I liked most of the tips in this post: Ten common database design mistakes

I especially agree w/ the poster about the foils of using a single master domain table. Been there, done that, learned my lesson. I don’t agree w/ his recommendation that you use stored procedures where ever possible though. I know this is the party line out of Redmond, but I like to put as much of the application logic in the code & make things in the database simple and obvious. I think it makes things easier to debug, test, and generally maintain.

Overall I thought the list was great though, so I wanted to save a link to it here and recommend it. Enjoy.

I saw this page on the ETL integrator wiki & thought it was a great template to use when doing an initial write-up for a project. See: http://wiki.open-esb.java.net/Wiki.jsp?page=ETLSEOnePager

I might try to use something similar for my next side project. Perhaps w/ some enhancements it could even be used as the template for presenting a project to a architecture review board @ work.

I don’t generally like to post just links, but I thought this article on digital copyright law was worth reading for people interesting in the music industry transformation and history of ownership with respect to property. Enjoy.

History suggests copyright crusade is a lost cause

I just finished Producing Open Source Software by Karl Fogel. As the name would suggest, most of the time is spent on issues specifically relevant to open source. Often not at all relevant to teams that are not distributed and don’t communicate primarily via email and irc. For instance, it spends a decent amount of time on how to “manage” the mailing list. As in, coordinate your efforts behind the scenes and then present a consistent and unified face to the rest of the team on the mailing list. Interesting and reasonable for open source development, especially with corporate sponsorship, but not particularly applicable outside of that context.

I did find one particular piece of advice interesting. The suggestion that code ownership is detrimental, and that ‘author’ tags in the source should be discouraged. Most of us have probably struggled with this in the past; certainly I have. The obvious question, which Karl brings up, is related to when you add or remove author tags based on the amount of ‘ownership’ said author has maintained. The code usually changes quite a bit over time & the changes are made by many people other than the original author. How or when an author gets added or removed is not at all clear. Removing the issue entirely, by getting rid of the ‘author’ tags, would also foster more of a group ownership mentality. The only downside is that sometimes the code documentation isn’t sufficient to understand what is going on. (i know, i know, bad…. but it happens) In those types of cases it’s nice to have some sort of contact who can clear up any confusion. This is also a case where the balance between the pros and cons may be different for distributed teams working on open source software versus centralized teams working on proprietary software. But even for centralized teams SCCS logs may give you enough to go on to find the right contact. I think I will try this suggestion for my next project and see how it works out.

Overall a good book: Well written, an easy read, and some ideas to think about. What more could you ask for?

I play w/ persistence apis like some people play w/ languages. Yesterday I ran across a post about the persist project. In addition to being into orm tools, i’m super into _simple_ orm tools. The persist library certainly fits that bill. I played w/ it last night, and liked it. I still need to think about circumstances where I’ll use it though. Hibernate is deeply entrenched in many organizations where i work already and I have a pre-existing favorite library for unit test scaffolding (jdbi) which supports running ddl from a file. I find the sql file support helpful since I typically maintain all the create * statements in files in the source tree anyway for use during deployments. And given their existence it’s nice to re-use them to build and drop the tables during unit tests.

Anyway, kudos to jcamaia for a job well done.

Historically i’ve been a desktop computer guy. Until recently, I rarely traveled and wanted the extra power & lower cost of a desktop. But i’ve been spoiled by a nice fast laptop and I doubt I will ever go back. It’s so nice being able to haul it outside on nice days, write blog posts in front of the tv @ night, and otherwise be mobile. I whole heartedly agree w/ this computer world article. Laptops are going to replace desktops for nearly everything except servers in the next few years.

Retirment articles often seem like they say the same things over & over. Still, this one is as good as any of them. http://www.schwabinsights.com/2007_07/life.html

I thought our walking score would be really bad because we live in the ‘burbs, but it turns out to be a 66. (still not good, but better than i expected) I think it must be because there is a small grocery store near our house. Other addresses I tried that have a ton more places to eat & shop close had lower scores because they didn’t have a nearby grocery store. Just my theory…

Try it & let me know what you get.

http://www.walkscore.com/index.shtml

I ran across a post singing the praises of wordpress, so I thought I better give it a shot & see what I’m missing. This will be a personal weblog for me, as I have a work weblog elsewhere.