Primarily because it confuses tools that try to suss out the structure of the document by looking at tag types and then make decisions based on that (HTML-to-audio conversions can get really boned by overuse of tables, for example -- algorithms for sensible renderings of tables-as-tabular data can really mess up when presented with tables that aren't).
It's probably best to think of this as a conflict between people who think HTML tags should have inherent meanings (it opens up a lot of opportunities for writing useful software, and it was the original idea with HTML anyway), and people who simply choose HTML tags to achieve particular visual effects (because that's all that was possible for a long time). CSS only comes into this because it fixes that latter problem, so the former group has largely pinned their hopes on it for a resolution to the conflict.
Anyway, CSS2 does offer a solution for the table case too[1]-- you can assign the layout behaviors of tables, table rows, and table cells to any element. In principle you could do table-style layouts using only CSS and DIVs. Too bad browsers don't uniformly support that part of CSS. So, sometimes using tables for layout is a necessary evil. Just be aware of the downside. At least support for CSS tables has improved markedly in recent browsers, so perhaps they'll be a real option soon.
Not at all. Someone just sets up one of these "human botnets", and sells time/cracking capacity to individual spammers who might not have the resources to set it up on their own.
I know capacity on "real botnets" is resold to spammers (and other no-goodniks) this way; I don't see why people wouldn't be reselling capatcha-cracking resources too. It's a buyer's market.
The point was that the plane wouldn't have stalled because the pilot had just opened up the throttle -- by the time the plane was in the new attitude there would have been enough thrust. The software didn't take that into account, however, insisting essentially that the pilot wait until the engines had reached full power before beginning to make any attitude adjustments. Of course by then it was too late.
Hmm, how do optical interferometers used for astronomy work then? My understanding was that stars don't output coherent light.
On the other hand, the actual patent doesn't mention interferometry at all; it seems pretty hand-wavy about any image processing that might need to be applied.
This Kingdom Hearts stuff is a bit of a poison pill for fanart, isn't it?
Square's one of the most fanart-friendly companies out there, and suddenly they're collaborating with one of the absolute least friendly. It's inviting trouble.
Yeah, but most SVG implementations do halfway decent filtering on scaled images, and if you're embedding e.g. a JPEG in an SVG container, you can re-use the same base image (meaning it only gets downloaded and cached once) over a reasonable range of destination sizes.
It is also a pragmatic issue; heavy user reliance on proprietary drivers would make Linux development unsustainable.
Just to pick one obvious example: if someone's running a proprietary video driver, and their network driver is crashing, it's proven in practice to be extremely difficult to distinguish between a subtle bug in the network driver and a bug in the video driver scribbling over memory in the network stack.
Without visibility into the source code for everything running in kernel-space, issues like that turn out to be extremely labor-intensive to fix and diagnose. We don't have that kind of labor available. And of course if the issue's in the proprietary driver, the kernel developers can't fix it anyway.
A pragmatism that ignores consequences isn't much of a pragmatism.
Jumping out a window might be a "pragmatic" way to leave a building, but most people would agree the pragmatic solution would be to take the elevator to the lobby and use the front door instead.
Well, they were certainly emphasizing that they weren't aiming for top-of-the-line 3D performance. It's not too surprising that their downplaying of their 3D abilities could get misinterpreted that way.
Slashdot Mirror
Primarily because it confuses tools that try to suss out the structure of the document by looking at tag types and then make decisions based on that (HTML-to-audio conversions can get really boned by overuse of tables, for example -- algorithms for sensible renderings of tables-as-tabular data can really mess up when presented with tables that aren't).
It's probably best to think of this as a conflict between people who think HTML tags should have inherent meanings (it opens up a lot of opportunities for writing useful software, and it was the original idea with HTML anyway), and people who simply choose HTML tags to achieve particular visual effects (because that's all that was possible for a long time). CSS only comes into this because it fixes that latter problem, so the former group has largely pinned their hopes on it for a resolution to the conflict.
