To begin, congratulations to the Diaspora* team on their initial fundraising success. I fully appreciate the sentiment behind their project; they have touched a nerve among the thousands of people who feel jilted by Facebook’s recent actions.
That said, the idea behind Diaspora* (its implementation) is overly complex and won’t work for the large majority of Facebook users. That is not to say that there is no problem to fix, as we’ve seen online privacy is being eroded by Facebook at wicked speed. New Facebook users don’t understand how the site’s privacy settings work, and are thus overly exposed from their first day on Facebook on. Facebook does little to help them.
As Facebook finds its legs among the older and less technologically inclined, it needs serious competition to force it to abdicate its new system of assumed openness. When Diaspora* was announced, it garnered press, excitement, and over $100,000 in funding because people understand that. No one wants to let Zuckerberg decide the future of online privacy by himself. Given all of that, how does this Diaspora* work, and how well will it be received?
Technology Overview
Diaspora* gives its users privacy by forcing them to host their own node that contains their information. If the company does not have access to your data, it can hardly infringe on your privacy. In Diaspora*’s words:
Diaspora* aims to be a distributed network, where totally separate computers connect to each other directly, will let us connect without surrendering our privacy. We call these computers ‘seeds’. A seed is owned by you, hosted by you, or on a rented server.
The seed is an aggregator that will gather your content from around the internet into a central location, your server. Link your seed to my seed and we can share and converse. Exactly what the connection will do between the seeds is yet to be fully fleshed out.
In short, you host your own profile that is a amalgamation of your life online, and share with a select group of people.
This Is Too Complex
Not to say that hosting a personal FriendFeed and attempting to link it to my friends and their own FriendFeeds does not sound like fun, but it would be mind-numbingly tedious for most. The friction to get started is far too high for this project to ever gain real traction.
“But wait,” you may cry,” if the company has raised so much money why is its idea untenable?” It’s simple. People are mad at Facebook and they wanted to vent. They blew off steam by throwing dollars into the Diaspora* project.
Your parents probably recently joined Facebook. It must have been a hellish weekend walking them through the website and teaching them how it works. Imagine trying to get them to set up their own server to join Diaspora*. We technology minded people would find it a fun challenge, for everyone else it is a joke to even consider.
The Simple Solution Is Not
Diaspora* hopes to alleviate the tension that I just outlined by letting people rent space on communal servers. All the technological problems are therefore moot, and Diaspora* has a chance, right?
Not at all. By moving the data to a server that you are renting part of to host your seed, you are losing the whole point of Diaspora*: control over your data. Fine then, you might say, let the company rent server time to its users thus ending the fear of letting a sketchy company control your seed. That is another non-starter. Diaspora* does not want to control your data, and you do not want to afford them the ability to do so. If you want full privacy, you cannot let Diaspora* host your seed.
You either take the time and headache of setting up your own server, or you violate the core point of Diaspora* and pay someone else to handle that work for you. Either way, something is not right.
Will People Pay?
If they cannot handle the tech problem of setting up their own seed server, and are content to live with someone having their data, users still have to pay to have their seed hosted. In short, they have to pay for what they used to get for free, all the while not solving the problem they set out to do. Even more, if I am paying for hosting the more I use the host the more I have to pay. In short, the more I use Diaspora*, the larger the bills I might be on the hook for.
What is the value proposition?
All Together
Disapora* has the right drive, but the wrong vision. The good news is that the team will have more than $100,000 to work with this summer. They can fine tune and refine their idea to make it something that we would use.
Facebook needs to be taken down a peg, and these might be the guys to do it. With proper guidance their enthusiasm could become a social revolution. Good luck guys.
Totally agree. If you have to ‘set up’ a server, you’re done. The implementation has to be wrapped up in an application – think dropbox for activity streams.
Now *that* is an interesting idea.
So then is the real opportunity for Diaspora* or another company to find a way to make ‘setting up your own server’ as much of a no-brainer as setting up a playlist for your iPod? MP3 players used to be complicated, too.
Ditto-it’s not so much that you HAVE to use your own server as it is that you DON’T have to use Facebook’s. Would email be so great if we all had to use AOL? This presents the opportunity for choice in the marketplace.
