The presentation was basically an introduction to unit testing with PHPUnit. I've never done any unit testing, and to be honest, probably won't for a while. It sounds like a good idea, but at least for an app the size of IlohaMail, the kinds of bugs that unit tests will discover really don't take much time to debug anyway. What takes up the most amount of time for me are bugs that are tied in with the user's environment, and in such cases, unit test won't help (unless I send them tests which they run).

In general it went well, although only 10-12 people showed up, which was rather unfortunate (there was another presentation ini the PHP track during the same time slot, so I think I lost a lot of people to that --or maybe my topic was just plain boring). The presenter who was in my room before me went over time, so I ended up starting more than 5 minutes late, and had to trim my presentation accordingly. Not that big of a deal though (very few of the presentations I went to actually managed to get through all the material and still finish on time).

This talk was presented by Scott Ritchie, but he had to present material sent to him from Software Improvements. To his dismay, he found out at the last minute that SoftImp was moving away from OpenSource.

Notes from SoftImp's materials:

-apathy vs trust
	-electors don't care about code
	-trust is implicit
	-apathy
		-people don't want to know about flaws
		-people don't know how system works anyway
-paper trails useless because at the end, you'd have to count paper
	-> improve counting method/accuracy instead
-considerations
	-must be provably correct (i.e. not tampered with after deployment)
	-trustworthy developers
	-unreasonable deadlines
	-"controlled open source" 
		-read-only, 
		-only to those authorized, 
		-no copying, 
		-limited reads, self destructs
		-"hidden keys"

Notes from Scott's commentary:

-fork old GPL code?
-in CA, paper trail required (Diebold machines decertified)
-not a technological problem, a procedural, social problem
-What's the social construct that caused these issues?
	-in Australia, voting is mandatory

Talk by Ken Beck (creator of extreme programming, JUnit, and others)

For the most part, his talk was about how to make money while working on Open Source projects. He started by arguing that the Open Source model was unstable because it made no economic sense. He calculated the potential monetary value of his software (in terms of increased productivity, time saved, etc), and showed how what he actually got out of it was practically nothing in comparison.

Personally, I would disagree with his assertion that Open Source make little economic sense. It doesn't make sense in a monetary economy (or a capitalist/market economy), but it makes perfect sense in a reputation economy, and is therefore not inherently a unstable model. When I asked him about that, his reply was "alright, so we're getting paid in toothpicks... but that still doesn't pay my daughter's tuition."

His way of putting it added an interesting twist to my question, because then it would seem as if the Open Source "market," so to speak, is a market that has its own currency (reputation), and the real problem is that our currency has very little value compared to hard currency. That puts this problem back into the realm of a monetary economy, and so the problem we face might be similar to the kinds of problems many developing nations face.

Ultimately, I think the main problem is that our entire market segment, Open Source software and its developers, are currently severely undervalued. I think there's this bias against OSS that's preventing corporations and governments from investing in it, not entirely dissimilar to how developed nations may be reluctant to invest in a rising developing nation that faces considerable threats. When viewed in such a light, I think the ultimate solution is for the OSS world to make it clear that, 1) OSS is comparable, if not superior, in quality compared to commercial software, 2) OSS projects and developers need funding, 3) funding OSS projects has long-term benefits.

At the end, I doubt much will change, and we'll be left to beg, steal, or scavenge just to keep writing software that people can use. I think it's entirely possible that the real inherent flaw is in capitalism, but that's just my inner pinko commie speaking.
Show Rest of Post