To get these benefits, organizations need to clarify where they stand on diversity at every stage of the hiring process. Candidates who are most likely to improve the diversity of your organization will also be looking most carefully at how you talk about and approach diversity. 1. Assume many of your LGBTQ+ employees aren’t out at work The Williams Institute at the UCLA School of Law surveyed 935 adults who identified as “LGBT” in May 2021. They reported that , and 25% aren’t out to anyone at work. That’s likely because 45% of LGBT employees have already experienced unfair treatment at work at some point in their lives. Keep in mind that many employees will be reluctant to be themselves due to past experiences. 50% of LGBT employees aren’t out to their supervisor It’s your job to make your organization's culture so welcoming and authentic that employees feel comfortable enough to be themselves. Have leadership and management start putting pronouns in their bios. Don’t invest in anti-LGBTQ+ companies or donate money to anti-LGBTQ+ politicians. Bring in diversity and inclusion educators from the communities they teach about. Never stop actively working to improve your organization’s diversity and inclusion. 2. Understand no marginalized communities are a monolith Oppression and inclusivity are not simple things. Especially at work. A Catalyst report on inclusivity put it well: “ such as gender, race, ethnicity, religion, age, LGBTQ+ identity, etc. More granular views of employee experiences of inclusion are needed to truly make an impact.” The report advises organizations to avoid tokenizing employees and build inclusivity into the company culture so it goes beyond policies. Review your diversity and inclusion strategy and tactics. How is intersectionality addressed in the way you approach diversity and inclusion? If you don’t know the answer, bring in a dedicated diversity and inclusion educator to help you learn more and build intersectionality into your approach. Employees' work lives are often affected by their experiences as people with intersectional identities 3. Don’t justify why diversity is vital to your organization Organizations that trip over themselves to explain they’re inclusive because it’s good for business risk alienating potential employees. Research shows employees are more likely to view an organization favorably if they make it clear to potential employees that and avoid any urge to explain why. For example, instead of saying. “We value diversity because it benefits us as well as our employees,” you could say something like, “Honoring and respecting employee diversity is and will continue to be a core value here.” honoring diversity is at the core of the business 4. Take action To impact future and current employees, you need to do more than learn and communicate; you’ll also need to take action. Think in big and small ways, such as ensuring health insurance covers gender-affirming care and encouraging employees to add pronouns to bios and email signatures. Ultimately each decision adds up to a culture that either helps employees thrive or holds them (and you) back. You should prioritize diversity for your organization because it’s the right thing to do. If you can build an inclusive environment where employees like me can be their real selves, every part of the business will benefit, too. All this is not to say that Courier has mastered building a diverse LGBTQ+ team (or will be able to do so in the near future), but we’re definitely taking a whack at it and I feel confident that we’ll get pretty close. To be a part of this mission, come ! join the team

LGBTQ+ Diversity and How it Builds Better Companies

The Sims has become a battleground for the LGBTQ+ community, showcasing how games can bond and help marginalized communities. Sims represented a refuge for the LGBTQ+ community in Russia. EA's initial refusal to release an extension focused on LGBTQ+ seemed like oversight of the many individuals who would need that escape online. Thus, EA's initial stance justified as a protest on Putin's constitutional amendment forbidding same-sex marriages would only leave an abandoned community with less representation online - thus doing more harm than good. In this slogging thread, our community discusses the gaming representation of marginalized communities and how games can bring awareness to such groups. This Slogging thread by Mónica Freitas, Daniel Guzman, Jack Boreham, Sara Pinto, Abeer and Limarc Ambalina occurred in slogging's official #gaming channel, and has been edited for readability. Mónica Freitas Mar 2, 2022, 3:16 PM Gay weddings for Russia: How The Sims became a battleground for the LGBTQ+ community https://www.theguardian.com/games/2022/feb/22/gay-weddings-for-russia-how-the-sims-became-a-battleground-for-the-lgbtq-community Mónica Freitas Mar 2, 2022, 3:17 PM "When EA decided not to release a same-sex marriage story for The Sims in Russia, the LGBTQ+ community mobilised – and won an unlikely victory" Mónica Freitas Mar 2, 2022, 3:17 PM "Two weeks ago, Electronic Arts opted not to launch a same-sex marriage story in Russia for its hit life-simulation video game The Sims 4. At a glance, this is a simple story – a game publisher refusing to compromise its art in order to comply with . Yet to Russian LGBTQ+ gamers, EA’s blog post wasn’t a show of solidarity – it was a betrayal." Russia’s controversial “homosexual propaganda” law Mónica Freitas Mar 2, 2022, 3:18 PM "Since 2013, it’s been to minors. This means that any LGBTQ+ content must carry an adults-only rating. It’s a law that EA knows all too well, as back in 2014, The Sims 4’s same-sex relationships saw it in Russia." illegal in Russia to “promote” homosexuality slapped with an 18+ rating Mónica Freitas Mar 2, 2022, 3:19 PM "Despite putting out pre-release trailers for the My Wedding Stories expansion in the region, on the day before launch, EA suddenly announced that it wouldn’t be available in Russia. “We have made the decision to forgo the release of ‘My Wedding Stories’ where our storytelling would be subject to changes, because of federal laws,” read a attributed to The Sims developers. “The ability to tell stories – any story – is at the core of what we do at The Sims. Holding back [characters] Cam and Dom’s story meant compromising the values we live by. We are committed to the freedom to be who you are, to love who you love and tell the stories you want to tell.”" blogpost Mónica Freitas Mar 2, 2022, 3:19 PM "To Russian Simmers, the move felt baffling. They swiftly rallied to create the hashtag and a with more than 10,000 signatures. Because The Sims 4 already had an adult-only rating in the region, the Russian Sims community strongly believed that releasing the downloadable content (DLC) wouldn’t break any laws. For Russians in the already marginalised LGBTQ+ community, seeing EA pull a game that publicly endorses same-sex marriage felt like a kick in the teeth." #weddingsforrussia Change.org petition Mónica Freitas Mar 2, 2022, 3:20 PM “It makes me feel isolated from the rest of the world, not supported by Simmers, or the LGBTQ community from other countries. It’s like everybody matters except us,” says lifelong Sims fan and proud queer gamer, . “I’m quite sure that the 18+ rating gives [EA the] opportunity to release the expansion. Our government wouldn’t be bothered right now by a single game pack after years of publishing The Sims here – there are bigger problems for them at the present time.” Daria Sachsen Mónica Freitas Mar 2, 2022, 3:20 PM "To Daria and many others in the Russian LGBTQ+ community, games such as The Sims offer a welcome refuge from real life’s challenges. “The Sims is one of the most popular games among our LGTBQ community,” she explains. “Everybody loves games where you can be yourself, free from the boundaries of our reality and where you can love like your heart wants to.”" Mónica Freitas Mar 2, 2022, 3:21 PM "After a week’s worth of relentless pressure from fans, EA finally reconsidered its stance. In a new , EA announced that it had decided to release the expansion in Russia after all: “We’ve reassessed our options and realised we can do more than we initially believed, and we will now release The Sims 4 My Wedding Stories Game Pack to our community in Russia, unaltered and unchanged.”" blogpost on 16 February Mónica Freitas Mar 2, 2022, 3:27 PM This news article showcases how games can bond and help marginalized communities. Sims represented a refuge for the LGBTQ+ community in Russia. EA's initial refusal to release an extension focused on LGBTQ+ seemed like oversight of the many individuals who would need that escape online. Thus, EA's initial stance justified as a protest on Putin's constitutional amendment forbidding same-sex marriages would only leave an abandoned community with less representation online - thus doing more harm than good. Mónica Freitas Mar 2, 2022, 3:29 PM What do you think of this article? Can games impact outcasts communities? Have you ever heard of a similar situation? Does The Sims hold a bigger meaning than we've ever thought? Daniel Guzman Mar 2, 2022, 4:26 PM As a gay man, I grew up with little (or any) representation on media and that world. I just saw straight couples on games, series, movies, advertising, and somehow that's sending the message that those couples are what's normal. This kind of situation, like the sims and gay marriage, helps many people realize that they're not doing anything wrong, that they can love whoever they want and that their lives can be as normal as any other straight person. Video games, especially the ones mirroring our human daily life, are really a game-changer factor at the time of building a more inclusive society. Jack Boreham Mar 4, 2022, 10:18 AM Mónica Freitas, I think games can be a powerful tool in increasing diversity and representation. It is the biggest entertainment industry on the planet. It has a far higher impact than movies. Thus, this is no surprise. Games tell stories. Sara Pinto Mar 4, 2022, 11:17 AM Mónica Freitas, I can see how excluding this game from such a big country can affect people, especially one with little to no representation of the LGBTQ+ community. The Sims is a great game because you can create whatever you wish, so it can be a way to express yourself in a world that sometimes doesn't let you do it as much. Mónica Freitas Mar 4, 2022, 11:26 AM Daniel Guzman, I love everything you've just said! Is there any other big name in gaming that you know of that's big on LGBTQ+ representation? Mónica Freitas Mar 4, 2022, 11:28 AM Jack Boreham, that's true! It seems to be done in a low-key way at times, but it certainly sends a message. Do you know of any other game big on minorities representation? Mónica Freitas Mar 4, 2022, 11:30 AM Sara Pinto, it's all about social and cultural representation. You're a big The Sims fan. Have you noticed any other signs of inclusivity? Abeer Mar 7, 2022, 10:20 AM Mónica Freitas, games can definitely impact more outcast communities. It’s a medium with so much potential that I believe the mainstream media have overlooked for many years now. I remember many nonbinary friends using games as a tool to explore their gender identities by taking on roles that they weren’t able to in their everyday lives. They were able to explore safely without fear of judgment from those around them. Jack Boreham Mar 7, 2022, 2:48 PM Mónica Freitas, Star Wars: Knights Of The Old Republic. Although not the most diverse. It had a diverse range of characters and even a lesbian cat lady, for a game of this era (2003) that was a huge step. Mónica Freitas Mar 7, 2022, 9:32 PM Abeer, that's amazing! What game did they use? One of my best friends did the same: her sims storyline was all about LGBTQ+ and she was always really excited about it. Mónica Freitas Mar 7, 2022, 9:33 PM Jack Boreham, I'm not sure a lesbian cat lady was a read or not 😂 but for 2003, that wasn't too bad. Limarc Ambalina Mar 9, 2022, 2:29 AM This is such a great thread, a very important story, and a great example of how video games can enact change for the better. 🔥 1 Sara Pinto Mar 9, 2022, 12:44 PM Mónica Freitas, I can't remember a specific expansion, but as far as I remember, you could choose whoever you want for your Sim to be with, and you could choose to look like whatever you like. It's in the same lines as what Abeer said. It's pretty cool not putting up barriers on this subject. Plus, a few years ago all this was still a bit taboo. 🔥 1 Daniel Guzman Mar 9, 2022, 3:18 PM Yeah! Mónica Freitas for example Cyberpunk 2077 includes a lot of LGBTQ+ characters! Also Fallout Series Games! 🔥 1

