I think that the group needs to be aware at least of possible feature
creep problems. While it may not be a problem in a non-commercial
setting like this, it can at least cause diversions in the mailing
list. I have been guilty of causing this myself in another group, and
so am now more careful about comments that stray from the primary focus.
Having said that, I would like to throw out an idea. Keep in mind I am
not a programmer, so I may be way off base here. But I have often seen
problems with web sites having broken links. My idea is that a web site
with links would not directly embed the external link in the document,
would rather do it through a link table that is maintained external to
the document.
The link table would have the external URL, plus other optional info,
such as who maintains the URL you are linking to. This way you would
have some resources to track down links that disappear. It would also
make it easy to go through the link list periodically and verify that
all the links still work. Maybe this is so easy to do with something
like perl that a separate list is not necessary?
Anyway, I had the idea of having both an incoming and an outgoing link
list. By maintaining an incoming link list, you can move your own stuff
around, and if you keep your incoming link list up to date, you will not
break links to you.
On the down side, I think web browsers would balk at this, I imagine
they don't want to have to go through intermediate lists to get to a
destination site. However, there might be some way this concept could
be useful.
Sorry for the diversion.
Dave Morgan,
_______________________________________________
Genes-devel mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/mailman/listinfo/genes-devel
