> Okay, let's follow a standard. But wait... some here believe there is
> only one standard.
There is one standards group, ratified and approved by the committee
elected to draft and approve said standards. Microsoft is a member of the
consortium that the W3c is composed of, as are thousands of other technology
and non-browser companies. This is not just some "rogue" standards agency.
Our friends in Redmond regularly submit proposals to the W3C for approval.
> They're both arrogant and wrong; just because it comes out of W3C doesn't
> make it THE standard. Lacking any enforcement teeth, it's just a
> standard.
It's not there to be enforced, and yes, because it comes out of the
W3C, it _does_ become the standard. Microsoft supports their decisions, and
so do those from the Mozilla project and other projects and companies.
> What standards compete with it? Well, obviously chaos.
Chaos is not a standard, it's a condition that results from the
_lack_ of a standard (in this case).
> Also IE.
IE is not a standard, nor is it a reference platform for determining
standards-compliance. In fact, quite the opposite. It may be the most widely
used browser on a desktop machine, but that does not make it a standard. In
fact, the latest IE still fails to support a lot of things browsers like
Mozilla and Opera have supported for _years_ now.
The claim of IE being the most-widely-used is subject to great
skepticism these days, since a growing majority of surfers forge their
browser's UserAgent string to mimic IE (myself included), just so they can
make purchases online and use their online banking institutions.
Many places like to try to block non-IE browsers, because they can't
make a functional website that works in IE as well as other browsers, mostly
due to laziness. Instead, they require IE, the less-secure and less
functional client.. how odd. More bank accounts have been broken into and
liquidated due to flaws in IE, than any other browser.
> If you are going to stick to a standard, it reasonably should be the most
> common standard.
The lowest-common denominator? Sure, HTML 3.2 and CSS1 would work.
Oh wait, IE doesn't fully support either of those. Oh no, a paradox!
Where can I download a copy of this IE standard you mention? I'd
like to give it a read, if I may. The RFCs and the IETF exists for a reason.
If you'd like to rewrite the RFC and the IETF points to it, you can say
whatever you want about spaces in URIs. Microsoft tried, and failed.
> And the prevailing installed base is the most reasonable standard if the
> target is a tool to parse data tested for specifically that base.
The target client/audience is a Palm device, not an IE browser, in
this case. If Plucker works best with validated, standards-compliant
content, then that is what we should require or cater to.
Does it work with all pages that "look good" in IE? No, not hardly,
because IE lacks standards-compliance, and many pages that are designed to
"work well" in IE, are composed of completely invalid HTML. Also, many tools
like FronPage and DreamWeaver (the two largest offenders in this class) do
not generate valid HTML, and in fact, make up their own imaginary tags and
elements, which appear nowhere in the HTML specs, but which IE seems to
support.
> But wait... someone (me) tested those a while ago with respect to this
> issue and reported back that Opera going back several versions, IE going
> back several versions, and Mozilla all support URLs with spaces.
The issue isn't the support for URLs with spaces.. it's really what
to do _upstream_ about the problem, which is not a browser or client
problem, it's a web "designer" problem (a lack of knowledge about the
technology they're using, since spaces are _not allowed_ in URLs).
Browsers support them, only because it's an easy fix to a stupid
upstream problem. As I've said before, Plucker can support them too, it's a
one-line fix on my side of things, and I'm sure just the same on the Python
and Java side of things. However, this does _not_ fix the problem, the
server side, where the error is. A bandaid does not stop you from cutting
yourself again. Proper precautions in the future stop you from cutting
yourself.
No matter how much you want to believe that standards don't exist
for any sane reason, they do. Browsers use them to deliver content. Many
browsers support them more rigidly than others, and some not at all.
You can choose to ignore them as well, as Microsoft does in most
cases for browsers, but the dozens of other people working on browsers and
HTML tools don't. You see, if we follow the standards set aside by the same
group that the browser manufacturers use, we can be guaranteed that our code
and tools will continue to work with those products, regardless of whether
or not some other tool does not. If you want to make up your own elements,
tags, and other attributes, and try to get other people to use that
"standard", then you're more than welcome to try.
> If you're going to live and die by the standard, at least make certain the
> standard is accepted and recognized by the world you intend to interact
> with.
Agreed, which is why I personally use those ratified by the W3C and
the IETF, as do the thousands of other companies who write tools that deal
with the same content these organizations preside over the standards of.
This is really a moot discussion though. The standards aren't going
to go away, just because the most dominant browser in the desktop market
chooses to ignore them.
The goal is to not burdon the end user with the inadequacies of the
website or website designer. Putting things like spaces in urls and using
backslashes for paths in the URI can be easily fixed and compensated for,
but they are definately _wrong_ in the context of HTTP as a delivery
mechanism for that content.
d.
_______________________________________________
plucker-list mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-list