Gay Weddings for Russia: The Sims Edition

Dozens of advertisers instructed the company to not show their ads to people of “unknown” gender, meaning people who had not identified themselves as male or female By Jeremy B. Merrill Google’s advertising system allowed employers or landlords to discriminate against nonbinary and some transgender people, The Markup found. Companies trying to run ads on YouTube or elsewhere on the web could direct Google not to show those ads to people of “unknown gender”—meaning people who have not identified themselves to Google as “male” or “female.” After being alerted to this by The Markup, Google pledged to crack down on the practice. “We will be implementing an update to our policy and enforcement in the coming weeks to restrict advertisers from targeting or excluding users on the basis of the ‘gender unknown’ category,” Elijah Lawal, a spokesperson for Google said. Google’s forbid ads targeting or excluding male or female people from jobs, housing, or financial products, in order to comply with federal anti-discrimination laws. But until The Markup alerted Google, the company gave advertisers the option of keeping their ads from being shown to people of “unknown gender”—effectively allowing employers and landlords to either inadvertently or purposefully discriminate against people who identify as nonbinary, transgender, or anything other than male or female. policies The Markup found two such job ads on YouTube, which is owned by Google—one for jobs at FedEx and the other for Dewey Pest Control, a California-based chain. In both cases, Google’s ad targeting explanations, collected by New York University’s Ad Observer, indicated that the employer had targeted the ad based on gender but that the data did not specify which gender was targeted. In those cases, Lawal said, the advertiser had chosen to exclude people of unknown gender from seeing the ads. Upon further review, Lawal said, the company “identified approximately 100 advertisers out of many thousands” who had done the same for housing, credit, or job ads. advertisers, “out of many thousands,” that Google identified as having chosen to exclude people of unknown gender from housing, credit, or job ads. ~100 A FedEx spokesperson, Shannon Davis, said FedEx declined to comment “on our specific marketing or recruiting processes.” Dewey Pest Control didn’t respond to a request for comment. It’s not clear if the advertisers meant to prevent nonbinary people or those identifying as transgender from finding out about job openings. That would be “completely unethical and I think dangerous to society as a whole,” said Joseph Turow, a professor of communication at the University of Pennsylvania. Lawal said that Google’s existing written policies forbid excluding people on the basis of transgender or gender nonconforming identity, for any ad. But he acknowledged that the “unknown gender” checkbox had effectively allowed advertisers to discriminate by gender despite Google’s policies against that. “This is why we are working swiftly to implement a change,” he said. While federal law prohibits discrimination by gender or race in advertising for employment and housing, there’s little in the way of case law about whether restrictive online ad targeting qualifies as discrimination. Nor is there any case law about excluding trans or nonbinary people from such ads, said Pauline Kim, a law professor at Washington University in St. Louis. If a company’s intention was to exclude nonbinary or gender nonconforming people, Kim said, “you could possibly argue … that it is a form of sex discrimination” under Title VII of the Civil Rights Act of 1964, which, according to a , covers discrimination on the basis of gender identity and sexual orientation as well. recent Supreme Court decision The of employment discrimination against nonbinary and transgender people is grim. What makes advertising discrimination different is that, “you don’t even know what you don’t see,” said Kendra Albert, a technology law instructor at Harvard Law School who studies technology, law, and gender. prevalence Beyond gender, the primary categories Google offers for targeting ads to some people and not others—what it calls “personalized advertising”—are age, household income, and parental status. For ads for job opportunities, housing, and credit, Google’s rules say advertisers shouldn’t use these categories. The boxes start prechecked, meaning that in the default mode, gender isn’t taken into account in determining who will be shown the ad, but advertisers can uncheck them, effectively excluding certain categories of people from seeing their ads. That’s allowed for most ads, but Google uses other mechanisms—”internal processes” that Lawal declined to explain—to block job, housing, and credit ads that exclude people by gender, age, household income, and parental status. Meanwhile, someone signing up for Google or editing their account settings has four options for reporting their gender: “male,” “female,” “rather not say,” and an option to set a custom gender in a text box. Lawal said that the “unknown category is intended to refer to individuals where we have been unable to determine or infer the user’s gender and is not intended to allow for targeting or exclusion of users based on gender identity,” but said that people who choose not to identify their gender or write in a “custom” gender also fall into this category. Google does offer a way for users to see how they’re categorized for ads, on an ads preferences page. Google’s options for users amount to putting “a rainbow-colored Band-Aid” on “systems that were not really designed to include nonbinary people,” said Albert. “Really the question they should be asking is which gender are you, and which of these gender categories would you like us to serve you ads for,” and explaining how the ads system uses gender, Albert said. Allegations of race and sex discrimination have d Several years ago, civil rights groups sued Facebook for allowing discrimination in ads for jobs, housing, and credit; Facebook settled the suit and agreed to take those options away. The U.S. Department of Housing and Urban Development (HUD) filed a lawsuit against Facebook too. Last year, HUD that it had “worked with Google to improve Google’s online advertising policies to better align them with requirements of the Fair Housing Act.” After those interactions with HUD, Google job, housing, and credit advertisers from excluding either men or women from their ads, along with similar rules for age and other protected groups. ogged online ad platforms for years. announced banned While ads for jobs, housing, and financial products fall under special protections, it’s perfectly legal—and very common—to target other kinds of ads to one segment of the population, by age, gender, or other categories. Those categories are what digital advertisers want, Turow said, so they’re built into the heart of online ads systems. “The nirvana of advertising in 1994 has turned out to be a big mess in 2021,” he said. Originally published as " " with the Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license Google Has Been Allowing Advertisers to Exclude Nonbinary People from Seeing Job Ads

