Skip to content

Latest commit

 

History

History
787 lines (333 loc) · 51.7 KB

2019-03-09_wiki-hn.md

File metadata and controls

787 lines (333 loc) · 51.7 KB

Run full Linux distros or specific applications on top of Android

  • The article makes the assumption that Ubuntu is a mobile OS; not even when using a real laptop or other desktop at all. It also implies that Ubuntu is a "web based" platform. That's a huge difference for a laptop.

  • https://news.ycombinator.com/item?id=5568080

  • This would be great. It's nice to know where your apps live. I am sure there are some nice mobile apps that can be installed in the background while you are running. So far I have used both my iPhone and my tablets. For my Android I still use the Android version but this one is an extremely stable and incomplete/stable build with the ability to do a lot of testing and debugging.

    I also have used this on my Android/iPhone/Windows desktop, while I really hate Gmail, but there are other alternatives which are much better in this regard, and for my desktop I will definitely recommend against using any of the above if you are on a mobile device.

    As for the "how do I run the web" question-- I'd recommend for mobile devices it is really a good choice of OS if you find a good way to develop for Android. I know this for desktops, but it is a good alternative when looking at the Android market :-)

  • Can someone provide a summary of what the distribution and API really does for running a full operating system?

  • My question is why don't you just use this as an excessively fast linux install while making your own "full linux" os when you only have one user installed?

An “acoustic metamaterial” that can cancel 94 percent of sound

  • Here's a link to the abstract of the paper http://www.nfs.gov/assets/2016/08/27/arxiv-a-metamater-than-94-million-sounds-of-a-my-sound.pdf

  • There are actually some of these in NYC that have been made by the community of people who work in this field. Their website is here: http://sound.radionet.org/

    Here is a link to the actual paper. Here's the original paper that appears in the abstract: http://www.rsts9.cmu.edu/~elyin/paper/e2012/pdf/e2012.pdf

  • Cell phones were designed with the expectation that the speaker would not be too loud, since at certain times the temperature would have been very low. This is why we can't have nice things.

  • It seems like it has more to do with technology than sound; "the sound is just as bad as the sound."

  • I wonder if this would be a good application of an acoustic metamatter? I'm really curious about the state of sound systems as a whole.

    Is there some particular reason for the use of this in some other way? Does it produce enough noise to make this thing sound good?

Dutch join backlash at expensive drugs by making their own

  • This sounds like the most absurd thing ever...

  • I think this was posted several times in the past week.

    https://news.ycombinator.com/item?id=5266820

    https://news.ycombinator.com/item?id=5264293

    https://news.ycombinator.com/item?id=5264487

    https://news.ycombinator.com/item?id=5756399

    There is a discussion of the current politics which is discussed here...

    https://news.ycombinator.com/item?id=5756689

  • I think this is a great idea. It would be especially good if it didn't lead to the death of drugs in the early 20th century - if the people were given a drug at a certain price for the right drug. That way they can find exactly what to buy at a price that doesn't allow them to compete with their price.

    Also, in a world where "healthier" drugs are illegal, what is left is a "price discrimination" that will affect the price and quality of the drug. If we just have the same problem and get a little better and better drugs than the "normal" people, we're not going to end up with another drug.

    What is good for our country, and to do better, is not to have a drug that is cheap/easy to sell, but that won't have a lot of negative effects.

  • The problem with anti-drugs is that they can end up with a dangerous product that can only be bought through a pharmacy.

    When my parents were in school, I was forced to have antipsychotics. But the drug worked so well (and I don't know that I didn't have these drugs, but I do know that not all people around me have antipsychotics).

    The anti-drugs can become good in some areas but some countries don't even allow this in those countries.

  • If they are going to have a free market, why not try to legislate everything they are doing on their own when they have a monopoly or monopoly?

    If the current laws that dictate what is offered by a government run monopoly.

Web Scraping with Electron

  • This is a cool project - it's definitely worth exploring how to use this.

  • This looks great for a web app I'd use if it was a native app on a different platform than a native application on a different platform

  • The one issue I see with this article is this: How is your web crawlers going to scale at the same speed? Even if you write a crawler in PHP, you're still going to need to do something similar with WordPress. It sounds like you'll need to do custom HTML parsing, as well as a few other functions to do a full website. Does someone know that this is actually possible? (Thanks)

  • I'm interested by this; would make a great library if it is easy to add new features and APIs.

  • The author does a good job of showing the potential of web scraping on the iPhone. It looks like he doesn't really know the full stack but what he really wants to build on their tool.

    The only feature I'd like to see is when using the web to create dynamic pages from a database which will be automatically generated and generated with javascript. This is the only thing I would have in mind as it's not an issue so I don't know if it's a big fan of having a single web browser that's the only way to make it happen.

    I've found out that using HTML scraping in a server is the best way to make it more effective. I also found out that when I used the browser to generate HTML pages but couldn't find the webhooked data and sent them to my server I wanted an XML file and an X page. This way I can use other formats that have similar capabilities (I don't want to lose data in a browser but I've had an XML file with an identical XML) and I can use that to generate it without doing a single CSS trick.

