Re: SAP and MySQL to cooperate (?!)

Robert Treat wrote:
> On Mon, 2003-05-26 at 06:56, Jean-Michel POURE wrote:
> > Le Samedi 24 Mai 2003 09:53, Ricardo Ryoiti S. Junior a =E9crit :
> > > I think it's all about a marketting move. I don't imagine how SAP
> > > DB and MySQL could be "merged", and developing a new system from scratch
> > > doesn't seem to be plausible. I don't expect much from this at least in
> > > short/mid-term (regarding software, not media hype). Anyway, in two or
> > > three years we may se a new database that possibly will do what Postgre=
> SQL
> > > does now and have been doing for a long time.
> >=20
> > Agreed. MySQL will not merge with SAP-DB.
> >=20
> > Legaly, SAP needs to maitain installed SAP DB databases. Calling SAP BD t=
> he=20
> > next "MySQL version x" will allow SAP to drop SAP DB silently. In other=
> =20
> > words, SAP DB strory may be over.
> >=20
> <snip>
> > Furthermore, SAP-DB story shows the importance of the notion of "communit=
> y". A=20
> > Free Software project cannot survive without a real community of developp=
> ers.
> >=20
>
> It will be interesting to see what happens to the sap-db open source
> community, since ISTM they will be losing a lot of their corporate
> backing. You have to figure sap db will stop development of sap-db
> pretty soon, which means the community of developers that are involved
> will probably have to fork the code to keep working on it. While a
> strong commercial stewardship can work wonders for an open source
> project, theres no doubt that it can also be the death knell when the
> companies aspirations change. Still, I'd welcome any sap-db who want
> to join the postgresql project which can offer some solace in that it
> is not beholden to any one company.

It sure as though development contributions from SAP will be coming to
an end. If the code base were "more approachable," there would be the
possibility of someone else taking it on. But this code appears scarier
than Mozilla was, and it seems unlikely that anyone will want to use it
without being *intensely* committed to it.

>> MySQL is now coming to a cross-road : they can continue owning MySQL
>> completely without external contribution, or start building a
>> community with real CVS access and shared responsabilities.

> So what is the future of INNO-DB? Doesn't this screw the folks
> developing that table type? Will mysql really put forth the effort to
> develop two "real" database engines? Seems unlikely to me. I guess I'd
> also like to extend a welcome for those who are using INNO-DB and
> realize they too may be on the short end of things in the not too
> distant future...

I'd think it likely that if the MySQL people truly are interested in
integrating the code together, they would need all the capable hands
they can get, and the "INNO-DB guy" would doubtless be one of them.

But the process, however it shakes out, is sure to lead to a whole lot
of disruption for users of both database systems.

While it is surely an attractive idea, from a "press release"
perspective, to meld together the "popularity on the web" of MySQL with
the much greater functionality of SAP-DB, actually accomplishing this is
not as easy as slapping them onto the same CD and installing it. If the
developers had some really nifty code analysis tools, it *might* be
imaginable that there could be some shortcut to "make it pretty."

But more realistic is the prospect that the "integrating it all in"
project would be as daunting as "porting Mozilla to use GTK+ rather than
Motif," and the result would be that there wouldn't be _usable_ results
from the project for a considerable period of time.

This may represent, for MySQL AB, a case of "shooting themselves in the
foot;" to admit the need to integrate in SAP-DB-like functionality would
be something of an admission of inadequacy. And if it takes a long time
to get to "adequacy," that's a long term "admission of inadequacy." I
doubt that's what they wanted as a marketing plan for selling licenses
to MySQL Pro...

Those that were using SAP-DB aren't likely to wait for that project to
complete, what with the fairly massive risks.

I would expect them to leap three ways:
-> Those that were feeling a bit "iffy" about the whole "open source"
thing are likely to head back to Oracle or DB/2;
-> Some will doubtless knock on Interbase/Firebird's door;
-> Some will probably come looking at what PostgreSQL has to offer.

A "welcoming approach" would probably involve improving the German
documentation :-).
--
output = ("cbbrowne" "@" "ntlug.org")
http://www.ntlug.org/~cbbrowne/languages.html
"It used to be thought that you could judge someone's character by
looking at the shape of his head. Whether or not this is true for
people, it is generally true of Lisp programs."
-- Paul Graham, _On Lisp_