Hey Google, Show me Non-Discriminatory Job Ads

Hello, loyal readers! After republishing my first Women in Information series from last fall, I’m now ready to republish my second series from this spring. These interviews have proven to be really popular, and I’m honoured to bring them to Hacker Noon. Security I will do a third series for Fall 2017. If you’re a woman or non-male who works in , or you know someone who is, cybersecurity feel free to tweet me about it! I have who help to encourage my unpaid writing. Most of my writing for corporate blogs, and my writing for Peerlyst is paid, but this isn’t. If you enjoy pieces like this, my Patreon support helps make it possible. , even for only a couple of bucks per month, and I’ll greatly appreciate it. six Patreon supporters Consider becoming a patron This was originally published on Tripwire’s State of Security blog in April. My of interviews on women (and non-males) in information security was really popular. I spoke to some amazing minds in the cybersecurity field last fall. As spring arrived, I figured that there are probably a lot more professionals in our field who also have interesting stories to tell. last series Encouraging more non-males to enter the industry will inevitably benefit cybersecurity as a whole. In resuming my series, hopefully more readers will be inspired. This time, I spoke to . She owns , for which she blogs about news and ideas regarding the human element of cybersecurity. Human beings are the greatest vulnerability, after all! Since 2013, she has run her own firm, J L Barker Ltd., and she recently co-founded Redacted Firm. She is engaged by organisations of all sizes, from multi-national firms to SMEs. She has also appeared at important security events, such as and . Dr. Jessica Barker Cyber.uk SteelCon 2016 Infosecurity Europe 2016 So you run Cyber.uk. What compelled you to start that venture? Kim Crawley: I wanted a place to write accessible articles on cybersecurity that people of any level of knowledge or understanding could hopefully engage with and get something from if they are interested in the subject. I do research from time to time, where I generally look at attitudes and behaviours online. For example, I surveyed 1,000 people in the UK about their attitudes towards biometrics replacing passwords; 1,000 people about whether they know of 2FA; and 3,000 people in the UK, US and Germany about their opinions of Edward Snowden’s actions. Dr. Jessica Barker: I think that’s a great motive. Ignorance of cybersecurity is harming laypeople, and they need knowledge. KC: I totally agree. I think there is an onus on us in the cybersecurity profession to communicate more effectively about our work, the threats online, and what people and organizations can do to better protect themselves. JB: I try to write most of my articles with technical laypeople in mind. But when my non-IT friends read my articles, they say that they’re still difficult to understand. Educating end users can be a challenge. I really appreciate how you recognize human beings as the most significant vulnerabilty. KC: We all get used to the language of our profession and can become detached from the level of understanding that most people have when it comes to the internet, security and privacy. JB: Yes, I think you’re exactly correct. A lot of laypeople don’t even know the word “malware.” KC: You’re absolutely right about malware. I was on a pretty popular radio show in the UK (the on BBC Radio 2) talking about ransomware. Even with my perspective and determination to keep in touch with what the average person knows about security, I was shocked at how low awareness of ransomware is. We all know in the industry that it’s one of the biggest threats at the moment and has been that way for a while. But most people haven’t heard of it. JB: Jeremy Vine Show That’s incredible considering that I’ve seen ransomware in Windows clients since a decade ago. I had to deal with consumers freaking out about it. KC: I think the greatest misconception laypeople have about “hacking” is that they believe how it’s depicted in Hollywood. A “hacker” typing commands into a terminal at 300 words per minute overlooks the fact that social engineering is such a major attack method. Yes, I agree. I work with organizations on social engineering awareness, and it’s interesting to see how shocked people are by what we can find out about them via some OSINT and how that could be used in an attack. Most people still think hacking is about targeting technology and are not aware of the extent to which people are targeted. JB: Very true. Now let’s shift gears a bit. I notice that you gave a talk about impostor syndrome at SteelCon last year. Do you think women get it more frequently? KC: It’s tough. A lot of people in the industry struggle with it. I’ve had to work on my own issues with imposter syndrome, so that I don’t let it hold me back. Instead, I use it as a motivator. The danger then, of course, is burnout. I think it impacts a lot of people in this industry because cybersecurity is so diverse. It’s impossible to know everything, and we can become desensitised to how much we do know about our own area of expertise. So, you meet someone else and in conversation with them think, “They know so much more than me,” all the while forgetting that they have their own area of expertise, as you have yours. JB: I also think that most people in the industry have a personality trait whereby they want to keep learning and keep challenging themselves — which is great, but it can mean people sometimes feel like there is a lot of stuff they don’t know. There are also some cultural issues, which a lot of people commented about when I did a bit of research on it. We have a tendency in this community to tear people down if they don’t know something or get something wrong. People often say that women get imposter syndrome more frequently, but I don’t agree, to be honest. Studies have implied that in the past, but I think that could be because women are more likely to talk about it than men. From my research and conversations I’ve had with friends and colleagues in the industry, men can struggle with it just as much as women. That sums up why we all need to encourage each other in the industry. As much as there can be a culture of tearing people down, there are also a lot of people who will build others up, be supportive, and be encouraging. We need more of those people. Well said! It’s clear you’ve spoken at a lot of cons. What was your first con speaking experience? How did you prepare? How did you feel? How did the opportunity arise? KC: I think the first con I spoke at was Manchester BSides four years ago. I was speaking at events before that, on panels, and at more corporate events. But I think that was my first community con. I was approached by one of the organizers, which was great and really gave me the encouragement that I needed. I gave the closing keynote and spoke about the psychology of fear and cybersecurity. To be honest, I was really nervous speaking to a technical audience about something that’s so human-focused. The response was great, and I was pleased to find that people were genuinely interested in the human side of cybersecurity. Around that same time, I spoke at SteelCon about how to communicate cybersecurity messages more effectively, and at IRISSCon about gender and age differences in cybersecurity attitudes and behaviours. JB: I start prep for all my talks in the same way, which is to create a spider diagram on paper of all the key points I want to cover. Or sometimes I use a whiteboard or Post-It notes. I then flesh out all of the points before I start on the slides. I used to script my talks and then rehearse them until I had learned them. But as I’ve gotten more experience and more confidence, I no longer do that. I now prefer to think of the talks as more of a conversation I’m having with the audience. I will know the key point I want to make on each slide and overall for the talk. But I don’t script them, and I practice to some extent but without rehearsing so much. Did speaking at cons get easier as you got more experience? KC: Cons have definitely gotten easier as I’ve gotten more experience, more confidence, and more understanding of the fact that even technical audiences are interested in the human side of cybersecurity. JB: That’s very different than what I did. For BSides Toronto, I made a PowerPoint presentation from research notes, then I discussed each slide spontaneously. It was amazing that I kept to the 20-minute time limit. I was co-presenting with my ex-husband, and he was going to go on a tangent on a totally unrelated topic. I had to stop him and redirect the presentation to the subject matter. KC: Everyone has their own style I think, and after a while, they find what works best for them, which is why giving advice to others can be tricky. JB: Definitely. Now on to another gender-related question. In previous blogs, I mentioned that I’ve found I’ve been discriminated against less often in cybersecurity than in my earlier jobs due to my gender. What has your experience been? KC: I’ve found the same, too. I definitely experienced more discrimination when I was just starting out in the industry than I do now. I don’t know if that’s because things are progressing or because we are more likely to face discrimination when we are in more junior positions and less established in the industry. Or maybe it’s because we know the industry better now and know which people, companies, and events to avoid! JB: Yeah, I agree. Sexism is less of a problem, but it’s still there. I’ve been pleased by how many women and transgender people I’ve met in our field, and articles like these help. KC: Very true. JB: I have to say, I find the topic of your talk at BSides London last year interesting. The history of the word “cyber.” I wrote an for Tripwire on the differences, similarities and origins of the concepts of information security, cybersecurity, computer security, etc. recently. I think sometimes people lose sight of the “meta” stuff, the cultural stuff. KC: article Thanks. That’s an example of a talk I’d been thinking about for a year or so before I did it. I felt like I was putting my head above the parapet with that one because a lot of people in the community obviously hate the word “cyber,” which I understand but think we need to move on from. JB: I found that when I was talking to senior executives or people in the media or the general public, they related to “cybersecurity” so much more than “information security.” And although they technically have their own distinct definitions, we often use them interchangeably. So, I wanted to look at the cultural issues around the language we use and encourage people to address the fact that cyber is the word people outside of the industry relate to. I did some research and found that most people in the community refer to what we do as “information security” but most people in the public call it “cybersecurity.” In fact, more people in the public call what we do “e-security” than “information security.” When we’re trying to raise awareness and change behaviours, the language we use really matters. E-security? I’ve never heard of that term. Is it more common in the UK? KC: Not at all! It really surprised me, too, because I have never heard anyone actually use the term. JB: Wow. That’s amazing. I would have thought information security was more common. On another note, what advantages do you believe there are to people who don’t identify as male (women, nonbinary people) being in cybersecurity? How do you think we uniquely benefit the industry? KC: For me, it comes down to talent, representation and fairness. First and foremost, talent is equally distributed among the population; it does not discriminate. So, when groups are discriminated against or not equally represented in an industry, that industry is missing out on the talent that those individuals could provide. We deal with some really complicated issues in cyber security, and we need that talent to address those problems. Having diversity also facilitates the representation of different world views and different experiences. People with different life experiences will come at problems differently. JB: Diversity also breeds diversity, so from a representation point of view, the more diversity we have the more that sends a message that “you can do it too.” From a fairness point of view, this is a really cool industry to work in, and I want people who would be interested in the field and who would contribute to it to know they are welcome regardless of their gender identity or any other factor! That’s an excellent answer. One final question… What do you think the greatest challenges will be in cybersecurity in the coming years? KC: I think we will continue to struggle with the same problems we have for decades, like patching, legacy systems, shadow IT, and of course human behaviour. But with the pace of technological change (or rather, the pace of change in how we use technology) continuing to grow so rapidly, we face the risk of still being challenged with how to keep up. With more connectivity comes more vulnerability, and so, of course, the is a big challenge that will continue to grow. Engaging with the average person, enabling them to understand the security and privacy concerns with, for example, “smart” devices in the home, is a big challenge. Responsibility for security with regards to the Internet of Things is a tricky issue. We need security built-in from the start, but in a global market with no incentive on manufacturers to do this, we are placing too much responsibility on consumers. Responsibility for cybersecurity is not a new issue, like most of the issues we deal with in cybersecurity, but as our use of the internet continues to grow, the challenges will also grow. JB: Internet of Things Excellent! It’s been a pleasure chatting with you. I’ve learned a lot. KC:

