RE: Question about gnome-db...

["Olivier Nenert" <onenert condumex com mx>]

Hi RODRIGO :)
(I had already seen my mistake.. I guess it's because I have some friends
here that respond to
Ricardo) (even my father in law !!!) :) disculpame comoquiera...

I'm glad I caught the structure..
Sorry I didn't see the development environment for database apps,
but is it is the final goal, then Bruno was right (at least for how I
understood his question)
because I don't think he really really wants to use access database (of
course ALSO, but probably not mainly) but I understood he really wanted the
environment you are talking about...
(I was puting such an application in the final branch, when it should be the
top)
So, as I was telling to bruno in the next mail,
maybe we'll have some interest in glade...
there is a libglade now, so maybe we can use parts of glade to make a
specific builder.
the libglade is like a "runtime" lib.. I don't know if they plan to make a
"design time" lib that would allow us to manipulate "creations forms"
providing our own code editors and code generators or something like this..
but it would be probably interesting to see that with someone "glade" to
tell us..
Because we would else have the problem of trying to do something in a big
part already done just to integrate it better.
(or I didn't understand what is the development environment you mean, which
as always is very possible)

If I get it right, then it is a very interisting project (it was already),
andeven more exciting because we could share work with glade, so that one
branch of the project would be shared with glade..
Hmm.. in fact that would better be a new branch for the glade project that
we could use for our purpose, but as I don't think anybody else is currently
interested in something like that, we would have to promote the idea to
glade leader project and see which one of us can do something in that way.
(maybe bruno if I understood well his goal)
(I suppose the graphical compiler is something like that, right Bruno ?
so that I shouldn't be too far from the main idea)

>>  I mean, gnome-db by itself wouldn't evolve in a development
>> platform like access. (at least I dont think so... I don't see the
benefit
>> of it).
>
>Well, it will be a development environment for database apps, that what I
was
>talking about in my previous mail (app server stuff).
(I hadn't get it right)

>> (Ricardo.. if I am so wrong I appologize.. that's how I understood it so
>> far)
>
>Rodrigo, not Ricardo. You're not wrong at all, you did understand it very
well.
>The only thing you forgot is the development framework, which will sit on
top of
>all these:

Que verguenza... lo siento de veras.... es ridiculo

>
>|-> gda application interface (CORBA server)
>|->gda builder libs (CORBA client)
>|------> gda builder apps
>
>I havent't started too much on this, so if lots of people blame me because
they
>don't want any development stuff (apart from the basic structure Olivier
has
>very well described), I'll stop what I've done so far. But my idea is to
make
>gnome-db a complete framework for execution of database-oriented apps, some
sort
>of application server.

I don't think thuis can happen the only thing I would see is the point I was
giving to bruno
that I didn't see the interest of it. I was wrong...
At first, i was thinking we'd make all that is need to programers to create
easily database apps,
and I thought the components could be used with glade (if we could get to
solve the multiple-inheritance problem we were talking about) (I think we
have solutions, and we just need to chose the one we like best). So, in this
way i thought there was no need to provide a builder if there is already
one.
But I was forgeting the idea you are talking about can be much more than the
builder...
The application you are talking about needs a builder, but also needs
toprovide all comodities
to specifically work on the database(s).. it would not be an app builder
used to make database apps, but in fact a database apps builder. It makes
sense, I had not seen it.

So.. I didn't think much about the builder problem.. but I am pretty sure we
shouldn't try to make our
own from scratch. glade seems to be advanced and working fine with lot of
support,
so my idea wouldbe to try to use it for our interests.
This way, glade would also grow in a very interisting way.
The idea would be for glade to provide a lib, an api that allow its user to
manipulate widgets
"at design time" glabe-design-lib would handle drag and drop, would provide
special signals,
so that when a component is clicked, we have a chance to call our own code
editor and to generate part of the code, etc...
So.. we would be sure to provide always good building capabilities evolving
at the same pace glade would. (as they *are* into builder stuff, they
probably will have better ideas than us and make some cool improvments to
the interface that we would then inherit for free)
And of course, the big idea for glade is that it maybe also in its own
interest as it modularizes the development. (I have no idea how it is
structured now, maybe it is already like this.. maybe it is not possible to
turn it into this ??).
But the best of it is obviously that anyone would have the same
capabilities...
let's imagine... we use this lib to make a database oriented builder..
it is not the end !!!
many gnoe aplications use guile !!!
for now, it is very based on scripting... but if we had such a
glade-design-lib...
those application could very easily make people not only provide scripts for
their application s,
but whole form-based mini applicartions... that would be great..
gnumeric would get some benefit...
gimp would get some benefit...
anyone:)

oh well I'm geting excited :) it is not so much our concern as it is glade's
what do you think Rodrigo ??
If you agree, I'll contact glade team to tell them about the idea, after
all, it wouldn't take us so much time, and if one of us is willing to help
(I think perhaps Bruno) then maybe they agree too...
the idea makes sense don't you think ?

Cheers...
Olivier.