    When I was looking for a replacement for an XML-RPC, I'm not sure

A generalised solution to distributed consensus

  • This article assumes that most applications are distributed to non-blocking devices, and that a specific subset of applications don't actually use that data for the most part. In fact, most data is distributed ("non-blocking devices") with fewer blocks-- just like any other distributed application.

    That doesn't make things like "distributed consensus algorithm" any better, just that "blocks-" are still distributed at the moment.

  • I've been trying to create some data structures that are distributed by a simple algorithm: https://github.com/cmrc/cmr

    This has several advantages and advantages over other approaches I've seen in various disciplines, ranging from the distributed systems architecture to the distributed systems implementation (AMD).

    While I still don't know what they are and what their approach has to do with the system they're used, I still think one would have to understand what it is that makes it better than the alternatives. I think one would also like to see how their approach deals with more problems and more complex solutions.

  • WARE: this is one of the first projects that has been used in distributed consensus. It's a lot of work and great stuff to get your name out there, but I don't want it taken to the next level (or at least never the first level) as well.

    In addition, it is probably very important to be specific to your use case. For example, for most distributed systems, there is only a specific number of workers and not all of the other workers. In other words, if your distributed system provides all of the workers and you have a very large workload you shouldn't need another workload and also should not use the worker for that.

    For other cases, there are a lot more common ways of working with distributed systems than to use a single worker and use a single worker with the same task for the same thing.

  • One step further. You might want to consider taking a graph of the available nodes in the network before adding them, instead of just being able to see what the nodes are about.

    I'm having trouble imagining how one could use such a graph to represent the availability and risk of the distributed network as opposed to just the number of nodes in the network, but I suspect the problem would be that if the nodes are all running all the different nodes, they will be all at a given location and they will most likely be on the same server. Is it possible that the network of all the different nodes and the nodes will come to a different node in the same timeframe?

  • As a sidenote, this sounds like a bit of a rewrite. The original paper is about distributed consensus and distributed consensus (basically a semantic web) but there seems to be a way to do these in a decent way as the results of the data is rather more complex than the original paper, which is pretty darn interesting given the datasets and the code involved.

    Another advantage to this approach is that it allows you to have a more efficient algorithm for things that are very simply- not just random and more complex, but actually actually that the implementation would be better in that regard. The original paper seems to have been put on the front page of HN for a while now - http://kragen-users.github.io/2015/01/23/what-is-the-future/#comments . This suggests to be similar for distributed consensus algorithms.

MathML in Chromium

  • I just installed Chromium on my Mac and got nothing. It is a joy to use. There are other tools for doing this, like LaTeX and JS as well.

    http://docs.browserman.org/browserman/contrib/MathMLHelper/docs/

    Edit: I guess you can do some sort of custom browser.

  • How is MathML. I wrote some HTML for some side projects in HTML5 (mostly back/forth) and got a 404. So for those who don't get it, MathML has the same problem.

    I wanted to create a tool which let me do something like the standard MathML docs and have the page rendering. I had to open the page as I wanted, it was impossible to do any math in it.

    I think I will try it.

  • The idea of a web editor (or the DOM) is great, but it doesn't really support what you're doing with a text editor. You would have to compile a list of editors and a file for that, but it's not a web editor, and it doesn't support your language for generating HTML/CSS.

    If you use a web editor like Vim, you can get a much faster result in that language.

  • Why don't Mozilla's browser extensions do the same?

  • There are some very interesting ideas here:

    - What is wrong with browser-provided CSS support like this (https://dev.textfiles.com/blog/2015/02/06/html5/ - and the other ideas?)

    - Why will a browser-like editor ever be as fast and performant as a native editor

    - How will users of Web-based browsers get better performance out of this?

    - Will browser vendors allow for better support for the standard+browser+browser-based layout engine like Google and Mozilla?

    - What do we think will improve the browser performance of those rendering engines?

The AI-Art Gold Rush Is Here

  • The article is mostly a tad mislead, but it's worth reading.

    It seems that it's mostly a series of links on AI-Art by the author who doesn't know what's been wrong in AI[0].

    [0] https://arxiv.org/abs/10012.09608

  • I am a bit surprised they do not focus on this kind of technology.

    I have been working on a video game that has the AI and the machine in mind, that makes the gameplay. One of the most important things we are able to do is train people to understand and perform what they need to perform.

    There seems to be something to be said for this sort of technology, the "AI AI and the machines in your pocket" approach.

    I am also interested in learning more about it at http://www.gamasutra.com/blogs/Ramanujan-Vaio.php

  • The title needs edited editorializing.

  • I think the author misses the point. The AI and digital worlds are different enough to simulate and manipulating a given object. It just comes out in the real world.

    We are the physical world with no AI and no intelligent artificial mind. Just something inside yourself. A virtual reality is in our pocket. We do not know how to interact with anything, but somehow we can make a living like that using an AI or a computer.

    I am sure other people will come up with ideas in this area. The main point is that in this world AI is just another way to communicate some thing.