If you say ‘server’ too loudly you can clear a room of non-tech people.
I don’t think server, I think Boxee Box. This could have legs and sold in Best Buys if they want it to be big.
Yeah, so don’t say server too loudly. Say this: “easy to set up personal space on the world wide social network for $9.99 a year, includes 1 free WordPress blog and 5 e-mail addresses”
Seriously, it’s you that has an idea problem, or rather, lack of vision. Look how many blogs there are. Some are created by techies that first get hosting and then FTP their WordPress and theme, but many others are just accounts created on WordPress.com or Blogger.com
If the word server does not scare you, then your personal space (or seed) will be awesome, because you can tweak it, create custom plugins and what not (just like a PHP programmer or HTML + CSS guru can do on his/her own WordPress install). But if you’re not, then you’ll just have to get with the hosted seed option on GoDaddy with a choice of 5 themes and 10 of the most popular plugins.
But if you have a tech friend or family member, maybe he can set up a custom seed on a shared hosting account, split the costs of hosting with the whole family, and now if you want that kick-ass NFL-score-tweet-profile-updating-theming plugin, you ask your nephew to set it up and he gets a nice present for Christmas in return.
Serious… lack… of… vision. I see endless possibilities, and lots of people underestimate that.
I think you oversimplify the possibilities. It isn’t just about individuals. I might invite a bunch of my friends to join me in paying for a server. If I got 40 of them to join me, which wouldn’t be difficult, we could have our own VPS at slicehost.com for the princely sum of 50 cents a month each. A couple of us could administer the site, which once established would be a minimal commitment. The smallest slicehost VPS would be more than enough to host the social media activities of 40 people. It’s simply a matter of trust among people you know, rather than people you don’t.
Also, I think diaspora could be made into an easily installable program that could use a proxy server to just provide its open network endpoint to the net without requiring the user to set up port forwarding on their router, or leave their desktop machine open to the internet.
At some point, this kind of Social Network of One (to use Jeff Pulver’s phrase) must become a reality. It’s the only way people can feel safe with their social networking.
If it must come about, then work on it. In fact, help these guys do it. But from what I see so far, this is just not it.
Setting a server sounds daunting but it isn’t. Look at the whole bunch of services that run to your PC. A proper installer can do miracles.
Also, having diaspora hosted somewhere does not defeat the purpose: Facebook is able to exploit our data because it can read ALL the social graphs and all the interactions. In a hosted environment, the hoster might be able to read my graph but there is a high chance that none of my friends is also hosted there.
After I install my littler server app, what happens to my seed when I close my laptop and head downtown? Am I dead on the social network for that time?
And so forth.
I agree that the idea has problems, but you seem to be assuming a priori that these problems are (a) unknown to the people behind Diaspora, and (b) not solvable.
The latter is certainly untrue, and while it’s possible, it seems highly unlikely that no one from Diaspora was reading your post here.
Oh I am certain that a member or two of the Diaspora team has read this article. They’d be crazy not to. *wink*
Not once they have sorted out the distributed backups that are on the road map. You did read the presentation? Right?
You get the point of Diaspora and are able to explain it well, good job.
“By moving the data to a server that you are renting part of to host your seed, you are losing the whole point of Diaspora*: control over your data.”
This makes a false assumption: that the choice has to be black-and-white, all-or-nothing. Handing over all of your data to the privacy-free disaster that is Facebook is a far cry from using an instance of Diaspora* hosted on someone else’s server. Even if you don’t have the funds or the know-how to set it up on your own web server somewhere, the idea isn’t necessarily off the table: does wordpress.com work? Does it cost money? Is it known for horrendous invasions of privacy?
Clever thinking and *ethical* use of advertising (or maybe a small monthly fee, or whatever else) can solve these problems. Their largest concern should be the (remote) possibility that Zuckerberg/Facebook will have some kind of privacy epiphany and reverse their stance on these things — that would torpedo Diaspora* pretty thoroughly.
Either you have complete control over your data of you do not. Sounds simple to me.
Sounds oversimple to me.
