Open APIs and Open Standards

As Dave Winer has been pointing out in recent weeks, there is something quite interesting happening in the blogging/microblogging world.

First WordPress allowed posting and reading wordpress blogs via the Twitter API.

Then yesterday our portfolio company Tumblr did the same.

John Borthwick has been advising companies for a while now to build APIs that mimic the Twitter API. His reasoning is that if your API look and feels similar to the Twitter API then third party developers will have an easier time adopting it and building to it. Makes sense to me.

But what WordPress and Tumblr have done is a step farther than mimicing the API. They have effectively usurped it for their own blogging platforms. In the case of Tumblr, they are even replicating key pieces of their functionality in it, as Marco describes:

The really cool thing – because our following models follow a lot of
the same principles, we’ve been able to take advantage of a ton of
native features:

  • Retweeting = Reblogging
  • Replying = Reblogging w/ commentary
  • Favoriting = Liking
  • “@david” = ””
  • Conversations = Reblogs

And as Dave Winer points out, this effectively creates a standard that third party clients can adopt. And Dave ends his post with this highly provocative thought:

If Facebook were to implement the Twitter API that would be it. We'd have another FTP or HTTP or RSS.

I'm still trying to wrap my head around all of this and the implications of it. And I'm not writing here in my capacity as an investor in Twitter and Tumblr or a board member of Twitter. I just think its fascinating and worthy of discussion in this community. So let's get on with it.

Reblog this post [with Zemanta]
#VC & Technology#Web/Tech#Weblogs

Comments (Archived):

  1. Sean Kaye

    Obviously Dave is right, if the almighty Facebook followed suit you would end up with a defacto standard and more importantly, you’d lower the barriers to entry for developers to build really great cross-domain applications.I think Dave is doing what he’s famous for though, stirring the pot. Clearly Facebook won’t go down this road. They are doing an awful lot of work to lock people into their platform and then encourage others to feed off them. While I do agree with the merit of them doing it, I think this is just a case of Dave pointing out the obvious upside while knowing full well it won’t happen.That’s why Dave is such fun to read.

    1. dave

      Your crystal ball is getting a little cloudy. :-)I don’t believe you say I believe.Facebook is getting to be a big company and they employ a lot of people and they don’t all see the world the same way. My guess is that they’re giving the idea serious consideration and could well end up supporting the Twitter API.The reason is pretty simple — their API is all over the map and they need to simplify it so they can get some of the application momentum to flow their way. One way to get a huge dose of that is to support an existing API that’s widely supported, thereby enabling a huge base of developers to quickly get up to speed with developing for Facebook.At that point the question will be does Facebook stay within the confines of the Twitter API and that’s another story.I’m glad you find my writing to be “fun to read.” 🙂

      1. Ken Kennedy

        It’s an interesting idea, but part of the issue with bolting the Twitter API onto Facebook is that Facebook’s functionality is a pretty large superset of the Twitter one, and they already do a poor job of mapping one to the other with their existing tools (like Twitter/ apps for FB sync). A simple example is the fact that in Twitter, an tweet is a tweet is a tweet…but to Facebook, a tweet that syncs over with no link in it is a “status update”, which has one set of privacy settings, and a tweet that syncs over with a link in it is a “wall post with link”, which can have another set of privacy settings. Same thing with FF syncs, etc.This is already confusing enough; until FB cleans up some of the way it works, trying to bolt on another API would muddy the waters even further, IMO.

        1. dave

          Ken, that is indeed a serious issue.

          1. Ken Kennedy

            I hear your larger point, though, Dave, and in fact the confusion w/in FB over how things update is in a sense a separate issue. One that using the Twitter API as an goal highlights. Heck, maybe it might even be useful to flag issues and raise them to the forefront. A lot of users don’t even realize that a wall post and a status update are considered different types by Facebook.It is certainly true that the Twitter API is not only clean and simple, but has tools, bindings, and scripts all over the place that support it. Providing a clear and clean mapping to a least a portion of the FB API could be very interesting!

          2. David Semeria

            Alas, this is how most web standards emerge. Spurred on by adoption in unanticipated contexts, they are coerced into an unnatural and inelegant transition from the specific to the general.Plus ca change.

      2. MikeSchinkel

        @Dave, I agree. Facebook could embrace, extend (and extinguish?)

  2. Marc Brandsma

    this all sounds like an old emailing protocol à la X400

  3. reece

    Looks like “24 Reasons You’ll Love Tumblr” just became 25…Very cool.Marco, if you’re listening – please get Tweetdeck on board, too.

  4. christmasgorilla

    Twitter is a new standard. It is becoming a social utility as the founders used to say. I think that the big reason it’s working so well is the openness–which allows a real ecosystem to grow (, clients, etc).The biggest win for all of these web based services in toppling the old guard completely (telecom, yellow pages, etc) is to have an ecosystem of interconnected services that can all talk to each other with the same or interconnected clients.

  5. bojanbabic

    Twitter API is getting serious. Maybe soon we’ll have twitter:// protocol Recently I’ve been playing with twitter and blogger api in order to create Twitter Live Blogging service. Service would have starting hash tag ( i.e. #startliveblogging ), that would trigger live blogging . Since this tag, all twits would update blog post until closing hash tag ( i.e #endlivebloging ). This is kinda bummer because blogger will response and implement this feature very fast.But lesson learned 🙂 You gotta iterate and gotta iterate very fast if you want to swim along big fishes.Cheers

  6. Akshay Mishra

    An API standardization is a very critical step for some very serious development work to take place.If WordPress and Tumblr are doing that, then it is, indeed, the right step forward for web-app development involving user-content generation.Look at what OpenGL and DirectX (it’s not open, though) did for the graphics industry – the absolute deluge of content in terms of 3D applications after OpenGL was introduced is an indicator of the effectiveness of the standardization of an API.Also, having an open standard helps immensely – Microsoft let C# be an open standard – and thus was able to challenge (if not displace) Java in the managed-code development industry. It has also led to the development of the Mono framework enhancing the interoperability of .NET apps on platforms other than Windows.Facebook, though, is not going to adopt this anytime soon – because, semantically, it is a different kind of system.Aside: Disqus could really do with WYSIWYG comment editing! 🙂

    1. fredwilson

      i’ll make sure the disqus team sees your suggestion

      1. Mark Essel

        The power of comment following would also lead naturally to a twitter API

        1. fredwilson


  7. Mark Essel

    I commented to Jesse Stay earlier, and Dave Winer (I’d link to these posts, but I’m on weak iPhone interface) as well. This is how the web works now if enough people develop applications to adhere to a common standard, it virally spreads to other developers- transforming from an API, to a protocol. There’s just too much utility in morphing our tools slightly to communicate with all the other incredible utilities built on the original API.Twitter is going to benefit as much as it gives. Tumblr can see an explosion of new users as interfaces connect to the Tumblr database using the Twitter API. Twitter is going to be in front of 100million more bloggers at wordpress as they begin to use tweetdeck and Twitter to interact with their blogs.The web melts down walls, and shows us the power of open in unlocking value.Btw, I’m tweetie 2 tumblring at http:/ now.It hasn’t shown up on the the tumblog but it shows up in tweetie 2