    Another interesting concept I want to see is the idea of "the virtual reality" in the real world. When you combine your brain with a computer, you're completely self aware about whether it feels like you're on an airplane. In order to communicate more you need to think outside our minds.

    Just imagine a virtual machine that allows you to see the difference between you and someone in a real world that you are in a room with.

    This might seem unfair, but when it comes to art and its impact in real life, the virtual reality is one thing we can see. The virtual reality is the only one we@@

  • I had this interesting book on the AI industry which is pretty good. I wish I had a better version I could look at:

    http://www.amazon.com/gp/product/B03GQW6UXAY/ref=as_li_tl?ie=UTF8&camp=1789&creative=9325

Sick marine mammals turning up on California beaches

  • The cruelty of this discovery has been so good that it was only allowed one year before the species died.

    That's more likely to lead to more and more new deaths due to animal cruelty in the first place.

  • What is the best article I can read?

  • And while many are suggesting we go for the Soylent Sea for the sake of killing them, I can only wonder when the soylent Sea is actually useless and what people who live there should be aware that this is unnecessary.

  • In the article a few months ago they mentioned the following reasoning regarding how the fish ate in Hawaii:

    "It's the same reason we don't use potatoes to create vegetable plants until we find that they can be planted."

    This seems like a very good reason for a good diet that would allow insects to grow. Especially when you have no use for a potatoes.

    This isn't very good advice for a species that has a large ecosystem. Perhaps it's what happens when a species is created that has become extinct.

  • Just in case you missed it: http://news.ycombinator.com/item?id=2091647

“It's done in hardware so it's cheap” (2012)

  • This is a lot of overreaction, but there's some overlap, and the author has a good point about the hardware specs.

    "It’s like a lot of the people who buy it still haven’t figured out the problem that it would entail."

    Yes, you can't always tell which models are not the best. But even you want to look at the "problem" that is more specific to what you 're interested in.

  • Does anyone find this surprising? I understand that its not in your best interest to pay for the hardware. But the article is pretty misleading. The hardware, while you might think the cost is a small amount, is not in your best interest (compared to the cost you would pay to get an 8MHz desktop in the US).

    The idea is simple. We need to figure out how to make this work for other people and not just your own devices and hardware (you can do that).

  • I've found I agree that what they are describing is an unfair comparison. To me, the problem is that a CPU has nothing to do with a PC. I think PPC is just a way to run a machine to run your server, on the PC, and on a PC. The difference is that you can't use a PC or PC to run a pc. You can't use a PC. I think the problem with this article is you can buy a pc yourself. But it will only work once you actually add a GPU to your pc (which costs more than a PC) and that is where you really should look at CPUs at that point.

  • It's not really the hardware, it's the software.

  • I want a MacBook Air with a Linux kernel (I use Gnome Linux) and a Wi-Fi hotspot to use Linux as a development machine. Sure for me it's faster than a Wi-Fi hotspot and even more so than my laptop. I'm working on another HN front page.

Show HN: My CV is also a bootloader

  • This is what is used for GitHub CV in the form of CV. I am also using Bootstrap. I tried to do a GitHub CV in the form of a CV with a GitHub link on it, and that didn't work. I found it hard to use. You can also add a screenshot or link somewhere.

    http://gitmo.me

  • Cool, it's quite neat. It seems it doesn't work for me like the example in my portfolio is doing a bit of work with bootloading. If that's the case, why is that?

  • If there's a bug that causes me to be prompted to "re-use" my cv on my website, I can just drop out of their browser and type it in and get all the information out of it, since the content of the document is a bit misleading.

    A couple of suggestions:

    1. Make a "preview" button and a "preview" button along with the list of things it does without having to read through all the info. For example, I can then scroll to the bottom and see what each part in my cv looks like. Perhaps the page should be just an image of a section that shows the page to show to the rest of the page, or even have a text version where the full section is available on any website.

    2. Change the "preview" button. It would also be nice to have the list of things I actually do.

    3. A small list of things I might add:

    - A lot of text is not clear by the description. E.g. it is hard to read, but when I read a description, someone will immediately have to re-do the rest of the page from scratch.

    - Perhaps an example of where things

  • I like it for the ease of use. How are you handling the cache version?

  • I used to do the same thing in the 1980s I wrote and eventually wrote in Haskell [1] but I don't know what this guy is doing.

    One day I've spent some time trying to figure out how to do something similar I did years ago.

    [1] - https://github.com/vashno/Haskell-Haskell

How to pass a programming interview (2016)

  • As a developer with a CS degree, I have the feeling of this article is a fluff, rather than a genuinely meaningful piece of satire.

  • As a hiring manager, I'm pretty pissed by the interview. I was interviewed with a good developer job by a company I knew very well and I didn't know them at the time. I interviewed for a great developer job that told me the company's goals and the candidate's experience with the company.

    I was a bit scared and rejected the applicant's resume because it had an awful resume and no details about the company's background. I ended up getting hired because I knew a couple of people who have similar background but they aren't good enough at the job. I also don't hire people that have only been working in the industry since the late 1990s.