Women in Information Security: Dr. Jessica Barker

Last fall, I interviewed six women and non-males who have exciting careers in . Those articles were all published in . cybersecurity Tripwire’s State of Security blog Ideally, all people in our field, regardless of gender, race, ethnicity, age, nationality, and sexual orientation, would simply be regarded as “people who work in information .” Unfortunately, we work in a male dominated field, and sometimes dealing with sexism affects our careers. security I think it’s especially important to encourage more women and transgender people to consider careers in cybersecurity. So my interview series shines a spotlight on some of the brightest minds in our field — who just so happen to not be male. My series was very well received. So, as spring arrived I decided to continue it. As of this writing, most of those new interviews have been done already and their corresponding articles have been sent to my editor at Tripwire. You can look forward to them being published on in the coming weeks, probably starting in April. Tripwire’s blog Until then, I’ve decided to republish here. Please enjoy them! my interview series from last fall And if you can spare a few bucks, please consider contributing to . I don’t get paid for my Medium published articles, and the trickle of money here and there that I receive from helps keep me going. Thank you! my Patreon my generous patrons There’s also a way you can help me that won’t cost you any money at all. Click on the little green heart if you like my article, it’ll help with my visibility. Most appreciated! Women in information security, being a minority, deserve a spotlight. Previously, I’ve interviewed , a Scottish security analyst who is currently working in the United States, and , a woman who went from library sciences to infosec, who’s now a host of the PVCSec podcast, and who runs her own infosec business. Tiberius Hefflin Tracy Maleeff Recent years have brought greater visibility to people who don’t completely identify with binary genders, male or female. I think openly nonbinary people are even less visible in our field than binary women are. Isly identifies as a nonbinary femme, and she works as a penetration tester for the defense industry. : Hi, Isly! How would you describe your job and title? Kim Crawley I’m a penetration tester for the civil department of a defense contractor. Isly: That sounds pretty intense! KC: We’re all pretty laid back. I: My (now former) husband did pentesting for the Canadian Forces over a decade ago. Do you have any favourite pentesting suites? KC: If it wasn’t civil and was government, I’d be less hesitant to work there, for sure. We have some folks who are still active reservists. I: Suites? Hmm. Depends on the engagement, really. I just did an OSINT engagement, and we made use of Maltego. I mostly stick to Kali and sometimes use Metasploit. I’ve stuck myself with Linux. Are there a lot of plugins for that? Most of my familiarity is with OpenVAS. KC: Well, we don’t do scanning, really. We have another team for that. Mostly we do black box and fuzzing and then follow-up with Nessus for anything we may have missed. Some of us specialize in (ICS)/SCADA, so they have their own tools. We’re learning Dradis next week, in fact. I: industrial control systems I think the earliest versions of Kali were incomplete, package- and feature-wise. But the latest releases are much improved. KC: Yeah, that’s what I’ve heard. It’s quite useful. I: Have you done a lot of SCADA work? What do you think of some people speculating about SCADA IoT? I think SCADA systems should only connect to private internal networks, anyway. KC: I haven’t done any, but my mentors and colleagues at work have, and I’m sure I will within the next year. If they’re considering SCADA-based IoT, that makes me nervous. They’re just so uniquely vulnerable, and when one domino topples in an ICS layout, they all do. I: Those of us with knowledge of infosec are concerned, but the suits often won’t listen. KC: Also, a lot of people utilizing SCADA have everything on a lateral network connecting to it. That’s very problematic. I: Have you seen utilities experience downtime due to that? KC: I personally haven’t, but some of my friends have surrounding the need for engagements over the last few years. They’re people I work with, so I trust the source. I: It’s especially scary when it affects stuff like nuclear power and sanitizing water. KC: Yeah, those things are horrifying, I agree. It can be done safely, but it takes careful planning. I: So, how did you get into infosec? KC: I started out as a Linux sysadmin, and my company needed help with the abuse department. I kind of flourished there, then went into abuse department management. I: But then I enjoyed the technical guts of the work too much, so I moved into other fields in infosec. So, something must have gotten you into sysadmining in the first place. KC: I worked in a completely unrelated field for a decade. I moved and I wasn’t allowed to keep my job. I got laid off. After not having a job for seven months, I finally applied as tech support for the sysadmin job, so the lowest level. I: And it turns out I had a knack for it, moved up from lowest level to admin to tech management within five years. From a helpdesk background, I always thought was the lowest level of IT. I think sysadmining can be more stressful, but it generally pays better. KC: that Yeah, helpdesk, call centre, chats… it’s very low level. I’m glad my company then gave me the chance to earn my way into a sysadmin role. More complex problems with a bright side: fewer issues about billing and people yelling in my ear. I: Did your employer realize your gender identity? KC: Yes. There were fewer and fewer females the higher up I got into the admin department. I: Did you ever feel like you had to fight against sexism in IT? KC: I was the first non-male technical leader in the admin department, as a supervisor for admins, the company had ever had. So yeah, I did. I felt like I had to prove myself double. I worked extra hard. There were men in that role who didn’t pull their weight. I: Did you have a lot of private sector experience before you got into the public sector? KC: No, I had no private sector experience. I’ve had very few employers, as once I like somewhere, I tend to stick around. It’s kind of against what people do now, with changing jobs every year or two for greener grass. Now people expect to do that. I: I guess you’re kind of lucky that way. Aside from my long helpdesk gig, all of my infosec work has been contracting. KC: I’m a consultant, full-time, so luckily not contract, and a lot of the people I work with have been with the company for several years and enjoy it. It’s nice for family life. That can also let you travel. I: Okay. See, I was writing study material for CISSP and CEH programs and I had to self-teach. I learned those exams that way, and that’s backwards. KC: Same on self-teaching, the last job where I did some security sysadmin work was barely scratching the surface. And they didn’t wanna pay for anything but Security+, which is garbage compared to what’s out there. I: Do you think infosec employers are getting stingier like that? KC: I can’t really say. That employer wasn’t an infosec employer. They just needed some attention to security, so it wasn’t their primary focus. If you wanted the RHCSA/RHCE, they’d pay and give you a promotion. I: Here at this employer, they really do care. I got signed on with a ninety days paid OSCP lab because they recognize it’s a good certification. We all get one training or cert paid per year. They also pitch in for bigger security conferences. You should count your blessings. KC: I really do. One of my female friends on Twitter is required to speak at four conventions a year, but I don’t know about certs. I’m sure they also provide them, but they’re a huge company she works for. I: Do you think you’ve had a socially progressive influence? KC: Oh yes. That said, in all these years, I’ve been doing technical work. So there have been few females or non-males. I: That said, I have tattoos and some minor facial piercings and coloured hair. I leave the latter two visible but demure and wear long sleeves. I applied to about twenty things before I got the job I’m in now, and that was only because a friend who works there referred me. So yeah, it’s hard with how I am and how it is. You couldn’t just demonstrate that you could reverse engineer a piece of malware (that obviously isn’t Stuxnet?) KC: I don’t know how to reverse engineer! I applied as a SOC Analyst because I didn’t think I could do anything more complex after being just a sysadmin who read logs. But with this shop, I applied to a SOC Analyst position, and they felt I had the brain for pentesting, so here I am. I don’t have a lot of self-confidence at the end of the day. I: I think all that loganalysis can hurt your eyes if you don’t have a bunch of metrics monitors and log analysis software. But then, what if the log analysis software has a catastrophic bug? KC: I didn’t have the luxury of using software back at my old job. I got very used to using ‘less’ and ‘grep’ and ‘regex’ to pull what I found. I: Wow, you get extra respect from me. But back to the idea of women in infosec, do you think it’s detrimental to the field when 99% of us seem to be male? The lack of diversity? KC: Yes, but that can be spread to a large portion of the IT world. I know more femme and non-male developers than I do in infosec. I’m only one of maybe three non-males in a team of twenty-six, I think. The ratio was worse at my old job for admins. Even worse for tech support. Though there are of females in non-technical roles there, like sales and billing. I: way lots What would you say to a young girl who’s interested in an infosec career who might be reading this? KC: I’d quote Dual Core’s paraphrasing of his own rhymes, and say, “Find all the clues, hack all the things.” I: Make IoT Barbie say “Math is hard.” KC: not Ha, IoT Barbie! I forgot that existed. I: If you enjoyed my article, there are two ways that you can help me. First, you can click on the little green heart to recommend my article. Secondly, you can make a small donation to . Thank you! my Patreon