    1. fredwilson

      ‘Twitter is going to benefit as much as it gives’. I sure hope so because it is giving a lot

      1. Mark Essel

        I see the API copying as a form of followon investing. Fred invests in company A, I trust Fred’s judgement, and want to invest as well. As long as there’s a huge community supporting the API, developers feel much safer writing to inhabit that space. At least I do.And Twitter needed a nudge as it hit a user plateau over the last couple months. Being a protocol driver means additional real time competition, but it also means potentially more rich data going through their tubes as wordpress and tumblr users and users from X, Y, Z see Twitter in various interfaces.I mentioned it earlier in the year, and I’m still working on one monetization route, content hosts (like Twitter) need to leverage all the information going through their systems to provide ever increasing user value.(we finally got branded amazon ad widgets, I had to dig through JavaScript yesterday to guess what parameters were. Soon (this week) hosts will have a rich variety of choices. Still working on a visual search interface that users get value from)

  8. AndyFinkle

    I’m thinking in more simple terms …that Twitter will REPLACE RSS (the ‘S’ in RSS should stand for ‘sucky’ not ‘simple’

  9. whitneymcn

    I like, and agree with, this formulation: Twitter has de facto defined a standard.I particularly like it because over the past few months I’ve heard a number of people proposing that “Twitter is a protocol;” while I think people meant it figuratively, it’s so inaccurate that I just couldn’t get behind it.The other piece that interests me here: if Twitter’s API is becoming the de facto standard, it means that any API that doesn’t offer both read and *write* access is quickly becoming nonstandard and broken — and that gets a big hallelujah from me.

    1. christmasgorilla

      Agreed on twitter as protocol (it’s not), but could be interested if it replaced SMS.But here’s the question on an API as a de facto standard: what does that mean? Any REST API to any service that offers both read and write? Easy enough to implement from any service. Is the point that developers who’ve made client software can suck in other services without much work?I mean, I like tumblr and I like twitter, but I don’t care that much about being able to do both in the same place (and linking those two accounts is annoying, just look at Mike Hudack’s twitter feed). I’d be much more interested in Google voice adopting a congruency to the Twitter API so that I could pull in SMS messages into the twitter client of choice.

      1. Aviah Laor

        Great idea! Now tell that to AT&T…

      2. falicon

        I’m with you here…I *think* when people talk about an API becoming a standard or a protocol, they are generally speaking from a non-technical standpoint (ie. if you want to do X, you do it like Y — actual implementation may vary)…sort of like building a car, they all more or less have the same parts/features and follow the same patterns, but each manufacturer does it in it’s own unique way (so a sales person might say car X is much like car Y because they look and act alike while an engineer would say they are actually nothing alike at all)…For now, I’m just glad more and more systems are opening up and making it easier to build on top of themselves…we’ve got a long way to go, but it’s starting to get easier to glue bits and pieces of information from various services together for our own interests…and that really is exciting!

      3. whitneymcn

        I have a hard time seeing Twitter replacing SMS: I think it was (and to some extent still is) is key that Twitter was designed to piggyback off of SMS, the lowest common denominator mobile communication system we’ve got. Even now, trying to supplant SMS seems like it’d be a huge and uphill battle, and I don’t yet see why Twitter would want to take on that battle.For the de facto standard thing, to me it doesn’t have much to do with linking services (I’m with you on it being kind of annoying) and more to do with making it simpler to work with all of the services and open up opportunities for some cross-pollination.Looking at the snip Fred took from Marco’s post, what jumps out at me is that he’s listing some pretty core functionality for “social” web tools: create something, reply to something someone else created, reference the creating person, etc.Different services will naturally have some unique functions, as well, but agreeing on a core set of addressable features means that the cool little tool that I just built can *basically* work with Twitter, Tumblr, and any WordPress blog just by changing the endpoint I aim at. Instead of entirely custom coding for every service, the core pieces all work out of the box and I can focus on the more interesting stuff.

        1. christmasgorilla

          But it seems there have always been projects that try to be abstraction layers for many services. Why have they never taken off? Too expensive (enterprise)? Too much of someone else’s code?I mean, why does it make sense for all of the services to mimic a certain API pattern rather than all write to an open source abstraction layer?

          1. whitneymcn

            For #1 I think the difference is that we’re not talking about an abstraction layer here: rather than me using’s custom calls, which (they say) map to the appropriate Twitter, Tumblr, and WordPress API calls (until one of those services changes something), I’m just using the same calls with different endpoints, because Twitter, Tumblr, and WordPress have all agreed on how to address certain basic functions.The approach makes things simpler for me but it probably costs me money and –more troubling than cost — introduces a bunch of failure points: goes down or slows down, they have bugs in their own code, some other client of theirs could get the service blacklisted, and so on. That’s a lot to take on in return for the simplicity it gets me, so it doesn’t surprise me that it hasn’t taken off.The “de facto standard” approach gets me that same wonderful simplicity without adding any failure points that I don’t already have. It works because the services all say “fuck it…the way you guys are doing it makes enough sense for me.”For #2, why not write to an open source abstraction layer? Not to be glib, but I think probably because there wasn’t one: there wasn’t a plan to codify “core social web n.0 functionality” so that services could build to it, there were people doing stuff and seeing what worked. Because Twitter was pretty open from the get-go, I think that a lot of the “standards” just emerged there, and because their API has always exposed virtually all of the functionality of the service, a lot of people wrote for it. Then people at other services said “hey, it turns out that a bunch of what we do is functionally identical to what Twitter does, and people already know their API pretty well, so…fuck it, the way you guys are doing it makes enough sense for me.”To your point, though, it’d be great if people took the next step and moved this over to an explicit, service-independent standard.

          2. ShanaC

            As odd as this sounds, this is how most kinds of abstraction works…reference until you get it right…

        2. fredwilson

          And here’s a provocative thought. The key to monetizing social media is attaching value to those core social engagements we are now coalescing around

    2. falicon

      I think there’s a long way to go before something like the Twitter API can truly transform into a protocol…It’s interesting to note that even the Twitter API itself is still in a transition mode…with the new twist stuff and a few other API additions they seem to have moved to an ‘official’ version one and started to make everything a bit more RESTful (finally supporting more of the HTTP protocol such as PUT and DELETE).

    3. fredwilson

      ‘If its not read/write, its not a real API’ – Joshua Schacter

  10. philiphotchkiss

    This is fascinating and makes good sense. However, I think the one big issue with Facebook is the issue of trust and intent. WordPress, Twitter, Tumblr, etc. have, I believe, earned a much higher level of trust than Facebook. Facebook’s recent ‘privacy policy’ moves and general way of doing business has undermined its trust with many.I think the corporate behavior of Facebook strengthens Sean Kaye’s views posted below. My view is that it’s unlikely Facebook will move in this direction-both for the reason Sean cites, but also because it’s strategy, as evidenced by its behavior(s), doesn’t seem to fit the spirit of a common standard.

