Comments on: Myths about DATallegro, Ingres, open source, etc. http://www.dbms2.com/2006/09/20/myths-datallegro-ingres-open-source/ Choices in data management and analysis Thu, 09 Feb 2012 13:48:12 +0000 hourly 1 http://wordpress.org/?v=3.0.3 By: Stuart Frost http://www.dbms2.com/2006/09/20/myths-datallegro-ingres-open-source/#comment-6769 Stuart Frost Thu, 21 Sep 2006 03:49:40 +0000 http://www.dbms2.com/2006/09/20/myths-datallegro-ingres-open-source/#comment-6769 Thanks Curt, I hadn't heard that particular claim before. As you point out, we d indeed have our own proprietary software that sits outside of the Ingres codebase. I guess I'm a little puzzled as to why it matters all that much to a prospect. If the appliance works, performs extremely well and is available at a compelling price, what difference does it make whether the vendor contributes to open source - except perhaps that it could lower their development costs in the future. We feel we have a very good balance of contributing certain changes back to Ingres Corp where that makes sense. Most, if not all of these changes make it into the open source version. As a result, we get to stay in fairly close step with the open source version. At the same time, our proprietary MPP code is kept out of the open source domain. As for the Ingres optimizer, it does still exist on the individual appliance nodes, but we greatly constrain how it operates. Our own parallel optimizer is a separate layer on top of Ingres. Stuart Thanks Curt, I hadn’t heard that particular claim before. As you point out, we d indeed have our own proprietary software that sits outside of the Ingres codebase.

I guess I’m a little puzzled as to why it matters all that much to a prospect. If the appliance works, performs extremely well and is available at a compelling price, what difference does it make whether the vendor contributes to open source – except perhaps that it could lower their development costs in the future.

We feel we have a very good balance of contributing certain changes back to Ingres Corp where that makes sense. Most, if not all of these changes make it into the open source version. As a result, we get to stay in fairly close step with the open source version.

At the same time, our proprietary MPP code is kept out of the open source domain.

As for the Ingres optimizer, it does still exist on the individual appliance nodes, but we greatly constrain how it operates. Our own parallel optimizer is a separate layer on top of Ingres.

Stuart

]]>