The Diaspora that wasn’t, and the way into the walled gardens.

About a year ago, I deleted my Facebook account. This was just after Facebook introduced the “like” button on other sites. It didn’t take me long to figure out that what Facebook was doing was moving from beyond their little walled garden, and instead were moving into tracking my movements on various sites. I found this whole thing reprehensible, and decided that enough was enough. I deleted my Facebook account, and blocked at the router. The cut was clean.

Recently, JoDee did a search for someone that she lost contact with a while back. One of the results was a picture of this friend via Facebook. I asked her if she wanted me to lift the ban, temporarily, but she declined. That got me thinking about my Facebook account deletion.

Before everyone and their brother hopped onto Facebook, I got more comments on my blog from folks that aren’t the usual tech suspects (you’ll probably see a who’s who of the tech folks in the comments section on this blog entry). When I had a Facebook account, I would syndicate the blog feed over to the notes section, just as I had with Livejournal before it. What I noticed is I would get comments on the blog entry over on Facebook. Heck, I even got posts from family members that I hadn’t talked to in years when I posted that my Grandmother passed away. Facebook had given me another means of communicating with folks. Its’ not that the means of communication was unavailable to most people (after all, the blog is still here, and comments are open), but what Facebook  gave them was a convenient dashboard where they could easily read, review, and comment on a post. When I gave up Facebook, I essentially deprived those folks of that convenience. By making a choice between my own personal privacy, I made a choice that I didn’t want to use their platform for communicating anymore. Unfortunately, with the way that  Facebook is designed, once you leave the system, you can’t communicate via the system.

When I was in college, just as the Internet was making inroads into most systems, we had a VAX system from DEC. Every student was granted an account on the VAX machine, where they could do things like use WordPerfect, send e-mail, and so-on. As I became friends with some people, I was introduced to a mailing list called The Programmers League, or TPL. TPL was a social group on the VAX, where folks could send messages to each other. I actually was granted permission to add folks to the mailing list at one point. It was hugely fascinating, and very social. And, much like Facebook, once you left the confines of the VAX machine (via graduation and account deletion), it was no longer available to you. I’m not sure if it carried on after I graduated, since I’d moved on to the Internet proper via MERIT accounts, and the like. (e-mail addresses were like Pokemon: collect them all!) However, even if I wanted it to carry on, eventually at some point it wouldn’t. Eventually, the accounts for alumni would be purged.

Just about every person involved with Free, Libre, and Open Source Software (referred from here-on as FLOSS) recognizes the problems with Facebook: it’s a walled garden, sharing of personal information is opt-out (assuming you can find it), questionable practices regarding tracking for advertisements, questions of ownership of data, and so on. Even more folks recognize that Facebook is the 800lb gorilla in the room (What does an 800lb gorilla do? Anything it wants). What is less apparent is what the appropriate FLOSS response to Facebook should be.

Diaspora is one of those responses. They’re not necessarily the only response (there are others) but I think it’s indicative of the wrong sort of response to this problem. The biggest problem with Diaspora today is it solves the wrong problem. Diaspora is essentially a clone of Facebook with all of the privacy controls brought to the forefront. While this is indeed one of the problems with Facebook, the solution in Diaspora is misguided in thinking this is the only problem with Facebook. If Facebook were to adopt Diaspora’s privacy controls, there would still be problems with Facebook. Diaspora’s approach is fundamentally flawed. Unfortunately, they have enough mindshare from their campaign to get started that folks may think this is the best that the FLOSS community can do. They may settle for what Diaspora offers. That is absolutely not what FLOSS should do.

One thing that FLOSS gets right is open protocols., for all of it’s warts as a community, gets that the problem with Twitter isn’t that we need to have access to the code (although that is one problem). The problem with Twitter is that it too is a walled garden. In order to communicate with anyone on Twitter, I must have an account on Twitter. (and the underlying software, gets this right by allowing federation using OStatus. Federation via OStatus allows me to set up a instance wherever I choose, and allows me to follow folks on other instances. It’s a brilliant approach, and I hope it gains more momentum. Unfortunately, it doesn’t have enough momentum right now to make Twitter adopt it. And why should Twitter expend their energies to adopt OStatus? After all, they’re the ones with the larger community.

