All-

eWeek.com is showing an article talking about MultiValue databases, and mentions Revelation Software and OpenInsight specifically. Written by Charles Barouch, known to many in our community, the article makes a good reference piece and shows that our type of technology is alive and well.

Here's a link:

http://www.eweek.com/c/a/Database/PostRelational-Databases-Flying-Under-the-Radar/?kc=EWKNLINF060408STR1

I'd suggest that you rate the article as well- a highly rated article will let an author publish more, and Chuck is a staunch MultiValue supporter.

Mike Ruane

Read, enjoyed, and 'voted' highly…

This does raise a pet point of mine though - that we constantly undersell the capabilities of the multi value model… the names we use, 'multi value' and 'post relational' being the worst examples of underselling. 'post relational' may have had marketing 'zing' 20 odd years ago, when 'relational' was the most exciting thing in the database marketplace - but relational has no marketing zing at all now, and consequently neither does post relational. Neither of these terms can compete with terminology such as object oriented and XML.

Of course the multi value technology competes with these 'newer' technologies fairly well, but the technology competing effectively is irrelevant when most people never look behind the terminology.

I have been using different terminology for some time - that does compete with 'object oriented' and 'XML' for interest - and gets people looking beyond just the terminology to the technology itself. The terminology I use is 'multi dimensional'. 'Multi dimensional' has enough 'zing' to compete with 'object oriented', and sounds like it is related to 'XML'; and, multi dimensional better describes the capacities of this technology than multi value does anyway (multi value suggesting just one level of embedding, multi dimensional suggesting many levels of embedding).

So, is it possible to start a conversation about using new terminology, instead of the very old 'multi value' and 'post relational' terminology? Our old terminology does not help us at all, and cannot compete with the newer terminology used by our competitors who operate in the limited dimensions arena. I am quite resigned that our technology will never again have any kind of significant market share for as long as we continue to chronically undersell the technology, and when the very names we use for it are a quarter of a century out of date, and have no hope of generating interest over newer terminology, this is an area of chronic undersell.

Read, enjoyed, and 'voted' highly…

I read it too and enjoyed it. Charles always writes good stuff and I appreciate the way he translates our craft into the mainstream. This is also why I get a lot out of my conversations with Bob Carten. Both gentlemen have a unique gift of taking the (so-called) new and innovative and showing how our stuff has been doing pretty much the same thing for years.

I'll take this moment to encourage many on this board to read Spectrum magazine. First, Revelation has a regular article in this publication which is usually very good. Second, many of the other articles deal with general business and IT issues which apply to all MultiValued databases. Charles Barouch is a regular contributor.

Having said the above, I think I was disappointed with the way the article ended. It seemed somewhat abrupt, as if the final remarks were chopped off. I was expecting some kind of "final analysis" to everything that Charles had documented. I also believe there is a type on Page 2. Under the Modernization section, it says, "…relational database vendors are continuing to keep their products up-to-date." I'm sure the word post} was missing from the beginning.

'post relational' may have had marketing 'zing' 20 odd years ago, when 'relational' was the most exciting thing in the database marketplace - but relational has no marketing zing at all now, and consequently neither does post relational.

I don't have an opinion on this one way or another. However, I prefer "extended relational" over "post relational".

Neither of these terms can compete with terminology such as object oriented and XML.

Perhaps, but "MV _NEC OO" and "XML _NEC DataBase". Thus, it wouldn't serve us well to try and use either of these terms.

The terminology I use is 'multi dimensional'.

Yes, been using that term for a number of years myself…when it seemed like a profitable buzz word to use. However, I think it may have been ahead of its time. I find that I spend a lot of time trying to unlearn the typical listener from their traditional (i.e. relational) background. In the end I'm usually met with skepticism and then the questions of data reporting come up…in which case I must admit is an area that we do not do well.

So, is it possible to start a conversation about using new terminology, instead of the very old 'multi value' and 'post relational' terminology?

I'm in favor of this but we have to decide whether we want to extend the conversation with the rest of the MV world or keep it in-house. In truth, I doubt there will be much traction elsewhere. I tried to discuss the issue of dated branding methods with Nathan Rector at the conference where he was introduced as the new owner of International Spectrum. The response was quite underwhelming.

If we go rogue, then we might alienate ourselves from the rest of the MV world…which is something Mike has fought hard to rectify. For better or worse, we are tied at the hip with our cousins.

Just some food for thought.

[email protected]

SRP Computer Solutions, Inc.

View this thread on the Works forum...