Re: gnome-db/TOAD/gASQL (was Re: [gnome-db]Access-like prkect)

[Rodrigo Moya <rodrigo gnome-db org>]

> 
> >>etc) instead of 2 with redundant functionality, or with stuff
> >>supported in one and not supported in the other one
> >>so, should I shut up or is there something interesting in what I
> >>say?
> >No, I think this is worth thinking about.
> >When I started gASQL (before I knew about GNOME-DB), I had the same need
> >as you (and I still have). I decided to use GNOME-DB because it offers a solid
> >foundation and allows to connect to many databases in the same way.
> >About a possible merging, I agree with you about the fact that we should
> >not duplicate our work. I've thought about the respective positions of gASQL 
> >and GNOME-DB, and I think the two products are complmentary. Let me explain:
> >* about the archutecture: for the two, libgda is the common engine
> 
> >From the user's point of view GDA = ODBC,  a generic database connectivity
> layer.  But from my understanding GDA != GNOME-DB, yes?
> 
yes, GDA = ODBC

> >* AFAIK, GNOME-DB is composed of some low level widgets (to manage libgda 
> >connections, to connect to a database, a tabular view of a recordset, etc), 
> >and of some higher level applications like gnomedb-fe and the rolodex (which I
> >believe does not evolve a lot anymore). gASQL uses the GNOME-DB "low level 
> >widget" and is also a high level application (like gnomedb-fe).
> 
> My conception is of GNOME-DB as providing a GUI configuration process, a very
> simple tool for blasting out arbitrary SQL statements,  and a widget set (much
> like Borland provides for its development platforms).  Under the theory that a
> great DB-aware widget set + good development tools (GLADE, etc...) will yield
> abundant and good database applications.  I think this formula has worked very
> well for that other operating system.
> 
that's a very good point for concentrating on the GDA engine and the
widget set,
and just have gnomedb-fe be a tool for making things quickly.

> >* now, about gnomedb-fe and gASQL: in my point of view, the two do not
> >cover the same area: I see more gASQL like an "acces like" application, and 
> >gnomedb-fe like a swiss army knife for the databases. They do however have 
> >some functionnalities in common. I do use gnomedb-fe sometimes when I want 
> >to do some things without all the constrains gASQL imposes.
> 
> I don't view them as competing.  GDA + GNOME-DB provides infrastructure. 
> gnomedb-fe = "ODBC Driver Manager" with a few enhancements (SQL editor  /
> executor).  I can see someone wanting to run a GNOME-DB app (say an accounting
> package) and installing GNOME-DB,  using gnomedb-fe to setup the connection,
> installing the GNOME-DB app.  Maybe occasionally using the gnomedb-fe to
> diagnose/fix a few problems now and then,  but otherwise just using the app. 
> Said user may have no intrest in something like gASQL.
>
yes, if it is integrated into GNOME-DB :-) power users will have their
SQL
scripts, and this sort of things, but new users (or users coming from
Windows)
will find very attractive a tool like gASQL
 
> gASQL is an Access/Paradaox/dBase-ish tool.  For people who spend their day
> plowing about inside database, populating tables, searching data, etc...
> 
but this is what worries me. I mean, if you use databases, and install a
DB
app, you may want to just have a SQL execution frontend, but other users
may find this horrible.
And, for example, GNOME-DB is included in GNOME Office (whatever it is),
and when
there is a GNOME Office release, people will install this office suite
with MS Office in mind, and having GNOME-DB limited to power users
(users that know SQL) will maybe make a bad impression of GNOME-DB, and,
maybe of the whole suite. That's why I worry about having
gASQL+GNOMEDB-FE functionality merged

Of course, a solution to this would be to try to push gASQL into GNOME
Office

> >I agree that at the moment gnomedb-fe lacks some functionnalities that
> >gASQL has (or will have): the queries designer, the forms edition mode. 
> >The trouble is that to have these functionnalities, a lot of overhead
> >is needed (keep gASQL informed of the possible datatypes, functions,
> >relationships between tables' fields, etc) which is too much if a user wants 
> >to make a simple SQL query and get a result fast.
> 
> Exactly.  I had not really expected much more in the way of features to be added
> to gnomedb-fe.  Something like gASQL would pick that up, providing forms, a
> report composer, etc... using the widget set that GNOME-DB provides.  And, of
> course, GNOME-DB apps would allow users to execute said reports, etc....
> 
but, for example, Carlos is developing a report engine, and it seems a
nice thing to me to allow users to view their reports from gnome-db, not
by having to install or develop an application. The same happens with
any single feature we may add to libgda: GNOME-DB is the frontend for
GNOME of libgda, so it seems sensible to offer an UI for all the
functionalities. And, if some of those functionalities are also in
gASQL, we're, firstly duplicating effort (even worst in our case, since
Vivien is the gASQL maintainer, but he's also one of the most prolific
contributors in GNOME-DB), and secondly, doing a bad favour to gASQL.

> >So, to summarize, I would say that we need the two applications. However
> >what could be nice is:
> >* show the two applications as part of GNOME-DB: we could say that
> >GNOME-DB is made of two applications that are gnomedb-fe and gASQL and has a
> >GNOME-DB-devel package to  make your own applications using parts of 
> >gnomedb-fe or gASQL
> 
> Yes. I'd almost classify gnomedb-fe as a "utility" rather than an "application".
> 
this is nice for me, only am I thinking about end users, which may be looking
for someting similar to Access/Paradox/dBase in GNOME, and if gnome-db
just lets them use SQL, maybe they'll use another application/desktop,
or even worst, operating system.

This is why, so far, the best solution for me is to integrate gASQL as
it is (with
some review for not having duplicated code/features) as a Bonobo
component in
gnomedb-fe. And, the best solution, for me, would be to integrate it
into the same
source tree (as the component to be loaded by default in gnomedb-fe, and
have the current 'Database' one be renamed to 'SQL' or whatever, to
clearly note that it's for power users that know what SQL is). This way,
we offer all functionality in one shot, in one "killer" app, useful for
both power and end users.

cheers

--
Rodrigo Moya <rodrigo gnome-db org> - <rodrigo ximian com>
http://www.gnome-db.org/ - http://www.ximian.com/