“Social networking” and “having complete control over your data” are incompatible concepts. Either you are sharing some data with some people, or you are sharing zero data with zero people. Let’s ignore that and move on, I guess:
If a guy can’t host his own solution on a server absolutely nobody else has access to, he should just give up and use Facebook with all security options set to “everyone”? Of course not, that’s ridiculous.
There is a gray area, a middle ground. A service provider that agrees in writing not to mine your data for marketing information, and a social networking app that gives you intuitive, fine-grained security controls — that’s a good start.
Once again I have to ask why having your data mined for marketing information is a negative thing? You are going to be seeing ads on sites no matter what. Right? So what is so wrong about having ads related to what you are into. It would feel less insulting than seeing irrelevant ads.
I am not understanding what you and everyone else seems to be afraid of. What is the worst case scenario in your mind? Please enlighten me.
Thanks.
If your data is open for advertisement purposes, it becomes too easy for anyone to vacuum up what you read, what you like, inferences are made and the next thing you know!
Insurance denied
Job denied
other inferences both commercial and personal are made?
To avoid the worst case scenario that raycote stated, why is anyone on Facebook using their real name and date of birth anyway?? If you don’t want your private information used by the Facebook vampire Zucker, then why give accurate information in the first place? Use your first name, spell your last name incorrectly and pick a birthdate that is not your own and easy to remember and the privacy problem is gone.
I think this whole privacy issue is a storm in a teacup!
How can anyone use my false information for anything?
Complete control over your data is an illusion most of the time, because you can always lose it to hardware or software failures, which is why we compensate with backups.
The facebook problem isn’t about the code that manipulates the data, and the decisions it makes. The short term best interests of facebook are not those of the users… but facebook is paying for the code, so they make the choices, and we users either accept it, or leave.
If we are paying to run code in a server, we get to call the shots about the defaults we choose. I’ve paid other people to host my web services for years, nobody ever censored it, nor put ads on it… because they would cease to be paid if they did that. Same with hosting a diaspora “seed”… it won’t be messed with by the hosting service.
unfortunately you will never have control unless you are on stand alone computer, any time you show someone something, they can pass it on
Diaspora could open up a new server market if they develop a large enough following but I doubt it will happen really, unless the media really gets behind them. Still I would find it hard to believe that a significant number of people would pay to have their social data safely hosted somewhere. And like you say, if it Diaspora does manage to make a dent, Facebook would certainly do a public apology and return back to their original privacy settings which would obviously tank Diaspora. I said in my post below that this is probably what needs to happen.
Who knows, maybe Diaspora is just a secret experiment being carried out by Mark Zuckerberg to really see if people give a shit. I would not be surprised if this was the truth.
Good point. Once your device or computer is off, then your info is gone unless they make a system where your seed is constantly, or periodically, pushing your data to your friends’ computer. So basically your friends computers will store a snap shot of your profile (info, pictures, updates, videos, etc). Ok ok. I know. That might sound ridiculous now but in the future when the net gets faster and when we make a quantum leap in storage, this will be a snap. Take this for example. The memristor is any of various kinds of passive two-terminal circuit elements that maintain a functional relationship between the time integrals of current and voltage. It is said that the memristor will allow multiple petabits to be stores in one cubic centimeter of space. Source: http://highscalability.com/blog/2010/5/5/how-will-memristors-change-everything.html
Perhaps I just went way off on a tangent but I was seeking a solution to the problem. Anyways, the fact that your data will not be visible while you are offline could be a huge problem for some. then again, isn’t that more like real life? Its not like we get to experience our friends when we are not spending time with them. So this will bring us closer to physical reality. The issue is that everyone is so used to having access to their contacts’ profiles all the time. People will certainly not like the idea of not having access to their friends profiles.
I can see some people using a product like Diaspora as a way to back up their social profiles, but I do not see them transitioning over to Diaspora all together. The Diaspora team also admit that this is a service that will be complimentary to pre-existing social networks.
In fact I do not really understand what the problem is with Facebook’s lack of privacy. All it means is that you will have more accurate ads shown to you. It’s a whole lot better than having irrelevant ads shown to you. Additionally, a Facebook user will have a more personal experience across much of the web (as long as web sites/blogs implement the new social plug ins).
