HNHacker News
TopNewBestAskShowJobs

sansnomme

1,145 karma · joined January 22, 2019

submissionscomments
sansnomme··on What DoorDash pays, after expenses, and what’s happening with tips
There is room for Uber-app-in-cloud-emulator multiplexers. As single app endpoint for a variety of delivery apps so you will have more options to choose from when business is slow.
sansnomme··on The technology pushed into schools today is a threat to child development
The problem is that most schools do not provide a good education on how to be creators instead of passive consumers of technology. Learning to use tools such as Excel should be coupled with a tutorial on Visual Basic, beyond shallow here's-a-formula-here's-a-grid. Millennials and Gen Zs don't need the an entire class dedicated to how to use Khan Academy. They just need to know that Moocs and other free resources exist. Another flaw is that a lot of academic stuff are supposed to teach skills by proxy. E.g. learning how to use a search engine through writing a research essay. But most of the "academic search engines" and databases are frankly terrible. Take a look at most universities' library websites and see what they recommend. Most so-called academic journal tooling cannot hold a candle to e.g. Google Scholar. This is especially bad if you study humanities and is willing to just accept technology advice from your institution; you would be wasting hours on search tools with ancient UIs because your professor thinks Google is too "general".
sansnomme··on Adult site leaks data of cam models
What about fintech KYC APIs? Are those sufficiently compliant?
sansnomme··on Computational Power Found in the Arms of Neurons
Also: spiking neural networks. Doing useful stuff with them has been a work-in-progress so far though.
sansnomme··on DevDegree: Work at Shopify and get a free CS degree in parallel
GPT-2 trained on own Reddit profile and deployed to HN perhaps?
sansnomme··on Teen Saved from Seizure After Online Gaming Friend Calls Police 5k Miles Away
For constant-surveillance companies like Facebook, a good way to justify the data collection would be to add automated systems to detect such problems. They already have algorithms in place for suicide prevention, it would be easy to increase coverage to stuff like seizure, heart attacks, and other problems that may correlate with long usage of computers. Even simple heuristics such as sudden keyboard depression and subsequent inquiry dialogs that are missed could help. Of course, ML on webcam vis-à-vis cats accidentally walking across keyboard would help to rule out false positives.
sansnomme··on Building a Secure Electron App
For other types of security try https://github.com/tauri-apps/tauri

It is an electron-like framework written in Rust, and is designed to be as lightweight as possible.

sansnomme··on How to Get Vaccinated Without Parental Consent
And there it is. The first assumption is that my background is in CS and not biology, that I do not understand the nature of your deeply mysterious arts. You yourself hang out on HN, I hope the irony isn't lost. This whole attitude is exactly why there is no decent seed funding for biology. Too conservative, too risk-adverse. The ivory tower knows best. The obsession with "better to die from cancer than from an untested treatment". It took decades for Right to Try to become reality and still there is tremendous opposition. Let's leave the orphan drug rent-seeking to private equity funds. Individual researchers should "know their place" and thrive only in the generosity and largesse of pharmaceutical giants. It is no wonder people are losing faith in the biological sciences. When computing is bringing breakthrough after breakthroughs, what has biology given us? Some new materials? Marginally improved treatments? How many magnitude level improvements have you seen? Where is my iPhone? Where are my VR goggles? Most if not all improvements are slow incremental gains built on the blood, sweat, and tears of poorly-paid phds who will be lucky to get paid 6 figures after graduation. It is easy to "look back on the decade" and pat yourself on the back, celebrating the oh-so-vast gains. The gains in biology hardly holds a candle to how much computing has advanced. The field of electronics has gotten cheaper than ever. Yet in biology, just how many people do you think have access to say, cryogenic electron microscopy? This isn't on Stanford light source level of difficulty, but there has been no effort at all to democratise the technology. Your ML researchers have access to ASICs for computing mere months after they get designed, both actual hardware and in the cloud. Tons of funding and development being put into improving developer user experience. Can you say the same for biology? The amount of "data processing" for a cryo microscope is trivial by modern cloud standards. Where are the Raspberry Pis, the Arduinos, of biology? Where are the cheap microscopy equipment clones? Because there is no market. There is no demand for biology.
sansnomme··on How to Get Vaccinated Without Parental Consent
Science is about being falsifiable. You can prove many things with a battery, several magnets, and a coil of wire. Non-organic chemistry is similarly easy to demonstrate, even the most ardent of science-deniers would hesitate to dip their fingers into aqua regia, should you tell them what it is. But when it comes to biology and organic chemistry, things are completely different. Let's not talk about controversial scientific questions like microwave radiation effects on DNA. Simple stuff like "catching a cold". How many people do you know believe that it's a function of the weather? Sure, there may be a correlation with cold temperatures which has been proven, but the existence of viruses and the fact that they cause colds is basic science. We teach children the ideas behind splitting the atom and unleashing the metaphorical Death of the Bhagavad Gita. In other words, your average high school kid has a decent idea of what may cause Armageddon, the difference between a compressive and a "gun" type. But how many can tell you how influenza viruses work? There is a huge problem with our current approach to biology, bioengineering, and biotechnology education. Influenza kills a lot more every year, regardless of how often blog.nuclearsecrecy.com shows up and gets upvoted on HN.
sansnomme··on How to Get Vaccinated Without Parental Consent
I will just link to my previous comment: https://news.ycombinator.com/item?id=21799204