Women in Information Security: Isly

7 Surprising Ways People Became Infosec Professionals

Next-Gen FHE Toolchains for the Dev Multiverse: How TFHE Is Taking Us There

Analyzing HIVE Flaws and Part-Prototype Models: A Comparative Study

Feature Engineering for Machine Learning Models: Everything You Need to Know 

মেশিন লার্নিং এর জন্য ফিচার ইঞ্জিনিয়ারিং

Feature Engineering für maschinelles Lernen

Ingeniería de funciones para el aprendizaje automático

Ingénierie des fonctionnalités pour l’apprentissage automatique

मशीन लर्निंग के लिए फ़ीचर इंजीनियरिंग

機械学習のための特徴エンジニアリング

기계 학습을 위한 특성 엔지니어링

Engenharia de recursos para aprendizado de máquina

Разработка функций для машинного обучения

Makine Öğrenimi için Özellik Mühendisliği

Kỹ thuật tính năng cho Machine Learning

机器学习的特征工程

This paper is available on arxiv under CC 4.0 license.
Authors:
(1) Muhammed Yusuf Kocyigit, Boston University;
(2) Anietie Andy, University of Pennsylvania;
(3) Derry Wijaya, Boston University. Table of Links Abstract and Intro
Related Works
Data
Method
Analysis and Results
Conclusion
Limitations
Ethics Statement and References
Appendix: Toxicity Measurement
Appendix: Correlation Over Time
Appendix: Wikidata
Appendix: Hyperparameter Sensitivity Limitations Reliability of Word Embeddings The representation power and how much word embeddings can model our beliefs is an active point of discussion. This line of research is critical because it allows differentiating under which conditions word representations generate good abstractions to analyze. The critique can be summarized in two main lines: structural and statistical problems. Structural problems (Arseniev-Koehler 2021) relate to the Distributional Hypothesis (Harris 1954) that current word embeddings are based on. This is criticized by structuralism in multiple aspects. Primarily issues arise when the phrase a word is known by the company it keeps is interpreted as a word is the company it keeps. The representations of words or groups that we present are not to say that the content represents them precisely as the representations, but they are more closely linked to certain concepts if they are closer in the semantic space. The second important issue we must be aware of is that concepts are not always dichotomous or binary. They can be graded or related in multiple ways. We, however, in line with An, Kwak, and Ahn (2018) assume a dichotomous nature to a set of adjectives. The statistical problems mainly stem from the frequency of different words/concepts. Valentini, Slezak, and Altszyler (2022) shows that word representations tend to yield higher similarity scores between high-frequency words than between high and low-frequency words. Joseph and Morgan (2020) show that word representations produce similarity measures that better represent human beliefs in with higher frequency words. Loon et al. (2022) argue that word embedding methods like word2vec cluster high-frequency words together and low-frequency words together. They argue that people have an intuitive positive bias where positive words are more frequent and if a racial/social group(black names in their case) is less frequently represented in the corpus compared to their control group(white names). They are just likely to be related to negative words and this relation would not necessarily be an indication of bias in the corpus. Structural problems are related to how we analyze the results and make inferences from them. However, in this paper, we target statistical problems on two levels. First, we look at the distribution of the context words in the primary data where the word vectors were trained and observe that the context words for these groups are not significantly different, and on the frequency of names, we sample equal amounts of (word, context) pairs for both African Americans and White Americans and train the embeddings for these groups from scratch rather than using a pre-trained embedding that would depend on the frequency of names in the initial corpus. One specific limitation that we had to work around in this project was the issue of false positive matches. Even though we search for exact name+surname combination in books the resulting n-grams are not necessarily always about the same person that we intent on. While a person with the same name+surname combination likely belongs to the same racial group in our analysis, this might not be the same for other use cases making this a relevant limitation to consider. Another relevant limitation is the data size limitation. We have replicated our results with a manually curated name subsets however a smaller subset of n-grams was not a viable option. We have also observed that decades before 1850s return too few samples and the results are not meaningful for those decades. The question of how many n-grams would be enough for a new application is still an open question that we were not able to answer. This question would also extend the scope of this paper beyond a reasonable limit. This paper is available on arxiv under CC 4.0 license. Authors: (1) Muhammed Yusuf Kocyigit, Boston University; (2) Anietie Andy, University of Pennsylvania; (3) Derry Wijaya, Boston University. This paper is available on arxiv under CC 4.0 license. Authors: Authors: (1) Muhammed Yusuf Kocyigit, Boston University; (2) Anietie Andy, University of Pennsylvania; (3) Derry Wijaya, Boston University. Table of Links Abstract and Intro Related Works Data Method Analysis and Results Conclusion Limitations Ethics Statement and References Appendix: Toxicity Measurement Appendix: Correlation Over Time Appendix: Wikidata Appendix: Hyperparameter Sensitivity Abstract and Intro Abstract and Intro Related Works Related Works Data Data Method Method Analysis and Results Analysis and Results Conclusion Conclusion Limitations Limitations Ethics Statement and References Ethics Statement and References Appendix: Toxicity Measurement Appendix: Toxicity Measurement Appendix: Correlation Over Time Appendix: Correlation Over Time Appendix: Wikidata Appendix: Wikidata Appendix: Hyperparameter Sensitivity Appendix: Hyperparameter Sensitivity Limitations Reliability of Word Embeddings The representation power and how much word embeddings can model our beliefs is an active point of discussion. This line of research is critical because it allows differentiating under which conditions word representations generate good abstractions to analyze. The critique can be summarized in two main lines: structural and statistical problems. Structural problems (Arseniev-Koehler 2021) relate to the Distributional Hypothesis (Harris 1954) that current word embeddings are based on. This is criticized by structuralism in multiple aspects. Primarily issues arise when the phrase a word is known by the company it keeps is interpreted as a word is the company it keeps. The representations of words or groups that we present are not to say that the content represents them precisely as the representations, but they are more closely linked to certain concepts if they are closer in the semantic space. The second important issue we must be aware of is that concepts are not always dichotomous or binary. They can be graded or related in multiple ways. We, however, in line with An, Kwak, and Ahn (2018) assume a dichotomous nature to a set of adjectives. The statistical problems mainly stem from the frequency of different words/concepts. Valentini, Slezak, and Altszyler (2022) shows that word representations tend to yield higher similarity scores between high-frequency words than between high and low-frequency words. Joseph and Morgan (2020) show that word representations produce similarity measures that better represent human beliefs in with higher frequency words. Loon et al. (2022) argue that word embedding methods like word2vec cluster high-frequency words together and low-frequency words together. They argue that people have an intuitive positive bias where positive words are more frequent and if a racial/social group(black names in their case) is less frequently represented in the corpus compared to their control group(white names). They are just likely to be related to negative words and this relation would not necessarily be an indication of bias in the corpus. Structural problems are related to how we analyze the results and make inferences from them. However, in this paper, we target statistical problems on two levels. First, we look at the distribution of the context words in the primary data where the word vectors were trained and observe that the context words for these groups are not significantly different, and on the frequency of names, we sample equal amounts of (word, context) pairs for both African Americans and White Americans and train the embeddings for these groups from scratch rather than using a pre-trained embedding that would depend on the frequency of names in the initial corpus. One specific limitation that we had to work around in this project was the issue of false positive matches. Even though we search for exact name+surname combination in books the resulting n-grams are not necessarily always about the same person that we intent on. While a person with the same name+surname combination likely belongs to the same racial group in our analysis, this might not be the same for other use cases making this a relevant limitation to consider. Another relevant limitation is the data size limitation. We have replicated our results with a manually curated name subsets however a smaller subset of n-grams was not a viable option. We have also observed that decades before 1850s return too few samples and the results are not meaningful for those decades. The question of how many n-grams would be enough for a new application is still an open question that we were not able to answer. This question would also extend the scope of this paper beyond a reasonable limit.

