I finally gave the word to my friend, Charles Fry, to cut me loose from hosting nephi.org so I can take it over once more. I have to publically thank the guy for being so generous and hosting my site for what was probably three years or so, free of charge for almost the entire time.
The only problem that comes with someone else taking care of your baby is that you tend to neglect it, and now that I’m taking full responsibility for it again (as in, paying for hosting myself), it will probably motivate me to get to work on it.
There’s a lot of good projects and ideas that I’ve had for the site for years now, and I have a bad habit of putting personal projects off for an extremely long time. I’m getting better at it though, and hopefully soon there will be something to show for it.
Case in point, it was exactly a year ago that I setup trac and started filing bugs for my development tracker for MDP. I really need to get to work, and finish up these things I started so long ago.
I was reading Patrick’s blog this morning, and he mentioned something about QA in general that got me thinking. “I think most Open Source projects could use a code freeze and a long bugfix-only phase.” I have to agree with him 100%. Gaim is a perfect example of that, I think. Back in the day when they had a release every few weeks, they would always fix some bugs, but add new features and those would introduce new bugs as well. I kept thinking to myself that I wished someone would just freaking stop adding new features and instead just fix the bugs, and have a stable bugfix release cycle.
Then I got started thinking about my projects as well. Oftentimes the reason I get side tracked from working on them is because I have in my head all these features I want to add, which can be some genuine hard work and dedication. You really gotta get in the mood to work on stuff like that, and that’s hard to come by. However, if instead I just focused for a while on fixing the things I already promised, and half implemented, and are broken and annoying and not working properly, it would make me feel a lot better just getting that part done and done, and I’m sure the users would appreciate it too (GPNL comes to mind … why the crap don’t I have a search button on every page?!).
So, I think that’s what I’m going to do. I got pinged this morning on the LDS OSS mailing list because my contact email form was broken on nephi.org (and probably was for about, oh, three years), asking about a DB abstraction layer for the scriptures exports. I replied that I’d be getting to it in the next release, but I’d also be doing all this whiz-bang stuff to add new features like views and foreign keys and relationships and fix the data integrity and yada-yada-boom-boom. I’ve got a great idea. How about you *fix* what is broken instead, and keep the same state of features, but just cleaned up and working properly. That is much easier, and a much shorter task list than taking over the world. I tell you what, I think I can manage that.