Date: Tue, 13 Sep 1994 13:01:30 -0400 (EDT)
From: "Steven D. Majewski" <sdm7g@virginia.edu>
To: "Daniel W. Connolly" <connolly@hal.com>
Subject: Re: <URL:...> considered harmful
In-Reply-To: <9409130509.AA06650@austin2.hal.com>
Message-Id: <Pine.A32.3.90.940913123748.16880A-100000@elvis.med.Virginia.EDU>
On Tue, 13 Sep 1994, Daniel W. Connolly wrote:
>
> Yes, we should create a more reliable scheme. But we shouldn't
> write it in stone until we've got some experience with it. I have
> some experience with the <URL:...> scheme, and it's almost all bad.
>
> Until the benefits of the mechanism clearly outweigh the costs,
> we have no business standardizing on <URL:...>.
>
I agree, Dan.
Drop that section.
The proposed recommendation conflicts with (most) current practice.
( Both in code and in documents )
Neither mosaic or lynx will open a URL like:
URL:ftp//ds.internic.net/
( so what's the point of being able to grab <URL:ftp://...> out
of a document? )
I can think of better formats. ( Or at least ones *I* like better! :-)
But the point is, that if it's not an *enforced* part of the standard
( i.e. if there isn't something that will reject non standard usage )
then I think it will end up being widely ignored, and will only muddy
the issue more. It will not be THE STANDARD - just one more
representation.
BTW: *MY* better scheme is different from Dan's better scheme.
Again: I don't think there is wide agreement on this point.
I suspect that some of the disagreement hasn't been very vocal
because of the weak wording - i.e. it's recommended, but I'll
just ignore it. That's been my feeling. I wasn't eager to start
an argument about it, but as long as Dan has started one, I'll
state my support for his position.
It's not ready for standardization.
( not even as a merely recommended practice )
- Steve Majewski (804-982-0831) <sdm7g@Virginia.EDU>
- UVA Department of Molecular Physiology and Biological Physics