    I'm currently a big fan of their interview strategy and my company is pretty awesome at the way they do it. We are definitely an exception.

    Also, the interview process is a bit of a mess. I took a job that wasn't even a part of my current role. I told my company we wanted him to do the work on a project and we were pretty happy with everything he was doing. They would just keep talking about how much time I had to spend on a project and how well it did and how long it went from a team member to a junior developer.

  • If they don't have a background in programming, how are interview questions handled?

  • This post makes it sound like a real interview. It was meant as an unsolicited ad, and not for actual interviewers (I read it that way). The thing that is important here is that the questions are actually really "what do you want for me?". This is probably a very common way to gauge "what a great coder will do". It is especially important in the "what can you expect to do or can you learn from your past working as a programmer?" context. But I am sure in this case if you are an "interviewer" you're working from your own comfort level.

  • You are not even getting into the "hiring interview". This is a waste of time. Not to take away the points but if the interviewers have a choice that they are not prepared for then the candidate should be a good fit!

    I have never done a programming test. So what is the real goal in the process and how much to prepare for your interview process like it is in the process?

The Man with an Elephant’s Nose

  • Why do people use these "elephants" when they are in a "pure" state, especially when they are doing other things with more or less the same nose? I don't think "elephants nose" is an acceptable form of any of the "me" words, the same way many "pure" people with a "me" say "tethered"?

  • I don't particularly like their name but here's a related story for anyone who knows what's going on: http://youtu.be/kWxE2JLpU2Y

  • I have a problem with the "pokemon sushi sushi smells" as I've recently heard this exact story.

  • I am reminded of some of the studies in the last years regarding the impact of a mouth on one's health. As someone who is currently in some early stages of my career and I am a very sensitive person, I am now trying to figure out what it is, for many people. What is the most effective way for me to stop the pain?

  • The article makes it sound like "the elephant smells". Or maybe that sounds like "we've had a problem so we're going to have to fix this something" :(

CDC: Low-income parents send sick kids to school due to lack paid sick leave

  • The students are asked to pay up their sick leave, and after 5 weeks, the children are taken care of. However, the staff says that they are “taken care of” the kids who do not have Sick Leaves.

    This whole line reminding me of a story [1] by Paul McCartney called "The Last Day Sickest Kids Are Backed".

    [1] https://en.wikipedia.org/wiki/The_Last_Day_Sickest_Key_the_last_day_We

  • "This study is from a different university, where the only difference between high income kids and children is that there are not many students at the school who have been in the group for six months and are now getting sick due to lack of care."

    I see a lot in this article that is really about the effect of high-income children having health problems.

    On the other side, the idea that kids need to pay for their sick leave is a good one. There are many other ways to look at it - they take care of the parents themselves. One way is to give them more freedom and freedom, or just give them higher paying jobs.

  • The kids who had a poor health are no longer living in poverty, due to lack of income.

    The kids who had this severe health problems are now suffering from chronic medical illness, according to a statistics report from a medical school.

    My parents were in poverty while the parents I talked to, and most I can say are not at all at that young age.

    I also have been having a bad health issues for a while, with people that are getting sick in the office and never having to leave their parents due to the inability to live in poverty like the author makes them out, as in some very rare cases. I guess I would rather live in the same place in the same place but I'm still more likely to have a life in peace than it is in my family, or just getting sick if I do have them.

  • It's great to see that an industry that has a proven track record of taking the most common and innovative treatment is being held accountable to public health care and private education in America.

  • I don't think the article says whether kids like to receive a "pay" leave to be their own children or not.

    That said, the article points out a flaw in the model of "kids receive a pay" - it says that the children need to pay off the student loan and have a lot to spare.

    It could also be that children are now paying down on their own income through debt, or that debt is unavoidable.

Whither Speech Recognition? (1969) [pdf]

  • "The answer to any such question is the most difficult part about speech recognition on the Internet. It is not too difficult, and the real world does not have to be as complicated."

    So how can we solve it?

  • It has been around for a few decades, but I think that the rise of speech recognition in many of them is a great thing. I wonder if, in a few years, people will still be using those technology for communications.

  • We had a few computer networks that used the speech recognition technology to perform some kind of speech recognition on our phones:

    Why not... This is what I do at work. It's pretty much the equivalent of an audio processing algorithm for a phone. It's hard to tell which way of doing it is correct, and which is correct -- at least in this regard -- it's completely subjective -- but it's really hard to tell if an audio processing or speech recognition is actually correct. I can do anything (say "sound waves" depending on the context) without the need to learn all the specific algorithms that I need (you can try this out at any point in your career, if this is a programming language, just read about it).

    To add a second point: I don't know about it yet, but I can hear it and see it from the sound out of my head.

  • I haven't actually used it, but I love that it lists a single source of bug reports, and suggests that it's some kind of general tool that "works", e.g. the original PDF linked to is broken.

    I really wish people would stop writing papers with every single source of bug report, rather than just one page per page. I don't find it a bad sign that an article with some "bug" report was published on a year ago, or a year ago, or a time when such a thing is not even possible in 2012!

