4 ms·

This seems like something that should have been done long ago

by Zekio 10y ago

Jaruzel 10y ago
Alas, with so many emerging 'standards' across the web now, maintaining relevant RFCs seems to have fallen by the wayside.
- detaro 10y ago
  In the other direction, some people have gotten sick of working on standards that don't lead to anything (often because they are created in a vacuum, before actual deployment). Getting the balance between "standardize before everyone is doing completely different things" and "don't standardize before you know what the solution should look like" right is tricky. Especially in the web space, there are tons of fairly theoretical (parts of) standards that nevertheless cost the participants a lot of energy.
  - richardwhiuk 10y ago
    The problem is that technology on the web increasingly goes through an inflection point. Prior to the inflection it is too early to standardize, because it's not clear that how something will actually work, or whether it'll actually be use. Post inflection point, there is an established way of doing things (by sheer weight of running code), so standardization often offers little benefit. During the inflection point, those involved are generally way too busy to worry about standardization!
    - mstade 10y ago
      Also post inflection point, if standardization happens it often becomes a matter of just documenting de facto standards – warts and all. This can make bugs not just a curiosity of the standard but hard requirements, and a difficult corner to paint yourself out of. Case in point: HTML. (And in particular the WHATWG living standard.) Counter case: HTTP. Through my rose tinted glasses I think it's a beautiful after-the-fact standardization effort. It's not without warts of course, nothing is, but still a great example of carefully turning the state of the world into a solid standard.
- gsnedders 10y ago
  A bunch of people from browsers tried to get interest in registering a font top-level type around five years ago IIRC and the basic conclusion on the IANA side was that they did not want to add a new top-level media type and that they should just be under the application media type. For example, when WOFF was being standardised, nobody in the WG wanted application/font-woff, but it was the best that there was consensus for with IANA.

This seems like something that should have been done long ago

by Zekio 10y ago

This seems like something that should have been done long ago

Alas, with so many emerging 'standards' across the web now, maintaining relevant RFCs seems to have fallen by the wayside.

detaro 10y ago
In the other direction, some people have gotten sick of working on standards that don't lead to anything (often because they are created in a vacuum, before actual deployment). Getting the balance between "standardize before everyone is doing completely different things" and "don't standardize before you know what the solution should look like" right is tricky. Especially in the web space, there are tons of fairly theoretical (parts of) standards that nevertheless cost the participants a lot of energy.
- richardwhiuk 10y ago
  The problem is that technology on the web increasingly goes through an inflection point. Prior to the inflection it is too early to standardize, because it's not clear that how something will actually work, or whether it'll actually be use. Post inflection point, there is an established way of doing things (by sheer weight of running code), so standardization often offers little benefit. During the inflection point, those involved are generally way too busy to worry about standardization!
  - mstade 10y ago
    Also post inflection point, if standardization happens it often becomes a matter of just documenting de facto standards – warts and all. This can make bugs not just a curiosity of the standard but hard requirements, and a difficult corner to paint yourself out of. Case in point: HTML. (And in particular the WHATWG living standard.) Counter case: HTTP. Through my rose tinted glasses I think it's a beautiful after-the-fact standardization effort. It's not without warts of course, nothing is, but still a great example of carefully turning the state of the world into a solid standard.
gsnedders 10y ago
A bunch of people from browsers tried to get interest in registering a font top-level type around five years ago IIRC and the basic conclusion on the IANA side was that they did not want to add a new top-level media type and that they should just be under the application media type. For example, when WOFF was being standardised, nobody in the WG wanted application/font-woff, but it was the best that there was consensus for with IANA.