Diaspora also use “What if a social network you are on suddenly crashes (all your data is deleted) or they go out of business.” as a selling point to join them. What is the likely hood that this would ever really be a problem? I personally think it is very low but seriously, I would love someone who specializes in determining the likely hood of an event like this of happening. Sure an EMP from the sun or weapon could do it. But this is an ultra rare event I’d imagine. So yes, there is a possibility of another social network going out of business but if you were on that network then you would receive their notification and then proceed to back up your info manually. So in this case it would be nice to have a Diaspora seed running on your computer but once again, the likely hood of that happening is slim to none (unless someone could prove me wrong).
So the whole idea, is a nice try but, really when you really think about it, what they are fighting against is not a big deal. And if Diaspora does gain traction and becomes an actually threat to Facebook, then guess what? Facebook is going to take action and make everyones’ profiles a whole lot more private, like it used to be. And maybe that is what is needed. Maybe that will be Diaspora’s ultimate purpose/ Time will tell.
If I have $100.000 right now, I will tell everyone to use opera unite instead …
Totally what this article made me think of but i wonder if we’d have a hard time getting people to install opera? firefox plugin? ie tool? Dropbox is also an interesting idea I’d looked into, but they seem averse to having people build an apps around it (google code project has been dead for almost a year).
The emotional issue that sparked interest in Dispora is control not privacy. And the answer is not as black and white as Alex makes it out to be. All we really need it an alternative, or several alternative repositories of our data and the ability to move from one to the other without losing the connectivity to our network of friends. If Dispora only creates a model for the emergence of multiple hosted Facebook’s with lower switching costs, we will be able to participate in the medium and each of us individually will be able to vote with our feet to express our comfort or discomfort with the service providers hosting our data
I totally agree that the Diaspora team is getting a lot of attention primarily due to users frustration with Facebook. They aren’t the first to work on such a project either. I recently wrote about a new project called One Social Web that aims to do build a similar service by the proven developers that built Storytlr. This seems like a much more sound approach, but I agree that the chances that it will have any real impact and gain acceptance by mainstream Facebook users is slim.
Thanks Mark for the mention of Onesocialweb. I want to clarify one point: in our case, the target is not the end user. We do not expect you to setup your own box at home, just like you are not running your own email server. Instead, we want to create a protocol that enables many more providers, just like for email today. You may get an account via your ISP, your telco, your work, your school… and just like email you can use it to interact with others. Facebook may stay the ‘gmail of social communication’, but at least the platform becomes open, enabling new players to join the party and small startups to innovate on top of the infrastructure.
I have been working on a project similar to Diaspora for a few months now (http://hubbub.at) and I don’t agree with the sentiment that running your own server is totally out of the question. Consider SMTP, the email protocol. Everybody uses it, and it uses essentially the same paradigm. Most people don’t have their own mail server, they use some existing service. With Diaspora and Hubbub, this is essentially the same. We also have public servers where anyone can create an account.
The important idea here is to not entrust all your data with a single company that can decide pretty much anything they want. Instead, let’s use an open, decentralized protocol. Like we already do for everything else on the internet.
“The friction to get started is far too high for this project to ever gain real traction.”
Very clever of you, did you travel to the future?
Really, we don’t need cynical people in this kind of project. So go sit in a corner and believe that the work can’t change, while other go off and change the world.
I agree with Mike 100%, it’s so typical of all the naysayers to go “oh, it can never work of course” right up until it does work and then they turn around and act as if they’ve always been championing the idea from the start. I strongly believe at least one of these open projects will succeed and become the standard protocol, maybe even several of them. Social networking won’t stay monopolized forever.
“Imagine trying to get them to set up their own server to join Diaspora*. We technology minded people would find it a fun challenge, for everyone else it is a joke to even consider.”
Why don’t you imagine a profile server that is the size of a USB stick and plugs into your router.
You have no imagination, you only assume the worst. This site is what, “thenextweb, seems like you don’t even believe it can happen.
The discussion misses what I think is a crucial point – a decentralized network could allow multiple servers, but not necessarily one per user.
