4 ms·

Well, for one thing you'd have to form some kind of giant committee to decide on what characters actually exist, and assign them all unique numbers. Can you ima

by db48x 7y ago

Well, for one thing you'd have to form some kind of giant committee to decide on what characters actually exist, and assign them all unique numbers. Can you imagine how ridiculous that would be?

AnIdiotOnTheNet 7y ago
Said committee would be mired in politics of what should and shouldn't be included. Cuneiform? Klingon? Clip art? They are all technically human communication.
- tialaramex 7y ago
  The comment you're replying to is joking about the Unicode Technical Committee, which of course exists and really does what was described. So it's sarcasm. But the Technical Committee doesn't spend very much time worrying about whether it should include Cuneiform (yes), Klingon (no) or Clip Art (mostly no but there are some things you might think of as "clip art" that are in ISO-10646 / Unicode).
  - AnIdiotOnTheNet 7y ago
    Yes, I'm aware of this. My post is also made in jest along the same lines, albeit with a grain of genuine distaste for what is and isn't allowed in Unicode.
- db48x 7y ago
  So much better to have local control instead of ending up beholden to a committee of busybodies.
  - jancsika 7y ago
    Yeah, I think the local control argument would be quite effective against a number of technologies. In fact Linux mailing list comes standard with its own early argument from Tannenbaum against the monolithic design. So no alternate history necessary there. :)
jancsika 7y ago
Hehe. It's interesting to think about these: * unicode -- I think the naysayers would win here because there are way more people today who understand the security and complexity issues surrounding the topic than those who understand the details of why unicode and the various encodings work the way they do * gnu -- this would be a great challenge to the naysayers. But I think it could be done if the naysayers really studied the hardware of the time and could ask pointed questions that a Stallman impersonator couldn't answer. * fftw - "hey we're gonna build a library that queries at runtime for whatever SIMD instructions are available on the user's computer and choose the fastest algo." Yeah right. :) The possibilities are bounded but many.

by db48x 7y ago

Well, for one thing you'd have to form some kind of giant committee to decide on what characters actually exist, and assign them all unique numbers. Can you imagine how ridiculous that would be?

AnIdiotOnTheNet 7y ago

Said committee would be mired in politics of what should and shouldn't be included. Cuneiform? Klingon? Clip art? They are all technically human communication.

tialaramex 7y ago
The comment you're replying to is joking about the Unicode Technical Committee, which of course exists and really does what was described. So it's sarcasm. But the Technical Committee doesn't spend very much time worrying about whether it should include Cuneiform (yes), Klingon (no) or Clip Art (mostly no but there are some things you might think of as "clip art" that are in ISO-10646 / Unicode).
- AnIdiotOnTheNet 7y ago
  Yes, I'm aware of this. My post is also made in jest along the same lines, albeit with a grain of genuine distaste for what is and isn't allowed in Unicode.
db48x 7y ago
So much better to have local control instead of ending up beholden to a committee of busybodies.
- jancsika 7y ago
  Yeah, I think the local control argument would be quite effective against a number of technologies. In fact Linux mailing list comes standard with its own early argument from Tannenbaum against the monolithic design. So no alternate history necessary there. :)

jancsika 7y ago

Hehe. It's interesting to think about these: * unicode -- I think the naysayers would win here because there are way more people today who understand the security and complexity issues surrounding the topic than those who understand the details of why unicode and the various encodings work the way they do * gnu -- this would be a great challenge to the naysayers. But I think it could be done if the naysayers really studied the hardware of the time and could ask pointed questions that a Stallman impersonator couldn't answer. * fftw - "hey we're gonna build a library that queries at runtime for whatever SIMD instructions are available on the user's computer and choose the fastest algo." Yeah right. :) The possibilities are bounded but many.