A Novel Method for Analysing Racial Bias: Collection of Person Level References: Limitations

Mastering Maps in Go: Everything You Need to Know

গো-তে মানচিত্র আয়ত্ত করা: আপনার যা জানা দরকার

Maps in Go meistern: Alles, was Sie wissen müssen

Dominar mapas en Go: todo lo que necesitas saber

Maîtriser les cartes en Go : tout ce que vous devez savoir

गो में मानचित्रों पर महारत हासिल करना: वह सब कुछ जो आपको जानना चाहिए

Go でマップをマスターする: 知っておくべきことすべて

Go로 지도 마스터하기: 알아야 할 모든 것

Dominando mapas em Go: tudo o que você precisa saber

Освоение карт в Go: все, что вам нужно знать

Go'da Haritalarda Uzmanlaşmak: Bilmeniz Gereken Her Şey

Làm chủ bản đồ trong Go: Mọi thứ bạn cần biết

掌握 Go 中的 Maps：你需要知道的一切

This paper is available on arxiv under CC 4.0 license.
Authors:
(1) Jihoon Chung;
(2)  Zhenyu (James) Kong. Table of Links Abstract & Introduction
Review of Related Work
Proposed Research Methodology
Numerical Case Studies
Real-World Simulation Case Studies
Conclusion
References IV. NUMERICAL CASE STUDIES This section provides numerical studies to compare the performance between the proposed method and benchmark methods. The studies evaluate the performance by varying the correlations (ki in Eq. (7)) between the KCCs in nonstationary process faults. All the numerical case studies consist of 20 independent trials. The code of the proposed algorithm is implemented in Matlab 2021. The CPU used in case studies is an Intel® Core™ Processor i7-8750H. This section compares the proposed method with the following benchmark methods. • User Grouping Sparse Bayesian Learning (i.e., UGSBL) proposed in [2] is an SBL method that clusters the KPCs samples into groups sharing the same process faults in sparse estimation. • Spatially Correlated Bayesian Learning (i.e., SCBL) proposed in [8] is an SBL method to diagnose process faults by exploiting the spatial correlation of KCCs. • [10] proposed the SBL method that considers prior knowledge of the process faults in sparse estimation. • MSBL proposed in [32] is a basic SBL method that assumes the independence of KCCs and stationarity of process faults. Performance Evaluation: To evaluate the effectiveness of the proposed method, two performance measures are used in this paper. The first measure evaluates the process fault detection capability, and the second measure is related to estimation accuracy. Each measure is evaluated by varying the correlation coefficients of spatially correlated KCCs. Fault Detection Capability: First, the proposed method requires correctly identifying process faults whose KCCs have excessive variance. In order to assess the capability of detecting process faults under sparse conditions, the area under the receiver operating characteristics (ROCs) curve (AUC) is utilized [46]. AUC is widely used for evaluating the accuracy of binary classification. Specifically, AUC measures the quality of the classification accuracy between process faults and non process faults irrespective of the threshold of variance. In other words, AUC accounts for the trade-off between the type 1 and type 2 errors. When the method is capable of achieving perfect classification results, the resulting AUC value will be 1. However, the AUC with a value close to 0.5 denotes that the estimated variances of KCCs from the method are similar between the process faults and non-process faults. Table I represents that the proposed method always achieves the best performance of AUC in the various spatial correlation between KCCs. Specifically, the AUC of the proposed method is close to 1.0 at every correlations levels, representing the perfect classification between the process faults and nonprocess faults in terms of estimated variance. Since UGSBL clusters the KPCs samples that share the same process faults, the method shows a comparable result to the proposed method  when the spatial correlation of KCCs is relatively low and moderate. However, the performance degrades in the high spatial correlation since the method assumes independence between all KCCs. The performance of SSBL is robust to correlations. This is because SSBL utilizes the correlation of correlated KCCs in its estimation. However, SSBL performs worse than the proposed method since this method assumes the same process faults between all KPCs samples, which are unsuitable for dealing with nonstationary process faults. MSBL and [10] show poor performances since these methods assume the independence of KCCs and stationarity of process faults. 2) Estimation Accuracy: In addition to correctly identifying the process fault, the method needs to accurately estimate the variance of process faults. After the process faults are identified, the practitioners are interested in how large the variances are to make decisions for the maintenance of the process. To achieve this objective, this paper provides the normalized mean squared error (NMSE) between true and estimated variance. Table II represents that the proposed method achieves the best performance in the estimation accuracy of the variance in most of the correlations. It implies that the proposed method is useful not only for the process faults identification but also for measuring how severe the process faults are. Compared to benchmark methods, the effectiveness of the proposed method in the NMSE results is noticeable in the presence of high spatial correlations between KCCs. Unlike AUC performance, the NMSE of SSBL increases as the correlation increases. It implies that SSBL successfully identifies the process faults utilizing the correlation information but cannot accurately estimate the variance since SSBL assumes the stationary process faults over KPC samples. The performance of UGSBL, MSBL, and [10] represent similar trends to the performance of AUC. This paper is available on arxiv under CC 4.0 license. Authors: (1) Jihoon Chung; (2)  Zhenyu (James) Kong. This paper is available on arxiv under CC 4.0 license. Authors: (1) Jihoon Chung; (2)  Zhenyu (James) Kong. Table of Links Abstract & Introduction Review of Related Work Proposed Research Methodology Numerical Case Studies Real-World Simulation Case Studies Conclusion References Abstract & Introduction Abstract & Introduction Review of Related Work Review of Related Work Proposed Research Methodology Proposed Research Methodology Numerical Case Studies Numerical Case Studies Real-World Simulation Case Studies Real-World Simulation Case Studies Conclusion Conclusion References References IV. NUMERICAL CASE STUDIES This section provides numerical studies to compare the performance between the proposed method and benchmark methods. The studies evaluate the performance by varying the correlations (ki in Eq. (7)) between the KCCs in nonstationary process faults. All the numerical case studies consist of 20 independent trials. The code of the proposed algorithm is implemented in Matlab 2021. The CPU used in case studies is an Intel® Core™ Processor i7-8750H. This section compares the proposed method with the following benchmark methods. • User Grouping Sparse Bayesian Learning (i.e., UGSBL) proposed in [2] is an SBL method that clusters the KPCs samples into groups sharing the same process faults in sparse estimation. • Spatially Correlated Bayesian Learning (i.e., SCBL) proposed in [8] is an SBL method to diagnose process faults by exploiting the spatial correlation of KCCs. • [10] proposed the SBL method that considers prior knowledge of the process faults in sparse estimation. • MSBL proposed in [32] is a basic SBL method that assumes the independence of KCCs and stationarity of process faults. Performance Evaluation: To evaluate the effectiveness of the proposed method, two performance measures are used in this paper. The first measure evaluates the process fault detection capability, and the second measure is related to estimation accuracy. Each measure is evaluated by varying the correlation coefficients of spatially correlated KCCs. Performance Evaluation: Fault Detection Capability: First, the proposed method requires correctly identifying process faults whose KCCs have excessive variance. In order to assess the capability of detecting process faults under sparse conditions, the area under the receiver operating characteristics (ROCs) curve (AUC) is utilized [46]. AUC is widely used for evaluating the accuracy of binary classification. Specifically, AUC measures the quality of the classification accuracy between process faults and non process faults irrespective of the threshold of variance. In other words, AUC accounts for the trade-off between the type 1 and type 2 errors. When the method is capable of achieving perfect classification results, the resulting AUC value will be 1. However, the AUC with a value close to 0.5 denotes that the estimated variances of KCCs from the method are similar between the process faults and non-process faults. Fault Detection Capability: First, the proposed method requires correctly identifying process faults whose KCCs have excessive variance. In order to assess the capability of detecting process faults under sparse conditions, the area under the receiver operating characteristics (ROCs) curve (AUC) is utilized [46]. AUC is widely used for evaluating the accuracy of binary classification. Specifically, AUC measures the quality of the classification accuracy between process faults and non process faults irrespective of the threshold of variance. In other words, AUC accounts for the trade-off between the type 1 and type 2 errors. When the method is capable of achieving perfect classification results, the resulting AUC value will be 1. However, the AUC with a value close to 0.5 denotes that the estimated variances of KCCs from the method are similar between the process faults and non-process faults. Table I represents that the proposed method always achieves the best performance of AUC in the various spatial correlation between KCCs. Specifically, the AUC of the proposed method is close to 1.0 at every correlations levels, representing the perfect classification between the process faults and nonprocess faults in terms of estimated variance. Since UGSBL clusters the KPCs samples that share the same process faults, the method shows a comparable result to the proposed method  when the spatial correlation of KCCs is relatively low and moderate. However, the performance degrades in the high spatial correlation since the method assumes independence between all KCCs. The performance of SSBL is robust to correlations. This is because SSBL utilizes the correlation of correlated KCCs in its estimation. However, SSBL performs worse than the proposed method since this method assumes the same process faults between all KPCs samples, which are unsuitable for dealing with nonstationary process faults. MSBL and [10] show poor performances since these methods assume the independence of KCCs and stationarity of process faults. 2) Estimation Accuracy: In addition to correctly identifying the process fault, the method needs to accurately estimate the variance of process faults. After the process faults are identified, the practitioners are interested in how large the variances are to make decisions for the maintenance of the process. To achieve this objective, this paper provides the normalized mean squared error (NMSE) between true and estimated variance. Table II represents that the proposed method achieves the best performance in the estimation accuracy of the variance in most of the correlations. It implies that the proposed method is useful not only for the process faults identification but also for measuring how severe the process faults are. Compared to benchmark methods, the effectiveness of the proposed method in the NMSE results is noticeable in the presence of high spatial correlations between KCCs. Unlike AUC performance, the NMSE of SSBL increases as the correlation increases. It implies that SSBL successfully identifies the process faults utilizing the correlation information but cannot accurately estimate the variance since SSBL assumes the stationary process faults over KPC samples. The performance of UGSBL, MSBL, and [10] represent similar trends to the performance of AUC.