Anyway, CSS2 does offer a solution for the table case too[1]-- you can assign the layout behaviors of tables, table rows, and table cells to any element. In principle you could do table-style layouts using only CSS and DIVs. Too bad browsers don't uniformly support that part of CSS. So, sometimes using tables for layout is a necessary evil. Just be aware of the downside. At least support for CSS tables has improved markedly in recent browsers, so perhaps they'll be a real option soon.
[1] http://www.w3.org/TR/CSS2/tables.html#q2
He's a real aviation buff; I think he'd get a kick out of it.
Out of curiousity, have you ever had problems with things that other people have done with your work being erroneously attributed to you?
Nothing more intuitive.
Not at all. Someone just sets up one of these "human botnets", and sells time/cracking capacity to individual spammers who might not have the resources to set it up on their own.
I know capacity on "real botnets" is resold to spammers (and other no-goodniks) this way; I don't see why people wouldn't be reselling capatcha-cracking resources too. It's a buyer's market.
The point was that the plane wouldn't have stalled because the pilot had just opened up the throttle -- by the time the plane was in the new attitude there would have been enough thrust. The software didn't take that into account, however, insisting essentially that the pilot wait until the engines had reached full power before beginning to make any attitude adjustments. Of course by then it was too late.
I think I'd prefer to leave the explicit setting of application heap sizes behind with Mac OS 9, thanks.
I suppose it's the same logic whereby grandma must stand at the top of the stairs in order to be protected from the Terrible Secret of Space.
The main performance issue for Ruby right now (relative to e.g. Perl) is that it still uses an AST-walking interpreter rather than a VM or JIT.
That part's being worked on (YARV), though.
I've also been told that the parser is slow, but I haven't looked into that much yet.
Actually, from everyone I've heard from so far who's tried it, the Wiimote is surprisingly precise and works great for FPS.
Actually, you can't get high off of industrial hemp; the THC content is negligible.
I was thinkin' Lavos originally, but ... dang, yeah.
Before or after birth?
Cool, thanks for the clarification.
Hmm, how do optical interferometers used for astronomy work then? My understanding was that stars don't output coherent light.
On the other hand, the actual patent doesn't mention interferometry at all; it seems pretty hand-wavy about any image processing that might need to be applied.
Rather than (or in addition to) having a tiny lens on each sensor, you can apply interferometry to get a giant "virtual" lens.
This Kingdom Hearts stuff is a bit of a poison pill for fanart, isn't it?
Square's one of the most fanart-friendly companies out there, and suddenly they're collaborating with one of the absolute least friendly. It's inviting trouble.
er, s/embedding/linking to/
Yeah, but most SVG implementations do halfway decent filtering on scaled images, and if you're embedding e.g. a JPEG in an SVG container, you can re-use the same base image (meaning it only gets downloaded and cached once) over a reasonable range of destination sizes.
That and multiplayer are what separate IF from MUDs.
Or people could start using a scalable vector graphics format... like, say... SVG?
Nah, just spyware installed by organized crime using W2K holes that MS will no longer offer patches for.
I'm not sure I see them going quite that far, but that sounds quite a lot like what they did for OS9-in-OSX, doesn't it?
It is also a pragmatic issue; heavy user reliance on proprietary drivers would make Linux development unsustainable.
Just to pick one obvious example: if someone's running a proprietary video driver, and their network driver is crashing, it's proven in practice to be extremely difficult to distinguish between a subtle bug in the network driver and a bug in the video driver scribbling over memory in the network stack.
Without visibility into the source code for everything running in kernel-space, issues like that turn out to be extremely labor-intensive to fix and diagnose. We don't have that kind of labor available. And of course if the issue's in the proprietary driver, the kernel developers can't fix it anyway.
A pragmatism that ignores consequences isn't much of a pragmatism.
Jumping out a window might be a "pragmatic" way to leave a building, but most people would agree the pragmatic solution would be to take the elevator to the lobby and use the front door instead.
Well, they were certainly emphasizing that they weren't aiming for top-of-the-line 3D performance. It's not too surprising that their downplaying of their 3D abilities could get misinterpreted that way.