    1. fredwilson

      The leader often doesn’t want to play ball with the others. Which often leads to them not being the leader in time

  11. Aviah Laor

    We have two contradicted models here:1. “Napster style”: Social API is used to pull content into a single main place (Facebook)2. “BitTorrent style”: Social API is used to exchange social data freely among scattered websitesFor sure, the second model will kill, in the (very) long run, Facebook. But it’s more natural to the way the web works and how people want to own their data. And once the social functionality is distributed across all websites, you don’t need the huge and costly server farms. Matt Mullenweg said that a lot of times (although currently talking about Facebook killer seems insane), and there is already

    1. Henry Yates

      I agree – this is the deportalisation argument. When everything eventually links well I dont think people will want a “portal” profile, they will want personalisation. However, if this starts to happen Facebook is in a good position to enable this as well.

      1. awaterman8

        Is it me or are both the Tumblr and WordPress APIs currently setup for your #1, but not #2? Both their blog entries talk about enabling *submissions to* WordPress/Tumblr from external clients like Tweetie (your #1). But I don’t see where they’ve enabled freely pulling content *from* WordPress/Tumblr into other external clients/sites (#2). Please let me know if I’m wrong about this though.It makes sense that every “node” in the social content space would adopt APIs to let other nodes feed content into them. But it’s a bigger risk — and has potentially much bigger benefits to users and the ecosystem — to enable full exporting of content out of one’s node. Twitter did this; Facebook will never do it, and I don’t see it being in Tumblr’s interest to do so either.

        1. Aviah Laor

          WordPress is trying to achieve #2 with BuddyPress, that will let “word press nodes” to communicate. and you can install WordPress where ever you want.You have a good point: once social nodes are independent, we will need to protect them. So it’s not only a social protocol and a web server, but also a social firewall. The road is long (but the prize is high).

    2. marko

      great point aviah. i think these models are not contradictory per se, its just about evolution. whether we look at it from the evolution of a service (ex. facebook, twitter) that opens into a platform, or we look at it as the loose collection of standards that result in the internet.of course, in this process companies try to position themselves as a control point in order to extract profit. but in an open ecosystem like the web, there is a very limited time before some layers are commoditized (i.e. your cheese gets moved) and a business is forced to innovate or die. seems like twitter now needs to decide quickly if ownership of this API design is an important business piece for them, or they can ‘let it go’ for free because they can extract more business value if it becomes standard and they will extract value elsewhere (btw: i think they should standardize).these dynamics are also the reason that google and facebook scare lots of people….they are gaining positions so central to the infrastructure of the internet that they could become the gatekeepers and tax collectors that threaten the internet as a truly open platform.decentralized models always win over decentralized models. ask the kings & queens of 18th/19th century europe 🙂

    3. fredwilson

      A friend told me yesterday that ‘twitter is just feeds’ and ‘photos and games” are more engagingI pointed out that there can be photo feeds and game feedsI’m with you on the decentralized model. For the most part, our portfolio speaks to that

      1. jeffreymcmanus

        we’re actually working on a game designed to be used with Twitter, we will launch it in the next week or two.

        1. fredwilson

          Email me when it launches please

  12. dudumimran

    Well, someone should build a bridge that provides an external twitter like API and internally a plugin based integration into Facebook and of course others. Maybe this can get them into thinking about it if new applications will sprout around this capability. I’m sure it won’t be as natural as their own implementation but it can be a start. This bridge can also standardize and normalize the future changes that each company’s API will take and will help developers cope with these changes.

    1. fredwilson

      Hacking facebook?

  13. John Stack

    I love that Tumblr is doing this and will try to figure out how some of my clients might use this. As a side note, I wanted to say that I appreciate you and folks like Bijan Sabet from Spark are being the board members that startups truly need by promoting their portfolio company’s values and messages. It takes a Team and the VC is on the team. I believe the challenge is convincing other VCs to roll up their shirtsleeves do the same. I think it should be included as part of the deal flow. The whole one-dimensional board member thing is so ’20th century. Thanks and happy holidays!

    1. sweller

      It’s rare to see VCs that can service their LPs, the fund and at the same time assist their teams in the same way that Bijan and Fred do.

      1. fredwilson

        I would argue that you should not try to do the first two. If you do the third well, the rest happens

    2. fredwilson

      Thanks for your comment jack. You can’t fake it or phone it in. People can smelll authenticity, values, and beliefs

  14. William Mougayar

    All this openness, interlinkage and seamless integration is great, but we have to be careful that we’re not re-creating unnecessary duplication. For e.g. I recently got suckered by FB to link my Twitter account to it, so that my tweets automatically show-up in my FB newstream. Then, I received push-back from my FB friends that I was “flooding their streams”. Obviously, Twitter streams are fast and furious compared to FB’s. What we need is curation or filtering capabilities where I can specify which types of tweets get facebooked or not for e.g., or vice-versa.

    1. Aviah Laor

      this exactly why, from a user point of view, it’s better to implement the social API on each individual website/blog and let you do whatever you want.Technically, today, it’s hard – but things can change.

    2. Aaron Klein

      You can use selective Twitter on Facebook. Add a #fb to the end of the tweet if you want it on Facebook.I don’t use it myself now, but I don’t have that same problem…

      1. fredwilson

        I just run all of my tweets to fb. I also run all of my blog posts to fb. I never post anything on I use it like friendfeed

        1. Aaron Klein

          Yep I do exactly the same thing.Interestingly, Facebook appears to have just copied Disqus and implemented “reply to e-mail” commenting. But it lacks critical context…the e-mail includes the comment from your friend, but not the post they were commenting on.

          1. fredwilson

            They’ll get it right eventually. It took disqus a while to get it right and they can still make the email experience betterI can honestly claim to have inspired this feature in disqus so if facebook was inspired by disqus, then I am a proud father of this feature

          2. Aaron Klein

            It’s the feature that pushed me over the edge to become a Disqus user.I’m just glad it seems not to post my out of office message. They’ve done a great job implementing it.

          3. fredwilson

            That was one of the quirks they had to work out

          4. Aaron Klein

            For us BlackBerry users who have signatures auto-appended to e-mail, one of the best ways to make it better is to recognize your sig (or maybe let you tell them what your sig is) and slice it out. My first few Disqus comments posted by telephone number and business e-mail address.My workaround is that the comments get auto-forwarded to a different e-mail address, also on the BlackBerry, with no signature on replies. It works but it’s not as elegant as it could be.

          5. fredwilson

            Daniel claims they do this but I don’t think theve figured out how to do it well yet

    3. fredwilson

      Or at a minimum de-duping

  15. Tim Ogilvie

    I love what twitter is doing, but an API that is widely used isn’t an open standard.If Twitter were to wake up tomorrow and decide they want to limit usage, change the allowed uses, or simply shut out a competitor, that’s a viable alternative. Anyone building a business on top of those APIs now has a supplier with a lot of power.