Sparse Bayesian Learning: Numerical Case Studies

This paper is under CC BY-SA 4.0 DEED license. available on arxiv Authors: (1) Juan Mera Men´endez; (2) Martin Bartlett. Table of Links Abstract and Introduction Initial Application Performance Tests Best Practices and Techniques Combinations Discussion Related Work Conclusion and References IV. BEST PRACTICES AND TECHNIQUES Below are the proposed techniques and approaches to reduce the cold start of Java functions and improve their overall performance. Along with their measurements in terms of performance tests. The proposed practices will be compared in the VI.Discussion section. A. Better suited configuration Increasing the allocated memory for a function is probably the most well-known way to reduce cold starts and also the simplest. The allocation of memory to a function is related to the amount of CPU assigned to that function. A greater amount of CPU generally makes the initialization of the function faster. However, it’s not always the case that more memory equals better performance. It’s also necessary to consider that higher memory allocation will result in higher costs, but if the computing time is reduced due to increased performance, the overall cost will decrease as well. To properly balance the memory configuration of a Lambda function, the Lambda Power Tuner solution [19] can be of great help to profile the function, displaying graphs such as the one shown in Figure 2 for Catalog lambda of our solution. In our case, chose the right memory value for the functions configuration suppose an improvement of about 70% in terms of cold start going from an average of 16307 ms to 4867 ms. In addition to a noteworthy 46% when the functions are warmed passing from an average of 512 ms to 216 ms as can be seen in Table II. For the rest of the versions that do not have an appropriate memory configuration, the default value of 512 MB is used. B. Snapstart Snapstart [16] is a feature offered by AWS to mitigate the cold start of Java functions. It involves initializing the function at the moment a version of it is created. Lambda takes a snapshot of the initialized lambda’s runtime environment, encrypts and caches it. When initializing new functions, it does so from the snapshot rather than starting from scratch, thus improving performance. A notable drawback of this feature is that it does not support many other features offered by AWS Lambda, such as custom runtimes, the arm64 architecture, Amazon EFS, and other managed runtimes that are not Java 11 or Java 17. In this case, enable Snapstart on our functions allow an enhancement of 16% at the cold start (going from an average of 16307 ms to 13736 ms) and an improvement of 21% attending requests when the functions are warmed making a difference of over 100 ms in each request answered (Table III). C. Arm64 architecture The instruction set architecture within AWS Lambda, either x86 64 or arm64, determines the processor that the function will use. AWS’s official documentation suggests that the arm64 architecture [18] with its Graviton processor can offer significantly better performance than the x86 64 architecture. Regarding our solution, the impact of using arm64 on the latencies was positive, getting over a 14% of improvement on cold starts and a 13% when the lambdas are warmed. Reducing the average latency by over 2000 ms and approximately 70 ms, respectively. The warmed lambda functions data can be seen in Table IV. D. AWS SDK v2 for Java The AWS SDK for Java 2.x [15] is the second version of the SDK provided by AWS for Java. This version is designed with a modular approach, allowing you to include only the service-specific JARs you need, reducing the size of your artifacts. That’s a key point of reducing the cold start, loading less number of classes when initialize the function. Furthermore, it boasts several other performance improvements such as asynchronous nature, connection management, concurrency support, resource management, and so forth. Due to that, in terms of performance its use is highly recommended. For us, migrate the lambda functions from sdk v1 to sdk v2 suppose an enhancement of about 40% in both cold start and warmed functions. On first going from 16307 ms to 9926 ms and on second one passing from 512 ms to 298 ms of latency as is visible in Table V. E. GraalVM GraalVM [20] is a platform that enables ahead-of-time compilation of Java code, leveraging the Native Image tool to produce a standalone binary. As a result, this binary offers significantly better performance compared to traditional Java code that runs using the JVM and is compiled at runtime. The binary is produced for the OS of the environment the code is built. It’s important to note that AWS Lambda is only compatible with GraalVM through the use of custom runtimes. Other disadvantage is that the ahead-of-time compilation sometimes would be tricky to make it work properly adding even the necessity of dealing with low level issues caused by AOT compilation. For us, create a native image of our code and uses it generates a great improvement of 83% on cold start, going form 16307 ms to 2800 ms. When the lambdas are warmed, it goes from and average of 512 ms to 231 ms which means a 55% of enhancement, visible in Table VI. F. Environment variable: JAVA TOOL OPTIONS AWS Lambda provides some customization options [17] for the managed Java runtime, which can enhance overall function performance. This is achieved through the use of the JAVA TOOL OPTIONS environment variable, which allows access to JVM features such as tiered compilation or garbage collector behavior among others. In this case we use only tiered compilation, fixing the level to 1, enabling the C1 compiler which means optimized code for fast start-up time. The exact value for the mentioned key is ”-XX:+TieredCompilation - XX:TieredStopAtLevel=1”. Java 17 managed runtime is default set to level 1 for AWS Lambda, while in Java 11, you need to set it manually. Level two can be used for an overall better performance, not only focus on the start time. In our test system, enabling that option translates to a 40% improvement in cold start times and a 31% improvement in response times for the system with warm Lambdas (Table VII). G. Others In addition to the previous techniques and approaches we propose and have tested, there are many other best practices [6], [14], [21]. Whether they are more dependent on the use case or more general, below, we mention some of them: • Reduce the size of the bundle to enhance the initialization phase. As less classes to load as fast the function will initialize. • Initialize the connections of third party out of the handler and same the static resources on /tmp to reuse it from one execution to another. • A framework, if needed, should be as lighter as possible. Due to that we suggest replacing spring with other smaller frameworks like Vert.x • Try to avoid reflection, that feature makes some JVM optimizations impossible to be done as it works with dynamic types. • Initialize all necessary dependencies and classes during initialization time. These approaches, combined with some of the previous ones we have tested, for sure will greatly reduce the impact of cold starts and increase the performance of Java functions by a really significant percentage.

