HNHacker News
TopNewBestAskShowJobs

frivoal

404 karma · joined August 17, 2010

http://florian.rivoal.net/
submissionscomments
frivoal··on Browsers' bid for relevance is turning them into time-bombs
EME is an API to talk to a Content Decryption Module, I.e. a DRM system. EME is easy and free to implement, but entirely useless without a CDM. There is no specification for CDMs. If you want one, you need to license it from Adobe or some other Hollywood approved vendor (or become one your self by playing by their rules).

You cannot just implement the royalty free web standard specifications and expect the web to work. You also need some corporate agreements.

Arguably this isn't really worse than needing flash or silver light. But at the same time it looked like these were on their way out and that the battle for a 100% open standards web was being won. And Flash & friends could be installed by end users, while CDMs need to come bundled in the browser. Also, in web standards circles, flash and co were considered problems to be solved, while EME is presented as the solution.

So if you're a new browser vendor, and have an inovative (and legal) business model, either Hollywood likes it, or you will not be able to support the whole web.

frivoal··on Lenin was a mushroom
My guess: It is also a cute character from Disney. In French they are not homophones, and people seem to care a lot less.
frivoal··on Twitter to Stop Counting Photos and Links in 140-Character Limit
http://but.you.can.also.do.it.with/url?so&what&the#point
frivoal··on Fantasy Math Is Helping Companies Spin Losses into Profits
You still need to go through GAAP (or some other ruleset) to figure out what goes into the E I T D and A of the EBITDA.
frivoal··on Sharp Agrees to $6.2B Bailout from Foxconn, Nikkei Says
They're not zaibatsu, but the point stands.
frivoal··on Sharp Agrees to $6.2B Bailout from Foxconn, Nikkei Says
Nice to see that economic reason prevailed over protectionism, but at the same time, sharp would probably be a good company to split into various smaller businesses. It's doing to many barely related things.
frivoal··on LinkedIn shares drop 40%, erasing $10B of company's value
On the other hand, you got the raise at the point where I started hoping you'd do great, even if you hadn't yet.
frivoal··on The last new subway line in Japan
Isn't most of Japan's debt owned domestically? Making local companies and retired people go bust by defaulting on debt they own doesn't sound like a good way out.
frivoal··on Japan’s Parliament Approves Overseas Combat Role for Military
A few couple of quotes from wikipedia:

"The Supreme Court is generally reluctant to exercise the powers of judicial review given to it by the constitution, in large part because of unwillingness to become involved in politically sensitive issues. When decisions have been rendered on such matters as the constitutionality of the Self-Defense Forces, the sponsorship of Shinto ceremonies by public authorities, or the authority of the Ministry of Education to determine the content of school textbooks or teaching curricula, the Court has generally deferred to the government."

"The Supreme Court of Japan has been described as the most conservative constitutional court in the world, and for good reason ...Since its creation in 1947 [the court] has struck down only eight statutes on constitutional grounds. By way of comparison, Germany’s constitutional court, which was established several years later, has struck down over 600 laws."

So what's the chance? Not very high.

frivoal··on Japan’s Parliament Approves Overseas Combat Role for Military
Also gathered constitutional experts, tasked them to research whether this would be constitutional (expecting them to say it was). When they concluded it wasn't, the gvt apologized for hiring the wrong folks, and moved on.
frivoal··on Jet Lag ‘Cures’ Aplenty, but None That Work for All
It's (a). If you get substantial food while you were starving, your body will guess that this time of the day is a good one to be awake at, in case the food source is available every day at the same time.

So a solid breakfast after a short period of fast resets the internal clock.

frivoal··on Vendor Prefixes and Market Reality
user-select does need some improvements in browsers, but this MDN page is somewhat out of date.

It is now being standardized: https://drafts.csswg.org/css-ui-4/#content-selection

Browsers haven't fully converged yet, but at least we're starting to see what they are likely to converge to.

frivoal··on Vendor Prefixes and Market Reality
There were some good reasons behind the intent for vendor prefixing, but it was a flawed solution to a real problem. As vendor prefixed made it to production builds, we ended up not with experimental properties that we could still change willy nily, but with strangely named properties that we have to support forever (since content depend on them working correctly), and without even the benefit of being able to evolve the design of the unprefixed version anyway we like, since "best practices" mean that css authors very often writen the unprefixed property in their stylesheet along with the prefixed one for future proofing.

What browsers are mostly doing (and the CSSWG is in the process of formalizing) is that experimental features should be included in nightly/preview/developer builds, but not in production builds (or be off by default, to be turned on via user settings). That way we can still experiment, but we don't use experiments in production, and avoid creating a legacy compat problem on things that were not meant to be permanent.

frivoal··on Greeks Reject Bailout Terms
It's a pretty good video. One thing it doesn't cover (or only implies) is that the debtor's countries spending plays a role in fueling the richer countries economies. Germany is in a better situation because of many fundamentals, but exporting to Greece (on borrowed money) didn't hurt.

Borrowing to spend without considering whether you can repay is irresponsible.

Lending money to someone so that they can buy from you, without considering whether they can repay is also irresponsible.

As much as Greece is guilty of the former, Germany is of the later.