And that’s what FLOSS needs to build. We need to build communities and software around open protocols. We need to make it so that social networking is a protocol, much like e-mail is a protocol. We need to have sites and services that can communicate with each other seamlessly, so that I don’t have to care if someone is using Facebook or Twitter, nor do they have to care if I’m using or Diaspora, yet we both can communicate with each other. We need to create critical mass around open standards and protocols. We need to implement sites with protocols like One Social Web. Every social media application from FLOSS needs to implement these, so that no matter what you use today or tomorrow, you’ll still be able to communicate with each other. If we can solve the problem of making social media as ubiquitous and easy to use as e-mail, Facebook and Twitter will want to implement these protocols or be left behind.

I presented these ideas to folks in my local Ubuntu channel, and they were skeptical. Indeed, taking on the behemoths of Facebook and Twitter will be challenging. Getting folks away from the social crack of these services will take time and energy. Much like lecturing a junkie that they’re ruining their life with drugs, FLOSS will not win by lecturing folks that they’re giving away their privacy and freedom to corporations. FLOSS needs to build a community and momentum around open protocols, and build sites that use these open protocols. The Internet e-mail system got AOL and Compuserve to open up their e-mail systems back when, and I think we (FLOSS) can do it again with Facebook and Twitter.

And then we can have a true diaspora.


  1. nony says:

    so do it already. I’m not talking about a huge start up, just a platform to use. Accept it will be a money pit, get financing based on that. If there exisisted a uniform place where I could be ‘social’ that would be there tomorrow, be consistent, easy privacy, well, you get the idea, I would use it. and over time wouldn’t just about everyone else?
    As users drop by and see for themselves wouldn’t they tend to migrate away from the platforms that treat them solely as ATMs?

    Last note, we have all become addicted to internet time, patience is a thing of the past?

  2. craig says:

    That’s essentially why I wrote this post. I think there needs to be a compelling reason for sites like Twitter and Facebook to adopt such changes. That compelling reason would be a bunch of other sites that can add value via open protocols. It’s not going to be just one effort on the part of one startup that will make this happen, though; it will be the continued vigilance of FLOSS projects to ensure that we adopt something, and keep using it. That’s the gist of the discussion.

    Thanks for reading and for posting!

  3. Raugturi says:

    I like the idea, but it will definitely be a slow road. Even after you have all the pieces in place to convert users, they are going to still want to communicate with existing Facebook and Twitter users. If you don’t build modules that allow them to hook into existing accounts on these services, people won’t want to switch because then they either have to turn off the existing services or use multiple sites that they can’t link. The obvious solution, and what will most likely happen, is that modules to post to these services and receive updates from them via their APIs will be developed, at which point Facebook and Twitter still have no reason to adopt the protocol.

    Only after a significant number of people have started to migrate would it reach a point where Facebook and Twitter have to open up to it. It seems like Twitter would be the easier conversion since they don’t have ads and therefore don’t lose anything by not having page clicks. For that same reason, Facebook might never come along. Keeping people on the site to generate ad revenue is their only goal. The only way opening up to something like this is good for them is as a last ditch effort to keep whatever user base is dedicated to their UI and want to stay but still be able to talk to all their friends who have left it behind.

  4. Iain says:

    Thought provoking post. I was on but I found it just didn’t offer anything over Twitter, other than openness, which is not enough.

    Diaspora at least tackles an important problem while being open. This _may_ get it the traction it needs.

  5. Rick Harding says:

    The fault I see with this idea is I can’t seem to find an example of a protocol winning out over an existing mass market/mindshare solution. Examples like email, http, dns and such are nice and all, but they were the early ones to get the mass adoption. Things like jabber, opensocial, etc are great and open protocols, but that alone doesn’t make something win. Anyone still messing with that google wave stuff :)

    Unfortunately it’s a “first to mass adoption or bust” out there. Especially once you start getting the kinds of users that facebook is getting. The ones that are never apt to change (AOL says thank you for that btw) and slow to adopt in the first place

  6. Elliott says:

    Diaspora is a federated service like They’ve said that they plan on releasing a spec (after all, the program’s open source, so anyone could create a compliant program). You seem to be complaining that they didn’t nail down the spec before they released testing software.

  7. craig says:

    @Elliot: Diaspora is not obvious in their intentions to federate with other systems. I’ve seen several folks ask how to federate between and Diaspora, with the response being cryptic at best. I hope I’m wrong, and they’re working hard on getting federation nailed down, but for now what I’ve seen is yet another service requiring another account. My complaint is that Diaspora didn’t solve the biggest problem of Facebook, which is the walled garden; instead they created disparate walled gardens.

