Dreaming in digital

[foxfirefey]

There's an exciting project I'm a part of called Dreamwidth. Some of you already know about it, some of you don't. Basically, Dreamwidth is a bonafide LJ fork best described as "a bunch of ex-LJ staffers rebooting LJ from the inside." Dreamwidth is not:

1. An LJ clone with shoestring resources, where updates and improvements rely mostly on LJ's development pathway and contain dangerous and unforeseeable issues.
2. A grudge project, where the only motivating factor for its creation is that it's somewhere that's not LJ.
3. A fandom project, although it's fandom friendly.

Why I like the idea of Dreamwidth

So, around a year ago, one old school ex-LJ employee, on the managing side, contacted another old school ex-LJ employee, on the coding side, and asked "What if, way back when, we'd gotten the chance to do LJ right?" After all, LJ mostly got sold to Six Apart because brad hated management, not because it was about to die.

While very different people, they both found the idea tantalizing, which led them to ask, "What if we made that chance for ourselves now?" They made spreadsheets with calculations until they found a decent business plan. Then they got to work.

And a lot of people joined them, me included.

I'm excited about Dreamwidth because it's making improvements to LJ that LJ will never get around to or be able to do--and I'm a part of that process, and coming up with good ideas, and doing work that makes a real difference for the service. I love how diverse the group of people working on Dreamwidth is, just like LJ itself.

I'm also excited about Dreamwidth because it aims to be a small but most importantly sustainable business focusing on paid account revenue; I doubt LJ's Russian media conglomerate owner can turn this goat into a draft horse of revenue, fear what they'll do while trying, and worry about what will happen to LJ if they decide it will never make a worthwhile enough profit to continue investing in. I also wince at putting the sentences " What kind of premium features are you most interested in? Should we do more with v-gifts?" next to each other in an entry on March 9th seeking feedback for an advisory board meeting on March 10th, posted in a neglected, by the wayside official community.

Dreamwidth won't be perfect, but it will feel much more like a home to me, while LJ feels more and more like a hotel whose owners don't visit much.

Stages of Dreamwidth

Currently, Dreamwidth is in closed beta. That means the only people with accounts on it are testers and other regular volunteers--although signing in with OpenID works. They'll try and find the most obvious landmines and test out the new features to make sure they work. Hopefully around mid-April, we'll have fixed and implemented enough to go into open beta (edit: starting April 30th) and open the site up to others through the invite code system or the purchase of paid accounts. Things will probably be a little bumpy at first. When they smooth out and the site is ready to move from dedicated hosting to a colocation site, the site will have officially launched. You can read more details about this process here.
What are the changes from LJ?
As long as this list is, this isn't a comprehesive list of changes, but ones I feel are notable as well as truly intended. By open beta

By the time the site goes into open beta, here are some of the (well, I consider them) improvements that should already exist:

• Your friends list has been split: you now can subscribe to people and/or give them access. That means you'll be able to watch somebody without giving them access to your locked posts, or give others access to your journal without adding them to your reading page. Subscription filters will be separate from access filters, and you'll have more of them than 30.
• A total revamping of journal styles, with standardization of options and CSS markup across all styles to make them easier to manage and modify. This will make it more easy to customize your journal with CSS and not have to learn S2. There will be some brand new, just for Dreamwidth styles.
• Increased limits on username, entry, and comment length. Usernames will be able to be up to 25 characters long; entries around 5x as long and comments around 4x as long. That username you thought was nifty, but was just a couple characters over 15? You could have it now. Writers? You'll be able to put longer stories in a single post. TL;DR commenters (me included) or thorough critiquers? You will now be able to discuss moar TL;DR or do more critiquing without splitting things into multiple comments.
• Importing your journal from other LJ-based journal services, including comments. Those comments will hook back up to the original person who made them on LJ. If that original person ends up getting a DW account they link up with their LJ, those comments will then convert to being from that original person. There will also be ways to crosspost back to LJ from DW, so you don't have to do it yourself. They'll also import interests, tags, and friend groups as access groups.
• Speaking of styles, the navigation strip is a viewing preference over a style preference: you'll be able to choose when and where you see the nav strip without it being overridden.
• Free users will be able to use the site directory search and create new syndicated accounts.
• Paid users will be able to use Google Analytics.
• The site navigation was decided by a card sort, making it much more natural and understandable. It will be easier to find what you're looking for.
• More site scheme choices: there will be different color and layout schemes for the site scheme. If you're not crazy about the red (it's not my favorite), it's okay, they won't force you to use it. There'll be better CSS classes, so it will even be easier to apply user side styling to it, as well. Additionally, the menu system is consistent across all site schemes.
• Staff accounts have their own account type, with a different head. You'll be able to tell at a glance when somebody from staff is talking.
• Improved icon (userpics) interface. The ability to upload multiple icons at a time. Additionally, icons now have a comment field and a description field, so you can use one to give credit and one to describe what the icon depicts, for usability and accessibility. The icon display page will also be paginated for larger numbers of them.
• Improved adult content warnings. Posts with adult content will be NSFW or 18+ instead of the nebulous "adult concepts" and "adult content", and people won't be able to flag you. You will also be able to describe why you applied that setting on something, for better warnings. Dreamwidth won't let users flag other people's posts for adult content, either.
• Nudge is dead. That's the feature that let you bug somebody who hadn't made a post in 2 weeks to post.
• Snap.com popups are dead. Those are the popups when you hover over links. (If you like them, you can install the browser plug in.)
• Dreamwidth will not have ads. An adectomy has been performed. Instead, there will be two levels of paid accounts and growth will be controlled through the use of invite codes. There will be an IPO-like event of 400 permanent accounts being sold during open beta, giving DW the capital to seriously run for the first year. This isn't some kind of knee jerk idealism--see Why Monetizing Social Media Through Advertising Is Doomed To Failure, which is an essay written by one of Dreamwidth's founders.
• Invite codes. While this may seem like something that's annoying, in the long run it's good. It helps DW not expand past what it can handle at a given time. It helps DW grow organically based on communities and personal relationships. Additionally, it deters spam--LJ has been having horrible spam and bot account problems lately. (Had any Russian bots friending you?) Invite codes seriously hinder spam account creation. There will be regular distributions of invite codes to different parts of the userbase, and to users who request them, as long as they can handle the growth.
• The ability to do LJ-name type links to other LJ/DW sites. They've extended the syntax, so you can go site="livejournal.com" or site="insanejournal.com" and it will work just dandy.
• Better OpenID integration. If you don't want to get a Dreamwidth account, you can log in with OpenID and have access to your friends who do. That includes getting comment notifications if you give a verified email. (LJ would do this, but you probably wouldn't have known it very easily.) There are many, many little usability changes we're making so that the site doesn't suck for OpenID users.
• On the backend, the code has been updated to run on new versions of Apache. Work is being done to improve the installability of the code by people who are not us. The code is being cleaned up and standardized. That is, if you wanted to and have the technical chops, you could rent a VPS for probably around $30 a month and run your own (lightly used) version of the Dreamwidth code. What's more, you'd find a small but healthy open source community around it. Features that didn't come with vanilla LJ code have now been open sourced, like support for paid accounts through PayPal and complete invite code support.

