4 ms·

I'm generally not in favor of preserving any sort of non-computing history, but in this particular case: > she discovered B.C.'s registry doesn't permit chara

by caslon 6y ago

I'm generally not in favor of preserving any sort of non-computing history, but in this particular case: > she discovered B.C.'s registry doesn't permit characters that are not part of the Roman alphabet. It makes sense. There's no reason not to allow Unicode characters, and Unicode already allows that alphabet to be used.

incrudible 6y ago
> There's no reason not to allow Unicode characters, and Unicode already allows that alphabet to be used. No reason besides the fact that Unicode characters can cause countless errors in legacy systems downstream. Perhaps the money that could be spent on supporting arbitrary renditions of indigenous languages in non-standard latin-derived characters would be more wisely spent on supporting those communities directly.
- Ekaros 6y ago
  And I would be lot more worried about humans in the systems... You can't honestly expect all of the people involved to learn to input all the possible chractersets that some groups use. You could train them, but I think spending fraction of money would do massively more good in other ways...
- caslon 6y ago
  > No reason besides the fact that Unicode characters can cause countless errors in legacy systems downstream. All systems worthwhile have been using UTF-8 since the 1990s. There is no excuse for legacy systems handling business applications not to handle UTF-8.
  - incrudible 6y ago
    > All systems worthwhile have been using UTF-8 since the 1990s. That's irrelevant. You can rest assured that in any larger government system some information must travel through the equivalent of a chimpanzee trained in Baudot code.
    - caslon 6y ago
      In which case the problem is with the government (that already supports French symbols, mind), not with the person hoping to have sensible support for names.
      - PowerBar 6y ago
        I'm guessing the system was built to handle extended (8-bit) ascii, which contains the accented French characters. Upgrading a system of this size (and potentially many other systems it needs to trade information with) to support UTF-8 would be a very large undertaking. It should have been built to support UTF-8 in the first place and should be upgraded, but the time it will take to do that will not help this person. With any luck, it may help someone else in the future though.
      - incrudible 6y ago
        Of course the problem is with the government. That's not my point. My point is that there are plenty of reasons why this kind of name is not supported. Besides the legacy computing reasons, having a name that only a handful of people in the whole world could spell out is going to cause problems. Fixing this problem would cost real money that could be used on other things. Perhaps the indigenous community should be consulted on their most pressing issues before enshrining the right to have these characters in a business name.

by caslon 6y ago

incrudible 6y ago

> There's no reason not to allow Unicode characters, and Unicode already allows that alphabet to be used. No reason besides the fact that Unicode characters can cause countless errors in legacy systems downstream. Perhaps the money that could be spent on supporting arbitrary renditions of indigenous languages in non-standard latin-derived characters would be more wisely spent on supporting those communities directly.

Ekaros 6y ago
And I would be lot more worried about humans in the systems... You can't honestly expect all of the people involved to learn to input all the possible chractersets that some groups use. You could train them, but I think spending fraction of money would do massively more good in other ways...
caslon 6y ago
> No reason besides the fact that Unicode characters can cause countless errors in legacy systems downstream. All systems worthwhile have been using UTF-8 since the 1990s. There is no excuse for legacy systems handling business applications not to handle UTF-8.
- incrudible 6y ago
  > All systems worthwhile have been using UTF-8 since the 1990s. That's irrelevant. You can rest assured that in any larger government system some information must travel through the equivalent of a chimpanzee trained in Baudot code.
  - caslon 6y ago
    In which case the problem is with the government (that already supports French symbols, mind), not with the person hoping to have sensible support for names.
    - PowerBar 6y ago
      I'm guessing the system was built to handle extended (8-bit) ascii, which contains the accented French characters. Upgrading a system of this size (and potentially many other systems it needs to trade information with) to support UTF-8 would be a very large undertaking. It should have been built to support UTF-8 in the first place and should be upgraded, but the time it will take to do that will not help this person. With any luck, it may help someone else in the future though.
    - incrudible 6y ago
      Of course the problem is with the government. That's not my point. My point is that there are plenty of reasons why this kind of name is not supported. Besides the legacy computing reasons, having a name that only a handful of people in the whole world could spell out is going to cause problems. Fixing this problem would cost real money that could be used on other things. Perhaps the indigenous community should be consulted on their most pressing issues before enshrining the right to have these characters in a business name.