Monday, May 05, 2008

Stovepipe Specialties

3 on a match - a trifecta of Ninja goodness!

Interestingly, I have just read a post that states the writer has never heard another programmer utter a phrase that I know would be well-worn to anyone reading the Pragmatic Programmers first book.

Kenneth Down, in his May 4th post in "The Database Programmer" blog, states that he has never, in 15 years of programming, heard another programmer say anything like the phrase "Minimize code, maximize data". Yet this has been a well-known adage from the aforementioned book, as part of their general manifesto. And their book was published in 1999!

Now, while I have not read Mr. Down's blog long enough to know much about him, the fact that the blog is "The Database Programmer" lends credence to the idea that it's the narrowness of his field that has kept this technique from common use. Database programmers may not be reading the same books as the general programmer population.

If this is the case, we need to break down the stovepipe specialities that are keeping us programmers from using the knowledge our bretheren mine from the field.



Technorati Tags --
, , ,
HTTP

2 comments:

KenDowns said...

Hi folks. I should have clarified. The idea is not popular in the blogosphere, and no colleague I have worked with has ever heard it. Or perhaps I should say common practice indicates nobody knows it? Anyway, my boss gave me a copy of The Pragmatic Programmer, a great book, but that made it no easier to sell data-centric ideas. I would further clarify and continue to claim that the idea is almost if not completely unknown to the younger generation, especially those heavily influenced by ORM and who have only done web work.

Dixie Software Ninja said...

Ken,
You win the "Frist post" prize for the inaugural comment on this blog!

I do find it odd that you are seeing little evidence that programmers are taking the data-driven approach. Now, it is difficult to extract all the data-y goodness from a program and make it minimal, but at least some data-driven work should be knwon