I'm just going to briefly re-visit some [important] things that were said before returning to the issue at hand.
See 1st bullet point by a then-developer discussing the seemingly forgotten philosophy of LiveJournal. See first part of entry, on attitude. Or evan's comment here.
Halcyon days here and yes I realise that he no longer works in any role for LiveJournal, plus see also here. To which one can only say wow. Large amounts of people wanting to help; yes the site was trailblazing then - c. 30,000 users but now there're more and amongst them people like those here: still interested.
Or the LiveJournal social contract of: Work with the community, for the community: We promise to keep you informed of changes to the best of our abilities without being intrusive. We promise to run our business based on feedback from the LiveJournal community, and with the LiveJournal community's best interests in mind.
Or this gem in which Brad explained... patch submitters will *love* to read that #74 one!
Time was when there were 5% paid accounts, including the inactive accounts. If you looked at active accounts, about 10-12% of users were paying, and that was perfectly acceptable. Those 10-12% paid for everybody else just fine, and the site didn't have any financial problems. Not to say it does now of course but the active accounts to paid users ratio is now only about 6.5% by my calculation. Many changes.
Over and over again we read of the central theme of an open source community with volunteer contribution. But that was then, not now.
I was sorry to read here that one genuinely interested volunteer developer was told on IRC that Danga didn't need anymore help from volunteers, it must've seemed a real kick in the teeth. Plus, if that's the sentiment going around it's not surprising external developers deserted in droves. The problem of external developers not knowing what's going on, or what they should be doing, their patches being left seemingly ignored, the apparent erosion of the LiveJournal open source methodology, the v. poor system documentation - these are all big problems for LiveJournal and therefore big problems for its members.
There seems to be a paradigm shift going on from the type of site and goals expressed in those links above - to Danga Interactive who develop the LiveJournal code. A project that was once a community effort but is now carried out by salaried primarily in-house programmers. A project where more minor bugs go ignored and part-completed projects languish because the small development staff simply don't have the time to do everything (how could they). A site where there is a lack of corporate identity with LiveJournal; having it seems no idea of WHAT it wants to be - grandiose plans for Trackback implementation with Pingback but no timescale available & RSS feeds implemented; but yet a site so far from 'professional' well-documented solutions like Traction, pMachine and Movable Type (or the hosted TypePad with its well implemented photo albums). LiveJournal.com has been allowed to fall into some weird limbo between a site like DiaryLand (which knows what it is) or the openly advertising-supported GreatestJournal and the 'serious' solutions like those aforementioned. So when someone's looking for a weblog solution to administrate - like ecereal who commented a few entries ago, less and less of them will look to LiveJournal and visit lj_everywhere. Or even for the LiveJournal.com hosted journals due to the erosion of the developer community, they too shall suffer; because there simply ISN'T enough time in the day for a small amount of employed developers to do "everything".
So one solution is to scrap the external development community concept and heck, decide it's flogging a dead horse. But I don't want that, I dare say mart doesn't want that and the folks who posted in those links above didn't ever want that. At least not back then. Even remaining out-of-office developers are getting disillusioned and it's not surprising if the message received is that work is not needed (read: appreciated). So what to do? Well, I don't have all the answers and there's been a lot of brainstorming already. But I don't think all is lost or that it isn't possible to arrest the decline.
My thoughts for the future
- There needs to be an office to external developer liaison person. Period. I disagree that the person can be outside given that the Danga culture has been allowed to develop into sharing information with others they can physically see only. Someone needs to be there to act as a conduit to the community. Forget every now and then an existing staffmember (who all work very hard) taking time out to do it. They can't do that and whatever job they're doing now. The argument it was never needed before is pointless. LiveJournal has around 1.3m active users now (source evan's lj_research community). Some of those are potential and willing [code] contributors; with that sort of amount of people a dedicated person is necessary. bradfitz wouldn't be happy in that role no more than he would be in a 100% biz management role, as he's said in the past. But it is a role which needs to be filled.
- The system documentation is ostensibly an organic resource which is contributed to, checked, up-to-date etc. For external developers to make progress plus for potential external LJ administrators it needs work. One person cannot sit down and write it all and some co-ordination is needed. Because of the high degree of technical skill needed plus familiarity with the code someone in house would be better. jproulx cannot be expected to do all that on TOP of everything else he has to do. System documentation is not the same as user documentation discussing how to use an lj-cut. It just isn't. There needs to be someone who has that as a main duty, co-ordinating.
- Collaboration overseen by the liaison mentioned in 1), I was trying to think of ways to do this when I decided an online community like a journal... then I thought hello? lj_dev?! This could be so much more than it is? At least that's what they say. Blogs at the top of rankings/those used well for education & projects what-have-you have side panels a top panel etc. Frames/CSS, whatever. Maybe that could work. There could be the link to recent entries by title with discussions. An RSS feed for all the out-of-office-developers to add to their own journal - from the liaison guy/girl; links of top issues ranked or sorted into alpha, beta stage or whatever. Current projects and the stages/timelines, a variation on the project listing Brad mentioned maybe. Definitely a link to Zilla, perhaps part of it embedded in. All this whiteboard stuff - well, the liaison person can transcribe stuff and if there's really a need for occasional pretty diagrams get a graphics tablet and put the results behind a cut. Not hard. Tap into the userbase for help putting it all together.
- Zilla. Re-integration of it into projects and daily use of it, mess around with custom fields if necessary. Zilla. Zilla.
I've mentioned bradfitz a few times, yes. I realise it's a long time since he was solely responsible for running things; LiveJournal has grown. But he remains the head honcho and retains overall executive authority. I believe it is necessary for him to take the bull by the horns and act on these problems. Somebody with the power has to. Yes, two new staffmembers would cost money; if it's not there then do what you have to do - sell some perm. accounts, raise the paid account price by $2 USD - whatever.
Don't have LiveJournal be like Chicken Little yelling 'the sky is falling what'll we do!' while you're surrounded by users who can help, who are LiveJournal. I once watched a program set in the Aussie outback called The Bush Tucker Man, he went around pointing out all the outback food - anything from honey to grubs/plants etc. He mentioned a plane crash there where all the survivors had died of starvation - surrounded by food. The resources for Livejournal - the nourishment for it to flourish is there; all it needs it a little work to find it, to work out how to get to it.
As a final note, when I joined LiveJournal I came via an invite code - they aren't around any more as they were always temporary while the site grew. The site one way or another'll be around for quite a long time but if things don't change the open source ethos where external volunteers could help build it will like iamo said go away too. Let's hope some of the core values of LiveJournal are remembered.