(no subject)

[dr_tectonic]

Spent all day today (between dropping off and picking up my car for maintenance, which required me to get up far too early) at a seminar about software engineering at our institution.

I enjoyed it, possibly more than I expected to. What I took away from it is that I'm doing pretty well with my development process, given that I'm operating as a team of one. The points people brought up were things that I had identified as important, and nothing really leapt out at me as a revelation.

Mostly, I think, it was just kinda fun to hang out with people that were all on generally the same wavelength.

Star Wars this evening; we surprised Jeff by deciding not to fly the skiff over the mountains, or follow the droids on foot through the mine tunnel, but to follow the droids through the tunnel in the skiff. (Yay for 6-meter wide borer tunnels.)

Comments

[nehrlich.livejournal.com]

Because I'm a cult member with no thoughts of my own, I point you to the Joel test if you're interested in tips about the software development process.

[navrins]

Wow. Wish I'd had that link a few years ago.

[dr-tectonic.livejournal.com]

I'm pretty sure I've read that before because, hey, it's Joel On Software. Having reviewed it, I'm happy to say that I can answer 'yes' to all of the questions that are meaningful for my situation.

[melted-snowball.livejournal.com]

Wow. I think 4 of those questions are actually relevant to my work. But I tend to get them all pretty close to yes...

[nehrlich.livejournal.com]

Yeah, it's definitely more tilted towards "if you are doing commercial shrinkwrap software development", but there's some good points there. As usual, he's evangelistic and single-minded in his belief that the way he wants to do things is the one true way, but that's how you make a good cult leader :)

[melted-snowball.livejournal.com]

Sure--and I've long since accepted that rules for software design generally don't apply to people designing and prototyping mathematical algorithms. Really, the top 10 commandments all boil down to "prototyping is good", "rapid prototyping is better" and more of the same.

[nehrlich.livejournal.com]

Oh, totally. I used to work on developing software for new instrumentation, so everything was all about prototyping. This is my first experience at a commercial software company, so it's a bit weird. Although I still think more prototyping is better, in pretty much everything.

[melted-snowball.livejournal.com]

My usual objection to thinking of life in this way is that commitment is a great thing (both in relationships and as a way of having that wonderful experience of being an expert at something, which is just an awesome thing).

[nehrlich.livejournal.com]

Yeah. I think I'm still 12 years old at heart. As mentioned in another Beemer post, I don't feel like I'm anywhere near an adult yet.

[melted-snowball.livejournal.com]

For better or worse, it's a hell of a lot easier to do my job if I am, so I seem to have become one. I'm even dressing more like one, which scares me.

[melted-snowball.livejournal.com]

I find most things that people say about how the software development process works irrelevant to my life as a researcher writing software (even, sometimes, reasonably large software) as a way of testing out and changing my own ideas.

[dr-tectonic.livejournal.com]

I think we avoided that, because basically none of the Software Engineers here are CS majors; it's all scientists who became programmers through the side door, many of whom are writing software as a way to do exploratory research.

[melted-snowball.livejournal.com]

Huh. Then that sounds very interesting indeed; I never get to see talks like that. Any summaries/content I can look at?