Atte writes: | On Tue, 5 Feb 2002, Erik Ronstr=F6m wrote: | | > One of my fears about abc is that is will go the same way as HTML. | <snip> | > The keywords are "backward compability". The browser developers don't | > need to follow any standard, and of course they don't: that would make | > 90% of all pages on the Net illegal. So they continue using their old, | > inconsistent ways of doing things. And since they work and the browsers | > allow it, people will continue using them when building homepages. | | Totally off topic, but the problem with html is that one browser dominates | to the extend that nobody cares what the standard says, they just write | code for that browser.
Good comparison. Of course, there are about a zillion other web programs out there, with various specialized uses. They often act differently out of necessity. Few of them pay a whole lot of attention to standards designed for a flashy graphical browser, for very good reasons. Consider, for example, a web browser for use by the blind. How could it possibly correctly implement any formatting standards? | > I think there is a large risk that abc develops in exactly this | > direction. | | I fear it's too late already :-( I hope it's not, though. I'm not up to | date with the work on the standard, is there still a commission | working on what to include in the standard? I really think this work is | extremely important if abc is to have any future. What seems to have happened is more or less consistent with the past work on abc. The (semi-official) standards committee started with the idea that what it needed was a clear formulation of abc version 1.6 as a standard, and has worked on codifying that. New features are to be put off until the current standard is established. Of course, this is of little relevance to people who need things not covered by version 1.6, so those of us have continued on our merry way inventing random extensions for our own use, and wondering if the standards folks will ever catch up. Actually, this is also pretty consistent with most industry standards which, most of the time, end up making official what a few of the major players have done and imposed as a de-facto standard. Some have argued that this is the best way to make standards. Their argument is that separate standards efforts always bog down in exactly the way that abc has bogged down, in long legalistic discussions of arcane details that confuse most of the readers until they go away. This leaves only the pickiest of the legal types, who then produce a standard that nobody can understand or implement. So implementers do a lousy job of trying to follow a standard that they can't understand while users wonder what they hell the standards folks were thinking when they came up with such an ungodly mess. Meanwhile, those who just want something useful continue to build what they think is useful, and check back with the standards effort now and then to see if there's anything there they can understand well enough to actually implement. There are a lot of such people, and what they do isn't always consistent or compatible. We're continuing in a long tradition ... (As you may have guessed, I've been involved in a few computer industry standards efforts. ;-) To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
