From amk at amk.ca Tue Sep 12 19:45:40 2006 From: amk at amk.ca (A.M. Kuchling) Date: Tue, 12 Sep 2006 13:45:40 -0400 Subject: [Soc2006] Producing a final report Message-ID: <20060912174540.GA13857@rogue.amk.ca> I haven't seen any discussion or announcement of the results of the PSF-mentored SoC projects. Which projects succeeded, and what did they build? Does someone want to work on a wiki page with the results? --amk From jimjjewett at gmail.com Tue Sep 12 22:35:11 2006 From: jimjjewett at gmail.com (Jim Jewett) Date: Tue, 12 Sep 2006 16:35:11 -0400 Subject: [Soc2006] final evaluations needed; better tracking Message-ID: http://code.google.com/soc/psf/open.html shows 4 final evaluations still needed. (Mailman, Mozilla plugins, Neural Nets, and pypy/ctypes). I hope this is just an oversight, but ... I know I procrastinated because the pay/no-pay decision was not easy. I wouldn't be surprised if at least some of those mentors were also less than thrilled. And since I can't see the other surveys, I don't know how many of the final-submitted said either failures or marginal. I assume that the two students dropped at the midterm were not successful. What can the PSF do differently next year? My suggestions, which may well have problems of their own... (1) Having a common mailing list for status, that all students should write to at least once/week. Hopefully, writing their own public assessments would make it harder to get into an "I don't understand, and don't want to admit it, maybe things will be easier tomorrow" loop. A common blog location (such as "everyone post status to planet-soc at least weekly") might work even better, but I think it should be something more "permanent" than a wiki, and more public than just talking to the mentor. (2) Is there something in common between the successful projects? The unsuccessful ones? This is particularly hard to judge since we can see all the applications, but can't easily see results for more than one student each. But maybe someone else saw things that I didn't. -jJ From l.oluyede at gmail.com Tue Sep 12 22:45:17 2006 From: l.oluyede at gmail.com (Lawrence Oluyede) Date: Tue, 12 Sep 2006 22:45:17 +0200 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: References: Message-ID: <9eebf5740609121345o160c08e2y6e147abb86495b52@mail.gmail.com> On 9/12/06, Jim Jewett wrote: > http://code.google.com/soc/psf/open.html shows 4 final evaluations still needed. > > (Mailman, Mozilla plugins, Neural Nets, and pypy/ctypes). > > I hope this is just an oversight, but ... I know I procrastinated > because the pay/no-pay decision was not easy. I wouldn't be surprised > if at least some of those mentors were also less than thrilled. My mentor didn't complete the evaluation yet because of illness. I'm the owner of pypy/ctypes project -- Lawrence http://www.oluyede.org/blog From jimjjewett at gmail.com Tue Sep 12 22:49:09 2006 From: jimjjewett at gmail.com (Jim Jewett) Date: Tue, 12 Sep 2006 16:49:09 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <9eebf5740609121345o160c08e2y6e147abb86495b52@mail.gmail.com> References: <9eebf5740609121345o160c08e2y6e147abb86495b52@mail.gmail.com> Message-ID: My apologies -- I sent this to the wrong list. oh, for the cancel button of trn... On 9/12/06, Lawrence Oluyede wrote: > On 9/12/06, Jim Jewett wrote: > > http://code.google.com/soc/psf/open.html shows 4 final evaluations still needed. From barry at python.org Tue Sep 12 23:15:58 2006 From: barry at python.org (Barry Warsaw) Date: Tue, 12 Sep 2006 17:15:58 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: References: Message-ID: <755FB50A-A8AA-4E0B-890A-E935BE880622@python.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sep 12, 2006, at 4:35 PM, Jim Jewett wrote: > http://code.google.com/soc/psf/open.html shows 4 final evaluations > still needed. > > (Mailman, Mozilla plugins, Neural Nets, and pypy/ctypes). > > I hope this is just an oversight, but ... In my case, the evaluation questions were sent to my gmail account, which I never read. ;}. Brett pinged me today, forwarded the questionnaire to my python.org address and I just sent it back to him. - -Barry -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (Darwin) iQCVAwUBRQcjjnEjvBPtnXfVAQK15AQAnau+dJUI8f5zK6DuTWtLHkiEpo2ZXnrd fwVU+3wNzRly2SZt3jlGbBnZP7vHw+svTvZdXL9VmZNIj4VTXqsWRp8PP8HfoxIM R3hwZ3oc1ldmtOIZEuDzzLE8v9n4lUvPknrgNMA4l/Fy6C4TZAPndYatGueSGaOt Y6DMjuJK8E0= =r0Pm -----END PGP SIGNATURE----- From i at mindlace.net Wed Sep 13 07:17:03 2006 From: i at mindlace.net (emf) Date: Wed, 13 Sep 2006 01:17:03 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: References: Message-ID: <4507944F.3010103@mindlace.net> Jim Jewett wrote: > I hope this is just an oversight, but ... I know I procrastinated > because the pay/no-pay decision was not easy. I wouldn't be surprised > if at