Thread: Re: [pgsql-advocacy] PostgreSQL.org Design Proposal

> -----Original Message-----
> From: pgsql-www-owner@postgresql.org
> [mailto:pgsql-www-owner@postgresql.org] On Behalf Of Mitch Pirtle
> Sent: 28 October 2004 14:41
> To: PostgreSQL www; PostgreSQL advocacy
> Subject: Re: [pgsql-advocacy] [pgsql-www] PostgreSQL.org
> Design Proposal
>
> I'm one of the people that enthusiastically offered to help,
> and have since jumped to 'lurker' status.
>
> There's a couple of reasons for this, and I know enough
> people to be confident that this is a common theme.  Most
> folks willing to help out here are of a technical nature, and
> when these people find out the new site is a homegrown effort
> they lose interest immediately.  Why in the world would
> someone want to build their own when there are hundreds,
> HUNDREDS of portal tooklits and CMS to choose from?  To an
> experienced developer this is like slowly removing your
> fingernails with a dull razor that was just dipped in fresh
> lemon juice.

We looked at a number of solutions - and found none that met our
requirements for easy translation and mirroring of virtually all
content. We cannot replicate database backed content to our mirror
network - that problem alone discounted the vast majority of CMS's out
there.

> Second, it seems there are a couple people that insist that
> everything is going fine, and that (more or less) everything
> must be done their way(TM).

I assume you mean me and a few other people? Well yes, most of us have
been here a long time, are heavily involved in not just the website but
all the infrastructure, and most importantly have been involved in all
the dicussions of the various technical issues we face. So indeed, I
think it's fairly safe to say that it should be done our way unless
someone else with a similar understanding of the overall picture has a
better solution.

In particular though, we are *not* receptive to suggestions regarding
decisions made long ago - for example, we are not going to consider
using a new framework, when 95% of the existing development is complete.

>   There are other people that are
> convinced that nothing is getting done, and appear so
> desperate that they will take anything that they can get.
> The disparity to me (a relative outsider) indicates that
> whatever effort I put into this is at risk of either being
> duplicated or dropped altogether before ever seeing daylight.

Not at all, if you work on the 'official' project.

> I joined the Mambo CMS team beause I realized it was an
> interesting-but-pointless exercise creating my own CMS, but
> mainly to also wire in support for Postgres ;-)  Drupal, and
> others could also fit the bill.  I just can't see that the
> website for a database is so special that it requires a
> ground-up approach of cobbled-together libraries and proprietary code.

Err, yeah. Ever heard of PHP & Pear?

> Third, the 'web presence' of the Postgres community as a
> whole is a mess.  Gborg, pgfoundry, www, advocacy, how many
> other sites are there?  And in what state?  Sheesh, this is
> as bad as python. (slaps
> forehead)

Precisely my point - we know this, and its part of the longterm plan.
Catch up on the previous discussions first, and then you'll be in a
better position to object to doing it 'our way (TM)'.

> We just had someone submit a graphical layout proposal for
> the new site (which I considered brilliant), hoping to help.
> They were
> immediately told 'thanks, but no thanks'.   ?

Yes, because we (pgsql-www), and the core have accepted a design
already, and we (in particular Alexey) been working to bring it live. I
assume you have looked at the project site and have seen all the commits
to the CVS tree? What would you have us do - start over every time
another nice design came in?

> I would love to help out, but after a couple weeks of reading
> the threads I'm not sure how to help, or if it is really
> welcome.  Who is in charge, what is the plan, and how can I
> (realistically) help?

This is open source - there is no 'in charge'. However, the technical
lead in the www project is Alexey. Marc Fournier, Robert Treat, Devrim
Gunduz and I look after most of the day-to-day web *and* other
infrastructure, however that is mainly operational stuff - we do little
actual coding on the websites.

If you want to know what needs doing, then you only need look back in
this thread - there's been the urls of the test site and it's todo list
posted.

Regards, Dave

On Thu, 28 Oct 2004, Dave Page wrote:

>> Third, the 'web presence' of the Postgres community as a
>> whole is a mess.  Gborg, pgfoundry, www, advocacy, how many
>> other sites are there?  And in what state?  Sheesh, this is
>> as bad as python. (slaps
>> forehead)
>
> Precisely my point - we know this, and its part of the longterm plan.
> Catch up on the previous discussions first, and then you'll be in a
> better position to object to doing it 'our way (TM)'.

Stupid question ... do we have somewhere a layout of the long term plan?
For instance, eventually we want to merge Gborg -> PgFoundry, so that we
have one projects related site ... I believe(?) advocacy is meant to merge
into www itself?  And there has been talk of better integrating developer,
at least in so far as layout?

----
Marc G. Fournier           Hub.Org Networking Services (http://www.hub.org)
Email: scrappy@hub.org           Yahoo!: yscrappy              ICQ: 7615664

Dave,

Thanks for the background.  I've been on this list for two months now
and have not seen this information - and you explain the needs well.

Some quick explanations, the easiest ones first:

I'm not pimping Mambo.  Currently it would not support what you need
anyway - it has just recently been integrated with ADOdb to support
postgres, and i18n hasn't had the time to get integrated with the new
core.

II am clearly offering to help (and not just complain), and wanting to
understand if my efforts would actually be used if I did.  I'm also
busy working on other FOSS projects, with significant time involved.
So if I spend the additional time away from my wife and kids then I
want to know it was worth it - that's not too much to ask, is it?

Again, I am not complaining, I am pointing out some pretty alarming
issues (even if my observations are/were incorrect).  Dave pointing
out several things has made a much bigger impact on my opinion than
petty insults and misguided sarcasm.  Yes, I've heard of PHP, if you
had been doing this (solely web application development, with either
php or python) for over a decade then I'm certain you would know me by
now...

My objection to a do-it-yourself approach is that anything beyond PEAR
and PHP is proprietary.  That means that you wrote it; and as such
then you have to support it, and you have to document it, and you have
to improve it, and you have to upgrade it to keep compatibility with
changes to HTML_QuickForm and DB and so on...  With a CMS, you'd
typically be using a finished system that had been developed by dozens
of people, with years of experience, with a lifespan that supports
itself (free upgrades).  Such a waste, IMHO.

I agree that mirroring is a huge problem, and anything less than the
heavyweight systems (Zope/Plone, for example) will have major issues.
Mirroring a dynamic site in general is a major issue, and switching to
a homegrown one just adds to the complexity and effort, no?

And I wasn't being rude IMHO, I've already picked on Dave for his
browsing preferences.  He says he has a huge monitor, but surfs the
web like he is on a Mac Plus.  Having a design that supports 800x600
to support the handicapped and comply with accessibility laws is great
in my book.  I just think Dave is being a weirdo for using such a
little browser window ;-) Sorry that wasn't so apparent in my previous
email.

So we are 95% there.  It definitely makes sense to get the new site up
before making any drastic changes, and give that approach time to
prove its mettle.  Knowing the amount of effort that has been
invested, and why things were done the way they were, I'm happy
helping where I can and not trying to change direction.  My intent was
never to disparage or discourage, just to clarify some things that are
not apparent to someone that has been lurking for the last two months.

-- Mitch