dw_biz | Saturday Open Q&A

Do you have a rough idea how many accounts you're expecting to be created, say, in the first week of open beta? How did you get to that number? Will things slow down quite a bit if you get a lot more than that?

*is not so much worried as curious*

You just want to make me make predictions so I look like an idiot when they don't come true! *laugh*

We've got around 37,000 accounts created right now, but most of them are OpenID accounts created by people importing their journals and aren't actually active. The stats page is going to get redone to break that down more clearly, so people will be able to see how many accounts are OpenID, personal, community, and feed accounts.

We're designing our network architecture for around 50,000 active users at first, which is sort of a magic number pulled out of our asses but which we think is going to be about right for our first six months or so. That requires about 40 servers, of differing setup. If we go much over that predicted traffic, it will be a bit slow, but likely not cripplingly so; we will have the "paid accounts jump the traffic queue", so paid users will go to the head of the line, and we'll be able to add more servers as necessary pretty quickly (thanks to the fact that we'll be using dedicated-server instead of colocation at open beta launch).

(So, if it turns out that we suddenly have an influx of 800,000 people all banging on the doors on April 30, yeah, things are going to be very slow -- but we'll be able to react to it quickly.)

Thank you for the detailed answer. 50,000 is a very impressive number, but I can believe it will happen. Someday I'll be bragging about having been in the first four hundred accounts. :)

Minor correction, 'dedicated server' means something else. We're using what are called VPS, aka, virtual private servers. But same net result, really, so it's just me banging on lingo. ;)

You've been very open about the financials for paid accounts and open beta onwards. I must confess to slight curiosity about how you're funding Dreamwidth at the moment, as you're not accepting payments yet. Are you using up your personal savings to pay for bandwidth, or do you have a small business bank loan, or has some kind rich friend donated a lot? And I totally understand if this is one question you'd prefer not to answer; I'm just mildly curious.

Mark and I have been paying out-of-pocket. We both contributed money from our savings to get it off the ground, and (for instance) our monthly server bill (which is around $500 a month right now) is going on personal credit cards. We'll pay that off when we start taking in payments.

We haven't taken in any money from anybody on this one but us! No bank loan, no rich friends. We're really serious about going this without any outside interests. We may decide that a short-term small-business loan makes sense when it's time to move to colocation, but we'll have to see what things look like when we get there. (We want to colo at around the time we decide that the site's coming out of beta, if it makes sense.)

Will you be open for donations once the payment system is in place? I'm not much of a blogger anymore, but I love the ideology behind dreamwidth. I'd like to contribute above and beyond the paid user levels regardless of how little (or how much) I'd use the service. I'm also fairly sure I wouldn't be the only one!

We won't accept donations, per se, because the word 'donation' implies that you're making a (often tax-deductable) contribution to a nonprofit organization, which we aren't. We want to avoid anything that could even give people the impression that we are, because there's already been some confusion between us and the Organization for Transformative Works, which is a nonprofit. (We're not affiliated with them at all, although some of our volunteers also volunteer with the OTW and vice versa.)

However! We are certainly not going to say no to you giving us money. ;) When we start accepting payments, we'll accept payments as gifts, so you'll be able to buy paid time for your friends. Then, shortly after open beta, we'll be adding in a "sponsor a free user" feature -- which is listed out in bug 211 -- so that you'll be able to make a payment to us and we'll apply it to a random active free user of the service. (Both sides of the transaction will be able to choose to be anonymous or not, and the random recipient will be able to choose to decline the payment and pass it along to someone else if they don't want to accept.)

Awesome. I also particularly like the "pay it forward" idea of the sponsorship; talk about good karma. :)

If I could suggest, when you open up the seed accounts at launch, you might want to include language about the sponsorship/gifts options? People buying seed accounts are the ones who are invested strongly, so it'd be a reasonable assumption that they'd be the most willing to occasionally pitch in some funds even after the initial cost.

Oh yes! We'd like to strike the right balance between "we are awesome, give us money" and not being too pushy, and we're likely to err on the side of not being too pushy, just because it can get really annoying otherwise -- but we will be very clear that people can buy paid time for others, absolutely, and we're going to be doing some frontend work to make it easier to send gifts.

Related, and while I'm thinking of this, could users sponsor/gift paid time communities as well as opposed to individual users?

And for the record, I'm pretty sure people would cave at "we are awesome, give us money." ;)

Yes, people can make payments for communities just like they can for other users! Although right now, there isn't much in the way of reward/incentive for paying for a community (other than the warm fuzzy feeling, of course) -- the paid features don't really apply much to community accounts. We're going to be doing some work to make paid communities more appealing, in terms of adding more features that are community-specific and not just the same features as a paid personal account gets.

