Cargill: food and … intelligence gathering

The WSJ has an article “Cargill’s Inside View Helps It Buck Downturn” with good insight into Cargill’s commodities trading. They collect and use an unprecidented amount of information on global events that will affect food prices, leading to success in commodities trading even in poor market conditions. A great example of a modern intelligence gathering operation in action. And a reminder that fancy analytics can be beat with good old knowledge.

A story from the DBMS wars

In 1982, George Byron Koch did a highly detailed technical review of many vendors of a shiny new kind of software: database management systems. Actually, DBMSs had been around, in one form or another, for a while at that point. But there had recently been an explosion of DBMS products on the market. The short story of this event, in the preface to the Oracle Complete Reference, recalls just how far the DBMS market has come since the time when the technology was as shiny and new as the compact disc.

If Google Books doesn’t give you the linked page on the first try, click refresh and you should get it.

The Modelers’ Hippocratic Oath

An insightful modeler’s manifesto on Wilmott.com that can be applied well beyond finance models:

MODELERS OF ALL MARKETS, UNITE! You have nothing to lose but your illusions.

The Modelers’ Hippocratic Oath

~ I will remember that I didn’t make the world, and it doesn’t satisfy my equations.

~ Though I will use models boldly to estimate value, I will not be overly impressed by mathematics.

~ I will never sacrifice reality for elegance without explaining why I have done so.

~ Nor will I give the people who use my model false comfort about its accuracy.
Instead, I will make explicit its assumptions and oversights.

~ I understand that my work may have enormous effects on society and the economy,
many of them beyond my comprehension.

This Manifesto can also be downloaded here (for logged in members of wilmott.com)

CEP is a marketing device. So what does that say about CEP products?

Following up on the recent post on Seth Grimes blog about how vendors of “Complex Event Processing” do not really care about the term CEP – they use it just as a brand to attract your attention, there is no usable underlying “theory of CEP” that all these vendors are following.

So what does this say about CEP products? Not much.

Most vendors of “CEP” products are selling a tool that that provides a language and run-time that can be used to code and run applications which process data in real-time. Maybe that data is messages from an ESB, stock ticks, web clicks – whatever. And each vendor has a different approach. Each tool is different, so each tool might be great for one kind of problem and terrible for another.

If you ever believed in the need for software to ease the development of soft real-time applications – well then that need would be there whether or not it’s called CEP. And if you never believed in the need for this kind of thing, then you should never have been interested in CEP products to begin with.

Bottom line: if these vendors can fill a fundamental software need (and IMO some of them do), then they will sell software and make money. If you have a need for soft real-time processing, go see if CEP vendors have some software that you’d like to buy. Don’t ever buy anything just because of a technology acronym, CEP included.

A classification scheme for CEP technologies? Ha.

While I’m thinking about it, I wanted to comment on something Paul Vincent said about 2009 CEP predictions.

Quoting Paul:

…the CEP vendor community, in the guise of something like the EPTS, probably needs to step in with a classification scheme for CEP technologies, providing certain prerequisites for the term CEP over “technologies that can do some aspects of CEP”.

Never. Gonna. Happen.

Why? Because CEP is just a brand to 90% of the EPTS (see my words on Seth Grimes’ blog). No vendor will be able to resist the urge to use any classification initiative to show how they are, being clearly in the best class, the most CEP product in the history of CEP.

The best part about this prediction is that if I’m wrong, I will be happy!

It’s funny to watch how the tug of war around this “shared brand” of CEP is disguised in technical terms.

Not to degrade the efforts of the EPTS. They have a positive direction. But let’s just say that I don’t think a classification scheme will come about in 2009. Or ever.

Guest post on Seth Grimes’ blog

Leave it to me to jot down a quick post to a Yahoo group that is more compelling and clear than the posts that actually make their way onto my blog.

Anyway, Seth Grimes reposts some comments of mine in an article on his Intelligent Enterprise blog here. Thanks Seth.

Readers who are interested in the CEP-as-brand discussion might like this other post of mine on the topic.