    1. falicon

      This is the biggest downside to anyone trying to build something off any of these services…I *think* the key is to building value on top of these services and position yourself so that you are really only initially dependent on them…What I’m thinking is stuff like stocktwits…building on top of Twitter gave them lower barrier to entry and adoption, but they also made sure that their own actual service added a ton of value and now they really aren’t so directly tied to Twitter (though they of course continue to benefit from being on top of twitter, if twitter were to go away for some reason for them, I don’t think it would be 100% crushing to their business)…We are trying to follow much the same plan with our concept…all of our apps and features are built on top of ‘social data’, but we are doing a ton of aggregation and (hopefully) value add to that data within our system as well…so right now, as infants, we are relying on Twitter, Facebook, LinkedIn,, Google Social Graph, etc. to survive…but as we mature into teenagers and adults the plan is for it to become much more of a peer relationship where we can at least survive on our own if/when we need (but will likely continue to live in the basement so we can save on rent and have great meals!) 😉

      1. fredwilson

        True. But when you move off of fb or twitter your ability to grow and acquire new users slows. Just look at stocktwits on compete or quantcastWe need to fix that problem. FB connect is their solution and it works well. Twitter needs one too I think

    2. dave

      Not true. These guys aren’t calling code on Twitter’s servers, so Twitter can’t shut them off. The only way they could, theoretically, interfere with Tumblr and WordPress, is by suing them in court.

      1. MikeSchinkel

        Agreed, it doesn’t need to be standards-body standard to be open, a defacto often standard works just fine… (Anyone remember “Hayes Modem-compatible?”)

      2. marko

        fully agree. great discussion on this topic but everyone needs to separate this into two distinct threads:1) using twitter as a ‘messaging’ platform (i.e. writing on Twitter’s API)2) mimicking the twitter API implementation on your own system to make it more digestable for developers (i.e. standardize around a common design interfaceIf 2) is going to evolve into a standard, Twitter needs to come out and take an official position on this….are they happy that others are doing this? or are they going to pursue their IP on these companies? I think Twitter has way more to gain by putting this IP in the public realm and driving a standard, but they need to make their position known.

        1. dave

          Why should Twitter make a statement?And what if Twitter said they didn’t want people to do clone the API? Whatthen.Elsewhere someone pointed to the example of Hayes-compatible modems, adefacto standard that the manufacturer didn’t ant.Then there was the IBM PC that IBM eventually tried to obsolete with OS/2and ended up obsoleting itself.The history of companies that accidentally spawn standards that want to”take it back” isn’t very positive.The market forces put pressure on Tumblr and WordPress et al to conform tothe limits of Twitter. So the negative aspects of standardization aren’t allon the originator.But you gotta wonder where the Compaq of Twitter is. The Ben Rosen ofTwitter. Who is Ben Rosen? Heh. An important guy for the people on thisblog. Imho.

          1. marko

            twitter doesn’t have to make a statement. they can say nothing, keep their options open, and make people wonder if they will get a call from lawyers. but i think it would be in twitter’s (and everyone’s) best interest for them to just say ‘hey, this is great, we support it, let many flowers bloom’. end of story.if twitter did decide to pursue IP….it means nothing to smaller start-ups as they have no money anyways. but it is a huge issue to bigger players (ex. FB, goog, etc)….they would need IP clarification agreement up front or they wouldn’t implement (even if they wanted to). which effectively stalls any open standard movement.i like the compaq analogy…..i think we’re at the same first steps of the whole identity & social media space as the PC industry was then….many more twists and turns coming 🙂

          2. kidmercury

            totally agree marko. when these companies stay silent, and when they have millions of users, raised millions of dollars, and clearly have a need to be profitable, it makes me wonder. it is like wtf, are you trying to rope me into your system and pull a fast one on me, if not why don’t you just come clean….yes of course they can always bluff but at some point that is going to be a PR disaster for anyone who goes that route.

          3. fredwilson

            Or maybe they are just trying to think this stuff through before acting. There was all this debate about the twitter firehose and then eventually twitter figured out how to give it away intelligently. Look for more of that thinking

          4. kidmercury

            cool, i simply think that the bigger you get the more trust you need to earn, and the way to earn trust is through transparency. simply announcing a “we’re thinking about it” helps to be transparent. ultimatley they want adoption, transparency is an effective tool in getting it IMHO. and at their size users should be very cognizant and ask lots of questions, they are a nation of millions charging virtually no taxes, living on deficit spending, so how the bill gets paid is an important question for all involved in the ecosystem.

          5. fredwilson

            Good points

          6. Prokofy

            AND history may not repeat itself and “make somebody the Compaq” but they may in fact do the opposite and prove that had Compaq not succumbed to the technocommunism of the 1990s, the newspaper and music industries might not have collapsed. It’s all interrelated.

          7. Prokofy

            This is a standard meme, that we “can’t” repeat the “horrible experience” of the “walled gardens” of Hayes Smartcom, Compaq, AOL etc. in the 1990s.Well, why not? Who says? this isn’t some immutable law.A) nothing says history has to repeat itselfB) opening up APIs everywhere so that widgeteers can get rich by poking into and scraping everybody’s data through click addiction doesn’t pay them, but pays the widgeteers (and the Google ad agency)C) If you don’t have walled gardens, proprietary code, and DRM, you cannot MONETARIZE. Nobody in web 2.0 has done this except the walled garden of Second Life, have you noticed? The income FB makes from gift sales is abysmally small given the population. D) the future is not in forcing opensource open APIS open standards (read: standards that opensouce extremists dictate) but in walled gardens that innovate because they get not only coders paid but users, too.

  16. Ed

    What can be, is all in how you see it.”possible” is made possible by those with the visionto imagine, then make the so.”If Facebook were to implement the Twitter API that would be it. We’d have another FTP or HTTP or RSS.”-Dave WinerWhen I first saw Twitter 3 years ago (back when twitter was so small, that the former ‘Public Timeline’ didn’t change – not a single tweet – for hours at a time),I said “This is the next http”. People laughed.Long before TechCrunch leaked internal Twitter documents showingTwitter’s vision of 1 billion users by 2013, I tweeted that they’dsee 1 billion users by 2011 (with no access to data).Facebook is huge.Google is so big, that they’ve become too far reaching. But with further stability improvements, clarity and unity of vision,and bold execution, Twitter can become the foundation for an extraordinary amounthuman communication. Think: Google-sized discovery + Facebook-sized networking,less the encumbrances. Then add the staggering growth left to occur in the coming years, bothby new internet/platform users, and greater adoption by existing users.The use cases are just so abundant as evidence that exactly this mechanism (platform)for interaction, discovery, business and life management, was waiting tohappen for every human (and machine) with the means to plug in.This is bigger than RSS (sorry Dave). And I cannot legitimately come at this from a technical perspective.But I think everyone from the VC’s to the C-Level team at Twitter,continus to greatly underestimate what they have here.What can be, is all in how you see it.

    1. Aviah Laor

      FB does not have a better way to commit suicide than to adapt an open social API, such as twitter. It reminds the internals emails that Bill Gates wrote against “too much WebDAV love” in the Exchange team

      1. marko

        +1 …which is exactly why everyone else should unite around open standards

    2. fredwilson

      We don’t underestimate it Ed. Its not easy to go from a killer idea to a killer platform to a killer business. We are trying hard, working hard, and getting there

      1. Ed

        I agree that a lot of very talented, passionate, and brilliant people are committed to the grapevine that is Twitter. I do not think it’s easy, and I don’t devalue the effort neededfor the hard won successes to date.But- I truly believe that every person with access to the web, especially the mobileweb, would embrace twitter if they just knew how, why and what.How many people around the globe will be online by the end of 2010?(I don’t know the newest projections)If one service offers reasons to be used by every one of them, it is twitter.