Well, some of the paid features (as they stand) would be worth it as a community maintainer, such as the increased tags. I have a community that would likely max out the numbers on a free account pretty fast. :)

Anyway, thanks for the info. I'm totally looking forward to giving money to you awesome people. :D

We're going to be doing some work to make paid communities more appealing, in terms of adding more features that are community-specific and not just the same features as a paid personal account gets.

This is good news. I'm on a few newsletter comms on LJ, and sometimes people are so nice and, say, buy us more icon slots. Except that LJ comms can't use icon slots, can they? The icons can be uploaded, but not used. (Or at least not so far as I can tell.) So it would be nice to have more paid benefits for comms, too, beyond "not having to see yucky ads" which of course won't apply on Dreamwidth. :-)

people may assume you are nonprofit because of the .org in your domain name?

You're right, that's a danger of using the .org. We do own .com and .net, so we could use any of those, but honestly I prefer .org, even with the potential confusion. It just feels right for who we are and what we're doing.

i'm totally in favor of it too and it feels right to me. but if people are under the impression that you're a nonprofit that might be one source of that impression.

anyway. all the kinks will get worked out over time as more and more people get to come in and kick the tires. you guys are awesome.

Hello, Denise.
I'm new, searching around, and have run into what I hope is an easily fixed problem: am trying to get a permanent account, but can't get past the 'paypal' button. When I hit the paypal button, either signed in, or not signed in, I get a string starting with

"[Error: must have one %LJ::CAP class set _account_default to use the payment system at"

which disappears behind a large blank spot and reappears as

"727 Apache::BML::bml_block('BML::Request=HASH(0x7fb3a0f639e0)', '_CODE', '\x{a}use strict;\x{a}use vars qw(%GET %POST $body);\x{a}\x{a}################...', 'HASH(0x7fb3a0d6e2c8)', undef) called at /home/dw/current/cgi-bin/Apache/BML.pm line 991 Apache::BML::bml_decode('BML::Request=HASH(0x7fb3a0f639e0)', 'SCALAR(0x7fb3a0a63db0)', 'SCALAR(0x7fb39a4b0160)', 'HASH(0x7fb3a0d6e2c8)') called at /home/dw/current/cgi-bin/Apache/BML.pm line 322 Apache::BML::handler('Apache2::RequestRec=SCALAR(0x7fb3a09806b0)') called at -e line 0 eval {...} called at -e line 0 @ dfw-web02]"

This happens both in Safari Version 3.2.1 (5525.27.1) and Firefox Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.8) Gecko/2009032608 Firefox/3.0.8

I don't know how to read that, so I don't know what it's telling me, so I don't know what to do next. But I would like to buy a permanent account (am actually hoping for a chance at a seed account, but that's a ways off, yeah?)

I hope this goes here? I'm still finding my way around. Here via dw user princessofgeeks' invite. (Don't know how to make DW-specific HTML tags yet, either. Sorry.)

The payment system isn't currently working yet! We're still finishing up the last bits of it; you won't be able to pay for an account until April 30, when we open to open beta. In the meantime, if you want to test out what the paid features will be, you can manually upgrade your account.

Permanent accounts will go on sale the evening (EDT) of April 30, along with regular paid accounts. We'll sell 200 of them in one batch in the evening of April 30, and then another 200 of them 12 hours later, to account for our users in other time zones and make sure they get an equal chance as well. If they sell out before you can buy one, you can buy a premium paid account instead, which has exactly the same features and limits.

Thank you so much!

I'll keep my eye open, then. I work nights and sleep days (so have some empathy with your sleep schedule). Is there a taggable post to watch, so that in case I'm both awake and at the computer I can try my hand?

We'll post to

dw_news well in advance, don't worry!

I've seen a lot of talk of invite codes and 'when you can give us money'. Will free accounts ever be available without an invite code or a lapsed paid account? Will openID accounts ever have their feature set upgraded to basic free status?

No, invite codes are here to stay.

Because of the reasons of having invite codes--one of which is to control site growth--I imagine that OpenID accounts will not ever have their feature set upgraded to full fledged basic free status. Instead, at some point it should be possible to upgrade an OpenID account to a Dreamwidth account using an invite or paying for an account, or attach an OpenID to a Dreamwidth account.

Do you have some idea when we'll be able to attach OpenID accounts to Dreamwidth accounts? And will it be possible to attach more than one OpenID to an account?

