Date: 2010-03-29 08:53 Security: Public xpost http://soph.livejournal.com/200038.html Mood: worried Tags: big posts, dreamwidth, google, google summer of code, livejournal, programming Subject: DW and the Google Summer of Code

I've been hesitant to post about this, as I know that both Mark and Denise read this journal, but this issue is important enough to me to post it.

So, the application phase for students to apply to organisations participating in the Google Summer of Code begins today. Dreamwidth is one of these organisations.

From the start, I haven't liked the idea of DW taking part in the GSoC, and that's because GSoC, by definition, focuses on the code. Students enter the GSoC to code, and to get money from Google for doing it. They're not there to spend ages learning about the culture.

One of my favourite parts of the DW development culture was nicely summed up by Kirrily Robert, in a presentation called Standing Out From The Crowd that she made at OSCON last year:

From the very earliest days of your project, recruit the diversity you want to see. The first two or three members of the project will set the tone for what follows. People will look at the project and think, “Is this somewhere I can see myself fitting in?”

If you’re working on a desktop app, recruit desktop users. If you’re writing a music sharing toolkit, recruit music lovers. Don’t worry about their programming skills. You can teach programming; you can’t teach passion or diversity.

This isn't to say that coding experience doesn't matter, but it does place a lot more emphasis on passion - someone's love for the idea of what they're doing. Passion, in a lot of cases, also implies experience and involvement, perhaps with similar projects, or even use of the app itself, if it's not being built from scratch.

And so far, every single one of our developers on Dreamwidth are here because of that passion. We're all avid users of the site itself; none of us started developing for DW before they started using it. Even when the project started and there wasn't a "site" to speak of, all of us originally came from LiveJournal - and since then, we've gained new developers from people using the site and wanting to help build it.

However, students who apply for GSoC may know nothing about Dreamwidth other than what they've read in the Welcome, Google Summer of Code students! post in dw_dev. Now, this isn't criticising that post at all; on the contrary, it's an excellent post and I'd advise every GSoC student applying to Dreamwidth to read it. However, one cannot know the community and the culture from one post. Even if they take the time to read other posts, maybe from the Latest Things page, you're not going to have that passion, or know the culture; that can only occur from actual usage of the site, having friends on the site, and taking part in things.

I want to emphasise the importance of having friends on the site here. Having a friend on the site - or at least someone you know - is an excellent way to become adjusted to the culture, because you'll be reading your friend, they'll be reading you, commenting on each other's posts, and possibly other people's too. It's one reason why I loved the invite code system on LJ, and love it on DW; it preserves the culture.

But again, GSoC students will probably not know anybody on the site before they apply. As such, their viewpoint will be mainly centred on blogging, not on community - which could explain why we've been seeing so many students wanting to write desktop clients. It makes sense that if your priority is writing your own posts, it's far easier to open up a client and use that. But if your focus is on community, the only way you can really get involved in that is by using the website itself, as very few clients will let you post comments to other journals. That said, the Dreamwidth API doesn't allow for that at the moment, so if there were any clients that commented on other journals, they'd need to web-scrape. (We are developing an NNTP server which might help with that in the future, which I'm looking forward to.)

Now, I trust Mark and Denise to not mess this up. I believe they know all this already and probably had a long discussion before even applying to become a mentoring organisation at GSoC. I believe that the candidate they pick will most likely already have been an LJ or DW user in the past. But that doesn't stop me from being afraid in case they aren't. They may be a fantastic coder, stick to all our Programming Guidelines, and wow us with how elegant their code is, but without that crucial knowledge of the community, and a passion for it, how will they be able to design features that we as a community want?

4 Comments | Post A Comment | Add to Memories | Tell Someone | Link

User: foxfirefey Date: 2010-03-29 09:29 (UTC) Subject: (no subject)

GSoC isn't just about what Dreamwidth can get out of some summer coders--it's also about engaging with the broader open source community and giving something back.

I think the reason so many people are into clients is more that many applicants don't know Perl well, and writing a client doesn't require them to know or learn Perl and lets them cater to an area of their strength--much like the person who wants to revamp the RTE is catering to their strength. And well, the server codebase is pretty damn intimidating.

Dreamwidth doesn't have to use the code made by any of the students. There isn't anything to fear. I'm not saying there isn't anything to lose, per se--I mean, time and effort goes into mentoring a student that could be spent elsewhere--but we don't have to absolutely maximize the utility of our students, which is what you appear to be describing. This is more about them than us. And if they're outsiders, it's okay. We can welcome outsiders. They'll know what features the community wants because the community will tell them.

And if some Google Summer of Code student writes a client that nobody wants to use, well, that's okay. It's not going to destroy our way of life! But y'know, we really could use clients. They're nothing to sneeze at.

Reply | Link