The role of magma on the Pharo

[Miguel Enrique Cobá Martinez]

El mié, 12-08-2009 a las 14:39 +1000, Stuart Herring escribió:
> 2009/8/12 Miguel Enrique Cobá Martinez <miguel.coba at gmail.com>:
> >
> > This is my personal view/opinion.
> >
> > Magma it is an *impressive* piece of software.
> > Magma it has more potential in the web arena than in the desktop arena.
> > Magma hasn't exploited all the potential it has.
> > Magma hasn't been marketed correctly.
> > Magma needs good documentation.
> > Magma needs to show off what it can do.
> > Magma needs to easy the seaside integration.
> > Magma needs a good looking/professional website.
> > Magma should provide alternatives to integration with seaside and the
> > pros/cons of each one.
> >
> > Now, what can we do in order to give Magma a relevant position in the
> > seaside/pharo/smalltalk arena?
> >
> > I could help with several of these points, but Chris, you must lead this
> > effort. It shouldn't appear tho the potential or real contributors
> > (fixes, addons, helpers, whatever) that you only mind of Magma and
> > nothing at all about every else that is not Magma.
> >
> I think it's a little unfair to say that Chris _must_ do anything.

I have never said that he *must* do anything besides to *lead*. This
means something as basic as a line stating what he would like to see
Magma trying to aim for. Or maybe stating that he will agree to have a
facelift to the magma project (but that he doesn't have
time/energy/whatever, that's ok). Or in general that he won't let the
efforts to run alone (answering questions/supporting as he always kindly
does) for years before the volunteers knew that it was not the vision
that he has for Magma and they lost the initiative.


> I think that if you really want all those things, then the best thing
> to do is to start doing it, submit what you have for discussion, and
> then see where to go from there.

Yes, that is what I am trying to do. I have offered to host and write
the site and documentation (at least part and subject to review), but I
don't want to end in the position of Keith that after several years of
work, it learned that is was pushing its own independent way, when he
intended to add value to Magma in an integrated way.

> It's easy to sit around saying someone needs to do something, but the
> best way to get it done, is to just do it.
> That's the approach I'm taking with my Pharo port of Magma (about
> which I attempted to send an email to the list earlier, but
> unfortunately did so from the wrong email address so it never got
> there - I'll re send it soon).

Well I'm trying to do my part too. I just wanted to start a discussion
not a flame war.

> 
> Coordinating a list of things that long - even if other people are
> actually doing the work - takes an enormous amount of effort, and
> Chris is just one man.  Personally I'd rather he continued doing what
> he enjoys doing, as that's what's most likely to ensure he continues
> to work on Magma in to the future.

Of course, he, as we all, do this for pleasure and joy, not for money or
by pressure. I don't ask Chris that work extra hours, just that he marks
a direction/goal/way. Then the people that agree with him, will try to
do it if it wants and have free time.

> Its up to the rest of us in the community to step up to take ownership
> of the issues we believe are important, and to not wait for someone
> else to organise things.
> 
> To address some of the items individually:
> > Magma needs good documentation.
> The best way to improve documentation is to describe exactly where its
> lacking.  I'd say create a page on the wiki where you can list the
> things that you think need more explanation, that way anyone who knows
> can help fill in the gaps.  The biggest problem with documentation is
> often knowing what to document - so a list of specific points or
> questions can go a long way towards improving it.

The format of the wiki of squeak it is really ugly. We all know. I want
to use a better platform, I don't care if it is smalltalk or not. Just a
better platform for the documentation. And integrated. And that don't
use *numbers* for the name of pages! 

> 
> > Magma needs to show off what it can do.
> Again, a wiki page with links to examples or real-world usage would be
> the best way to do that.

Yes a single demo in the site will help to sell this technology (I am
not speaking about $$$) to new users.

> 
> > Magma needs to easy the seaside integration.
> If the existing information for the Magma seasideHelper, and the older
> Magma-Seaside stuff isn't enough, then maybe someone can contribute
> some more examples of how to integrate it?
> In general Chris is right - there really isn't much required to
> integrate them, it's no different to using any database system with
> any other web platform.  You need to get a connection, manage
> sessions, and handle transactions. If seasideHelper doesn't fit your
> requirements, writing something yourself isn't much effort, but maybe
> there could be some general high-level examples of the sorts of things
> that need to be done.
> This particular point is very reminiscent of the way that people keep
> asking for Persistence options on the Seaside list.  Seaside-Magma
> integration really is in the realms of an external project - like
> Magma seasideHelper.  Both the Seaside and Magma projects can point to
> it as a way to provide the integration, but it's really not in the
> scope of either of them to actually maintain it.

Yes, because the Magma/Seaside integration it is very confuse in the
current state. If we, *we*, can improve this, a lot less questions will
appear on the seaside list and magma will be the first option to
persistance on seaside. A position that Gemstone is trying to take right
now. 

> 
> > Magma needs a good looking/professional website.
> That's nice-to-have, but the Magma pages on the wiki are currently
> better than a lot of projects' dedicated websites.  Having a nice
> looking website won't make Magma easier to use, as It's the content
> that matters the most.

Of course I agree. but the wiki page doesn't invite to contribute. It
looks ugly, it doesn't register the user that contribute (an important
ego-related point).

> 
> > Magma should provide alternatives to integration with seaside and the
> > pros/cons of each one.
> That's again just a case of improving the wiki page a bit.
> Really at the moment the alternatives are:
> * Use the old Magma-Seaside
> * Use Magama seasideHelper
> * Write your own solution.

I'm not asking for a new alternative. Just documenting the existing ones
so that new users can use it correctly.

> You could probably get an idea of the pros and cons of each of those
> approaches by going back through the Seaside and Magma lists, as the
> topics have come up a few times.
Yes, maybe I could, but what about someone that for the first time
arrives to smalltalk and to object databases. They want to know a list
of pros/cons without reading the complete archives of magma, seaside,
pharo and squeak list just for evaluating which alternative to go for.

As a last point, I didn't want to be critic on Chris, but I wanted to
improve the visibility of Magma (and Pharo and Seaside) by having a easy
platform to build webapps as GLASS is doing for the Gemstone folks.

I strongly believe that 

Pharo + Magma + Seaside

can be a *real* alternative to GLASS.

Cheers

-- 
Miguel Cobá
http://miguel.leugim.com.mx