Advogato: Edgar F. Codd, father of &quorelational database model&quo left us...

Edgar F. Codd, father of &quorelational database model&quo left us...

Posted 29 May 2003 at 04:49 UTC by gilbou

Edgar F. Codd, an IBM computer scientist who created the "relational database model" died of heart failure at home Friday in Williams Island, Fla at age 79.

Edgar F. Codd had published, in 1970, a research paper that would introduce the concept of relational database model:

http://www.acm.org/classics/nov95/s1p5.html

He explained that when presented with existeing data management systems he felt rather disguted by the lack of serious work and theory to have it work properly. Using mathematical logic heavily, he worked on a relational model where all information would be represented by values in rows and columns in tables and no longer pointers and connections between records up to the infinite.

The first relational product was announced in 1981 as SQL/DS. Two years later, DB2 was announced for larger machines.

Codd's death, posted 30 May 2003 at 09:21 UTC by schoen » (Master)

Just to be clear, Codd died on April 18 (last month), not last week. For example, here's a Mercury News obituary for him.

Can't edit :p, posted 30 May 2003 at 15:37 UTC by gilbou » (Journeyer)

Yes I know. I wanted to update the article and fix the &quo things but there's no edit link to fix the posted article... raph, could we have an "edit link" available to the original poster of an article to fix or update article ?

the 'edit link', posted 30 May 2003 at 18:50 UTC by sye » (Journeyer)

Question for gilbou , if the 'original' poster can fix or update article, how will you prevent your 'friends' to delete your article, together with others' contributions to the original article as well? Even if it is a desirable feature from one end user's point of view, i bet it can bring in unnecessary complexity to the code base and possibly some degradation of over all system performance to boot. Besides from the above consideration, i like the fact that on advogato, if one jabbered for instant public attention, his ( also his network link's) shortcomings are here to stay.

Back to the topic of relational model of data, Ted 'defined the key notion of essentiality in:

"Interactive Support for Nonprogrammers: The Relational and Network Approaches," Proc. ACM SIGMOD Worksop on Data Description, Access, and Control, Vol. II, Ann Arbor, Michigan (May 1974)

"This paper was Ted's principal written contribution to "The Great Debate". The Great Debate - the official title was Data Models: Data-Structure-Set vs. Relational . The great contribution of master Codd besides the assertion that "the entire information content of a relational database is represented in one and only one way: namely, as attribute values within tuples within relations" is his activism in promoting one theory of "relational completeness of data base sublanguage". A political consequence of that data model and its sublanguage is that it granted nonprogrammers the holding of an asset that they never had before - namely an environment where information and data can live separately from programming codes and being shared across the boundaries of institutions. Relational data model is a good fit for from simple to complex transactional business processes. The trouble as i see it now is that with the proliferation of that relational data model in banks, in governments, in educational institutions etc. etc, social interactions are cutting off all its edges so they can be fit into relational data model. That is most absurd. Several times, i asked credit card customer service representatives and their managers if he has ANY procedure in place to lower the interest rate for one customer from 24% to 19%, the answer is always: I am sorry but i can't do that for you no matter what. You have to pay 24% interest rate for another 6 month before you can call us back to have the system re-evaluate your credit rating.

Never had anyone asked the inventor: have you regretted letting out the genie? And i know the answer is we must break the bottle to think outside the box.

the 'edit link', posted 30 May 2003 at 18:50 UTC by sye » (Journeyer)

Question for gilbou , if the 'original' poster can fix or update article, how will you prevent your 'friends' to delete your article, together with others' contributions to the original article as well? Even if it is a desirable feature from one end user's point of view, i bet it can bring in unnecessary complexity to the code base and possibly some degradation of over all system performance to boot. Besides from the above consideration, i like the fact that on advogato, if one jabbered for instant public attention, his ( also his network link's) shortcomings are here to stay.

