Duane Clark wrote:
>
> Sigh.... In my browsing through bugzilla looking for this (I was looking
> for "link" rather than "slash") I came across several other cases like
> this where the solution has been "don't fix because the link is broken".
> I can understand how that might sound like a tempting solution, but I
> really believe it is excessively pedantic.
>
> In this case it is even worse, because according to the bug report, the
> behavior exhibited is in fact a deliberate attempt to fix a broken link.
> If the URL had been typed by hand, the solution implemented might be
> reasonable. But for a link imbedded in a document, I think I can
> confidently predict that 95% of the time this is the wrong fix.
>
> I would hate to see all these people put an incredible amount of work
> into Mozilla, and then only the hardcore people use it because it
> "doesn't work". It really doesn't matter whether the fault is with the
> web page creators or Mozilla, if Mozilla is the only browser that
> doesn't work.
You mean there are browsers that actually will follow links that only
have one slash after http:? The fault lies with the authors who wrote
such obviously wrong URLs. I look at that and think, "The obvious
intent was to have two slashes", not, "The obvious intent is to go to
/somedir on the same site." It's utterly unclear, and wrong in either
case. The only way to successfully try to work around the invalidity of
the link would be to pop up a dialog with something along the lines of
"The link you are trying to follow is not constructed correctly. Do you
want to attempt to resolve it on this domain or as an absolute URL?"
with buttons to those effects.
(My personal preference would be to break backwards compatibility here.
One person's pedantic is another person's flat wrong. :) )