  17. sweller

    For Twitter to really become the defacto standard for “OpenMe”, it will need to become portable and decentralized.

      1. Aviah Laor

        Wow. We will have mod_twitter

      2. ShanaC

        now that is awesome

    1. fredwilson

      What does portable mean in this context?

  18. BmoreWire

    I think it’s really a shame betaworks pulled down switchabit. All of this stuff is great but for the consumer they need a control panel to connect and disconnect all of these services and switchabit made it easy (though it was buggy as hell and i had to explain to all of my facebook friends why every 10 minutes for 2 days a bunch of code was being updated in my status) but that said if it worked properly the UI was brilliant and incredibly usable. I would love to see them bring that back from the dead or give it to Tumblr or someone as a tab in their dashboard.

  19. jeremyx

    …are any of these companies actually making any money?

    1. Phil Mck.

      This is what worries me about the proliferation of Twitter. At some point the overhead gets too big, the VC funds dry up, and the absolute requirement for an income generating business model has to be found or Twitter *really* transitions into an open standard like RSS/FTP/SMTP etc.

      1. fredwilson

        Watch arrington’s interview with dick costolo at the real time crunchupI believe he addressed your concerns very nicely

    2. fredwilson

      Facebook is. Can’t talk about twitter unfortunately

  20. Eric Lunt

    One simple thing that clients can do to speed this along is to make the URL of the Twitter API endpoint configurable. Tweetie does this right now, so you can specify an account that uses something other than as the API root. It might encourage web applications to adopt the Twitter API if there was already a rich body of clients that could instantly work with their app.

    1. fredwilson

      I was going to point that out eric and forgot to. That is the key. Thanks for pointing this out

  21. Kris

    This is an example that the most useful technology standards are derived from the marketplace, and sometimes even just one person, but rarely from a standards body.Maybe a standards body can get the process started, but it’s up to the entrepreneurs to make the connection between standards and a value proposition. Great minds see these things before they become “things”.

    1. fredwilson


      1. Prokofy

        Could you explain what sort of “marketplace” is involved in pressure from…companies that are sucking in VC investments but have not profited yet and haven’t even monetarized anything about themselves?! (except a few gifts on Facebook).I mean, in what way is this a MARKET Fred, all these freebie social media sinkholes taking all your money?It’s not market pressure. It’s strong-army from people with heavily ideological beliefs about “free” and “opensource” that are so zealous, they will sink money into these ideas. That doesn’t make them any more true than they were in the last century, and doesn’t make them any more profitable. The Germans sunk A LOT of gold into Lenin.

  22. Michael Campbell

    There’s quite a difference between HTTP (et. al.) and this mysterious Facebook API (“FBP”?). HTTP, FTP, RSS, are all protocols to allow *OTHER STUFF* to occur. FBP *IS* the other stuff. It’s the ends, not the means. *TP is the means, not the ends.

    1. Aviah Laor

      similarly you could say that HTTP is the “other staff” over TCP/IP. The (still to come) social protocol is another layer on the stack

      1. marko

        +1aviah is nailing all the key issues on this discussion.

        1. fredwilson

          Not surprising

  23. Rob Freeborn

    This is openly exposing my own ignorance, but wouldn’t assuming the Twitter API as your new standard and having it replace RSS be like assuming all email in the world runs through Gmail servers?Who, in their right mind, would let 100% of their business focused traffic run through a company as well known for “fail whale” as anything else.Isn’t that, in the end, the difference between an API and a standard/protocol? API = attached FOREVER to one company while a standard = portable.r.

    1. Rob Freeborn

      And here’s an oh so timely example of just what kind of problems you can expect when latching onto an API instead of a standard/protocol…

    2. ShanaC

      I’m going to second that idea. It might even help twitter in the end, because if they can sit down and come up with a standard based on what is in use with thier supporters, they might be able to focus on doing what they want to do faster.I don’t think it would truly replace RSS. Certain things, you need a different protocol simpler, and the web has its corners that still stays the same and doesn’t die off. So RSS will still need to be around just to support those corners. I mean, technically you can log into a GOPHER server… (I think there are something like 15 remaining): someone has to keep the standard going, because there is currently a lot more stuff running on RSS that won’t be disabled for a while…

    3. thewalrus

      need to separate two different issues here:1) using twitter as a messaging platform 2) mimicking the twitter API implementation on your own system to make it more digestable for developers (i.e. standardize around a common design interfaceyes, doing 1) makes you dependent on twitter system (not saying that is good or bad but certainly exposes the potential issue you raise). doing 2) does not, but exposes you to the possibility that twitter may see this as infringement on their IP (again no idea how they will respond to this).point being these are ‘apples and oranges’ issues and need to be separated

    4. fredwilson

      When you use the ‘twitter api’ to pull from wordpress, you are pulling directly from wordpressThe api is just a standardization of the calls you can make and how you make themThink of it more like a language than a service

  24. Jesse Stay

    Before we see this happen I really think Twitter’s going to have to take a leadership role in it. Right now they’re not making it easy to copy their API. They need to open source it, standardize it, etc. otherwise everyone will be reinventing the wheel every time they do. Right now it’s a very closed process. (See my article here from yesterday:

    1. fredwilson

      I read your post jesse and it raises some interesting points. Thanks for sharing that link here

  25. daryn

    I’m all for adopting a defacto standard versus trying to formally create one from scratch on paper – but now that we’re this far along, it may be time to create a baseline of the interface and create an RFC so that it does become an actual standard – not just what everyone does.But whether it is twitter or some other api, the more we keep things similar – if not “standard” – the easier it is to get adoption. I believe Vimeo’s API was a direct copy of Flickr’s – when they first launched there were still “flickr” references left in the documentation :)Twitter’s API doesn’t fit for everyone, but it’s a great fit for tumblr, and I’m so glad to see this!

    1. fredwilson

      Interesting that flickr and vimeo could use the same callsMaybe there are classes of social services and we need open standard apis for each class

    1. fredwilson


  26. RacerRick

    We use the Facebook API and Twitter API for posting to WineTweets.comFacebook’s API does alot of same things. 3/4 of the Facebook API project was learning it – so if they used the Twitter version, it would have cost me 1/4 to implement.

    1. fredwilson

      That’s the point. Nicely said

  27. Norbert

    Sounds like McDonalds to me.:) nmw

  28. Andrew S

    It will be exciting to see what happens when social media can flow seamlessly between the current “pools”. I’m reminded not of any historical protocol analogies, but of what happened when Cisco (and Bay Networks) figured out how to connect together different computer networks. When people can asynchronously publish and subscribe to any content, anywhere, what will media consumption look like?

    1. fredwilson

      It will explode as if it hasn’t already

  29. Carl Rahn Griffith

    What Twitter has reminded us all of is that ‘Simple Is Good’ …

    1. fredwilson


  30. Steve Farrell

    I love where this is going… but the query/response API is only half the story… how do I get on the wordpress, tumblr, etc streaming APIs? Where their equivalent of the firehose? And how will that work in a distributed cloud?