The barrier of entry to biology is artificially high due to all sorts of reasons, mostly regulatory. Biohacking is often frowned upon on the federal level and the equipment is expensive. Even with stuff like OpenTrons, we are talking about several magnitudes higher costs than e.g. Arduino and a breadboard kit. What exists of open biology is often tied up with drug culture, both the legal and illegal variety. The latter needs no explanation and the former is stuff like Four Thieves Vinegar that tries to break the FDA-endorsed establishment. It is an embarrassment to the field that what exists of amateur biology mostly stands in the shadow of illegitimacy.

The pioneer of Crispr application in humans has just been thrown into jail and vilified in the press. Most of the scientific community, while not publicly praising, are silently in agreement with the decision. The few condemnations comes from bioengineers, your classic biologists are more than happy to play the "wait and see" game. Nobody is willing to actually engage with society at large with regards to the issue. There is no Space X, there is no Google in biology. Your established pharmaceutical companies are more than happy to play gatekeeper and rent-seek instead of shattering regulatory capture. They are not interested in pushing back the state, they would rather play the revolving door game. How many "prominent" biologists have you heard of whom built a single company that achieved unicorn status? Look at the biology "entrepreneurs". Classic academia-private partnerships. There is no fresh blood, no vision, no Tesla. More public relations effort is being put into "immortality tech" companies like Ambrosia and Calico than existing state of the art technologies. Apparently it is more "publicly acceptable" to inject young people's blood into your veins to live longer than to perform an experimental procedure on infants to potentially cure HIV.

Your average university graduate with a degree in the sciences can finish their biology component through memorization. This is the way the current system is structured and it is not likely to change soon. The talent in biology that does not go to medical school is being lost to software engineering. It is easier to make money in ad-tech than trying to cure cancer. Why bother working in the field when "shipping" can take decades because of red tape and experiments on non-human mamallian animals are tightly-regulated till almost the point of uselessness? Agriculture gets more leeway with animals on an unimaginably large scale and magnitude than your average university lab. Lab-on-chips and organoids are great for press releases and glossy magazine covers, but they are far from production-quality. Even lab-grown meat is not production-quality yet. The mush you get isn't even worth putting into meatballs, forget lab-grown filet mignons. The current tech is essentially the same old rehash on plant protein, bean curd/tofu has been used to make vegetarian meat substitutes in places like Asia for centuries. Impossible food made it "hip". To translate it into HN-speak: Impossible Foods to vegetarian meat is what Docker is to BSD jails. The tech community is able to reason through the costs and benefits of backdoors in encryption, but when it comes to animal rights there is nothing but emotional-fueled hysteria about perceived abuse and mistreatment. The next time a family member suffers in hospital from an incurable disease, just remember the donation you made to the local animal rights campaign helped hinder biological research and development by decades. Let's hope the animals' "thoughts and prayers" are with you. After all, they certainly are conscious enough to benefit from legal protections and funding.

You lament about the state of public biological understanding? Talk to the powers that be. How many people believe in free energy and try to put it into practice? Perpetual motion machines are easily debunked. You don't see rampant exploitation of the ignorant and vulnerable by free energy believers on the same scale as the anti-vaccination movement. The same can't be said for biology. How many craft beer brewers actually understand what goes on inside? The next time you go to your local bar, ask around. That will give a pretty clear idea of what is the current level of biology education.

