[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Addressbook Database
From:       Andreas Pour <pour () mieterra ! com>
Date:       1999-04-08 3:26:14
[Download RAW message or body]

Peter Harvey wrote:

> > Hmm, from what I have seen ODBC seems a bit much for the average user. For
> > example, I only have 20, maybe 30 addresses I want to store. To do this, I
> > would need to download a driver manager, a driver, mySQL, kab (or what
> > have you) and then compile/install/configure all of them! At that point I
> > think I would rather just store them in a text file written with vi! I
> > think we need to come up with a simpler solution. It might be nice to have
> > ODBC support in KDE, but it seems like a lot of over kill for an address
> > book.
> >
>
> If the usefullness for ODBC were to be evaluated one application at a time then
> many people would probably find it a bit of overkill. When evaluated with all
> of KDE in mind (many apps can use it) then it becomes much more usefull.

Most importantly, how is one supposed to write serious office/etc. applications if
there is not a database available?  It is ridiculous for every program to write
its own database.  If someone wants to write a credit card processing software for
Linux, should they write their own database for that?  Or a law firm's biling
system?  Or an inventory tracking system?   Etc., etc.  A database is an essential
part of a programmer's API.  ODBC, for better or worse, is becoming the standard;
just like CORBA is becoming the standard for distributed objects.  Just like a
programmer'modern s API should assume the availability of a distributed object
broker like CORBA, so it should assume the availability of a database interface
like ODBC.

This is about much more than an addressbook.  It is quite true that for most
address books you don't need a database API.  The reason this discussion is coming
up, at least from my point of view, is that the addressbook database is being
discussed, and it would be nice to agree on standards before everyone goes off and
writes yet another database that will work only for one particular program.  The
advantage of Open Source is supposed to be reusability of components, no?

Of course, each programmer can do what he/she wants.  But, I do not see how
KDE can ever grow in importance in the enterprise/business market if there is not
a default database API available for programs.  While the back-end for the API can
be linked to a flat text-based file reader or what-not, the possibility must exist
to plug a heavy-weight DB into the architecture so that businesses which actually
process lots of data can use KDE apps.

For example, how large do you think a medium-sized companies' address book is?  It
can have millions of records.  If kab started off using an API that could tap into
MySQL and PostGRESQL, then these large address databases could be used with kab as
well.  This is just by way of example, obviously each volunteer does what he/she
wants :-).

Regards,

Andreas Pour


> Add to this the idea that KDE, just like the O/S it is based upon, should be
> scalable... I personally keep many more addresses than 30 and my Wife (the
> ultimate test) keeps thousands of addresses (source for contruction materials
> etc).
>
> When address books get this large a work group tends to want to share them.
> Again, ODBC is usefull for this.
>
> And this is just the address book example...
>
> Peter Harvey

[prev in list] [next in list] [prev in thread] [next in thread]