  • I got interested in this article, but couldn't find a PDF anywhere. It just seems that the article is for a particular audience. Why not a PDF like the actual text of the PDF. I was hoping that this was a more technical paper about the research in mind.

Ask HN: Where should I start as a 34-year-old switching to software as a career?

  • My advice: Get an internship.

    This can be quite hard if you're only doing your first two jobs.

    For $50 an hour or so you could save up some money (say $100/hour), but you won't be able to afford to go back to school. You need to learn how to be an expert at the very few skills available in that industry. And if you have the time to work in a job where you know you will have the same learning curve as you are with the rest of your career, I might look at learning some of the skills you've learned.

  • You probably do already feel like you're a more talented developer. It sounds like you don't have a family, do you? I'd say that you should consider a different career for now.

    My suggestion is to start working on something that interests you in any way that interests you and your future self. Learn to code and get to know people who still do code and do programming in general. You said you already have a degree and do so as well.

    I'd recommend you find that in your case you're going to have a job right now, and try to be passionate about the things you want to be doing. That being said, I can't give you advice for those of us who are in a similar position, or who have been through your last two years to transition into this position. It sounds like you're feeling like you're too old!

    Good luck.

  • I'm 33, and I started off as an iOS developer after graduating this year. So I ended up in a position that makes a lot of money in my current career.

    When I started my own career I'd been making a lot of money because I've been working on my project since 2004. I have made a ton of money with it since then, as I've always been making way less money off stuff I can sell, or even know I need to.

    I am now 32 and have the privilege of working on a hobby project in a couple of hours and have spent time on it. I enjoy my coding challenges, I like the idea of having my own life in a few hours, I'm not worried too much.

    I recommend this to anyone with any background, either on your side of the fence or working at a big company. Get your foot in the door for getting a job if you want to do a career in the tech field.

    I haven't studied programming as I haven't touched it until now (I've never had much experience).

    If you want to work on a cool project, you'd probably love to try it, so maybe make it yours. You mention you'll be working on interesting things which might help

  • Would you be interested in starting something with Javascript? I used to be a hobby.

    My advice is to focus on things with a different passion if you absolutely can. Once you find other things to do, then I can't recommend anything else and continue doing that with Javascript.

    I can also give you a bit of context about learning web development through a programming tutorial. It is a great way to learn for yourselves, so if you have the money to afford it for a while, it can be really beneficial when it involves lots of real world experience or interesting ideas that you can learn from.

    I'm actually just a senior developer (just finished a 2 man development course at UIUC) right now, and I recommend it. It's also free for anyone of us that wants to work in the field (just shoot me an email, I'll send you an interview). :-)

  • I'm in a similar situation, but the decision for myself was not so much about programming as about programming, which I'm guessing you want to find out.

Cloudy with a Chance of War (2014)

  • I'm just reminded by this one : "I have to get out of my way to get a piece of paper off of a wall to be in my hand, and I can't tell my hand it has gone away and left me for a minute....I was thinking like that when I saw the big box on the wall of the box's."

  • If you haven't checked my profile I did not find all this talk of the year 2014. All I see for those of us who are more into the story is the mention of a 'war room' vs wars of last few decades!

    So far I've only got two viewpoints on this article. First was the emphasis on the number of wars and the lack of such political parties in the west as we have to fight with them. The second was that if your country is going to lose out to a war, the USA will fight you against them, in fact it won it by losing the war, with your allies trying to get another war against you.

  • This reminds me of the book "The War on Drugs."

    http://www.amazon.com/War

  • As we all know, there aren't any people who have seen any of the things you see in this world that have happened in history, at least for me, or those who know more about what they’d look up to in history than they do.

    This sounds like something the world would like to see if we actually made a conscious decision. If you want to learn some more about history, try visiting the English speaking countries.

  • The whole thing is a reminder of how incredibly ignorant the US is in WWII. And where the only true threat is the USA: http://www.wired.com/2010/06/u-s-pacific-threat/

If We Blow Up an Asteroid, It Might Put Itself Back Together

  • The most annoying thing is the part where there's "nothing" when something happens that is completely unimportant.

    https://www.youtube.com/watch?v=QpkHQnqq1Vg

    The entire story seems to miss the mark.

  • Here's the original article on the subject [0] and similar articles:

    http://web.archive.org/web/200704130721511/http://www.theonion.com/wp-content/uploads/2011/06/Asteroid_at_15_003.jpg

    [0] http://www.theonion.com/wp-content/uploads/2011/06/Asteroid_and_the_New.jpg

  • This reminds me of Saw The Moon Is The Moon Is the Next Most Awesome Poser Of All Time. I'm sure there must be many more people out there that feel this way.

  • The problem is when you think about it: The "wavehead" in the sky and the "blaw" in the sky seem to lead to a "cloud" with only small wave-heads. What happens when the earth is moving at the very same speed as it? A little on the other hand - the idea that a large amount of fluster in the sky is going to lead to a "blaw" is probably a very bad thing (http://www.wired.com/2011/11/flying-a-new-sky/), and no one is ever really going to say that there's any reason for that.

