[theforum] Rebuild progress and this week's actions

Morgan Kelsey morgan at morgankelsey.com
Wed Dec 1 09:15:14 CST 2004


All,

>> > [martin said]:
> We already agreed that we'd strive to drupalise evolt.org by the birthday.
> We've already agreed the significant points of principle where votes are 
> of
> course essential. But my problem is: At what level do we stop having 
> votes?
> On every single line of code? Obviously not, but at some point we have to
> move into operational mode, and I think that time is now.
>
> I put forward a project plan that will deliver by 13 Dec, and proposals 
> for
> the high level functions and design and gave opportunity for
> comment/disagreement on all 3 together with a deadline for doing so. Did I
> boob by not putting 'vote' in the subjects?
>

nobody's calling you a boob martin, i think we all appreciate you taking the
project manager helm. and i think aardvark can make his own waffles.
i don't see us making the deadline with all the details addressed, but i
hope i'm wrong.

>> those of us that survive as freelancers are very busy during the holiday
>> season with our client's demands,
>
> You're not the only one - I have a significant deliverable due on the 
> 17th.
>

note: i said "us"

>> and may not have time to respond to every email.
>
> Sorry, but if you're too busy, you have to let evolt.org get on without 
> you
> and trust the organisation's intelligence to do so. That's been the
> evolt.org standard for a *long* time now, and in a volunteer organisation
> is really the *only* way to deliver.

of course, but right now we seem to be depending mostly on martin.

>> have we agreed the URLs are to remain the same?
>> (i believe we did last year some time)  has there been any progress on
>> *that*?
>
> Ah, now you're into the detailed requirements discussion. Fine. We've not
> touched that level of detail yet. Bring it on. Make proposals.
>

please re-read my article-length dissertation on evolt's pagerank
which came up the last time we discussed a rebuild:
http://lists.evolt.org/theforumarchive/Week-of-Mon-20030811/000298.html

i reassert that it would be a tremendous disservice to past and current
authors to change the current URL structure, which may or may not be
reproducable with mod-rewrite (i'm not qualified to answer)

>> > For this week, we need to:
>> > 1) Prioritise requirements in detail and analyse against what we
> already
>> >   have as standard in drupal and built in the PoC
>>
>> so the PoC is now our new direction? i seem to have missed that vote too.
>> is there anyone else out there who has another idea for a PoC?
>

i was simply giving the rest of the community a chance to chime in,
in case anyone else has been playing with a PoC.

(i must say "PoC" is starting to have the same effect on me that
"smurfy" did in the 80s.)

> Nope, we're not going to do another PoC - we're doing the final site now.
>

yes sir.

> However, we *can* discuss (and there has been discussion on #evolt) the
> next level of detail: whether we just modify the PoC, take bits of it, or
> start from fresh. Bring it on.
>

#evolt is not a satisfactory forum for developers scattered across
different timezones (imho).

>> > 3) Start HTMLising the design
>>
>> this looks to me like a bigger problem than some of us may think.
>> drupal seems to think that <div id="sumthin"><div
>> id="sumthinelse">stuff</div></div> is good HTML/CSS. sure, it may
> validate,
>> but so does lots of other over-written garbage.
>
> Detailed HTML design: bring it on!
>

i just did. not being familiar with drupal, i don't know how much work
it will take for it to produce HTML that is "evolt-worthy".

> The criterion is 'is it good enough to launch?' which may or may not have
> reference to the existing site. With a packaged system, and a much larger
> group of developers, it will undoubtably get much, much better than
> whatever we launch with in the days, weeks and months following.

i strongly disagree. (well, i'm kinda wimpy, but i *can* make strong odors)

that is not criteria sellable to a client, and i think is a slap in the face
with a wet tuna to those that have developed for evolt in the past.
i'm sure none of us tell clients, "don't worry, it will get better....soon!"

if new evolt is not better or comparable to the old by the proposed
launch, we as a community should not launch it.

therefore, i think 3 key requirements missing are:

1. SPEED -- i don't know where or how userfrenzy is hosted
(it could be at john's house which would explain), but the PoC is SLOW.
did anyone else notice the old "8 second delay" went away
when we dropped our DEDICATED linux/oracle setup for a
SHARED win2k/MSSQL one? does anyone want that slug back?

2. URL structure. the current system has a bizarre yet useful page address
system. namely, the random section of the URL which can be
made into anything.
can we reproduce that with mod-rewrite? (i think Garrett is
addressing it). we're not going to have to add a line to htaccess for every
past and future article, are we? or every permutation that currently exists?
YUCK!

3. quality HTML. again, i'm unqualified as an undrupalized individual to
participate, except to say, please view the current source.
drupal left to its own devices makes some really horrible markup.
i dare say dreamweaver/frontpage-horrible kind of markup.

and yes martin, i agree, we can't vote on everything.

these three points shouldn't even *need* a vote, as they seem
to be bred into evolt's bones. yet, they haven't been discussed,
and aren't in the current launch criterion (that i can see).

my....uh....3 cents.


nagrom
http://www.morgankelsey.com



More information about the theforum mailing list