[CLUE-Admin] Site Maintenance

[Jed S. Baer]

On Tue, 12 Oct 2004 06:34:55 -0600
Jeff Cann <president at clue.denver.co.us> wrote:

> I think we should talk about it again.  The reasons Jed gave last year
> was essentially that MUG was 'huge' - [2.7 MB] download w/1200 files and
> that it's not a 'clean' distro.

And, taking a look again (no I didn't download the whole thing, I am on
dialup -- just quickly browsed the CVS tree), I appears that what we'd get
is a site that looks pretty much like BYU's. Yeah, we can replace the top
banner, change the stylesheet, etc. But making real layout changes -- is
that going to mean modifying their php and html?

I we're going to go with a CMS, why not find one that's designed to be
customized? i.e., one which uses templates and plug-in modules to create
blocks within the layout we choose.

Our site focus has always been to have the meeting info be the main
content. Do we want to change that? Can the MUG main page be easily
modified to conform to what we want our site to look like?

Our poll results:
http://clue.denver.co.us/pipermail/clue-dev/2003-April/000180.html

> At that time in Apr '03, I thought OK, then let's code the features we
> want.  I also had the same 'code it ourselves sentiment' when I was the
> web master and didn't build anything for 3 years [the current web site
> is from 1999, BTW].  I did write a vague design doc, but no points for
> no code...

And what are the features we want? Shouldn't we address that before
deciding that MUG is the solution?

I know some things I want. But when has the discussion on CLUE-Dev
occurred where people have commented on the new design documents?

> Now, it's been another year and a half since we last broached this topic
> (since Apr 03) and we haven't gotten any new features on to the CLUE web
> site.  I do mean 'we', not Jed, BTW since no one in CLUE is helping Jed
> and like the rest of us, Jed doesn't have 20+ hours a week to write new
> code for CLUE.

Well, in point of fact, I have, and still have time. That isn't the issue.
The issue is maintaining momentum. More participation would help. Now that
Collins is on board, I have some fresh hope. I have a few more ideas, but
I'm not going to be able to maintain enthusiasm by myself.

> The original motivation of creating a dynamically written web site was
> to make updates and events easier.  So easy that the webmaster (me, now
> Jed) wouldn't have to touch his keyboard to update the CLUE web site.

And that's still the primary focus. But I will add that, although at times
tedious, continuing the manual updates hasn't been exactly a strain.

> Now, with members, it also would be convenient to give them a profile 
> automatically.  We also had Paul Billie providing a static calendar.  
> We do not have a calendar that we can maintain.  I also set up polling
> software which we used once - another manual effort.

And this interacts with the work being done by Todd Eason how?

I agree that we need a well integrated set of utilities for maintenance of
the server, including the web site and memberships. How do those two
things interact. I've asked this question before.

And how, too, does MUG interact with our membership model? (Logins and
'create an account' on the main page?)

>  All in all, our current operation is much more labor intensive and
>  there are 
> *still* no member pages (this was supposed to be Lynn's responsiblity,
> but he was not able to do that either).

http://clue.denver.co.us/~eraugust/ for example?

The question is management. Again, Todd's work? Further question: what do
want with member pages?

> Since MUG has features we've wanted to implement, such as member logins,
> member pages/profiles, polls, calendar, RSS feed, it seems like we
> should just use it.  This would also make migration to the new server a
> snap, since we just install MUG there.  We can copy the 'old' web site
> and provide links to it if necessary [email archives will be copied].

What's the purpose of member logins? IIRC, when we talked about
memberships last year, we were thinking members could update their CLUE
homepages by logging in via a restricted shell. Lynn set this up.

Would we really use polls?

Migration to the new server is already just a matter of moving the
repository and issuing a CVS command. Actually, this sounds easier than
adding old links to MUG.

> Finally Jed said, 'The primary goal, of course, is to make things as
> easy on myself as possible. ;-)'.
>
> It seems that an already-built user group software package would make it
> easy on Jed - the admins could post directly to the web site, instead of
> Jed doing it by hand.  Jed has much more time to fulfil his dream of
> being Donald Trump's protege :D  [or whatever Jed dreams about] 

Easy is good. So is maintainable, in terms of code base. So is the ability
to make the site be what we really need it to be, especially where that
diverges from being a clone of BYU.

I note that the MUG sourceforge page mug.sourceforge.net list "3" MUG
sites. One is BYU, one is the official demo site, and one is a site that's
currently running GeekLog.

Some other things that have come to mind for the CLUE site:
1) Schwag list. Acknoledgment of sponsors donations. There used to be one
(I think I found the page once, can't find it now) Would be maintained by
the sponsor coordinator, could have links to sponsors websites. Or, just a
way for the Sponsor Coordinator to maintain the sponsors page.

2) Ability to create 1-time logins for people to use to upload their
presentation slides.

3) Maintenance of news items.

4) Web interface for Membership Coordinator to maintain membership roster.

Other stuff pops into my head from time to time -- usually when I'm not
prepared to right it down.

But really, the main point is to establish what we really want/need from
the site.

Collins and I are planning on getting together next week to discuss
further progress on what I've already started. I still believe it's the
best way to go. It isn't really hard, it's more just a matter of getting
some traction, and then not losing it again.

I'm much more interested in tools, such as paTemplate, which will make the
work easier, than a one-size-(doesn't)fits-all pile of code.

jed
-- 
http://s88369986.onlinehome.us/freedomsight/

... it is poor civic hygiene to install technologies that could someday
facilitate a police state. -- Bruce Schneier