Back to the topic of relational model of data, Ted 'defined the key notion of essentiality in:

"Interactive Support for Nonprogrammers: The Relational and Network Approaches," Proc. ACM SIGMOD Worksop on Data Description, Access, and Control, Vol. II, Ann Arbor, Michigan (May 1974)

"This paper was Ted's principal written contribution to "The Great Debate". The Great Debate - the official title was Data Models: Data-Structure-Set vs. Relational . The great contribution of master Codd besides the assertion that "the entire information content of a relational database is represented in one and only one way: namely, as attribute values within tuples within relations" is his activism in promoting one theory of "relational completeness of data base sublanguage". A political consequence of that data model and its sublanguage is that it granted nonprogrammers the holding of an asset that they never had before - namely an environment where information and data can live separately from programming codes and being shared across the boundaries of institutions. Relational data model is a good fit for from simple to complex transactional business processes. The trouble as i see it now is that with the proliferation of that relational data model in banks, in governments, in educational institutions etc. etc, social interactions are cutting off all its edges so they can be fit into relational data model. That is most absurd. Several times, i asked credit card customer service representatives and their managers if he has ANY procedure in place to lower the interest rate for one customer from 24% to 19%, the answer is always: I am sorry but i can't do that for you no matter what. You have to pay 24% interest rate for another 6 month before you can call us back to have the system re-evaluate your credit rating.

Never had anyone asked the inventor: have you regretted letting out the genie? And i know the answer is we must break the bottle to think outside the box.

pandore box, posted 30 May 2003 at 21:24 UTC by gilbou » (Journeyer)

>Question for gilbou , if the 'original' poster can fix >or update article, how will you prevent your 'friends' >to delete your article, together with others' contributions >to the original article as well?
If only the poster of the news can edit his news entry, how would "friends" edit or delete that entry ?
If all the poster can edit is his news why would that let him edit or delete other posters news ?
The "great debate" is indeed interesting content. Reading stuff about it :-)
Being able to edit your (and only *your*) entry would fix that double post. Or did you that on purpose ? ;-)

It's a problem of automation, not relational data, posted 31 May 2003 at 03:57 UTC by johnnyb » (Journeyer)

"The trouble as i see it now is that with the proliferation of that relational data model in banks, in governments, in educational institutions etc. etc, social interactions are cutting off all its edges so they can be fit into relational data model."
This is a general problem of automation, and has nothing specific to do with relational databases. In fact, relational databases are probably the least susceptible to this problem. The problem is that people make processes that do not have flexibility. It takes longer to make something that allows for exceptions rather than something that doesn't. Therefore, programmers don't. It really has nothing to do with relational databases and everything to do with automation and the costs of doing it right.

I don't get why people object to the ability to edit, posted 31 May 2003 at 05:41 UTC by jbuck » (Master)

Advogato looks like crap because people can't edit. There are so many duplicate posts and other goofs. If the worry is that people will damage the historical record, do as some other sites do and have a time limit: a user can modify or delete an article or comment for up to, say, 30 minutes after it's posted, then it stands. That way people can remove duplicate posts and minor goofs not caught during the preview.
As for "complicating the code base", I thought that some of claimed to be Masters.

just to be clear ..., posted 31 May 2003 at 05:43 UTC by jbuck » (Master)

Under my proposal, a user could modify or delete his/her own article for 1/2 hour, not someone else's.

Indeed, posted 31 May 2003 at 09:13 UTC by gilbou » (Journeyer)

Yes, a small amount of time would allow entry to be fixed for errors. The poster should only be able to edit his entries only.

Again, i state it is unnecessary as we already have those features..., posted 31 May 2003 at 13:48 UTC by sye » (Journeyer)

The "preview" button before submitting an article, draft your posting as a diary entry first etc... suppose serveral people are about to submit their original articles at the same time, for a proposed feature such as "a user could modify or delete his/her own article for 1/2 hour", the trade off is that no other person can submit article during that time period. The reason that one can edit diary entries is because your own account acts like an indexing for all your entries. There is no such point of reference for the front page articles besides the time it was submitted. Of course, raph can implement a buffer zone to queue all drafts of articles and let people vote or edit that queue, sort of a grand community diary that everyone are allowed to have access to it. IMHO, such polishing of draft is actually taking away resource and time that ought to be devoted for sharing ideas and individual unique accent of speech.