Decade

[liv]

I officially started using DW as my main journalling / blogging home 4 years ago. That's not when I created the account; I was helping with documentation and testing and a teeny-tiny bit of development when the site was still in closed beta, so I had one of the first small handful of accounts back in January 2009. But I didn't want to "move in" here until it was opened for people who weren't connected with developers to use as well.

Dreamwidth celebrates its birthday on 1^st May, so this seems like a reasonable point to count from. Four years on, I'm still pretty happy here. DW provides pretty much everything I want from a blog host. Most importantly, it's stable, both in terms of near-impeccable server uptime and as a business which has a solid financial plan but no grand ambitions. The WTF system and the reading list make it extremely easy to meet interesting new people and avoid obnoxious people. I like having control over the appearance of my journal, including the option to tweak it with CSS. Threaded comments and a decent, non-obtrusive comment notification system. There's no advertising, which is such a big thing, I really really want to be using services with precisely this business model, that 10% of core users financially support the site for the 90% who don't pay but do create content.

Is DW perfect? No, I could certainly find things to quibble with. In particular, after the initial honeymoon phase, it doesn't really seem to be in active development any more. There are a bunch of really quite key features that have been floated around as being about to happen any time soon for more than half DW's lifetime. They never seem to get beyond spec, or at best, a half-hearted implementation that gives the desired functionality but isn't fully usable or documented at all. I mean, this is better than the other extreme of constantly rushing to add new features that detract from the core purpose of the site and break things, and certainly there is active, meaningful bug-fixing, the site's not just being left to rot.

In some ways this really does slot DW very tightly into the niche of being perfect for people who really miss LJ as it was before brad sold it to a company with serious VC investment. I am one of those people, certainly, I like that it's small and quiet and a bit old-fashioned and text-centric. I like that it's properly in the spirit, not just the letter, of Open Source, even though that does have the disadvantage that the only way to get new features is to get someone from a very small core team of experienced volunteer developers passionate about the same things you care about. And I like that the site is not trying to grow beyond its capacity, it's trying to be sustainable. But obviously with any site like this there's going to be a huge network effect and I think at this point I have to admit that DW is never going to be big. It's always going to have the dream of being the absolute ideal blogging / journalling / networking site that's run for the benefit of its users, and I like being part of that dream, but it's an ideal, not a reality.

The thing I don't love about DW is its employment practices, it's seriously underpaying most (all?) of its paid staff. I suspect that DW isn't actually worse than a lot of other tech start-ups, it's just that the culture where the staff don't just dogfood but actively use the site and are fully involved socially means that nosy people like me can keep their ear to the ground and find out that people are working for considerably less than the market rate because they're passionate about DW and for other reasons connected to their personal circumstances. It's a vicious circle, too; without being able to offer competitive rates to attract programmers, DW will never be able to develop enough features to grow to the point where it can afford to pay said programmers properly. It's little short of a miracle what DW has been able to achieve with volunteers, including a whole heap of volunteers who don't otherwise have backgrounds in programming or Open Source. But even that I suspect isn't quite enough.

Anyway, while I'm marking time passing, I have been blogging for most of 10 years; DW opened just a couple of weeks before the 6^th anniversary of starting my LJ. I've made just over 1500 posts in that time, and I reckon that probably puts me close to 2 million words since most of my posts are long-winded. Certainly a couple of million if you include comments. And those words have described getting a PhD, several different romantic relationships including the one with the person I eventually married, three jobs, living in Scotland, Sweden and moving back to England again, a bunch of travelling, reviews of several hundred books, and some pretty major shifts in my thinking about topics such as politics, feminism and others. I've met any number of new friends and got glimpses into the lives of, oh, a good several hundred people, many from very different backgrounds to me and whom I might never have imagined if I hadn't been on LJ/DW. I don't know if I've achieved the kind of competence that is supposed to come with writing a million words of crap, but I do think I'm a better writer than I was in 2003. All in all it's been a blast, and I'm certainly looking forward to the next ten years.