There are a few things that are in the pipeline, but might not be ready by open beta:

• You'll also have better ways of making sure that your style is the only one you see, if that's what floats your boat.
• The ability to have your profile in your journal style...and the ability to NOT view other people's profiles in their style.
• Community maintainers will have some better tools for managing posts. They'll be able to add cuts to posts, select more restrictive security levels, or set a more restrictive adult-content warning.

Plans after launch

Within the first year, we are looking to implement:

• Better photo hosting: we want to spec out and design, with the community, photo hosting that's designed to work with the ways people use Dreamwidth.
.
• Draft posts and scheduled posts, so you can work on multiple posts onsite at once, and have them appear in the future, like when you're on vacation.
• A kind of main account/alternate account system, so you can (privately) designate one of your accounts as the parent account and all other accounts inherit that account's settings unless overruled. This will also improve the "work as other user" drop-down, so you can select userpics while commenting as one of your other accounts, post to a community more easily with a sub-account while logged into your main account, etc.
• The ability to export your journal as a .pdf, although it might have some small cost associated with it to not kill the servers.
• Complete overhaul of memories, to make it much much more like del.icio.us--you'll be able to save your or other people's posts as public or private, browse all public bookmarks from a specific user, browse all public bookmarks from all users using a specific tag, etc. This will let us make a better explore page than the one LJ has, so you can find more easily find interesting new content on DW.
• A user to user classified system.

Other things that DW wants to be able to do:

• Implement killfiles, so you never have to read X's annoying posts and comments if you don't want to.
• Make an address book, so you can share your contact information with specific groups of friends, instead of having to make hacky where-am-I and screened where-are-posts all the time.
• Make community management suck less. Nebulous, but important. Ideas for doing this include doing things like making a built in "mod hat" flag for comments. We're also splitting up maintainer/mod capabilities into smaller bits and letting you defined groups, so that, for instance, letting somebody change the style on your community won't be letting them delete posts.
• Implement community chatrooms in DWTalk.
• Make happy little officially supported tools, like an IM-to-DW chat log converter.
• Ways for you to subscribe to only certain posts of your friends. We're discussing opt-in filters and opt-in tags, trying to figure out a good way to implement this.
• Cleaning up the URL structure.

Things that won't be on DW, at least not at first

• S1 -- it's old and not even officially supported on LJ anymore.
• Photo hosting -- the current implementation is clunky and not well integrated; we would want some serious rehauling when get to that stage.
• Phone posts and TxtLJ -- these cost extra money and use code we don't have to hook up to external services we don't have partnerships with at this time.

What does this mean?

I'm not leaving LiveJournal. I'll still be logging in and reading my friends page here and making comments. I might even be crossposting, depending. But my main journal home is going to be on Dreamwidth; my journal's already been imported, and I'm just waiting for it to hit open beta.

And if you want to come with me, or even just poke your nose around, I want to make sure you can. I'll start giving invite codes I get to people on my friends list starting with open beta, to anyone who is excited about it. If you're just curious, I recommend waiting for open launch to get invited, when things will be more steady. And if you don't want to wait for me to give you an invite code, you could get onto Dreamwidth by buying one month of paid account for $3.

If anybody has any questions, curiosities, concerns, or criticisms, let me know! Lots more information can also be found here.