[Theforum] [---Dev] RE: [---tent] Article cleanup issue

.jeff jeff at members.evolt.org
Tue Jul 23 17:36:16 CDT 2002


martin,

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> From: Martin Burns
>
> > well, it's been that way as far back as the code that
> > walker set us up with.
>
> OK, but I'm not sure that "It's always been that way"
> is the right reason for it still being there
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

agreed, but that sort of change requires a decision by committee, something we're not real good at around here.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > for one i think it's useful because you may know
> > approximately when an article was published with
> > respect to when another article was published.  so,
> > you go back through the pages of approved articles
> > from the homepage, page by page until you find it.
>
> I'd question how many people actually do that. I really
> suspect that you'd page through the 1st page or 3 and
> then search.
>
> It works for maybe 10 pages, but for 100? I really
> wouldn't have thought so.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

agreed.  i don't personally use it that way, but i suspect some do (though not very many pages deep, for sure).

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> Can we grep the log for accesses to page x of articles?
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

i'm the wrong person to ask that though i can help supply the strings to look for if necessary.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > it also is a really quick indicator to our visitors of
> > how many articles are currently approved and available.
>
> Excuse the SQL idiot here, but wouldn't a simple count
> run quickly?
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

sure, but depending on how you do it that'd end up being an entirely separate query.  creating and tearing down that connection are the slowest things you can do and should be avoided.

at the same time you wouldn't want to run a count() once for every row that's returned from the articles query.  that'd be bad too.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> I have the same questions about the categories too.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

fair enough.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> We don't have a strong taxonomy that people can
> hierarchically browse through - our site really is a
> "view 1st page or 2 then search"
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

except the search sucks so they're stuck browsing for some articles.

fwiw, the changes you're talking about really shouldn't be discussed with regard to how the current cms works.  instead, they should be addressed towards the redesign and how the next version *should* work.

.jeff

http://evolt.org/
jeff at members.evolt.org
http://members.evolt.org/jeff/




More information about the theforum mailing list