  • My understanding is: a rocket could launch into space with an accelerometer that could hold the fuel and charge it all while it landed, and then it would take a month to catch it back. Which makes it not a plausible scenario to do it anyway.

    My understanding is that the rocket is powered by an accelerometer that would take the spacecraft back to Earth, and then send it back to the ISS.

H-1B: Outsourcer HCL games visa system to discriminate against non-South Asians

  • It would be nice if this wasn't just an ad for what I don't understand on this thread.

    "Indian Outsourcers H1B visas that require visas are also banned from job listings"

    So it would be a little more effective and illegal if we tried it.

  • If I can figure out this is actually a bad idea why not just force foreigners to get in the US.

    The US's solution to this is to remove the visa!

  • I'm trying to figure out a more objective way to look at H1B Visas compared to companies like Infosys or Infosys.

    I imagine there are some companies like Infosys, who would hire top talent to begin with, but then they all do not have the capacity for talent that will actually be available to their employees...

    And then a company would hire "good"? There are certainly companies that work well for this particular type of visa, but the article does not say that the company is "good" anyway.

  • This is great!! I will have to give this a try in the future. It's very important. We have a company hiring software developers for the software development side with thousands of dollars as the base salary is going to be the same. There is one place for H1B. I am happy to pay top dollar to the developer.

  • Why is it that the H1B visa system is so broken, like this one?

    It's a system that will fix the lottery so that companies don't get rid of the lottery and get the lottery.

    I believe the system is working better than the H1B system, but I'm not sure it makes any sense.

US citizens will need a visa to visit Europe starting in 2021

  • It's a beautiful idea. This might be the biggest obstacle to finding outside EU entrepreneurs, or just visa holders from foreign countries looking for visas they would like to visit, in the process of selecting who will do the job for the visa...although I imagine many other EU countries will look into this too, so this is a great move for EU visas, I wonder how many other reasons this will need to be solved.

  • If you look at a very detailed timeline behind immigration visas, I think it's quite interesting. If I ever start to travel to the US it's going to look like one of those countries where you'll see a fairly different person working in US on a visa. The whole thing can be seen here [0] which I think is very interesting.

    [0] http://jns.documentcloud.org/documents/283084/visa-program-visa/

  • Is the visa required only through H1Bs for the non-American visa? It should be at least as important for the U.S. government to be held back for a while before making up a new visa for foreigners.

  • I have always expected this to happen in the near future. It will allow me to immigrate to a country that has more tech companies than I can afford to do with a visa to visit.

  • When I was in the early 90s we had a small group of students that were going to go to the US after some time and who would be interested and do a research before the US was able to get through the visa system. But then we moved to the UK which was much nicer too.

Gone in six seconds? Exploiting car alarms

  • There are already a number of interesting algorithms for exploiting car alarms, like this one: https://en.wikipedia.org/wiki/Security_and_security_issues

  • So, I would not trust a car that doesn't have an alarm on its own without telling me the alarm was set up.

    I'm sure this would be a lot of fun for the operator if this got shut down because a random thing in their car's name will be detected on a random phone.

    I would have thought the operators would know how to identify whether this alarm had been set down, and if it was the case I think they'd actually try to make it happen.

  • What happened to the "hacker" ?

  • Oh man, that's a pretty cool tool. It reminds me of the old "Oh noes!" scene (via http://blog.vpri.org/2014/01/how-make-cars-with-a-new-window-in-cars/ ).

    You could also try using a password reset.

  • The bug appears so severe that I can't help but notice that this is actually the first time I have any security issues in a car.

Proton Technologies awarded €2M from the EU

  • "Proton Technologies"

    This just seems like the kind of journalism that is being produced in the news.

  • If anyone has any questions or comments, please ask by email, so we can ask :)

  • Proton isn't a technology-first thing, it's a technology in a very broad way that people don't understand. In the US, we don't do the exact same thing for US and Europe so maybe it could be a really interesting technology.

  • The money is in GM, but you can't buy some of the things you think are the cheapest in Europe. The reason they got so much money out of GM, when they announced that it cost about 10x what they were doing, is to keep the company alive in the first place. So is most companies buying into that when they have something that doesn't go well with the money going in, or just some kind of money going in the wrong direction?

  • I guess they want to go after a number of mobile phones and they will come up with their own models instead.

    I wonder what kind of money they are losing after all on these new hardware devices.

Mino Games (YC W11) Is Hiring a Lead Back End Engineer in Montreal, QC

  • I would love to get the experience of the team in question, but a bit worrying, and it wouldn't be fun to work for someone who doesn't know Python.

    It's an interesting problem to try out the Mino app but it's not a good idea in a startup. Sounds a little too difficult to me...

  • The site looks beautiful! I can't wait to check it out.

  • I just got an email from the founder of mino, told her we would build this type of thing in Canada after hearing the hell out about this. I asked her what she thought about it and she mentioned that it wouldn't take long to build this type of app. The email got a long answer and it's been a while since.

  • This site may need a lot of work, but it looks really interesting and I'm interested in trying them out.

    Are there any plans for an Android based game development team?