  31. Jitle

    I think this is an amazing move for other online content creation and delivery companies to align with Twitter’s API. If you are not doing this in the near future, you will be considered “closed off” and resistive.I have always pushed start ups to building APIs because the content of your site could possible be made more valuable that what you yourselves may use it for. Ultimately, that may make others using your content more “profitable” but, the value of owning and creating the content becomes much more valuable when more people are creating and using it.

  32. andyswan

    I just can’t help but think the Twitter guys must sit back every once in a while with a quality bourbon….turn to each other and laugh. With no words said, it is understood that the conversation is something along the lines of “holy shit I can’t believe people are talking about our dumbass-why-not-last-shot idea as a web standard now.”I love that scene.p.s. before you start thinkin I’m makin fun of them…..please…..I’m just keeping count of all the inflection points where I would have screwed twitter up, and loving that they haven’t! Get that $2b guys!

    1. David Semeria

      Black swan, Mr Swan.

      1. andyswan

        Love it.  The best I can hope for is a mild tan.

    2. fredwilson


      1. andyswan

        I hope twitter gets at least 2 billion….and I really hope it’s Sprint, RIMM or Verizon that does it!  Think “OWN mobile and online identity!”Stupid as this sounds, twitter-handle could replace the phone number.  Admittedly the mind wanders….

        1. fredwilson

          I hope they never sell to be honest

          1. andyswan

            Even better!

        2. Ed

          It’s not weird at all. In early 2007 I suggested on Techcrunch, that Twitter build a standard where @Username would instantly hyperlink to the twitter account (as hub) from anywhere on the web similarly to the way http does (including email body). I suggested starting with blog comments. The TC staff flamed me worse than the other commenters.Funny, TC and so many others now have @username fields built inand they’re embraced above all grip points in their interface.I also suggested to MattCutts that Google Voice talk to Twitter about@username dialing the user (from authorized contacts).I think some of the disqus architecture could be the foundation for connecting@Usernames to a hub. It’s been 18 months since I suggested a look,and I just don’t know the technical possibilities. (Disclaimer: My understanding of twitter is deep, but it is in the ways humans do and can be served by it, and the monetization opportunities.I’m clueless about the code behind the curtain)

          1. MikeSchinkel

            Ed, for the record I think your ideas are SPOT ON.

  33. Norbert Mayer-Wittmann

    Sorry to like, um, self-follow but maybe I should clarify my point of view.Has anyone ever heard of the term “commoditization”.Generally when that term gets used then you should know that if you’re still in that space then you sorta missed a good exit.Not saying that happened to twitter (at least not yet). I expect twitter will remain a part of the English vocabulary for some time hence.But if you’ve got a meaningless string of characters, you should be looking for the exit ASAP.:) nmw

    1. fredwilson

      Meaningless is the key wordI don’t think facebook, twitter, tumblr, or wordpress are meaninglessThey are stock full of meaning every single minute of every single dayIn fact I am not sure there have ever been a collection of services containing more meaning in the history of the web

      1. New Media Works NMW

        The content may be meaningful – or it may not be meaningful (I will not argue about that).However, IMO content devoid of context is not meaningful to me. Twitter is meaningful (it’s part of the English language). I think it’s too early to call whether Facebook, Tumblr and/or WordPress will enter the lexicon. I would say that blog HAS entered the lexicon – it has taken a place next to diary and journal — and these words may change their meaning slightly… blog being a more “up to the minute” and “latest” log entry / account. Perhaps “news” will become completely eradicated.These terms (blog, diary, etc.) are what are known as “form” and “genre” — and even though they are crucial to information retrieval (as they provide something like “situational” context for the data they contain), they are actually quite difficult to define.An obvious example of the important of this kind of meta-data is that the significance of the statement “thou shalt not kill” depends crucially on the context. If this is a commandment in a sacred text, the significance is very different than if it were spoken by an actor in a fictional movie. Likewise, the words in the Bill of Rights have a special significance BECAUSE of the context (namely as part + parcel of the Bill of Rights). The same cannot be said were these words posted to Facebook or some other context. And yet again: content appearing at has a different signicance than content appearing at — and may in fact be more closely related to content posted at that the domain name is true metadata, in sharp contrast to on-page tags. Note also that Google and other search engines have figured out that such on-page tags aren’t really meta-data, even if they were INTENDED to be meta-data. Indeed, true meta-data CANNOT exist on-page — that would invalidate the “meta” qualification.The domain name is the most significant piece of metadata — it is the medium (and the medium is the message — or at least a very significant part of it 😉

  34. ShanaC

    the API would have to be handed off. Could Everyone deal with the idea of not being in control of the API and being instead bowing to a standard. What makes this all so interesting is that the standard might develop from user behavior and then companies responding. On top of that, what I really wonder about, how does this handoff of developing into a real protocol like object happen (beyond say that beautiful explanation in the APACHE website) Who gets to be in charge? How do we decide how its going to work so that all the stuff build on top doesn’t break? Right now there is too much stuff (and there will be more stuff) that is being build on top of that api… How do we make sure those objects are supported in the past and in the future? A company alone can’t do that. It’s not very economically viable. Nor a safe way of supporting the internet…it works best decentralized, and right now we’re all stuck on Twitter (sorry but at some point, they need healthy competition that is not Facebook in the US)So someone needs to eventually break off and just sit there and think this through, I suppose…

  35. tedroden

    What strikes me as the most interesting part of tumblr’s addition is what it means for tumblr’s dashboard. After several years of guarding the dashboard, offering no RSS/ATOM/API/any access outside of the web page, they’ve essentially opened it up.More so than using the twitter API as its interface, I think opening up the dashboard could be huge. I’ve got my fingers crossed they’ll start building real APIs so developers can get to the rest of the tumblr data and start building some really great apps… a feature that tumblr currently lacks.

  36. JFK

    Twitter API as an open standard?Twitter is a company and though they don’t currently make money, they have private designs and no incentive to give up their position to donate as a benevolent open standard. They would need to be ordered to do so by the Fed Trade Com.But the U.S. Gov has no interest in arresting control, because Twitter isn’t a monopolistic threat. They are perceived as nothing more than a fun broadcast tool. No financial or security concerns there.Like all APIs, Twitter isn’t a standard, just the flavor of the month and useful as a short term protocol.

  37. MikeSchinkel

    BizDev 2.0?

    1. fredwilson

      More like 3.0

  38. stevenwillmott

    Great post and great discussion – this kind of convergence is indeed a big deal since it makes using APIs easier (and more robust) for everybody. There are certainly different classes of services which could converge to very similar conventions (in fact in some ways the various actions can be seen as a mirror of the REST operations, but for specific content types + the relationships between objects), there are only a limited number of ways to build a well organised messaging API.Twitter has also done a great job on it’s API so it’s no wonder there’s a push to re-use the specification. It would actually be great to see open source / copyright free API interface specs for common types of services so that companies could avoid copyright and lock-in risks + contribute to the same pot of standard API operations.

  39. Lloyd Fassett

    I think the main issue is that RSS and HTTP are transfer protocols and this is a standard of human actions. A standard of human expressed actions is a different thing because there’s more intelligence in the action. Where The Goog looks at who links to who to rank things, with a standard in actions you’ll be able to get similar aggregate results of what is important. Twitter can do it in their own ecosystem, like showing what URL’s are most tweeted, but link other ecosystems together and you’ll get better results.There could be some interesting new findings too since this type of opinion/vote on data is easier and faster than sites linking to each other. Perhaps this data would allow for personalized algorithms to not just find what’s popular, but what’s predicted to be uniquely interesting to me.

