- From: Jukka K. Korpela <jkorpela@cs.tut.fi>
- Date: Thu, 10 Sep 2009 19:12:22 +0300
- To: <www-validator@w3.org>
Ville Skyttä wrote: > On Monday 31 August 2009, Jukka K. Korpela wrote: > >> Attribute "foo" is not a valid attribute. Did you mean "onkeydown" or >> "nohref"? >> >> I could not have imagined that they invented something _that_ absurd >> in trying to make the misguided heuristics "smarter". > > This is nothing new nor intentional. I'm fairly certain you knew > that and just wanted to insult someone. You're wrong then, no matter how certain you were. I had not seen a confusing message of _that_ kind before, and what else could it be than a failed attempt to fix the feature? > I don't know what you're > trying to accomplish with that; just reporting the issues is much > more productive, and I think you know that too. The issue has been reported many times already - I mean the basic issue of making suggestions that don't make sense on the average. I'll leave it to the universal board on absurdity to decide which is wrose, suggesting attributes that are not valid in the context or suggesting that you use "foo" when you have actually used "foo". I have already proposed that the misguided heuristics be removed, at least until a reasonable version has been created and tested. I have not seen any response to this proposal, which seems to be the only sensible approach now. > However, this reproducer and the others in your message I'm snipping > here are good in the sense that they are the first ones I've seen > that don't seem sanely fixable with the current implementation, so > I'll look into ripping the feature out for the next validator version. The experiment has proved out to be a failure long ago, and removing it should have been the _first_ action. Then you'd have time to think and experiment whether it could be re-introduced. -- Yucca, http://www.cs.tut.fi/~jkorpela/
Received on Thursday, 10 September 2009 16:13:13 UTC