least some of those mentors were also less than thrilled. I have no real idea, but I admit that I was less than thrilled with my own output; as much as I'd like to lay the credit at the feet of five-zillion-and-one web frameworks, I think I'm partially responsible. > What can the PSF do differently next year? I never heard from the PSF. Perhaps if the PSF wants the students selected to succeed more, they should communicate with them. > (1) Having a common mailing list for status, that all students should > write to at least once/week. I think this is a grand idea, if people write back; my wiki posts received no feedback. (I received voluminous and excellent feedback on the Mailman list.) Anyway, I think maybe a meta-mentor who checks in very frequently - daily or every other day - and just see what's up with the student(s) would be a good thing. I'd volunteer for that next summer, if the PSF decides to do something like it. ~ethan fremen From arc at xiph.org Wed Sep 13 08:40:54 2006 From: arc at xiph.org (Arc Riley) Date: Tue, 12 Sep 2006 23:40:54 -0700 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <4507944F.3010103@mindlace.net> References: <4507944F.3010103@mindlace.net> Message-ID: <20060913064054.GF19751@xiph.org> On Wed, Sep 13, 2006 at 01:17:03AM -0400, emf wrote: > > Anyway, I think maybe a meta-mentor who checks in very frequently - > daily or every other day - and just see what's up with the student(s) > would be a good thing. I got a better idea. Have a requirement be that all mentors and students be present on IRC while they work, checking in at least daily, with a bot tracking activity. Of course, I think every student should have two mentors - not just a backup, but a 2nd who's tracking the students progress etc. That way there's some redundancy in place. From j.lacour at gmail.com Wed Sep 13 16:38:52 2006 From: j.lacour at gmail.com (Jonathan LaCour) Date: Wed, 13 Sep 2006 10:38:52 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <20060913064054.GF19751@xiph.org> References: <4507944F.3010103@mindlace.net> <20060913064054.GF19751@xiph.org> Message-ID: [emf wrote]: > Anyway, I think maybe a meta-mentor who checks in very frequently - > daily or every other day - and just see what's up with the student(s) > would be a good thing. [And Arc Riley followed up]: > I got a better idea. Have a requirement be that all mentors and > students be present on IRC while they work, checking in at least > daily, > with a bot tracking activity. Call me crazy, but this seems totally excessive. The student should be able to act like a grown-up if they are planning on participating in the Summer of Code. This isn't babysitting, and these students should be able to stay on track without an excessive amount of supervision. If a student is not motivated and disciplined enough to do their job with a reasonable amount of interaction (1-2 times a week) with their mentor, then it indicates that they should be dropped from the program and not receive payment. It doesn't indicate that they need to be watched like a hawk and have their hand held the rest of the way. I look at this as an opportunity for a student to learn how to be a responsible member of a development team, and coddling them is not going to help to teach them responsibility and discipline. Just my 2 cents... -- Jonathan LaCour http://cleverdevil.org From i at mindlace.net Fri Sep 29 01:24:41 2006 From: i at mindlace.net (emf) Date: Thu, 28 Sep 2006 19:24:41 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: References: <4507944F.3010103@mindlace.net> <20060913064054.GF19751@xiph.org> Message-ID: <451C59B9.20007@mindlace.net> Jonathan LaCour wrote: > and these students > should be able to stay on track without an excessive amount of > supervision. There is no "track" as far as I can tell. I spent a large amount of time in the brambles. Perhaps other SoC projects were so clearly demarcated they were 'just' a matter of filling in the code. Mine certainly seems like it; redo the web interface of Mailman. I still found myself at many points needing guidance, and I definitely wasted a lot of time that could have been avoided with some interaction of more knowledgeable people. Looking back on it, I probably should have been posting more to the python list and other places like that, but instead I tried to find my own answer. That's generally a good principle in the open source world, but less sound when time is short and there's expertise available. > I look at this as an opportunity for a student to learn how to be a > responsible member of a development team, While I think your view of SoC is itself flawed, let's take your premise for the moment. I've worked in two firms that did software development; in neither of those did communication happen once or twice a week between team members. I suspect that