Here's an other article putting a different spin on this: http://www.telegraph.co.uk/finance/comment/ambroseevans_prit...

frivoal··on Things Rust shipped without
Nice. Simple and efficient. I am going to read more from that code base. Looks like this might be a nice read as well: https://ocaml.org/docs/papers.html#BytecodeCompilerandByteco...
frivoal··on Things Rust shipped without
An offset from the first label, or some nearby alignment point? Makes sense.

Do you have a pointer to the relevant part of OCaml's source? The whole thing is probably worth a read, but don't think I can set aside enough time for that soon, and it could take me a while to find the right part.

frivoal··on Things Rust shipped without
One of the things Servo (the web browser engine written in Rust) has not decided to write from scratch is a (high performance) js engine. If they did, they might have more feedback on this kind of things.
frivoal··on Things Rust shipped without
Hadn't thought of that, but it sounds interesting. I guess this means that opcodes in your bytecode need to be large enough to store a pointer, this could involve some tradeoffs in terms of minimizing instructions vs fitting in the cache, but I suppose it could fall on the right side of things often enough.
frivoal··on Things Rust shipped without
And they say C is a simple language... :)

To be fair, I've seen code like this exactly once.

frivoal··on Things Rust shipped without
This is a fairly mechanical conversion, so suppose a sufficiently smart compiler could do that.

But assuming a sufficiently smart compiler when you depend on your code being fast tends to cause issues. Especially if you expect it to run on a bunch of platforms with compilers of varying qualities. Last time I saw code last this, we certainly couldn't rely on compilers being particularly smart, and we did care about speed.

frivoal··on Things Rust shipped without
A bitecode interpreter is another place where it's nice to have gotos.

Here's the base code without gotos:

  typedef enum { ADD, MUL, ..., END } opcode;

  void run() {
    opcode ins;

    while (1) {
      ins = fetch_next_inst();
      switch (ins) {
        case ADD:
          perform_addition();
          break;
        case MUL:
          perform_multiplication();
          break;
        ...
        case END:
          wrap_up();
          return;
      }
    }
  }
You have 3 jumps on each loop. From the break to the end of the loop, then from the end to the top, and one from the switch to the right case. The first one might be optimized away, but let's remove it explicitly.

  typedef enum { ADD, MUL, ..., END } opcode;

  void run() {
    opcode ins;

    start:
    ins = fetch_next_inst();
    switch (ins) {
      case ADD:
        perform_addition();
        goto start;
      case MUL:
        perform_multiplication();
        goto start;
      ...
      case END:
        wrap_up();
        return;
    }
  }
Assuming a non lousy compiler, we haven't improved anything yet. But now the fun starts. We can go down to one jump for each iteration.

  typedef enum { ADD, MUL, ..., END } opcode;

  #define NEXT() \
  do { \
    ins = fetch_next_inst(); \
    goto *jump_table[ins]; \
  } while(0)

  void run() {
    opcode ins;
    static void *jump_table[] = { &&add_l, &&mul_l, ..., &&end_l };

    NEXT();
    add_l:
      perform_addition();
      NEXT();
    mul_l:
      perform_multiplication();
      NEXT();
    ...
    end_l:
      wrap_up();
      return;
  }
Voila! a single jump every time around. Now, depending on what kind of architecture you're running on, the size of the cache, etc, this may or may not be faster.

Granted, this is not the kind of code you write everyday. But sometimes speed matters, and good luck writing this without gotos.

frivoal··on Meteor Raises $20M to Build the One JavaScript Stack to Rule Them All
agree on 1 and 3.

As for 2, I think it makes sense. You can pre-render your GUI on the server side, so that clients get served proper html rather than an empty pile of js, and then have the same codebase take over on the client side, single-page-app style, just talking to the backend to fetch data.

frivoal··on Accounting for Developers
That's the easiest way to deal with it. If you can remember that (which is not hard), and where one type of account falls on the left/right spectrum (which isn't hard either, pick whichever you like), then you can figure out the rest.

Until you're used to these two words and they speak for themselves, assigning any other meaning tends to lead to more confusion than anything else.

frivoal··on Show HN: Get pinged when standards hit 90% browser support
Done: https://twitter.com/CSScommits/status/587951029661990912
frivoal··on Germany's trade surplus is a problem
Just curious, who said that?
frivoal··on The 500-mile email (2002)
I don't know about what "units" support, but if you ever need picolightseconds in your web design, CSS got you covered:

http://dev.w3.org/csswg/css-egg/#astro-units

frivoal··on Show HN: Get pinged when standards hit 90% browser support
I got a moment, so I did drop you a mail:) Thanks for looping me in.
frivoal··on Show HN: Get pinged when standards hit 90% browser support
CSS is device neutral and works on things that don't have shift keys or context menus, so we can't specify the mechanism. However, we can make it allowed, or even recommended for browsers to provide a way to select anyway, even if the spec leaves it up to the browser what UI it wants to use for that.
frivoal··on Show HN: Get pinged when standards hit 90% browser support
If you have suggestions on how to mitigate that (making the feature less useful as a copy protection without affecting (too much) the intended use, I'm happy to update the spec.
frivoal··on Show HN: Get pinged when standards hit 90% browser support
If you care for "user-select: none", you might be surprised to know that there wasn't an official spec for it to help the browsers interoperably converge, and you might be happy to learn that I just made one: http://dev.w3.org/csswg/css-ui-4/#content-selection
← PreviousPage 3 of 4Next →