Look at how blogging works. Sure, you can set up your own personal server with Apache, MySQL, PHP, and install WordPress. Or you can go to wordpress.com (or blogspot.com, or vox, or Windows Live) and set one up.
(Or, for that matter, look at SMTP – very few people set up their own email servers, but the open decentralized standard means you can host your email at any provider and still interact with the rest of the ‘net).
What if Diaspora used this model, allowing people to choose the provider they trust, rather than assuming every individual user must set up with own network?
My concern is that we have all these disparate efforts: Diaspora, OneSocialWeb, The Appleseed Project, Hubub.at, GNU Social, NoseRub, DiSO (and I’m sure several others I’m leaving out). But I don’t see anything pulling them together towards a common standard.
Perhaps a role for the Open Web Foundation?
As I posted earlier, I agree with you 100% about the perceived barrier of entry for individual users. But I don’t necessarily see anything wrong with the disparate efforts, because this is a chance to test several approaches in the wild, before we converge on one or two.
Sure, a lot of energy is spent on products that nobody will ever use, but we’re not sure which ones they are yet and there are many ideas worth exploring. I don’t feel any particular hostility between our groups and as long as we can avoid unproductive flame wars of any kind, we’re probably going to be OK.
Diaspora has introduced a concept, now is the time for innovative architectures and solutions!
@John: Why the rush for an immediate set of common standards? Let the “best” solution win out, and then form standards based on what worked and what didn’t. Hasn’t that generally been the way of the web?
I agree with the notion that social networking could work as a mass-distributed choose-your-own-host P2P solution. How about social networking as a web service? Profile data could be shared over an encrypted connection to various requesters based on the rules set forth by the profile owner’s privacy & sharing settings and the credentials of the requester. Each individual’s social network dashboard could reach out to its saved relationships and connections on a periodic basis to update the “feed,” while employing some update/caching mechanism to locally store and archive retrieved data, thereby limiting network congestion.
Just a note – using your own PC as a server in this type of web-service-based environment would not work too well. Using the email parallel, what good would an email server do you if it was down half of the time? It’s the same here, what good would a P2P social profile hub do if it was down half of the time?
The difficulty posed to non-technical users is a valid concern. The solution lies in creating a 1,2,3 bang-bang wizard offering ease-of-setup to new users. Each company involved in offering this hosted service would probably have to be responsible for billing, hosting, and installation, with the end product taking the form of a web-based wizard for getting started. Once running, the software would need to be robust, easy-to-use, updatable, and maintained.
An interesting issue would be the cataloging of people’s profiles. I think a practical solution would be for a central server to catalog all people available in the global network. Information stored would include an ID or URI and basic information such as first and last name – however, the data shared in this cataloging scope could be completely up to the privacy settings of the individual, even up to the willingness to be listed publicly. Now that I think about it, a URI would probably be preferable over an ID (as it’s the web, not a centralized service), and if people didn’t want to be listed publicly they could still make connections by privately giving their friends their URI.
Note: The “cataloging of people’s profiles” would be important for finding people on the network. Hosted hubs would probably reach out to some central catalog server for the purpose of finding people to connect with.
We can look at Usenet as a somewhat useful model. Not only could an individual’s ISP devote servers using decentralized network protocol, but various institutions, such as schools and non-profit organizations, would probably jump on board as well. Another possibility is that individually owned websites would be designed to “plug in” to the Diaspora (or other) network. These sites might charge a fee or provide ads, but others could very well operate freely too. Finally, it would not be overly difficult for individuals to jump into the mix using hosted space online and do-it-yourself websites geared towards interfacing social networking pages.
A lot of jumping to concusions and sour grapes here I see, lets wait and see, I have some reservations on feasability myself, but am not running them down without all the facts.
Dear sir you are the winner of the “missing the point award”. Diaspora and all federated networks IS the solution and thus are the future. No – your mother and sister do not have to setup their own servers, but your ISP, University can do so and provide you with an account. Diaspora can host a copy of this and people who don’t care or understand about security can use it. The rest of us will use our own. Diaspora will most likely allow each of us to have a local copy of our data or an easy way to retrieve them. Please go read a bit more on how the Google Wave protocol works and you will get the idea. The future of the web is decentralized. The sooner you understand this the better. Thanks.