Comments

[ceb]

Unrelated to post, but have you seen this from fjm?

http://fjm.livejournal.com/1245600.html

[denise]

so, this post has been seriously nagging at the back of my head a lot, and i hope you'll forgive me both a) coming back to it three weeks later, and b) commenting despite that awkward tension between private person and site-owner that happens sometimes. (i'm commenting as denise because this is more of the site-owner sort of thing, but in case anybody else stumbles across this comment, liv and i mutually read each other as private-people and have known each other for ages, which is how i came across the post in the first case; i don't usually make it a habit of jumping into posts where people are discussing DW!)

you make some totally valid observations, don't get me wrong, and i'm not going to argue that there aren't some (many) things we're falling down on, but ever since i read this post, two things have been really bothering me, and i'd like to address them.

first, we are absolutely not underpaying people, at least not in the way you mean it. obviously i can't discuss individual financial data or specific salaries, but we actually only have one fulltime staff member: everyone else is being paid a retainer to be on call for specific situations, or is being paid with the expectation of only a certain number of hours per week. we pay at or above market rate for the location in which people are employed, the number of hours people are expected to work, and the nature of the position, and we've done so since the moment we first took on paid staff, even in the months when we had zero income. we are absolutely not taking advantage of people's idealism and commitment to the project. if we hire somebody, we pay them at least what they could make in an equivalent job. if we were willing to compromise on that, we could have more paid staff. we're not willing to compromise on that. we don't (legally can't without making some sweeping changes, due to many factors that are boring and tedious to get into) offer benefits, and there are fairly complex situations revolving around the fact the US tax code is a nightmare, and we'd love to be able to pay people above market rate, but the only people being dramatically underpaid are me and mark. (i'm making below-entry-level salary; mark isn't being paid a regular salary at all.)

second, i think that saying the site's not in active development is a vastly unfair assessment. yes, there are a lot of big things that haven't been finished yet, and many of them are very noticeable and wanted very badly, but specific features not being finished yet doesn't mean nothing gets finished at all. if you look back at news posts, you'll see that we do release new features (at what i think is frankly a pretty impressive rate, given that we essentially have one staff programmer who not only does her own projects but also spends up to half her time per week reviewing and committing other people's code) -- the last release had the reply-by-email feature, for instance. sure, it's not huge and showy, but it was a lot of work (that represented around five to six weeks of person-hours due to the technical debt involved) and it's a feature that a lot of people had been requesting.

we try to have something new, not just ("just" being the wrong word but you know what i mean!) bugfixes, in every push. sometimes we fail! sometimes the things are tiny usability improvements that some people might consider bugfixes, not feature development! sometimes the things are so tiny that they're barely noticeable! but we do try. we've got another few major features that are on-tap and close to being releasable. we've implemented over half the suggestions that have been accepted through the dw_suggestions process. we've resolved over 250 bugs since the beginning of the year (and that number's probably low, since bugzilla broke their search-by-date with the latest release). development moves slowly in some cases, but it does move.

the problem is, yes, that we're understaffed, and (as i just said in a comment to another one of your posts) it's very hard to release major, sweeping features with a primarily-volunteer team. yes, if we had a million-dollar budget, we could do a lot more. we're limited by a number of constraints, including staffing, budget, the difficulty of working entirely remotely, the fact everybody's doing a bunch of different jobs all at once, the lack of a fulltime technical manager, the lack of someone dedicated solely to frontend development instead of doing frontend development as one tiny bit of her job, the lack of anybody (volunteer or not) doing UI design, the crushing technical debt we're still making payments on, and (quite honestly) the disability considerations of both the site ownership team and many of the volunteers. (i've been limited to about two hours a day of typing time for at least the past six months, for instance; i violate that limitation way too often, but my orthopedist would be much happier if i'd shut my laptop and locked it in a drawer for that six months. it has curtailed a lot.)

likewise, as i also said in that other comment, the fact that we have so many newcomers also slows development pace. i do not think this is a bad thing! i think it's great that we have so many beginners (to perl, to working on a webapp of this size, or to programming entirely), and i think it's probably one of the more significant things we've accomplished! especially since newcomers are the best people to spot the "missing stair", where we've worked around something for so long that we just gloss straight over how sub-optimal it is, or to spot the holes in our process documentation or the gaps in our tools; we get at least as much benefit out of the talents and perspectives of newcomers as we put in. but that approach does slow things down, in how much time experienced people have to spend coaching and tutoring, and it does mean that some weeks and months, the bulk of our person-hours comes from people who don't have the experience necessary to implement the really showy things.

are the things we're still lacking really lacking in some cases? yes, absolutely. have we been unable to get to many of the things i really, really, really want, that i think we need in order to attract users, and that people desperately miss when they're using the site? yes, absolutely. are there a double dozen workflows that could be improved so goddamn much? yes, absolutely.

are some of those missing features, usability disasters, half-finished projects, and giant gaps in our featureset quite frankly bloody embarrassing sometimes? yes, absolutely. yes. absolutely, wholeheartedly, a thousand times over, and i cringe sometimes that we can't get things done faster, and i could keep a team of fifty people busy for the next five years with all the things that need to get done. but i really don't think it's accurate to say "slow to release major features or fully complete some open projects" equals "not in active development".