interacting practically every day is more the norm for software development teams. > and coddling them is not > going to help to teach them responsibility and discipline. I hadn't realized it was the Summer of Responsibility and Discipline. Since the PSF has a finite amount of slots, once it has accepted proposals it appears to me that it's in everyone's best interest to maximize the amount of Code produced during the summer. I feel like I would have been more productive if I could have had a place to talk with other python-coders interested in the summer of code; some sort of water-cooler environment where I could ask quick questions and that sort of thing. immature-yet-full-of-sloth-ly yours, ~ethan fremen From amk at amk.ca Fri Sep 29 14:28:51 2006 From: amk at amk.ca (A.M. Kuchling) Date: Fri, 29 Sep 2006 08:28:51 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <451C59B9.20007@mindlace.net> References: <4507944F.3010103@mindlace.net> <20060913064054.GF19751@xiph.org> <451C59B9.20007@mindlace.net> Message-ID: <20060929122851.GC4884@localhost.localdomain> On Thu, Sep 28, 2006 at 07:24:41PM -0400, emf wrote: > I feel like I would have been more productive if I could have had a > place to talk with other python-coders interested in the summer of code; > some sort of water-cooler environment where I could ask quick questions > and that sort of thing. Any forum that Google sets up for SoC will not attract very many people and therefore the odds of getting any questions answered would be poor. The place to ask questions would be on the existing mailing lists/newsgroups/IRC channels, where the people with the relevant experience *already* hang out. --amk From jimjjewett at gmail.com Fri Sep 29 14:32:26 2006 From: jimjjewett at gmail.com (Jim Jewett) Date: Fri, 29 Sep 2006 08:32:26 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <451C59B9.20007@mindlace.net> References: <4507944F.3010103@mindlace.net> <20060913064054.GF19751@xiph.org> <451C59B9.20007@mindlace.net> Message-ID: On 9/28/06, emf wrote: > Jonathan LaCour wrote: > > and these students should be able to stay on track without > > an excessive amount of supervision. > There is no "track" as far as I can tell. That is typical for research projects, and not atypical for all software development. Surprising, but common. > > I look at this as an opportunity for a student to learn how to be a > > responsible member of a development team, > > and coddling them is not > > going to help to teach them responsibility and discipline. > Since the PSF has a finite amount of slots, once it has accepted > proposals it appears to me that it's in everyone's best interest to > maximize the amount of Code produced during the summer. Yes and no. If I had to choose between more code and more learning, I would choose learning; I believe google supports this priority, which is one reason they funded student interns rather than generic bug bounties. That said, for a given student, more/better code and learning are highly correlated. > I feel like I would have been more productive if I could have had a > place to talk with other python-coders interested in the summer of code; > some sort of water-cooler environment where I could ask quick questions > and that sort of thing. This is clearly something we need to address. I think some of the smaller organizations did a better job, in part because the focus was tighter. The connection between mailman and pypy and pycells will always be weak, but I think we (PSF) could do better than we did. -jJ From i at mindlace.net Fri Sep 29 15:18:40 2006 From: i at mindlace.net (emf) Date: Fri, 29 Sep 2006 09:18:40 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <20060929122851.GC4884@localhost.localdomain> References: <4507944F.3010103@mindlace.net> <20060913064054.GF19751@xiph.org> <451C59B9.20007@mindlace.net> <20060929122851.GC4884@localhost.localdomain> Message-ID: <451D1D30.8040002@mindlace.net> A.M. Kuchling wrote: > Any forum that Google sets up for SoC will not attract very many > people and therefore the odds of getting any questions answered would > be poor. The place to ask questions would be on the existing mailing > lists/newsgroups/IRC channels, where the people with the relevant > experience *already* hang out. I mostly was thinking of the PSF's communication with students, rather than Google's. I did use existing channels, though in retrospect I probably should have used c.l.py more. There's a difference, to me, between a channel/list where all the people on it are agreed to help out the students and the general cant of OSS fora, where underinformed or vague questions are often left to lie (at best). Irrespective of the consensus on the above, it would be nice for "communication from the PSF to mentored students" to surpass nothing. An orientation email from the PSF at the