    And yes, they are working on federation, but this post ( seems to underly another problem with Diaspora in that they’re trying to create both the code and the protocol at the same time. I think they need to step back and get some help from the community, but that’s another post for another time. :)

    Thanks for the response!

  8. craig says:

    @Iain: We miss you over on :)

  9. craig says:

    @Rick Harding: You’re right about not too many protocols getting mass adoption, but I think this one is one where we have to try to get the protocols adopted first. HTTP got traction because the protocols that were in place (Gopher, FTP, Archie) sucked, and the browsers were way better. I think if we get the protocols out there, and get some decent proof-of-concept services together, there may be enough push to get them adopted more. Ultimately, I’m convinced it’s not going to be just the FLOSS community that drives these protocols; we’ll need something like Google or another large and FLOSS-friendly company to adopt them to really make a standard. Maybe this is a windmill to be tilted at, but frankly I don’t see much alternative other than waiting for Facebook to do the Myspace implosion. We need to take this shot.

  10. l.m.orchard says:

    The reason I pick up things like / and then wander off from them later is that all the people I want to talk to are on Twitter and Facebook.

    If I limit myself to the FLOSS solutions I’ve seen so far, I miss out on a lot of daily connections. Even my Mom has stopped sending me email, and goes straight to Facebook. And, though it makes me feel ooky, I can’t blame her because I answer and she gets no spam. (Or, at least, the spam is for Farmville instead of penis pills)

    What that leads me to think about, though, is that we need solutions that embrace and extend platforms like Twitter and Facebook. I think we need to build apps that let you use what’s popular now, but with escape hatches. Build a better Twitter client and Facebook app, but with OStatus / Salmon / PubSubHubbub hooks.

    Then, we can tell a better story than trying to lure people over to parallel universes where their friends aren’t. Maybe the next time Facebook craps all over privacy, there might be an app to promote as an internet condom to use in-between as protection.

    Eh, still thinking about things. Maybe that’s not the best message to promote :)

  11. l.m.orchard says:

    (And of course, I wrote that comment without refreshing, so I didn’t see what everyone else wrote in the meantime :) )

  12. craig says:

    @lmorchard: Oh absolutely. When I deleted my Facebook account, it wasn’t a decision I took lightly. I haven’t heard froma lot of the folks that I was in contact with since I deleted it. That’s why I think we (FLOSS) need to come up with the open protocols to make it so we don’t have to care what system someone uses.

    Essentially it’s a selfish plea on my part. :)

  13. craig says:

    @Elliot: Also, Diaspora is creating a protocol that appears to federate disparate Disapora instances. What I’m talking about is more of a protocol that doesn’t care about platform, and just sends the message across. It’ll be up to the platform to decide what it will do with it.

  14. l.m.orchard says:

    Oh, and I have another blog post rattling around in my head about this, but I have a suspicion that free hosting with easy signup has a lot to do with this stuff.

    I suspect we have the parts & specs to put together some interesting apps, but they all need servers. Maybe that calls for friends-and-family hosting by magnanimous geeks, maybe at-home hosting with a PogoPlug / SheevaPlug appliance.

    There’s talk at Mozilla back and forth about this kind of stuff, but Mozilla is good at scaling mostly because Firefox users supply their own hardware.

    I guess Thunderbird has been catching some grousing lately because users are increasingly expecting the app to give them a free inbox when installed. Madness.

  15. oiaohm says:

    Really we need to get a clear meaning of what a wallgarden is.

    Diaspora really fails and is a wallgarden just not as bad as facebook.

    List of wallgardens questions.
    1) Can I run the software behind this service.
    Facebook no. Diaspora yes. So not as bad here.
    2) Do I retain ownership of my work.
    Facebook no. Diaspora yes again not as bad.
    3) Can a user backup there own data out of hosted provider simply and restore on a different server. Note the word user not system administrator or anyone else who would have direct access to the database.
    Facebook no Diaspora no.

    Of course once you have the means to backup. Differential would also be useful. So a person could maintain there master copy on there pc and sync with the service provider server. For the user this means if there service provider goes down no problem. Just get another service provider and sync. So user is free of the walled garden. Truly free without having to run there own servers with open access to the internet. Even can play into helping free up users options to host.

    Diaspora is not the only software failing to provide users with the means to backup what they have submitted. Like some forums you would like to be able to take a complete thread for offline reference but there is no export options.

    Remember not everyone has the means to run there own servers. What would happen if all of facebook disappeared of the internet. How much information would be lost because people cannot backup their information.

    Many blog sites are the same thing. Lack of backup functionality.