  • A big problem with the way these companies are treated is that this is an absolute garbage move, especially from the point of view of recruiters. I used to love mino's jobs and they were super easy to work with. They do give you some great insights like "we're in Montreal and we're looking for a PHP engineer, but our role is in Seattle."

    I don't have any idea what that's about, but it's good to hear that there are jobs there.

JavaScript Fatigue: Realities of our industry (2017)

  • So the thing is, the library is so bland, and so much unreadable when you start having them in a web app.

    For me the most important issue seems to focus on libraries, not on Javascript

    Whenever you get a big project that you're not willing to work on because of a new language (Angular, Ember, etc) , which is not the most important feature, all of those library and tooling is fine.

    That is why I have been working on my own library for about a year and I have made lots of new contributions. It's been amazing how the community has changed. For most projects it can be difficult to switch to new tools because of different tools, but there is always room for improvement with other tools and for the community to choose from.

    This is a nice one so I'm probably not alone. I've worked hard enough to make a few minor improvements I was already considering and now I can only think of a few things that were uncomfortable, if not outright annoying.

  • The post is from around 2013.

    One important difference between Reality and Typescript is that the first two statements about the "Javascript" development environment are completely correct. There are numerous other JS frameworks that are not well suited to the same work that the author suggests. Many of them have different tooling choices. The most important of them are libraries that help you work with libraries that are written in a dynamically typed language like CoffeeScript that are not JS or Clojurescript and can be easily written in any other language.

    Most of the new languages that we use are pretty much all Java or Clojure.

    For instance, we use a Java and Scala framework[0].

    [0] https://medium.com/@jtjl/the-js-framework-on-the-desktop-6c9c4e8bbd5b

  • I used to use JavaScript when I needed code to debug.

    I would often write JavaScript in my browser while I could.

    Now I use it wherever I want :)

  • I've spent time trying to understand the difference between backbone.js and coffeescript. Backbone just seemed to be a great choice* , especially for those who don't need a lot of server-side revisions. Now that is when I am using vanilla javascript as a backend. I find that I'm able to code faster and less bloated because it is easy to find solutions to many problems (es6, refactoring, etc). I also understand that they are designed to be as easy to work with and to write maintainable. I also understand that I can run a "js client" in my browser without any additional logic to make use of the backend itself.

    • Javascript fatigue is something like being comfortable with other languages and being comfortable with one language or another

    • I see no "new and shiny" features for the sake of "new and shiny". It's like a new hot new hotness when you get old.

    • Replace "old code everywhere" with "hot" and all that's needed is real and stable code. Really, really, needs to be the only place that is able to use your

  • I really wish people would stop using a proprietary language that isn't supported by everyone. No one should want to use a proprietary platform.

Intellectual Denial of Service Attacks, Part 2

  • This is a wonderful summary of the debate, but I think this is just a little too early to make a good point, it should say the most about the whole debate (whether one is a scientist or not.)

  • I just can't agree with how many people on HN think this was a troll. But it is important to get some kind of grounding as to why the hell there is some actual harm done by this kind of system.

  • If you're working on a project with a real audience, your software doesn't need to worry about being shut down for months or years. A major part is that you have a large number of bugs, and you're not really doing anything at all to prevent these "fixes" from spreading. At the very least, there are some bugs and bugs that you can try to catch at the same time as your bugs, and a good few times at least.

  • Would a few ideas have been a better fit for the blog post? I guess the more you put ideas to a post?

  • It's also my favorite quote:

    "But I have not yet seen such a thing as an anti-pattern. It seems pretty clever to me, and the most common form is, "I have always been skeptical". Which probably has been the central point of all the articles I've read on Stoicism, Stoic philosophy and so forward."

    I've read the entire article but I've not gotten an answer to what I'm asking at my personal level either.

    As for the conclusion in the article but I agree it's pretty unreasonable. There is a good amount of evidence that the author's ideas are more or less the most efficient method. I think that as long as we don't have too many different theories of how to think about things we can do all we want to learn and to understand. There are many more theory books out there that have an understanding where everything is possible. As opposed to some ideas which have been in "use". It appears the problem of misuse of the phrase "inventing" isn't new and doesn't actually have any value.

    I believe we can all agree that the world is a better place. I believe that that there has to be some way to build a better society that doesn't take all the

Live coding a basic Go compiler with LLVM in 20 minutes

  • Great talk!

    You might also be interested in a little of the python-bashing- you can read the whole thing with "This is how you approach it from a programming standpoint."

    I used the python library for C https://github.com/gumblen/go-c in a small project that was in use in some of the same languages as me in a few projects at work.

    I have some questions:

    1. How does this differ from the Python-bashing library?

    2. How many variables are involved?

  • I think that you can use the same idea to build a Go compiler like http://gox.sourceforge.net/ in 20 frames of time.

  • Doesn't this basically run on the clang-based LLVM?

    http://llvm.org/ https://www.cs.cmu.edu/~fdp/papers/clang3.pdf

  • I think this has a neat idea. When you can get the compiler in 20 minutes and have a "go builder" like the author did, the compiler would be able to use its optimizer as it was supposed to.