beginning of the projects indicating quality lists/newsgroups/irc channels for various python matters would be a great start. ~ethan fremen From jonathan-lists at cleverdevil.org Fri Sep 29 22:16:39 2006 From: jonathan-lists at cleverdevil.org (Jonathan LaCour) Date: Fri, 29 Sep 2006 16:16:39 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <451C59B9.20007@mindlace.net> References: <4507944F.3010103@mindlace.net> <20060913064054.GF19751@xiph.org> <451C59B9.20007@mindlace.net> Message-ID: <1A9B8AB3-4E1A-4589-8D76-70444A0DF749@cleverdevil.org> emf wrote: > While I think your view of SoC is itself flawed, let's take your > premise for the moment. I've worked in two firms that did software > development; in neither of those did communication happen once > or twice a week between team members. I suspect that interacting > practically every day is more the norm for software development teams. Interacting with your team members is certainly commonplace. But the role of a mentor is not generally one of writing code. Its that of offering oversight, guidance, and advice. Thats what "mentoring" means. I expect the students to be interacting on a very routine basis with the development team for their project. I just don't see the need for an hour-long discussion with the mentor every day. In my years out in the "real world" I certainly don't find myself being pestered by my boss several times a day. Rather, we talk about project status on a fairly routine basis a few times a week, and if I need any assistance or am having any issues, I contact him. I can't think of a job that I have had where my boss has had the time, or the desire, to hold my hand through an entire project. In fact, the best bosses I have had have left me alone as much as possible! > I hadn't realized it was the Summer of Responsibility and Discipline. Clearly we disagree on the benefits of the Summer of Code. If more of our students *and* mentors were responsible and disciplined, then more of the projects would have been successful. I hope that mentors and students improve in this respect each year. Sure, lots of code is produced in the Summer of Code. But, the number of active, responsible, and experienced open source contributors that are generated as a result of SoC is much more valuable. I think its a great thing to see young, energetic students grow into young, energetic, reliable contributors to open source projects. This is very exciting indeed! Once a student departs the Summer of Code, I would hope that they have all the tools they need to independently continue as a member of the project's community, and possibly even mentor another student down the road. > Since the PSF has a finite amount of slots, once it has accepted > proposals it appears to me that it's in everyone's best interest to > maximize the amount of Code produced during the summer. Maximum code is not necessarily maximum benefit. See above. > I feel like I would have been more productive if I could have had a > place to talk with other python-coders interested in the summer of > code; some sort of water-cooler environment where I could ask quick > questions and that sort of thing. This is a great idea, and I think we are on the same page here. I am not trying to say that the student should be locked in a room and expected to figure everything out on their own. I am just saying that I don't think that the mentor should have to watch their student like a hawk. Life just doesn't work that way. That being said, your point here is really good. Organizations would be wise to provide a mailing list, forum, or IRC channel to facilitate communication. agreeing-and-disagreeing-ly yours, -- Jonathan LaCour http://cleverdevil.org From jonathan-lists at cleverdevil.org Fri Sep 29 22:20:03 2006 From: jonathan-lists at cleverdevil.org (Jonathan LaCour) Date: Fri, 29 Sep 2006 16:20:03 -0400 Subject: [Soc2006] final evaluations needed; better tracking In-Reply-To: <451D1D30.8040002@mindlace.net> References: <4507944F.3010103@mindlace.net> <20060913064054.GF19751@xiph.org> <451C59B9.20007@mindlace.net> <20060929122851.GC4884@localhost.localdomain> <451D1D30.8040002@mindlace.net> Message-ID: <3477A410-626A-4B2A-9301-18B39DA1D588@cleverdevil.org> emf wrote: > An orientation email from the PSF at the beginning of the projects > indicating quality lists/newsgroups/irc channels for various python > matters would be a great start. +1. I doesn't even have to be a new IRC channel, mailing list, or forum necessarily. There are already a host of great places to find help, including c.l.python, python-dev, etc. This email should probably also include references to project-specific mailing lists to encourage the student to reach out to their specific community, whether it be CPython (python-dev), Zope (zope-dev), etc. -- Jonathan LaCour http://cleverdevil.org