With that, I've seen believable projections that the site will at least break even if one user in 20 pays for their account. Apart from the financial side, I am really impressed with the way that Dreamwidth is staggering the launch, with a real beta when the site is actually in beta, meaning that there's a list of critical bugs that have to be fixed before the deadline. They bought serious hardware in preparation for the launch, and have added 20, new, active accounts in 3 days without the site falling over.
They're thinking seriously about scalability. Yes, there was a major blip when they claimed they were going to launch last summer, but things got stalled. But between January this year and now, the site has gone from being little more than a domain name and a hope to being a very nearly fully functional site. Competence The Dreamwidth founders are Denise, aka rahaeli , and Mark, aka xb The former was employed as customer relations manager at LJ for years, the latter as one of the original programmers.
So they understand the technical and social aspects of LJ better than pretty much anyone else on the planet not excluding the current owners of LJ, the Russian media company SUP. This is why they're able to make a true fork of the code and then maintain it, not just clone LJ's code and hope that the parent site makes bugfixes available.
Dreamwidth has also managed to get a breathtaking level of involvement from some of the most engaged and knowledgeable LJ people, including some of the high calibre employees who got the chop in January, and people with years of experience as volunteers, developers and power users. This really distinguishes Dreamwidth from any of the clone sites out there. I think the only LJ clone that's big enough to be viable is InsaneJournal , and that has some major problems, including awful branding, and running an ancient version of the LJ code with some huge bugs up to the level of random data loss.
I don't want to speak badly about any of the people who have set up alternative sites; indeed, before Dreamwidth came on the horizon, I was just about ready to jump ship to Inksome. But the fact is that I don't really want to move my whole social life to a site that's being run as a part-time hobby by someone who may be a competent programmer, but has no experience in running a complex, commercial site or understanding of the ten years of accumulated kludges that make up the LJ code.
Community One of the ways that Dreamwidth has taken my breath away is precisely in getting so many people involved and excited about it. Yes, it's partly hype. But it's also dozens and dozens of people who are willing to put serious effort into building and maintaining the site. It's truly Open Source, not just on the level that theoretically the code is available to be used or modified, but because they actually do accept and commit patches from anyone who wants to fix a bug or introduce a new feature.
And they've gone to major effort to clean up the code so that it's consistent and properly commented and runs on modern versions of Perl and Apache. And they're putting time into training people with no experience of coding, including yours truly. When I went to look for a link to back this up, I discovered that LJ has hidden and obfuscated everything about its Open Source origins, so that basically you have to know exactly where to look to get the code repositories, and after that you're on your own.
But Dreamwidth is actually Open Source in spirit as well as in theory, and it's great fun to see and be part of. Shiny features Like I said, I'm not moving for the shiny features. But there are some really neat ones all the same. Basically, if you are used to LJ, Dreamwidth will be pretty familiar, except with a usable, intuitive navigation and interface, and a whole bunch of things that should have been possible on LJ but nobody ever got round to coding.
The thing about getting rid of "friends" and replacing it with "people who have access to my locked content" and "journals I read" is mostly cosmetic, but it's a very good sort of cosmetic. Navigation strip preferences are now site-wide instead of being connected to journal style, so if you like the strip you see it everywhere as a dashboard, and if you hate it you can avoid it without having to resort to Greasemonkey scripts.
You can preview comments without having to go to the extended options page. They're working on the ability to control multiple accounts with the same login. Interoperability This is the big one, for me. Dreamwidth isn't trying to lock people in to the site, quite the opposite, they're making serious effort to create a situation where it doesn't matter which site your friends prefer. They've implemented OpenID properly, so that if you don't want to create a Dreamwidth account, you get full functionality in terms of having a reading page, getting comment notifications, voting in polls, taking part in the trust system, etc; the only thing you can't do is make journal entries or join communities.
You can already import all your content from LJ or related sites, and if you import comments, the comment author can log in with OpenID and manage the copies of their comments on Dreamwidth. See FAQ 16 for more information. We hope to add a replacement for this at some time in the future. In the meantime, you could use ShoZu , or our mobile site. We have plans to release a low-commitment instant feedback system, but it will be very different.
Chat isn't part of our main site focus, so we decided to put our attention elsewhere. My Guests — This was released after we diverged from LiveJournal's code, and we'd like to do it differently. In the meantime, paid users can use Google Analytics for some aggregate statistics about their traffic. Repost button — This was released after we diverged from LiveJournal's code. S1 journal style system — S1, the oldstyle journal customization system, was outdated and hard to support.
We've put our work into improving the usability and customizability of S2 instead. Schools directory — Maintaining the schools directory took a lot of manual administrative work, and we would've needed to reproduce all the categorization and validation work LiveJournal had done over the years, because while the code for the schools directory is available for us to use, the curated data isn't. We decided that since the focus of Dreamwidth is more about finding new people with interesting things to say, rather than finding people you know offline, that effort was better spent elsewhere.
The site's commitment to open operations should make such abuses clear and thus almost vanishingly unlikely. There are a large category of people who will practically not be able to use the Dreamwidth site , whether they approve of it or not: those who do not speak English. LiveJournal has always placed considerable emphasis on the variety of languages in which the site can be used; this has resulted, among other things, in the site becoming one of the foremost, if not the foremost, Russian-language blogging platforms.
