Scott E. Preece <[log in to unmask]> wrote:
> From: [log in to unmask]
>| I have spent a fair amount of effort describing justification for META and
>| my problems with LINK in this regard. I'd appreciate a more descriptive
>| reasoning than just "makes more sense". Both are header-area elements; LINK
>| is a potential link, META is meta-information. Is the source of the
>| just-downloaded document a potential link (leading to confusion with the
>| URL used to download the document), or is a "bookmark" for the file
>I think it's a link - there can be many, intentionally distinct,
>versions of a resource, all pointing to the same source resource. Each
>has a separate, valid location.
>My own opinion is that we should *not* tell browser makers to substitute
>*any* meta-information for the opened-URI in hotlists, bookmarks, or
>location displays, without asking the user first. I know *I* don't want
>it guessing whether I want the "preferred" location or the actual
>location I opened to be the bookmark I use to reference it in the
I believe Dave Hollander addressed this issue:
Dave Hollander <[log in to unmask]> writes:
>In message <[log in to unmask]>, Scott E. Preece
>>When I display a document in a browser, I don't want it to show the
>>location as anything but what I specified and I don't want it to
>>remember in a hotlist an address other than I specified.
>I know I have told some of you I was going to duck out of this, but...
>It is a generally accepted axiom in information management that the
>closer that you are to the source, the more accurate the information.
>You hit a link from lycos or somewhere to an Hewlett-Packard page.
>Assuming there are multiple links, some deprecated some prefered,
>who do think is most likely to know the best URL to use for that node?
>By best, I mean one that is most likely to be successfully accessed again,
>next week and next year.
I suppose that browser makers could provide an option, which is basically
what you're demanding: not modifying your preferred behaviour "without
asking the user first." This seems more an issue of browser design than of
an HTML standard on where to store a "preferred" URI for a resource.
The URL used to access a resource may be invalid immediately after being
accessed. What purpose would there be in storing that in preference to the
URI recommended by the author, who presumably knows the structure of the
server better than the user and can recommend a long term URI for accessing
If you'll note from Ka-Ping Yee's message, each browser interprets,
stores and uses the URL used to access the document and the BASE element
URL in a variety of different ways. I believe it in our best interest to
resolve this issue with clarity so that browser makers can migrate toward a
 Ka-Ping Yee "Re: The BASE element dilemma", Tue, 15 Aug 95 11:33:04 EDT.
Murray M. Altheim, Information Systems Analyst
National Technology Transfer Center, Wheeling, West Virginia
email: [log in to unmask]