Will we be seeing any of the work that came out of the Google summer of code project and if so what and when? According to the Summer of Code FAQ:
All development must happen in the open and all code must
be made available publicly. You may mirror development on your personal
infrastructure at your option. You must also provide a copy of your code to
Google to be publicly hosted by us.
So where is the code? I haven't seen anything all summer.
I'd also like to make a comment on how this was run. While I noticed that some of the students engaged in seeking input in the forums here, I also noticed that others treated this just as they might have treated a class assignment, working in total isolation. Some of us asked questions or made suggestions on the project pages in the wikis and got absolutely no response back from the students. I didn't notice most of the students here in the forums asking about how things work in Moodle as a whole so as to make sure their contributions fit in with Moodle. Not all seemed to keep updating their wiki page with what they were doing. While they may have improved their coding skills over the summer, it doesn't seem to me that they learned much about the fact that good software developers need to be able to work as part of teams and that soliciting user feedback is a worthwhile part of software development. I would hope that next year the students be required to engage with the Moodle community a little more as I think it would be a worthwhile experience for them.
If you have a recent copy of the development version of Moodle (from CVS HEAD) you'll find that the absolutely marvellous reworking of the admin page (transformed to a block! works really nicely!) is already in there.
You'll also find a prototype of the global search facility in there, although I haven't tried that out (it requires PHP 5 and I'm on a PHP 4 box at the moment).
If you look at the wiki page Student projects you'll find that each project has a "notes" wiki page with lots going on.
Still think this has not been open enough? Google doesn't require active involvement in the Moodle forums, and nor should it. I expect there has been plenty of interaction with the nominated mentors, which is the best way for the students to develop good code without spending all their time in the forums debating the relative merits of SOAP vs XML-RPC (etc etc etc).
You'll also find a prototype of the global search facility in there, although I haven't tried that out (it requires PHP 5 and I'm on a PHP 4 box at the moment).
If you look at the wiki page Student projects you'll find that each project has a "notes" wiki page with lots going on.
Still think this has not been open enough? Google doesn't require active involvement in the Moodle forums, and nor should it. I expect there has been plenty of interaction with the nominated mentors, which is the best way for the students to develop good code without spending all their time in the forums debating the relative merits of SOAP vs XML-RPC (etc etc etc).
Dan Stowell への返信
Re: Question and Comment about Google Summer of Code Projects
- Martin Dougiamas の投稿
Correct, I've been spending a lot of time with these guys and they are overall doing a good job. I would really like to expand Moodle's involvement next year with more of the other developers acting as mentors.
The Database presets are also in CVS, as is the new AJAX course format. They still need some polish but it's all getting there nicely.
The Database presets are also in CVS, as is the new AJAX course format. They still need some polish but it's all getting there nicely.
One quiz-related idea that came up recently might make a really good SoC project. The idea is a quiz report that compares student's answers with each other, and tries to identify possible cheating (answers suspiciously similar, maybe also incorporating information from the log tables too in some way.)
Doing it as a quiz report makes it nicely self-contained, but actually devising (or searching the literature for) some appropraite statistical tests will probably take a couple of months for a really bright student.
Doing it as a quiz report makes it nicely self-contained, but actually devising (or searching the literature for) some appropraite statistical tests will probably take a couple of months for a really bright student.
Steven Levit, an economicist, had an algoritm for something like that. Maybe you can check his book, Freakonomics http://www.freakonomics.com/
Tim > The idea is a quiz report that compares student's answers with each other, and tries to identify possible cheating...
To me, this idea of "identifying possible cheating" seems hardly feasible from a technical point of view and hardly acceptable from an ethical point of view.
<rant>
What's with this obsession with "cheating" anyway? I am not being naïve, but the idea that students can and will cheat does not bother me much. From a number of messages posted in these forums, especially on the Quiz forum, it seems that a number of fellow-teachers have exaggerated expectations of what technology can/should do. They would like the technology to take care of supervising the students, making sure they do not cheat, etc. whereas in fact the surest (human) way to avoid or alleviate cheating is to devise quiz questions which ask intelligent answers of the student. Of course, devising such questions takes far more time, energy and intelligence than producing "dumb" questions which make cheating easy.
</rant>
Joseph
I agree on the ethical point: Teaching students to live in an open society and then mistrust them and observe them.. 1984
Comapring answers can be interesting from another perspectiv:
Why are the students making all the same (set of) mistakes?
- Wrong teaching?
- Difficult topic?
A nice way to identify topics for reteaching?
Hey, my code has been in HEAD since about the 3rd week into Summer of Code. You can check it out at cvs_root/moodle/search 
If I missed any questions on my wiki page, I apologise, I migrated the dicussion to the forum, and a few other developers and I expanded it there (wiki documents are a bit hard to discuss using the wiki software). I will be migrating all the decisions and bits of info back onto the wiki page in a couple of weeks.
As for the "summer" as such - I intend on staying on as developer/maintainer of the search code, there are still some features that need to be fleshed out (permissions, and making a better result processing/caching scheme). I think the 3 month period given to us is relatively short compared to how long some of the other developers have been around, so to jump in and be on good development terms with everyone is a bit tricky - I was quite intimidated in the beginning by the seniors devs
Thanks for your comments, good idea to start this thread!
If I missed any questions on my wiki page, I apologise, I migrated the dicussion to the forum, and a few other developers and I expanded it there (wiki documents are a bit hard to discuss using the wiki software). I will be migrating all the decisions and bits of info back onto the wiki page in a couple of weeks.
As for the "summer" as such - I intend on staying on as developer/maintainer of the search code, there are still some features that need to be fleshed out (permissions, and making a better result processing/caching scheme). I think the 3 month period given to us is relatively short compared to how long some of the other developers have been around, so to jump in and be on good development terms with everyone is a bit tricky - I was quite intimidated in the beginning by the seniors devs
Thanks for your comments, good idea to start this thread!