"Phillip M. Jones, C.E.T." wrote:
>
> Gervase Markham wrote:
> >
> > > The average consumer (and me too) will say it works with NS4.76 or IE5
> > > but not with Mozilla, and therefore Mozilla is broke. For perfectly good
> > > evidence of this, take a look at the bug page at how many times this has
> > > been resubmitted. It is the largest number I have so far stumbled across
> > > (though I have not looked at a lot).
> >
> > There's a bigger issue here. Who do we respect more? Old and broken
> > browsers, or W3C and other internet standards (such as RFCs)? If we "find
> > a meaning" for this sort of broken URL, what else should we guess at
> > interpreting?
> >
> > Computers have to read this stuff, not humans.
> >
> > Gerv
>
> The old Broken Browsers!
>
> Standards are wonderful; but, if they declare 99.999% of what done on
> the net currently wrong and unusable what good are the Standards.
This case is absolutely wrong. It's an abuse of a loophole in a
previous RFC that was fixed in a subsequent RFC to prevent something
like this from occurring. But people ignored that, and sadly older
browsers used it. I don't think there should be any leniency on such a
link; it's broken, and should be flagged to the user as such. There's
no correct guess, because according to the relevant RFC, this *has* to
be interpreted as an absolute URL. It's Just Plain Wrong.
> The only thing that was "really needed" was remove handling of Java
> through the browser, and creating a Plugin for Java that cause the
> Browser to use the current Java Standards simply just uses the latest
> Java as written (Transparently) and better handling of Style sheets.
Style sheets were really hard to support in the old codebase. Rewriting
was the only tractable option.
> New Standards shouldn't be created just for the sake of killing off old
> standards that work efficiently.
There is no old standard here. It's an abuse that was specifically
outlawed *by* a standard. The sad de facto standard of rendering these
links is as incorrect as the links themselves.
> This WC3 stuff is like the Sony Betamax and JVC VHS Tape system war. W3C
> being the Betamax system. You know what happened in the Betamax/VHS
> format fight. VHS won because more people bough the VHS.
This isn't a W3C problem. It's an RFC problem, and a lot more people
follow the RFCs than the W3C (the RFCs are much more open).
> Browsers should be only written for the end user in mind.
End users should be given the option of what to do in this scenario.
Not to interpret the link as a relative one, when logic tells me that
this would be wrong (despite the 95-99% figure thrown about here, I've
yet to see a *single* link in my years of Web surfing that uses this
broken syntax to indicate a relative link).
Life would be so much easier if HTML validation had been required for
publishing. Errors like this don't get past compile time in normal
programming languages.