We intend to implement the feature to attach your OpenIDs to your DW account by site launch (NOT open beta), so sometime in the late summer (July/August) time frame.

Yes, you will be able to associate multiple OpenID accounts with one DW account.

That sounnds fabulous! \o/

There seem to be an awful lot of people running around Dreamwidth who, somewhere else, would be seen as employees.
* Aside from yourself and

mark, what is the official status of these people?
* What, if any, compensation are they receiving?
* How realistic are the dreams of a volunteer eventually getting paid?
* How soon until you're hiring?

To focus your question a bit: are you familiar with the LiveJournal volunteer culture?

Let's assume I'm not.

Given Pau's question, and your answer, a little bit of background:

LiveJournal had a very big volunteer culture and still does, to some extent. There were a lot of people volunteering in various areas -- development, support, abuse, documentation, translation, miscellaneous. In the beginning it was almost all development. Over time, certain sensitive roles got converted to paid (systems administration first, some development, account support, etc etc).

Most of the people who have volunteered for LJ never received anything other than comped accounts and the satisfaction of helping out the community, of being part of something. Money is not the end-goal of many people, and it's certainly not the end-goal of Dreamwidth. While we expect the site to run in the black (net positive income), it won't be a lot of money.

I expect that Dreamwidth will have a similar volunteer culture. Lots of people who support the project with their time and energy, helping to build a better community for everybody to live in. But it will be on a purely time available basis. They come, they go, they help when they can, and if they can't, no problem. Do what they can, when they can, and we will love them for it.

There are very few differences between what an employee will do and what a volunteer can do. The real difference for me between an employee and a volunteer is that the former is expected to put in constant time. I carry a pager right now, so if a server goes down I have to fix it. 24/7/365. Denise is also on call, if there were to be an issue, I would call her on her cell phone and she'd have to get up and handle it. No choice in the matter, no matter how few spoons I have at any given point in time, I will have to fix the site if it breaks.

When we figure out how many users we're going to get and how many people are going to pay for accounts, then we will allocate money to pay people for helping. We've identified a few positions where we'll be looking for paid help early on, but that's not something we've finalized yet. Which means I don't know when we will be hiring, sorry! (And even when we do have money, I can almost guarantee it will be contract work, not actual employment, not until down the road a ways.)

Is it realistic to volunteer with an eye towards becoming paid for your time? That depends on a lot of factors, but generally - yes. Obviously when we need someone to fill a paid role, we will prefer to pay someone we know and can work with. Which gives volunteers a huge leg up. Does that mean it will always happen that way? Not necessarily -- Denise and I have a huge pool of really talented people we know and may draw from. But it is an advantage. "One foot in the door" and all that.

Of course, I must caution that it's not a good idea to volunteer only because you want to get paid. This project is about building a community, one I care about, and I want everybody who is involved in the project (volunteer, employee, whatever!) to care about what we're doing here. If your only motivation is money, you might want to consider finding a place more in line with your goals to spend your time volunteering for. :)

I hope that makes sense! Please let me know if anything doesn't, or if you have more questions.

Dreamwidth seems very much an example of "if you build it, they will come". Which begs the question -- where will they come from? You have import utilities in place for a number of popular journaling sites. What percentage of your userbase do you think will be transfers from those sites? After site launch how do you think the continued use of invite codes will continue to affect where new users come from?

We're pretty sure that a large percentage of our users are going to be migrants from another journaling/blogging service -- not just LJ, but many others -- and that's one of the reasons why we're so committed to the import and interoperability functions. We want to serve as many different uses as we can!

I do think that the invite codes are going to shape where new users come from, yeah -- we're more likely to see entire social networks moving over en masse, as one of the members of the group tries DW, likes it, and brings their entire network over with them. We also plan on doing some outreach to specific groups that we think will be happy on DW, and do things like giving out bulk groups of invite codes to specific groups and the like. (For instance, we plan on having "conference packages" -- someone who's going to a convention or a conference can ask us for promotional material, including flyers with invite codes, to hand out at their event.)

I'm coming to this comment a little late, obviously, but if you're interested in having anyone take materials to Dragon*Con in Atlanta in September, let me know!

This may not exactly be a biz issue, but, well, I saw it come up in someone's journal the same day I saw this post, and it does relate to business practices, I guess.

I've seen some deep suspicion of, and cynicism about, Dreamwidth on the grounds that you and Mark were members of LJ's Abuse team, and were therefore complicit (at the very least) in the Abuse team's behavior in such circumstances as the banning of users for having default icons showing breastfeeding (and retconning policy related to it). Were you guys in fact on Abuse? If so, how do you respond to people saying they don't trust you, your policies, etc. because of that history?

