Re: Database Selection

On 25/4/2006 6:47, "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> Shane Ambler <pgsql(at)007Marketing(dot)com> writes:
>> If you look at the history of PostgreSQL you will find that the development
>> at Berkley started with Ingres and after the code was used to start
>> Relational Technologies/Ingres Corporation the Postgres project was born.
>> A later version of Postgres was used by Illustra Information Technologies
>> which later merged into Informix and is now owned by IBM.
>> This gives you some some indication of the quality - two commercial
>> databases have been started with PostgreSQL code.
>
> Actually that's a misstatement --- AFAIK, Stonebraker and crew started
> from scratch when they wrote Postgres, because they wanted to experiment
> with a new system design based on what they'd learned while writing
> Ingres. So there's no code shared between Ingres and Postgres, and
> probably not much design commonality either, other than having sprung
> from largely the same group of people.

Bruce Momjiam say's "PostgreSQL's ancestor was Ingres" but I haven't found
anything concrete one way or the other on whether Postgres started from
scratch or from Ingres code
(the name comes from 'post' 'gres' - after Ingres)

> Illustra/Informix, on the other hand, is indeed a fork of Postgres.
> I don't know how similar that code base now is to ours, though. There's
> been enough time for pretty substantial divergence on both sides of the
> fork.

True but if the early code was good enough to make a commercial product
(regarded as one of the first commercially successful relational databases)
then it gives you some indication of the quality the project started from.
Starting from a bad bug ridden beginning could carry problems through to
today's version.