Do you honestly think the end product will be so complicated? Look at how simple it is for everyone’s proverbial Grandmother to make a web site nowadays.. everyone already thinks in terms of WYSIWYG, what makes you think Diaspora* will be any different? No, I say that you people who are claiming it will be too complicated, are reading too much in to the fundraising pitch. They had to explain how it works in detail or the more advanced users would’ve complained. You’ll see, it’ll be simple enough for everyone when it’s finally deployed.
Alex you are quite right in your assessment. Diaspora is plagued with problems, many of which the boys involved are either ignoring or have not yet encountered, but inevitably will. I question if they are really honestly disclosing the difficulties they are facing in delivering the product they sold the public when raising cash (hence their lack of updates), or are they just blindly moving forward in the hope they will solve any issues as they encounter them.
In their initial pitch the suggestion was to have each seed acting as it’s own server, presumably self-hosted, but this quickly evolved into remote hosting which sort of defeated one of the primary objectives, control. It was a neat card trick that not many seem to have noticed yet, but there are more issues looming on the horizon for Diaspora.
First is the naming system they choose. If they settle for anything other than standard URLs that fully resolve in the standard DNS (as they appear to be) then Diaspora is effectively nothing more than a propitiatory network as the seeds will will need to be reliant upon 3-party resolvers (just like the not-so-open OpenID naming system that is actually proprietary) to communicate to others. By doing this you are kissing goodbye to freedom and control in the process.
The only choice is standard URLs if the user is to have complete control over their IP and connections. The Diaspora boy seemed to have overlooked this point to date and are arrogantly ignoring any attempts to alert them to their mistake.
Second is that to achieve their stated goals the seeds would have to be self-hosted, but as they have quickly discovered (prompting their bait-and-switch) this is not a practical solution at present. Unfortunately most personal computers are not on and connected to the internet 24-7 which creates quite a problem for inter-seed communications especially as they are (apparently) having seeds push their content to others. The problem is compounded if the seed were located on a smartphone (as they have suggested already).
Third is the fact that most people own more than one computer which raises the issue of which computer to locate the seed/server or the spectre of hosting multiple instances over different machines.
Fourth, and most important, is that whilst they are opening up their source code (eventually) we are relying upon four young and inexperienced developers to dictate what is supposedly the answer to Facebook and it’s abuse of it’s power, to the rest of us. The reality is that this approach of ‘trust us we know what we are doing’ by what is effectively a group of dictators will not wash long-term with the public – especially if the claim is that the project is “open”. One must question if these guys are the best men for the job in the first place?
Whilst being a commendable project which certainly has resonance with the public, what is missing from the Diaspora equation is an intelligent DNS that ensures seamless connections by the use of re-routing and off-line request handling. The boys have already overlooked critical components which puts their credibility directly into question, which is exactly my point.
The boys need to realised that if they really want to have Diaspora become the successor to Facebook, they need to get really become “open” and collaborate immediately with the development community and the potential end-users at large, rather than naïvely believing that they should be exercising total control simply because Diaspora was their idea.
For any platform to have any chance of eclipsing Facebook, it needs to be launched as an organic open-source project where it is controlled by no single company or person but by all who contribute or use it. Problem for Diaspora is that at least one such project is underway already and has solved all the issues listed above but is also seeking patent protection which might leave the Diaspora boys with a social network that can’t quite function as promised without breaching patents.
Well, they were warned…
The diaspora project is interesting, but developing, social networking based
on open-source platforms is nothing new. There are several established social
network sites that offer robust tools users expect like blogs, status updates,
friends, photo sharing, etc. without the sketchy, arrogant, sometimes downright
creepy, privacy issues we have been forced to live with using FB.
Viewcaster.net is one example. ( http://www.viewcaster.net/ )
Unfortunately, the unbelievable reach that FB has makes it difficult for the
smaller start-ups to make much headway. MS is a good example of how we can
often follow someone backwards due to their influence (read $$) on an industry.
The Diaspora project is interesting, but the close collaboration they have had
concerning new accounts, etc. with the FB folks makes me think we should
approach with cautious optimism.