[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: What about non-DB articles?



> LDP
> OSS Web Pages on line (Apache SAMBA, etc)
> man pages (we can use man2html for this)
> /usr/doc (we can use text2html for this)

We use related links.  We may want to expand the table, add the capability to
assign keywords to related_links and so forth.  Then you stick the DNS HOWTO in as
a related link, give it the keywords... and there ya go... never more than one hit.

Basicly... I created the related links table for any "external" documentation,
meaning - not an article under the KB's complete control.

> Another thing I desperately want to do is us host all the data ourselves.

Thats fine.

> Ideally we'd sync once a week to a holding area and have it report on the
> pages that have changed/new.  That way we can go over whats changed and
> weed out broken links etc before migrating it to production.

Well... without having a firm understanding of what your discussing here... I'd
say, if we mirror these sites, there not going to radically change there structure
daily... so we build indexes of the files ourselves as well... then via cron we
compare those indexes to what is in our RelatedLinks Table, and spit the changes
out to be reviewed, have keywords modified, or whatever...

Thoughts on that?

Jason