    1. fredwilson

      This is a great comment lloyd. As I mentioned in another comment in this thread, I think we are coalescing around a common set of human engagements in social media and the key to monetizing social media is going to be attaching value to these engagements across multiple social systemsYour mention of page rank is interesting in this context

      1. Prokofy

        If could be just the opposite, Fred, that merging and linking all these social media venture capital sinkholes will make even a bigger black hole in cyberspace that will draw what little sunlight is available in commercial sites like amazon into the vortex.It really is a crap shoot. I don’t know how you sleep at night with that much paper out.

    2. Prokofy

      This is an interesting statement but innovative ecosystems thrive when they can grow in isolation (Galapagos). They don’t necessarily do well when they are all blended together and the kudzu spreads.

  40. kidmercury

    these conversations are starting to get too abstract for me.but i think we’ll see networks of businesses agree to standards. i think this relates to microfinance; microfinance funds will invest in companies that they can bring into their standardization system, and that will play by those others have noted this type of standardization is presumably outside of twitter’s control, unless they try to exert IP control. so, this sets the stage for small alliances of independent businesses to embrace and extend the standard as they see fit, and compete with each other as the next standards wars.but these types of conversations are always so abstract. not sure if i am being clear, if i am interpreting any of this correctly, or if i even know wtf i am saying. lol

    1. thewalrus

      abstract, yes….but this is how it already happens….coalitions forming within industries…sometimes informally around emerging technologies…sometimes more formally within standardization bodies.i think the twist you raise is that more distributed ownership models result in more widely distributed standards ‘owners’….which is kinda like no ‘owners’ at all….which is kinda why the internet has been so successful as a head hurts. a good weekend to all 🙂

    2. Aviah Laor

      But the bottom line is simple: To be able to host friends list, wall etc on your own websites.If you want a social application you install it on your own website and it interacts directly to other instances of this application on your friends websites.Why would we want it? Because you will get out of the FB sandbox. Freely use any language, tools and libraries of your choice. This will bring a plethora of ideas and applications which we could not imagine right now.

  41. Jeffrey McManus

    Much is being made of this. I don’t think it’s that big of a deal. This is going to look like a unified standard for a few months until one of the providers of the clone API needs to extend its functionality in some meaningful way (like adding the ability to provide file attachments to posts, say) and then it won’t look like a unified standard anymore. That’s not a bad thing — the notion of posting a message publicly is not a complicated operation and there would only be a need for an open standard here if it were really complicated.The one indirect effect I can see this having is that it will open up more products with APIs that are simple to adopt and program.

  42. Jan Schultink

    HTML was designed to access informationWe need something similar for social interactionMaybe the @username is the new “.com” where everything comes toghether, with different extensions and access permissions?

    1. fredwilson

      I sure hope so

  43. Hai Le

    My opinion, Open APIs or Open Standards is the trend of many websites. It’s foundation for web 3.0More information http://socialmediachannel.b… (2 last slides)

  44. Richard Bauer

    You are all much smarter than I am, and I understand the excitement surrounding the drama of todays new technologies. However, we have all seen the exponential growth of communication and media (new, old, social, etc) over the last 15 years. My question to you is this, have we seen this all before, or have we finally obtained that magical destination that we have been reaching for?I desire that we all accept that this is only what is happening at this moment. I suggest, there are two more kids in a dormroom, or in a hilltop village in (use your imagination) who will completely knock us off our collective feet again, and again, and again. I find this incredibly exciting, globally democratic, but also a bit sobering.My point is, I can read a P&L, I can read music, I can even read a little Hebrew, maybe even a palm or too, but I do not desire to read code. Just make it simple for me. Don’t make me have to invite anymore friends to anything. Don’t make me have to fight someone for a URL or Twitter handle. Fair enough?

    1. fredwilson

      You are right that there is no promised land. We just keep going

    2. MikeSchinkel

      URLs and Twitter handles are simply becoming part of the language. They have beauty in that they eliminate ambiguity. Even if URLs and Twitter handles fade something else that provides unique identifiers will take their place, and I believe that something is more likely to be text-based than not.As language evolves those who adapt best succeed the most and those who fight it fade into obscurity. My advice, embrace them, don’t fight them. JMTCFWIW.

  45. Guest

    I’m in!I’ll post something on my blog today.

  46. Prokofy

    Another, “everything for the widgeteers so they can make money off APIs that scrape our data” and “nothing for the people and their privacy”.

  47. Aviah Laor

    but once you can handle friends, messaging etc with simple API to and from your own website/blog, exchanging social info with others websites/blogs, you need FB less

  48. fredwilson


  49. awaldstein

    Insightful thinking. To compete with Facebook you need to create a ‘place’.If you look at Facebook data, it is more powerful than you think. Sucking in broadcast advertising dollars at astronomical rates with deep demographic targeting and an avg of almost an hour online per day per user!

  50. ShanaC

    I don’t think there is much of a difference in what you and Charlie is saying. it’s a matter of medium being the same, content being different through usage practices. We all hold the brush differently to do the exact same thing. Content here is content: I would say the computers are oddly social, even if we don’t think of them that way. They have to exchange to each other in a social manner, albeit in a non-human one.

  51. Aviah Laor

    Exactly – for now. But this is precisely why – when the tools arrive – you will have an incentive to move the social functionality/data to your own website/blog.Unlike Google search cache, the social group dataset is smaller, and there is no advantage from the user point of view to consolidate everything in one place. Unless you are the user Mark Zuckerberg.

  52. fredwilson

    Yes. A place we own is important. But I think we are all headed to our own space on our own domainFacebook connect is a realization of that

  53. awaldstein


  54. ShanaC

    Blogs have been around for a while, facebook got more power because of pictures and the fact that they were centralized and in fact, seen as a protected zone to hang loose.You definitely need both kids of spaces. As odd as this sounds, they may not really exist on the web, for we really can’t define space very well here.

  55. fredwilson


  56. Aviah Laor

    Yap. Disqus is in great position to extract the FB functionality back to the websites/blogs. We will probably see more “social plug-ins” to existing websites.

  57. awaldstein

    I’m a big Disqus fan and user. It’s my connector between places and how I keep tabs on my friends comments and travels around communities. My thinking on Disqus and communities

  58. David Meiselman

    I agree that can all be extracted for the moderately savvy user, but doesn’t FB keep the vast majority of people where they are for a long time to come? I might want to do what becomes possible with connecting social functionality to my own site, but the VAST majority of people I know who use FB would not and probably never will.

  59. fredwilson


  60. Aviah Laor

    thanks, i’ll read it

  61. Aviah Laor

    Agreed. But note that this is a tech issue -not a demand need. When building websites was hard, people needed GeoCities. It may sound crazy, but i think that FB is the GeoCities of social functionality. it will take time, sure, but once we have a decent protocol and API we will have the easy and usable tools for the mass (as mentioned above, integrating ope sourced “social HTTP” into the web server will be a huge step forward).