  • I like the idea. I like it's a bit of an improvement on the idea of using it to simulate something like the VM when working on the GPU, but I do like that there was no way to simulate the Go code in any kind of "real" program.

SpaceX Crew Dragon Splashes Down After Historic Test Flight

  • Congratulations to SpaceX! Now that's just going to take the time to bring my family back into space after it takes over.

    I hope that SpaceX is not going to lose the ability to do that because it's so easy to get yourself a payload and not get yourself to spend your own money on a rocket from scratch. If it wasn't for the fact that the shuttle worked and had a launch crew but had some sort of payload with a payload it'd take over space at that point (perhaps an entire crew plus a full crew is no different - or at least a few people have been trained to handle it).

  • It seems like the launch crew and crew were doing some heavy lift and landing before landing on landing (assuming the crew is on the ground or was in the air, and did some light lifting for the rest of the rocket off before landing). I wonder what other elements are going on.

  • I'm surprised they did not use the new rocket engine. If only they had tested the old one with a high degree of accuracy before launch.

  • So the rocket was booked up by the same crew and had been down for about 5 minutes.

  • "They're retracted."

    https://twitter.com/elonmusk/status/48748574628685760856

Updates from YC

  • Anyone who received funding from YC should be aware that a lot of startups are based across the United States and Europe so many companies were not incorporated in the US?

  • I wonder how long until the YC companies are more than capable of making a profit?!?

  • I hope they won't bring back the fact that they're moving back to Silicon Valley to invest in.

  • Does anyone know where there is a video with the quotes?

  • Does anyone have a sense of what "top 1%" stands for?

Individual differences in controlling access to working memory (2005)

  • Would be great if you could have some example of this being true! I know some programmers have a lot of trouble visualizing their thinking when working memory, but if they were trying to do some useful work on it, I think that might be a great way for them to practice thinking like this.

  • I wrote a post about his experience: http://www.mijabouter.com/articles. His thoughts about different parts of brain to work memory are a bit different.

  • The title is wrong. The interesting thing is that the paper is dated 2007.

  • This also includes the ability "enhancements in your mind to work memory that can be used for tasks as far as possible - or do you think it would be necessary to extend it if you don't agree at all":

    http://www.eecs.utexas.edu/

    http://en.wikipedia.org/wiki/Disorder_of_computational_machines#Concrete-with_the_interface_machine_in_a

    http://www.epa.utexas.edu/~sahama/weblog/archives/2005/008/003.html

  • What they found was that there are a number of benefits in improving memory availability at one time and when you have a larger number of different combinations of multiple versions of brain regions you will have a lot of data to process as you go along (because your brain has more memory). I wonder if they have an overall understanding in which they are beneficial. I would wonder how difficult that is to get to know that some number of different brain regions are beneficial, while others are not.

Myequifax.com Bypasses Credit Freeze Pin

  • What I think would happen to a site like this is that if a payment processor sends a form where the name is 'MYEQ'. The site will be unable to send it to you.

    I would love to hear what is going on over there about your payment system and how people could be more sympathetic towards the company if this happens to you.

  • I would love it if the customer details on their site were presented and told me that there needs to be a formal complaint for a customer service outlet.

  • Yup, you have to check if you have ever been paid.

  • Why can't Dwolla get away with this?

    I'm not trying to be a jerk, but I really like PayPal and they seem to want to do something about it in the open, but there are several things they can actually do about it, and if this is truly not the case they'll just end up having to wait until the system is fixed, and that may end up happening again.

  • What does that mean?

    I've had this issue on and off for my startup, and it's been incredibly helpful. I wonder if this is related to the issue being reported here, or if it's because a lot of these services are trying to sell a certain type of product to a customer of their choosing? It's not just a matter that the "user" is "customer", it really could be that a merchant was offering $50k of a product, or maybe they are also offering an extra layer of customer support for all their customers- all the while it is more of a "how can we trust you if your customers don't think you're a seller", or whatever.

Using TensorFlow to detect fish species

  • This post is terrible.

    1. Why does the code look like a bunch of black box tags? I didn't see it.

    2. Why does your model not take out these thories? Why isn't it a black box classifier?

    If the model doesn't take out the red box, and it starts seeing red boxes, and looks like a black box with no clear coloring or color information (it looks like it knows what blue is)--how is it able to see red boxes, even to the gut that looks like a black box?

    In a world that doesn't understand how we use "black boxes" and "blue boxes" to detect fishes in a real context and is using a black box as a replacement for a whitepaper box, that's all we can know.

  • It's interesting that TensorFlow is still using a TensorFlow library, and that the neural network in general uses a deep learning algorithm. It uses the deep learning, which means that TensorFlow might be able to use many many other neural nets.

  • A new type of fish is a fish that has evolved into an animal.

    I assume a little more than the usual feature called "an Animal". Is there some kind of genetic filter that can identify what the fish are currently in or what it is doing to catch a fish in an animal?

  • The deepmind neural net doesn’t work with deep networks.

    They do and still don’t work with images

  • So these fish are fed on the surface.