Performance Best Practices Using Java and AWS Lambda: Best Practices and Techniques

This paper is available on arxiv under CC BY-SA 4.0 DEED license. Authors: (1) Tomáš Jakl, Czech Academy of Sciences and Czech Technical University; (2) Dan Marsden, School of Computer Science University of Nottingham; (3) Nihil Shah, Department of Computer Science University of Oxford. Table of Links Abstract & Introduction Prelimenaries FVM Theorems for Positive Existential Fragments FVM Theorems for Counting Logic FVM Theorems for The Full Logic Abstract FVM Theorems for Products Adding Equality and Other Enrichment Conclusions, Acknowledgments & References Appendix A FVM theorems for coproducts Appendix B Proofs Omitted from Section III Appendix C Proofs Omitted from Section IV Appendix D Proofs Omitted from Section V Appendix E Proofs Omitted from Section VI Appendix F Proofs Omitted from Section VII APPENDIX E PROOFS OMITTED FROM SECTION VI Here we show FVM theorems for products of arbitrary collection of structures. Note that Section VI is stated in terms of binary products only but here we work with infinitary products, to support the claim preceding Example VI.6.

Appendix E Proofs Omitted from Section VI

This paper is available on arxiv under CC 4.0 license. Authors: (1) Mattia Atzeni, EPFL, Switzerland and mattia.atzeni@outlook.it; (2) Mrinmaya Sachan, ETH Zurich, Switzerland; (3) Andreas Loukas, Prescient Design, Switzerland. Table of Links Abstract & Introduction Formalizing the Group-Action Learning Problem Attention Masks for Core Geometry Priors The LATFORMER Architecture Experiments Related Work Conclusion & References A. Additional Details on the Model B. Additional Experiments and Details on the Experimental Setup C. Limitations and Future Work D. Deferred Proofs Abstract The Abstraction and Reasoning Corpus (ARC) (Chollet, 2019) and its most recent languagecomplete instantiation (LARC) has been postulated as an important step towards general AI. Yet, even state-of-the-art machine learning models struggle to achieve meaningful performance on these problems, falling behind non-learning based approaches. We argue that solving these tasks requires extreme generalization that can only be achieved by proper accounting for core knowledge priors. As a step towards this goal, we focus on geometry priors and introduce LATFORMER, a model that incorporates lattice symmetry priors in attention masks. We show that, for any transformation of the hypercubic lattice, there exists a binary attention mask that implements that group action. Hence, our study motivates a modification to the standard attention mechanism, where attention weights are scaled using soft masks generated by a convolutional network. Experiments on synthetic geometric reasoning show that LATFORMER requires 2 orders of magnitude fewer data than standard attention and transformers. Moreover, our results on ARC and LARC tasks that incorporate geometric priors provide preliminary evidence that these complex datasets do not lie out of the reach of deep learning models. 1. Introduction Infusing inductive biases and knowledge priors in neural networks is regarded as a critical step to improve their sample efficiency (Battaglia et al., 2018; Bengio, 2017; Lake et al., 2017; Lake & Baroni, 2018; Bahdanau et al., 2019). The Core Knowledge priors for human intelligence have been studied extensively in developmental science (Spelke & Kinzler, 2007), following the theory that humans are endowed with a small number of separable systems of core knowledge, so that new flexible skills and belief systems can build on these core foundations. Recent research in artificial intelligence (AI) has postulated the idea that the same priors should be incorporated in AI systems (Chollet, 2019), but it is an open question how to incorporate these priors in neural networks. Following this chain of thought, the Abstraction and Reasoning Corpus (ARC) (Chollet, 2019) was proposed as an AI benchmark built on top of the Core Knowledge priors from developmental science. Chollet (2019) posits that developing a domain-specific approach based on the Core Knowledge priors is a challenging first step and that “solving this specific subproblem is critical to general AI progress”. Further, he argues that ARC “cannot be meaningfully approached by current machine learning techniques, including Deep Learning”. An important category of Core Knowledge priors includes geometry and topology priors. Indeed, significant attention has been devoted to incorporating such priors in deep learning architectures by rendering neural networks invariant (or equivariant) to transformations represented through group actions (Bronstein et al., 2021). However, group-invariant learning helps to build models that systematically ignore specific transformations applied to the input (such as translations or rotations). We take a complementary perspective and aim to help neural networks to learn functions that incorporate geometric transformations of their input (rather than to be invariant to such transformations). In particular, we focus on group actions that belong to the symmetry group of a lattice. These transformations are pervasive in machine learning applications, as basic transformations of sequences, images, and other higher-dimensional regular grids fall in this category. While attention and transformers can in principle learn these kind of group actions, we show that they require a significant amount of training data to do so. To address this sample complexity issue, we introduce LATFORMER, a model that relies on attention masks in order to learn actions belonging to the symmetry group of a lattice, such as translation, rotation, reflection, and scaling, in a differentiable manner. We show that, for any such action, there exists an attention mask such that an untrained self-attention mechanism initialized to the identity function performs that action. We further prove that these attention masks can be expressed as convolutions of the identity, which motivates a modification to the standard attention module where the attention weights are modulated by a mask generated by a convolutional neural network (CNN). Our experiments focus on abstract geometric reasoning and, more specifically, on ARC and its variants, as they are widely regarded as challenging benchmarks for machine learning models (Acquaviva et al., 2021; Chollet, 2019). On these datasets, we aim to reduce the gap between neural networks and hand-engineered search algorithms. To probe the sample efficiency of our method, we compared its ability to learn synthetic geometric transformations against Transformers and attention modules. Then, we annotated ARC tasks based on the knowledge priors they require, and we evaluated LATFORMER on the ARC tasks requiring geometric knowledge priors. Finally, we performed experiments on the more recent Language-complete ARC (LARC) (Acquaviva et al., 2021), which enriches ARC tasks with natural-language descriptions, and we compared our model against strong baselines based on neural program synthesis. Our results provide evidence that LATFORMER can learn geometric transformations with 2 orders of magnitude fewer training data than transformers and attention. We also significantly reduce the gap between neural and classical approaches on ARC, providing the first neural network that reaches good performance on ARC tasks requiring geometric knowledge priors.

Stories, News & Mentions about NUBINARY

Nubinary’s Stories on HackerNoon