    First thing about breaking down the wall-gardens is scrappers to pull data out of them. Next is open source hosting systems that don’t require scrappers to get data back.

    Finally and most importantly proxy posting. So if I have to use somewhere like facebook to contact or talk to somewhere. I have the option of avoiding having to scrapper to get my information back. Proxy posting means a person never has a reason to return to the old interface.

  16. craig says:

    @oiaohm: Data portability is one part of the equation, but that’s not what I’m referring to when I speak of walled gardens. I’m thinking more of having to be on a particular platform or website in order to use their service. Twitter is a walled garden to me, even though they have an open API. is not, because they allow folks to federate the content to other sites. This is where Diaspora, Facebook, and Twitter get lumped with the same brush of walled garden-ness.

    Scrapers aren’t the answer. Scrapers are a hack to pick up the data from a site at a point in time. True openness can only come from commonly accepted protocols that are shared among many platforms.

  17. oiaohm says:

    Scrappers fall into providing a way for people to leave with there data. If you have years of data locked in a wall garden walking away is hard.

    Remember most people quitting a service only need there data from 1 point in time.

    Scrappers are only stage 1.

    Stage 2 is having something open source that can host the information scrapped that can be backed up by user so they will not have to scrap from it to get there data. Preferable with a open specification how its stored. behind fails my tests. Users are not provided with a way to extract there complete profile of data and to restore there complete profile of data.

    Problem is it possible to federate the content to other sites but still no provide user with a simple means to backup there content. is still weak.

    Yes we could have protocols that shared site to site that still don’t allow users to backup there data. Remember not all users of sites will be running there own sites. This is the problem with your wallgarden define. Its bias to people who can run there own sites so can federate on to there own servers.

    Open data formats for storing the backups are critical. Really the means to backup in a processable format is more important than commonly accepted protocols.

    Why if you have the data you can reprocess it and resubmit it elsewhere.

    In the time of PC we had wallgardens in the forms of applications file formats. But due to us having the data it was possible to reverse and get the contained information back when the program failed to operate. Yes painful but possible.

    Time of clouds we are looking if a site fails we lose all the data in them unless we are either the person running the site or have scrapped content. Or are on one of those rare services that are open source and allow backing up of what you have done.

    Scrapers are 100 percent not the long term way I know that. Classic example of why scrapers are not is when google bot went threw a particular site a long time ago hitting all the delete buttons.

    Sites that force uses to use scrappers to get there data are risking the users data due to the fact scrappers may have bad side effects on the site. Remember google is the biggest scrapper out there. So there are very valid reasons why people should want backups of there information from sites.

    Clouds simply threaten the fact the data was never stored on your hardware. And if they disappear you lose everything.

    True proper federation of data has to include a path for the person with just a browser to download and upload the stored information on the server and have the means to run the server elsewhere. Why is this important what if you federated server is off the internet due to fire-walling. Yes these do exist. Straight site to site federation can be impossible.

    The problem here is we don’t have a proper define of what makes a wall garden. With a proper list of what has to be provided not to be a walled garden by anyone measurements.

    My list is fairly simple.
    1) users must be able to backup there data.
    2) Backups must be a open format users could process.
    3) users must be able to restore there data. Preferable to many different services in a controlled way.
    4) User must not have to be disconnected to change over. is good here. twitter users swap over to identi people following on twitter can remain doing so.
    5) Must have the means to run my open server if I wish.
    6) Must not be required to run my own server or be on the internet to access the backups.

    Direct sharing of data between services you notice is not on there. Its a nice extra feature yes. But for my data its not critical. Reason if I have processable backups and I can upload data from backups into different services I don’t need common protocals between services.

    Basically common format of backup with common tools to decide what to send back online is all that is required to share the data between services the long way.

    Common protocols between servers are a niceness. Not require for it not to be a walled garden.

    Problem is open source and closed source web applications have got really good at building wall gardens. Very poor at building non walled gardens. Its about time we see the issue.

  18. craig says:

    I’m not sure what you call a walled garden isn’t better defined as data portability, which is a very important, but tangential concern. I agree, having the ability to move data from one platform to another is important ( is the most recent example of this need), but that’s not the most important concern for me.

    I want the ability to communicate in real-time with folks on other systems. Data portability doesn’t help me with that requirement.

  19. kumara says:

    What if Facebook/Twitter are the new Windows. Windows had/has many flaws, but it gained market dominance at the right time; another commenter made this point above.

