[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
FromNess, ToNess
- To: <acad!blitzen!rick>
- Subject: FromNess, ToNess
- From: Eric Dean Tribble <tribble>
- Date: Thu, 25 Jan 90 10:50:04 PST
- Cc: <xtech>
- In-reply-to: <Rick>,17 PST <9001251825.AA11996@xxxxxxxxxxxx>
Date: Thu, 25 Jan 90 10:25:17 PST
From: acad!blitzen!rick (Rick Mascitti)
btw, what happens to a (still useful) link whose home doc has
lost all sponsorship? I would assume that links with LinkEnds in
said doc would cause it to pass some dependency criteria and hence
be archived, but what if said doc is merely home to a link with
LinkEnds elsewhere?
Strike! The interaction of links and archiving is an open-ended
nightmare. All my archiving models are archived for the moment, till
I really dive in. I worry occasionally that our current archiving
model (sponsorship, et al) will interact counter-intuitively with
links. If we can comeup with specific ways in which links ought to be
archived, or prevent archiving of documents, that would give us a
perspective from which to view the sponsorship model.
The current story (and one we can make real clear) is that sponsorship
only flows in the forward direction. If sponsors aren't directly
putting money into links or the documents which contain them, then the
link can be archived. Note that if you want to keep a particular
link, you can copy it out and sponsor that document (or with
standalone links, you can (and must) just sponsor it dorectly). One
other way links remain around is that they appear in sponsored
backfollow documents.
Sorry for the disorganized response. If I organize, I'll think about
it too much :-) Your in a good position to invent scenarios because
your less familiar with the proposed mechanism. Any suggestions?
Thanks,
dean