However, the site's attitude to translation works better in some cases than others; as new pieces of text are added to the site in the native English language, they may not necessarily be added to other languages for some time, if at all.
There are some occasions where phrases may be combined in ways that make sense in English grammar, but may not work so well in other languages.
Lastly, LJ tags are designed to make sense and be easily memorable abbreviations in the English language, whichever language you may be working in. It works well enough in some languages for the site to be popular in some countries, but it's clearly not perfect. The Dreamwidth movement takes an alternative approach to this. The Dreamwidth site will only ever be in the English language.
However, the site's commitment to open source means that it ought to be relatively easy for almost? Accordingly, over time, we might expect there to be a Russian-language implementation of the Dreamwidth code, a French-language parallel Dreamwidth site, a German-language one and so on and so on, as well as likely several parallel English-language implementations of the code.
Again, this has the drawback of splitting locked discussions among many destinations. The Dreamwidth movement aims to solve this by a notion sometimes referred to as "federation", whereby users who have accounts on multiple Dreamwidth or LJ codebase sites ought to be able to see posts that they can see under Friends-lock on one site at all sites in the federation - and all this without compromising the security of the Friends-locked posts to the outside world.
This has got to be the true killer feature for Dreamwidth, if they can make it work well, for those of us who might not want to have to track several different sites. It's also a heck of a technical challenge, and currently it's unsolved - it's just "just"! However, it's listed as a bug that is blocking the open beta, so presumably a cunning plan is in place. The whole operation does rely on OpenID to some extent, which some have argued against, though I'm not sure how current the OpenID criticisms are.
As a thought experiment: could someone start a rogue site using the Dreamwidth code, attract users to it who start federating Friends-locked posts from other sites to this rogue site and then start publishing the Friends-locked posts in public?
Surely it's possible, but presumably it's no more of a concern that a site that might make Friends-locked posts that are posted there public. Iain once posted five tests , as was in vogue at the time, of ways in which he might have liked LiveJournal to develop. I'm fairly sure that some of the above properties of the Dreamwidth site will act as a disincentive towards his participation, but he may have more time for the movement at large.
The concept of federation, if well implemented, should go some way towards easing his concerns regarding the legal basis of the site and his concerns regarding the single geographical basis for the servers. It's also been publicised that another feature stopping open beta is the revamping of the journal style system so that people might change their styles by use of the web standard CSS system rather than LiveJournal's rather opaque and proprietary S2.
There is something of a common theme to some of the above: there are lots of developments that are currently blocking open beta, but not in place. You can see the bug counts on a frequent basis, but the race against time is a difficult one. Even when the site enters open beta, ready for prime time or not, there are a long list of plans for features to be implemented.
There's so much that's to be done and yet so much not yet done. But there's so much that has been done; see the bug counts, and see these fascinating descriptions of how much can be done in just one week. See how many people are working on the project! See how many people are volunteering on the project already, and how many more are signed up to volunteer once further infrastructure is in place.
The staff page shows that many of these have some fairly ideal credentials, too. This shows that the project has momentum and talent and dedication and love behind it.
Dreamwidth is making so much progress that it's hard not to love it. This is as fascinating and exciting as projects of this type come. That's why it's just damn cool , as far as I'm concerned, and that's why I want to associate myself with it, in whatever way, shape or form.
To an extent, I'm not even sure how much I care to what extent the Dreamwidth project meets all its goals, so long as the momentum remains so strong and the project develops in delightful ways, even if they aren't necessarily the delightful ways I'd have thought or requested. It's a delightful case of open source volunteerism against the establishment that LJ has become.
Part of this inevitably reflects a degree of cult of personality; something doesn't become cool unless there are cool people working there. One more reason why you might not love the Dreamwidth project is if you have something against its principals, but I absolutely do not.
It's true, and it may be relevant to some people, that synecdochic in another guise ran LJ Abuse for some years, at a time when they did some questionable things. To some people, that may be a strike against her ability to run another similar site. While I don't necessarily agree with all the things that happened, I think it would be blinkered to hold these past events against her; while the "I was only obeying orders" defence is stronger in some circumstances than others, I think that running the site and setting its policy is as good a chance for her to show how she'd like to run a site, given the relative lack of constraints at DW that were in place at LJ.
If you must, judge her on how she runs things by choice, rather than how she ran things by compulsion. Her reflections on her time in charge of LJ Abuse and how DW abuse might be different are a revealing starting-point, and I like what I see.
It's an oversimplification to say that I think Dreamwidth is cool because I think synecdochic is cool, but it would be wrong to say that that isn't part of it. I get the impression that she'll co-run the site with insight, love and an ingrained streak of wisdom that other people just won't have. She is a tremendous communicator and fiercely charismatic. Furthermore, she has major, major kudos just reflected from being held high in the opinions of some of those whose opinions I trust.
Voting for a place to focus your social networking is like voting for a political party. Anyone can promise anything, but at the end of the day, probably the most important factor is how people react to the developments which could not have been predicted. Due to their backgrounds and the way they present themselves in the rest of their lives, I'm far happier to put my trust in the Dreamwidth principals, their hopes and their vision than I am anywhere else.
In short, I vote Dreamwidth. You might not; there are some of you who, I suspect, will only really ever be happy to vote for themselves - or, depending on how you interpret the paradigm, not vote at all. Nothing is certain, but I'm convinced that Dreamwidth is as good a bet as anywhere.
0コメント