    Each time we get higher in the stack; ethernet won vs. token-ring, intel won vs. motorola, microsoft won vs. unix.

    I think this battle is done. Diaspora and its successors will remain in the margins like Linux has.

    Disclaimer: I’m not on facebook or twitter. But once grandparents start using it, the competition is over.

    My target would be the next layer in the stack; perhaps in-home streaming secured via pgp/aes using videowalls/holograms to make unified households across distance. Why read you sister’s tweet when you can just have her strolling past on the kitchen wall occasionally while you’re having dinner and she’s having breakfast. Instead of our data living on fb servers, have it secured at your place of residence. Then tweets & statuses become second rung to your life feed, much like Windows is now the ugly thing pretty girls use to get onto facebook.

  20. Human Plague says:

    You pretty much summed up why everyone flocks to Facebook in your third paragraph.

    The value of a social network is the society, not the network.

    A historical “nerd” equivalent of Facebook/Twitter was Slashdot. The reason it wasn’t Facebook or Twitter? Because it was Slashdot. It lost its lustre over the years, but the point can’t be lost on those of us who read a blog post that mentions VAX? Fidonet and Usenet also qualify. Anyways…

    Facebook has everyone. (Note: I also deleted my Facebook account. I’m not being factual here, just trying to make a point). The trolling and spam that plagued Slashdot are destroyed by convention on Facebook. That is, on Facebook, you’re actually you. Not a handle or an alias. Your friends are people you probably know and don’t want to piss off with clever ranting. In 1997 this would be absurd, I would never join “some forum” and allow people to know who I was from the get go. Fast forward to 2010, the idea of being you is not that crazy anymore.

    Twitter has advertisers and celebrities; that they have the concept of “Verified Account” should speak volumes. Twitter’s character limit forces everyone to be intellectually the same. It’s a great leveller. Combined with star power and constant mention in media news broadcasts, the rest writes itself. “Your not on Twitter?!” /advertising trick to make you feel like a loser. Quick join!

    For historical posterity: MySpace was a legitimate social network until bands cannibalized it with spam. Friendster was a legitimate social network too. Common theme: exodus.

    Protocols only solve the following problem: If something turns out to be a fad, at least it wasn’t locked in.

    The FLOSS community can’t solve Facebook. We’ve had our own communities for decades. No one wants to join them because they are “lame.” As much as geeks have an aversion to centralized data, everyone else wants to be part of something cool whatever that thing is.

    PPS: Richard Stallman doesn’t use web browsers, he sends a link to a demon that uses wget to fetch the page and sends it back to him. He doesn’t seem to have a problem communicating with the world at large?


  21. craig says:

    @Human Plague: I think we can build communities, but yes, if we build communities that are strictly tech-oriented, we’re no better off than we were beforehand. However, if we build the infrastructure so that someone else can build the community, but still communicate with our communities, we’ll be much better off.

    And yes, the spam and bad-manners of some can really ruin any community. That’s partially why some folks left from I maintain that there’s still enough good there, and I hope that folks come back, but I think I’m in the minority. :)

  22. suitti says:

    Brilliant post, BTW.

    My wife turned off her Facebook, i’m pretty sure, because of Farmville spam. There’s a way to turn off Farmville from your ‘friends’, which i’ve done. One of the problems with FB is that once you’ve let an app into your life, there doesn’t seem to be a way to get rid of it. Once you’ve entered information, you can’t get rid of it. So, i don’t, in generally, allow FB apps to do anything. I may have slipped, though, and i have no idea how to find out.

    I’d never have started FB. Some of my friends demanded it. And, i’ve built a little community there. But, it’s not ‘friends’. And the whole emotionally laden nomenclature bugs me. You ‘like’ a post, or even a poster. But ‘unlike’ turns off your like back to neutral. It’s not negative. Really, by default, you ‘unlike’ everything. It’s stupid.

    And i’ve built a community on FB because i was already there. I’m sure there were better choices. If there were a clearly better long term choice, maybe i could get my communities to migrate enmasse.

    It’s been a year or so since you dropped FB. Any word on if any of your FB data is still somehow available?

  23. @suitti: Frankly, all of my Facebook information was either crossposted from Twitter or this site, or was comments from other people on Facebook that wouldn’t make much sense outside of the conversation there. So I’m OK with my data being erased from Facebook. It served it’s purpose, and I’ve moved along.

    I do miss some of the conversations and pictures over there, and that’s why I’m so hot on creating more open communities. I’m tired of everything being behind a wall.