This post originated from an RSS feed registered with Java Buzz
by Russell Beattie.
Original Post: Coworker's New Blog and O/R Mappers
Feed Title: Russell Beattie Notebook
Feed URL: http://www.russellbeattie.com/notebook/rss.jsp?q=java,code,mobile
Feed Description: My online notebook with thoughts, comments, links and more.
Heh. I'm blogging this from work - one of my coworkers named Chris at my current contract has just started up a weblog. It was actually thanks in part to the fact that I'm at this job because of my weblog and the fact that we share a similar background - we're both techies who went to college for Journalism (though he actually went on to get his Masters). He's been thinking about starting one for a while (isn't everyone?) but I gave him a little nudge. And here's another. :-)
Chris just posted his third post which is about Java O/R Mappings and I figured I'd give him a little link love as incentive to keep it up. A quote:
I was telling a coworker about my current database persistence engine, Solarmetric's Kodo JDO implementation, and received just
a bit of a suspicious look. The suspicion quickly turned to astonishment when I told him the reason I use an O-R mapper
in the first place, which is speedy development. I think I may have used the word 'agile' in a moment of weakness, though I know
it's bad form to drop buzzwords like that in a friendly discussion.
Anyway, he correctly pointed out that JDO adds yet an extra layer (and a pretty thick one at that), slowing down the build and
adding more complexity.
...
One thing about using O-R mapping tools: Developers who can't easily compile and run Java programs don't much like the idea that the database should be controlled through objects rather than directly with SQL. I can't really blame them, but the impatience around the office has been building lately. My boss, the Skipper, spotted the O'Reilly Java Data Objects book on my desk, pointed at it, and exclaimed: "The Devil!"
Heh. That would be *me* ranting about the complexity and OR mappers above and happily I seem to be in with good company here (the boss). LONG LIVE SQL! Like I've said before on this weblog, for every layer between you and your data you better have a *damn good reason* for it to be there. Otherwise it just leads to useless complexity and bugs.
Go leave him a comment on his weblog and tell him how much you agree/disagree. You know, I've been evangelizing weblogging quite a bit since I arrived. I need to write up a short "Russell's Guide to Weblogging" or something...
Previous Topic