Oh wow, someone remembers the short period I pretended to be on the Abuse team back in 2002/2003? I'm pretty impressed someone would go that far back. I didn't do so well, as one can imagine by my LJ account only having ~500 support points. At the time most of the severely controversial decisions (strikethrough, nipplegate, etc) happened, I was no longer affiliated with the LJ Abuse Team in any way and was part of the Six Apart Operations team working on the Vox.com product.

Anyway, all of the large issues that came about on LJ were motivated out of the business model they chose to monetize LJ with, which required advertisers to be happy. If the advertisers weren't happy about something, then 6A would have to choose between "do we want this money and piss off some users?" or "do we not care about this money?" How much do you suspect the latter happened?

Dreamwidth's business model is user-supported. We believe, with 5% of users paying, that the site can sustain itself. We do not have the same pressures that 6A had and SUP have. There are no advertisers or investors to keep happy -- it's the users we have to keep happy. With our priorities already aligned along that axis, I hope that will immediately allay a lot of concerns people have about what kind of site this will be.

The other comment I'd make is that our policies are different. This is a different site. Now that Denise and I have a hand that is not bound by corporate policy we do not control, you can see what we really believe in and how we would have run LiveJournal if we had the chance. We have no default userpic rules (if it's allowed as a userpic, it's allowed as default), we have far more lax content rules in general, etc. Those aren't published yet, I know -- we're still writing them! -- but they will be published for feedback before Open Beta.

Trust is something that is earned over time. If someone looks at the past and doesn't trust us based on our affiliation with and actions for LiveJournal, I'm okay with that. A bit sad, but okay. I just hope that people who are not comfortable with that past will take the time to watch Dreamwidth and see how things go. Over time, I hope we can earn their trust through our actions on this site.

Actually, if the-shoshanna and I saw the same thing, the complaint being made was much more about

denise's history on Abuse than it was about you, with some fairly intemperate language and claims.

(I was coming to ask the same question, I'm glad I found an answer, thanks!)

If you're talking about the post linked on today's metafandom - and I'm sure you are - I found the point they were missing about migration was that yeah, D and Mark don't expect anyone to up and move because they're annoyed with LJ - they expect people to move if they think DW offers them a better service. Which is why DW is a code fork, not a clone, and why they're going to the trouble of making things like the WTF-list, crossposters, importers, talking about default hierarchical tags, scheduled posts, draft posts... all things that aren't part of lj, and therefore a real draws beyond "it seems cool!" or "I like the people running it!"

Ooh, sorry, totally off topic. I need to make a post about this, I think!

I think most of my response past the first paragraph apply equally for both Denise and myself. Had things been up to us, LiveJournal would have gone a very different direction than it did. Now that we have Dreamwidth, we can prove it.

I'm sure Denise will come through and write her own response to this, too! She's asleep right now, though.

+nods+ I believe you both, I like what I've seen, and I cannot wait for the end of the month. *bounces gleefully*

While I've got you though... I got

tea's response in my email. I did not get yours.

Have you gotten it yet? I just checked the worker queue, we don't have any pending emails. Well, we do, but all of them have failed for some reason or another and are waiting to be retried.

Also keep in mind that, especially with Gmail, we don't always talk to the same SMTP server on Google's end. So sometimes mails don't all arrive at the same time if we send a couple. (And of course, the threading in Gmail is broken for DW and we haven't fixed it yet, so the mails won't necessarily show up in the same thread.)

I did just get it. That was weird. the two showed up right on top of each other.

Thanks for the attention!

Of course, we might have seen totally different things, too.

Agh, hit post too soon.

I am neither Mark nor Denise, nor a Dreamwidth owner, although I've occasionally leapt in for Denise when she didn't answer something as fast as I thought she ought to.

I just wanted to clarify, since Mark's statement elided the point a bit, that Denise did indeed head Livejournal's Abuse Team for several years. She started working for LJ pre-6A, although she left shortly after Strikethrough.

I keep starting this answer, getting a couple of paragraphs into it, realizing it's not what I want to say, deleting it, and starting it over again, so let's hope the fifth time is the charm.

I volunteered for the abuse team from 2002-2003, and ran the team from mid-2003 to the point I left LJ in 2007; I've tried to be pretty clear about that, I think. (My duties on LJ weren't just running the abuse team -- my job description was longer than your arm -- but I'd say I spent about half of my work week on Abuse-related issues.) I'm still bound by the terms of my employment there, and I can't and won't provide any details on specific incidents.