Edit: why the downvotes? If biology makes you uncomfortable, then take time to learn it. After all, isn't this the infamous HN response to a lot of problems? "Learn to code"?

sansnomme··on Ask HN: Why Shouldn’t I Use Lisp?
Grammarly?
sansnomme··on The FDA Announces Two More Antacid Recalls Due to Cancer Risk
If the temperature in your car is 140 Celsius, you have bigger problems than carcinogenic medicine.
sansnomme··on Deploy your side-projects at scale for basically nothing – Google Cloud Run
It goes up of course. Nothing's for free. Don't expect DigitalOcean level of pricing, the premise of cloud is on efficiency and scale, not necessarily being the cheapest raw computing provider when compared to traditional hosting. In other words it increases the elasticity of computation, but the price per unit of computation goes up. (Of course there are cost-savings like ops, security, uptime etc. But if you are talking about one-off side projects or something ML-related then the raw compute/bandwidth costs far outweighs everything else.) For PaaS the free tier is subsidized by your enterprise and other paying customers.
sansnomme··on YouTube Gives Us Love Without the Messiness
Because early death allows you to quantify the economic impacts. "Quality of life" isn't exactly a strong economic indicator.
sansnomme··on Rete: JavaScript framework for visual programming and creating node editor
See also: https://github.com/jagenjo/litegraph.js
sansnomme··on HTML attributes to improve your users' two factor authentication experience
For low-stakes auth or simply duplicate user prevention SMS is sufficient. E.g. dating apps, sharing economy services, e-commerce sites.
sansnomme··on A Croatian fisherman caught a new NASA device while fishing
Try using OpenCV on some images of stretched uncoiled rope and attempt to segment out the rope if you can. You will get the idea pretty quickly. Think ultrasound, that sort of thing.
sansnomme··on Gandi loses data, customers told to use their own backups
Replication is explicitly designed to correct issues associated with bitrot and add redundancy.
sansnomme··on Pursuing a Business I'll Love
And yet companies like Autodesk are pivoting towards subscription too. They are about as "hard engineering" as you can get, literally.
sansnomme··on Apple's auto layout and visual format language for JavaScript (2016)
Reminds me of https://gss.github.io/guides/ccss
sansnomme··on How scientists across three continents produced an Ebola vaccine
Were you allowed vein finder?
sansnomme··on Which Machines Do Computer Architects Admire? (2013)
There was a company called reconfigure.io that came the closest to an accessible HDL (they compiled Go to VHDL/Verilog) but seems to have died and their founder is now with ARM.
sansnomme··on Which Machines Do Computer Architects Admire? (2013)
That C like language is called "Verilog". (Yes I know it's a HDL but the point still stands. FPGAs are commodity these days.)
sansnomme··on Technology Will Soon Give Us Precise Control Over Our Brains and Genes?
Do you honestly want to get dopamine boosts from work? It is an easy question to think about when working as a code monkey can bring in 6 to 7 digits in compensation but what if your job is flipping burgers? Retail?
sansnomme··on Always Review Your Dependencies, AGPL Edition
Package managers need to automatically derive properties of end builds based on licenses. E.g. Eclipse License 2.0 without the presence of another more liberal license means it cannot be used in copyleft software, any dependency that is copyleft is also infectious etc. Of course it won't account for every single legal property but the basic checks should be done. To prevent work duplication a single binary/library written in Go/Rust can take care of this problem and be used across all package managers. It is in the interest of GitHub to sponsor such a project.
sansnomme··on Show HN: Blook – Helping US and Foreign entrepreneurs setup their U.S. company
It would be great if you can integrate features of clerky with international tax law/accounting advicr. US tax laws are notoriously complex. You won't be able to differentiate from e.g. Stripe Atlas with just the current set of features alone.
sansnomme··on Show HN: WeTube. Sync YouTube player across multiple devices
What's the stack? You should do a blog post!
sansnomme··on NHS gets £40M to cut login times on its IT systems
Instead of IBM, just use a Bloomberg terminal. It is as secure if not much more than all those security checklist-ticking healthcare software. For good or for bad, a lot more money is on the line for Bloomberg than e.g. a personal data breach in healthcare.
sansnomme··on Ask HN: What Technologies to Learn in 2020?
Sorry I meant Molecular Biology of the Cell. I think it's meant as a reference more than anything else. Bio is like ML in that it moves very quickly, so there is the need for constantly updated evergreen texts.
sansnomme··on Masterminds of Programming: Chuck Moore (2009)
Consider the current lawsuits regarding APIs, I think it's probably safer for wikimedia to defer to whatever the corporate PR department believes to be correct lest they be compelled to testify in court.
← PreviousPage 5 of 21Next →