Media

  1. Jan 29

    Both parts of this statement, that this has been addressed and that State Farm does not "engage in this sort of activity", appear to be false? Elsewhere in the HN thread, other people note that the exact same thing has happened to them.

    https://news.ycombinator.com/item?id=22180379
  2. Jan 2
    Replying to

    Cool! Am I hitting some cached CDN assets that haven't expired yet or is the page now 39MB?

    Waterfall plot of page load. Some assets seem much smaller, others seem to be roughly the same size as before.
  3. 30 Dec 2019
    Replying to

    Google Design's "Best of 2019" page is 64MB, takes 98s to load the "2019" that's the main visual element above the fold (from a Cable modem in Australia).

    Waterfall graph of page loading. The giant above the fold gif is the first thing to start loading, but it's overwhelmed by the other huge gifs that load in parallel.
    Timeline of page updates, it's 98 seconds before the above the fold gif goes from showing the top half of an outline of "2019" to all of an outline "2019".
  4. 15 Dec 2019

    Interesting indirect argument for pair programming

  5. 10 Dec 2019

    Huh. I guess this is why it was so much work to maintain a working Octopress/Jekyll install back when I used Octopress (speaking as a non-Ruby dev who didn't maintain a Ruby dev environment and relied on system packages as much as possible)?

    Twitter's alt text limit is too short to quote the first entire sub-bullet of this with any context, see https://lobste.rs/s/6ame3m/developers_shouldn_t_distribute_their#c_ozfhi8 for the comment I've quoted and https://lobste.rs/s/6ame3m/developers_shouldn_t_distribute_their for the entire set of threads.
    Show this thread
  6. 8 Dec 2019

    BTW, if you want to try CPUID spoofing without virtualization and have a VIA processor, Agner Fog wrote this little utility: Performance delta on benchmarks varies, here's an example of a ~50% gain (47%):

    Score with normal CPUID: 1845
Scored with spoofed CPUID: 2721

(higher is better)
    Show this thread
  7. 7 Dec 2019

    This kind of thing is why the CPU startup I worked for allowed users to change the CPUID: you get huge performance gains from putting GenuineIntel in CPUID(0), but it would be a copyright violation to distribute our CPU with GenuineIntel in the CPUID

    ... The short story is that Intel checks for "Genuine Intel" CPU's when it's numerical library MKL starts executing code. If it find an Intel CPU then it will follow an optimal code path for maximum performance on hardware. If it finds and AMD processor is takes a code path that only optimizes to the old (ancient) SSE2 instruction level ...
    Show this thread
  8. 6 Dec 2019
    Replying to

    A system for coordinating U.S. nuclear weapons moved off of 8-inch floppies earlier this year, 2 years late. Coincidentally, the original deadline was around when Google Cloud's product roadmap PM said that 1 year deprecation should be enough for anyone.

  9. 3 Dec 2019
    Replying to
  10. 20 Nov 2019
    Replying to

    Ok, I read the rebuttal in detail and it's as bizarre as the original paper. Just for example, consider the screenshotted quote. What could "We were the first to correct our work immediately after noticing this very issue" even mean? This makes no sense.

  11. 20 Nov 2019
    Replying to

    Yeah, it's better, but IMO the methodology is fundamentally bogus. But even if the methodology could work, the errors in the paper are bizarre. Confusing memory safety and implicit coercion, statements like the screenshot, etc. None of the 4 authors noticed or knew better?

    "Advocates of dynamic typing may argue that rather than spend a lot of time correcting annoying static type errors arising from sound, conservative static type checking algorithms in compilers, it’s better to rely on strong dynamic typing to catch errors as and when they arise.", see https://danluu.com/empirical-pl/ for more.
  12. 20 Nov 2019

    How secure is hardware disk encryption? "In order to to recover the data from a locked MX100 drive, we connect a JTAG debugging device. Then, we use it to modify the password validation routine in RAM so that it always validates successfully"

    Show this thread
  13. 2 Oct 2019

    One thing I think is interesting is that increased scale doesn't seem to help here. Despite Zuck's bluster, I get about the same amount of spam in FB as I get in Twitter and Gmail sends me more spam than either despite probably investing much more effort in spam prevention.

    ". . . It’s why Twitter can’t do as good of a job as we can. I mean, they face, qualitatively, the same types of issues. But they can’t put in the investment. Our investment on safety is bigger than the whole revenue of their company. [laughter] And yeah, we’re operating on a bigger scale, but it’s not like they face qualitatively different questions. They have all the same types of issues that we do."
    Show this thread
  14. 13 Sep 2019

    There are so many smoking guns I don't even know which one is my favorite. Maybe the time that Steve Jobs (CEO of Apple) got Eric Schmidt (CEO of Google) to fire a recruiter who accidentally violated the wage fixing agreement. The CEO himself made sure the recruiter was fired.

    Show this thread
  15. 13 Sep 2019

    Since the most common response is that companies have little market power and therefore must pay well: Remember the time when Google, Apple, and almost every major tech company other than FB created and wage fixing agreement?

    Twitter won't allow the full quote due to length, this is the section in the doc that ends with "long-term . . . right approach is not to deal with these situations as one-offs but to have a systematic approach to compensation that makes it very difficult for anyone to get a better offer."
    Show this thread
  16. 13 Sep 2019
    Replying to

    If it were just that quote, I'd say that he says "places that pay more are bad", but this is a common theme for him and he also says things like the following and generally complains that other companies are paying too much and ruining the job market.

    Sorry, Twitter won't allow a quote this long. Summary is that Joel Spolsky is saying that big tech companies pay too much have don't even have work for the people they're hiring to do.
  17. 16 Aug 2019

    The median chip I worked on had more novel design than the P99-novel software project. We were able to set and hit schedules. Ditto for my optics work. Software isn't really unique (no, not even in the grandiosity of its practitioners).

    Show this thread
  18. 15 Aug 2019

    "actually the seed is also a hyper-parameter"

    Twitter won't allow a quote this long, see https://news.ycombinator.com/item?id=20709020 for the full comment and response.
  19. 30 Jul 2019

    People say dynamic programming is a ridiculous name, and it is, but at least it's merely meaningless.

    Twitter won't allow long enough alt text, see https://en.wikipedia.org/wiki/Dynamic_programming#History for the screenshotted quote.
    Show this thread
  20. 29 Jul 2019

    Periodic reminder that basically every "RISC is obsolete" / "RISC is unscalable" article is based on a faulty assumption. RISC has never been about not having instructions like FJCVTZS.

    "I wish I could also be excited, but to me, this is just a reminder that RISC architectures are fundamentally unscalable, and inevitably stop being RISC as soon as they need to be fast. People still call ARM a “RISC” architecture despite ARMv8.3-A adding a FJCVTZS instruction, which is “Floating-point Javascript Convert to Signed fixed-point, rounding toward Zero”. Reduced instruction set, my ass."
    Show this thread

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·