  62. awaldstein

    Yes….but for many, Facebook is one of the channels for their businesses–journalists, yoga instructors, bloggers ;), consultants and on. Some with Fan pages some without. Ever one of our groups is like a community in NY–we have our variety of functionalities that we allow and are useful. That will be true as spurs to our own locations over time and people will follow what is interesting.

  63. fredwilson

    Never is a long time

  64. David Meiselman

    That’s funny about GeoCities..I have been calling FB the AOL of social. For the longest time, my mother though AOL WAS the internet, just as today so many people think of FB being all of social rather than just one example.

  65. fredwilson

    Nice comparison

  66. Aviah Laor


  67. ShanaC

    now you are making me feel so so strange…

  68. ShanaC

    It is what makes it a difficult medium to work with. very confining.

  69. ShanaC

    That’s an excellent metaphor. But I think he’s still right: Part of the process of doing this is deciding one is not the spout and the hose and the connector: It’s deciding what you are. Twitter needs to do that, and outsource the rest. Happened to the Netscape people: Majority of them went into Mozilla….

  70. ShanaC

    We all might be looking at the wrong problem: They might have to split offsome people because they are not yet the hose, the connector, nor a spigot. The choosing is the hard part in all of this, and the scary part. I don’tenvy them.

  71. falicon

    Right now Twitter is the well…it’s where the water is being stored and pulled from…historically, that’s a great ‘thing’ to be (new hoses, connectors, and spigots will be developed over time, but everyone will…ehm…keep going back to the well).

  72. ShanaC

    One words: Oil. I keep thinking I’m seeing oil here. You can’t have just one well, or else you drain it. tragedy of the commons…You can’t be the only well forever. People will want to cash in, and do better. Information is a weird sort of commodity item…

  73. anand

    I would agree with your centralization argument, but also would say they grew because of a lack of technical knowledge. It’s easy to setup a facebook page, but not as easy to setup a WordPress blog or general website.I would also argue that both FB and Twitter took off because they generally don’t require daily updates and ask for little in terms of writing. Whereas a blog post might require lots of thought, time, and creativity, a simple status update does not require as much energy.Sure, blogs and social networks can coexist perfectly well on the internet, but I think over reliance on FB and twitter can make us lazy, and pounding out an occasional blog post can really force us to think.

  74. fredwilson

    Space – there is an interesting concept. I’m going to get a bunch of architects and web designers together to talk about space sometime this year

  75. fredwilson

    I do that every day. I deny everyone that’s not truly my friend. I think that’s how you have to use fb to get value out of it

  76. ShanaC

    I started to learn how to do that only after I grew up with facebook. That being said, facebook is in some ways a testaement of the fact that I grew up with it: It isn’t socially awkward to randomly friend people if they are your friends in a network and the network is small. As were the early days of Facebook, when it was just college campuses.It still has elements of this, most people don’t realize that if facebook has a centralization issue, it is largely due to the fact that people change over time, and facebook has managed to change with them. So now you are getting push through about your more distant ties and well as close ties…people you wouldn’t friend now are still there in your stream. The only action is to defriend, but that doesn’t seem appropriate either always. Occasionally it is nice to know you can still contact that random person from five years ago, when you were close.So I wonder about social awkwardness- I wonder if it is just born out of our ability to change rapidly and computers, less so. At the end of the day, they are all still Turing’s machine, and I don’t think without our intervention, they would know these subtle differences.

  77. fredwilson

    It would take a lot money to do that. What’s wrong with a big profitable independent company?

  78. awaldstein

    Great idea. Consider sculpters and urban designers as well.

  79. ShanaC

    I would say also add in A Topologist (particularly if you can find someone doing some sort of topology as it related to computing) Best conversation I ever had about space was with a phd candidate who clearly knew topology: He sat there and got me thinking for a few hours about how it is impossible to place five balls in real space that are equidistant to each other.A sociologist or an anthropologist: Presentation of self in space. Most people never talk about the idea of what you wear, how you behave in different contexts and spaces. Having someone around who explains that well (and there are people who do specialize in this issue and the internet) would be helpfulSomeone in fine arts: Arnold mentioned Sculptors- I would say it depends on the artist and the medium. I’ve seen sculptures that are fantastical or have different realms of evidence that they discover beyond space- to ones that really deal with space issues and perception issues. I’ve seen drawings, paintings, and video instillations. I even know a website that deals with this issue in part (glitch art on the internet). It really depends on the artist and the work.I’m pro the architects. Just make sure the architect gets we are talking about humans in space and transgressing space. Not always clear with some modern architecture, and they don’t bad talk each other because of the difficulties of becoming an architect.Other than that, I’m in. Great idea, I would love to hear that conference.I want to meet a web designer who wants to shatter the perception of the box that I see every single time I visit a website. Who is going to be the DuChamp of webdesign and make up realize that just because monitors are squares like canvas/paper, we don’t have to lay everything out to match that square- and we can get people very fully involved by breaking the perception of the fact that monitors are squares…Ok, this is what happens when your background is in drawing, particularly nudes…

  80. awaldstein

    I agree. Facebook Connect as one connector; Disqus potentially as another.

  81. awaldstein

    I agree. Facebook created groups so that within Facebook each person could have multiple grouping of friends, channel partners, what not. The reason is that folks do want to keep them separate and all of these to Facebook are just a massive, powerful advertising datamining tool of immense proportions. More is better for their system.

  82. David Meiselman
  83. David Meiselman

    Agreed! It is a long time. Even people who were afraid to venture outside the walled garden of AOL back in the day eventually moved on to the proper web. But I was talking about the mentality (as well as sophistication) of most on FB. To a degree, extracting the FB functionality for use on your own site is a 1% writer-v-reader issue. I guess that was what I was really driving at… and in that sense, while the reader types might extract their commenting and do it across sites of their distributed social graph, I am not sure that the ability to extract the writer piece appeals to masses of FB users. Maybe so, though. Perhaps If tools like Tumblr and Posterous make it so easy and were integrated with something more like disqus… I guess never really is a long time!

  84. fredwilson

    Will do. Do you know a lot about this?

  85. fredwilson

    Twitter+tumblr+disqus is the anti FB to some degree

  86. awaldstein

    Not an expert.

  87. MikeSchinkel

    Off Topic, but it sure would be nice if the Disqus email notifications included not only the comments but also the comments they were in reply to. I’m wanting to follow this thread but am getting overwhelmed trying to decipher the email notifications.

  88. ShanaC

    You tell me- I blog irregluraly, but my usual length is about 2000 words.And you have to think through at some point 2000 words. You have to think of something you really want to say if you are going to write that much.

  89. David Meiselman

    I agree it would be to me if I was limited to that as my only medium of online social expression, but confining to some is easy and comfortable to others.

  90. ShanaC

    I was thinking more in practical terms. I was trying to run an art piece on Facebook, and I was having huge traction issues because of the way facebook is set up.If you want in, or know of ideas of how to fix it to be more narrative email me shana dot carp at gmail yada yada yada

  91. anand

    Exactly. It can sometimes be a challenging and time consuming task,..<del>

  92. Prokofy

    You need a lot of closed wells that “the people” can’t drain, so that people who own the deposits can get paid so as to drill more oil.