What I can say is that the LJ Abuse team is probably the most misunderstood and mischaracterized group of people I've ever seen, because of a couple of reasons: the process was never clearly explained, the system (as it was built) didn't encourage transparency, the culture that grew up on LiveJournal was very antagonistic and mistrustful, and rumor moves fifteen thousand times faster than fact. The system, as it exists on LJ, also prevents members of the Abuse team from clarifying misconceptions, correcting those rumors, and explaining what actually happened in the event that the facts of the matter were different than the way it's presented via rumor. I have never once seen a summary of any situation that presented what I would consider a whole and complete picture, and that was always the number one thing that annoyed, frustrated, and discouraged me while I was working for LJ.

Part of the root cause of every single major conflict that come to people's minds was that rumor mill; part of it was a less-than-perfect process; part of it was unreasonable expectations on the part of the people doing the protesting; part of it was the corporate business model being used; part of it was a fundamental tension between the desires of the userbase and the desires of the people making the business decisions; part of it was the fact that no matter what, in any conflict between two users, one "side" is going to walk away upset and pissed off no matter what happens.

We can fix some of those problems with Dreamwidth, and we plan to. Our process is going to be a hell of a lot more transparent, our policies are a lot more permissive, we're designing from the ground up to avoid a lot of the major pitfalls we saw happen on LJ, and we'll clarify any of our policies upon request (and change them as necessary when it turns out that we've gotten them wrong). Some things -- most things -- won't even be an issue on DW, because our policy on it is "we don't care".

Some things we won't be able to fix, because they're fundamentally intrinsic to the nature of any sort of Terms of Service enforcement effort -- someone's always going to walk away feeling like they've gotten shafted. There's absolutely nothing we can do to alleviate that, except build our policies via community consensus (as much as possible -- there are some things the law tells us we have to do, like our copyright policy, that we only have so much wiggle room around) and build as much transparency into the process as we can while still respecting the privacy of people who are bringing issues to our awareness.

Having the experience of seeing what worked and what didn't work on LJ -- and, specifically, seeing what led to that culture of mistrust and antagonism, and what made it worse -- is letting us design a system that (we hope) won't lead to the same culture of mistrust and antagonism. It's also let us see, for instance, where a technological solution is better than a social solution or some kind of Terms of Service enforcement effort.

I've actually deliberately held off on writing any of our enforcement policies other than the DMCA compliance policy (required by law), specifically because I know that Dreamwidth isn't going to be LJ and the issues that arise on DW won't be the same issues that I used to see on LJ; I don't want to pre-bias Dreamwidth's community development by laying down guidelines and policies we might not ever need. I figure that our first six months or so is going to be a time of growth all around, including in site policy. I will be writing up the proposed enforcement process before open beta, and we'll toss that out there for community review; I've been wanting to wait until we get more of an audience for that, though, rather than letting debate and discussion be dominated by just the voices of those already involved in development.

To those who are mistrustful because of past history, all I can say is: like

mark says, trust is something we earn (or don't earn) over time. This system we're building here (not just in Terms of Service enforcement, but overall) is considerably driven by "if we'd known then what we know now". A big part of that, as it specifically relates to Terms of Service enforcement, is building a system where criticism of ToS enforcement (and there will always be criticism of ToS enforcement) is grounded in fact, not rumor, and a system in which it is trivially easy for us to provide exactly the level of information that a user wants to know, so that people can educate themselves on our policies, their rights, their obligations, and what we expect of them without the debate taking over the entire service.

Thank you for the time and thoughtfulness you put into this answer, I'm glad I came back to find it.

I had a pretty good guess that "if we'd known then what we know now" was going to be part of whatever you said, and it's good to see people taking what didn't work and making it work in the future.

So, yeah. Thanks.

Since I was the one who started this thread, I wanted to point you to a post I just made that lays out my own take on this question. Just FYI.

Thank you very much for the link :)

I did just get it. That was weird. the two showed up right on top of each other.

Thanks for the attention!

Saturday Open Q&A

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

no subject

Free accounts, both native and openID

Re: Free accounts, both native and openID

no subject

no subject

no subject

Who works here? Who will work here in the future?

Re: Who works here? Who will work here in the future?

Re: Who works here? Who will work here in the future?

Re: Who works here? Who will work here in the future?

The Source of Dreamwidth's Userbase

Re: The Source of Dreamwidth's Userbase

Re: The Source of Dreamwidth's Userbase

no subject

no subject

no subject

no subject

no subject

no subject

no subject

..in the correct thread this time, please...

no subject

no subject

no subject

no subject

no subject

no subject

no subject