The Gillmor Gang - Gabe Rivera, Cliff Gerrish, Robert Scoble, and Marc Canter - talk Plan B and the Track endgame. Recorded Tuesday, June 3, 2008.

[music]

Gillmor: Hi. This is Steve Gillmor. Welcome to The Gillmor Gang for Tuesday, election day, the last day of the Clinton campaign day. Cliff Gerrish, welcome.

Gerrish: Hi. How you doing?

Gillmor: So, how do you think our Plan B thing is going?

Gerrish: [laughs] Well, so far I think it’s still probably a plan B. But it is sort of interesting that there’s been no response on Track from Twitter at all, from what I can tell. I haven’t seen it anywhere.

Gillmor: No. They’re stonewalling it completely, which would just be my point.

Gerrish: What I don’t understand is, do they not know what they have, or do they not want to call attention to it because they know what they have and they don’t have a good solution for it yet?

Gillmor: Well, I think it’s partially that.

I’m hanging up the video, and I’m calling you again, Jerry. Sorry about the crosstalk here.

I think it’s probably that they don’t want to respond to random requests from the peanut gallery.

Gerrish: [laughs]

Gillmor: Because they’re going to have to respond to everybody.

Gerrish: I suppose. But it’s clearly a feature of the service, or has been a feature of the service. And they talk about restoring the service to full strength, and yet they mention IM but not Track. So, in some cases, you can put messages into Twitter via IM, but you can’t see anything back — or at least I can’t at this point, anyway.

And then the next question would be, what about that feature, and why not talk about it? They’re trying to be transparent about some of these other things.

Gillmor: I don’t think they’re trying to be transparent at all. I think that they’re being forced to be transparent.

Gerrish: Well..

Gillmor: I think that they’re… [laughs]

Gerrish: It’s because Robert Scoble shows up with a video camera at their door. [laughs]

Gillmor: That’s not quite what happened. Robert, you want to say what happened?

Scoble: I’m in a conference right now. I’ll talk to you in about five minutes.

Gillmor: OK.

Gerrish: OK. They are starting to blog. They’re talking to Robert. They’re doing some things. But clearly, a lot of pressure was building up on them by not talking. And some of that has been relieved by talking. But the issue about Track is that there’s Twitter sort of uncovered something very powerful, and for some reason they’re not addressing that piece of it.

Gillmor: Yeah. I don’t think it’s…

Gerrish: And I don’t know if it’s just their architecture is not going to support it, or what.

Gillmor: Oh, I don’t know whether it’s going to support it or not, but I think it’s disquieting at the least that they refuse to talk about it.

Rivera: Hey. This is Gabe. Do we really know how many people actually use Track? I love it, I think it’s important, and I tell everyone I know who users Twitter to use Track, but I suspect they’re just prioritizing. The whole IM is down, right?

Gillmor: Yeah.

Rivera: So I think they’re just putting… Well, I don’t know why it’s down, but they’ve got a whole mess of problems, and I think they’re just fixing what’s at the top of their prioritized list.

Gillmor: Well, it makes sense that they’re bringing IM back last. I mean, it makes sense. What doesn’t make sense is that they won’t answer the questions about Track.

Rivera: Yeah. Which question is this?

Gillmor: It’s just “Yes, we’re going to support it. We’re going to bring it back as soon as we can figure out how to do it.” That would shut me up, at least for another couple weeks.

Rivera: Where have you posed this question?

Gillmor: On Twitter.

Rivera: OK. Have they replied?

Gillmor: No, they haven’t replied. They’re ignoring me.

Rivera: What’s your rate of having them reply to your Twitters? I’ve Twittered things to them, and I don’t even expect a reply; I just hope that they read it.

Gillmor: Yeah, I don’t expect a reply. In fact, I expect that they won’t reply.

Rivera: Yeah. I hope Track comes back. You ever use Summize?

Gillmor: Yeah, I use it all the time now. But the problem with Summize is, first of all, it doesn’t refresh in real time.

Rivera: Yeah. It does something interesting, though. You do a search in the page, it’ll tell you when the page has updated on your same results page. But then you have to click something, and then you get the update.

There’s like one more result. If you just leave the page there, your search result page, a message will pop up saying, “There’s more results.” I don’t know why they don’t just show you the results, but they say there’s more results. Then you have to click something and get it. Which means, even Summize, I bet someone there could provide an IM interface to that. But that’d be roundabout.

Gerrish: It seems like it could work the way Twistory works.

Gillmor: What is Twistory?

Gerrish: Have you seen that? It’s a very cool little service that works off of Summize and, well, Twitter data out of Summize. And it has a list of there’s like six or so verbs, like “think,” “feel,” “hate,” “love,” and so on. And what it does is it filters Summize flow for Tweets that contain those words. So, “I hate that Track is down,” “I love that Track is up,” whatever. And you click one of those words, and each of the Tweets that contain that verb start filling into your browser.

Gillmor: Right. But that’s like, “Gee, thanks. Why don’t I just follow ‘love’”?

Gerrish: Well, you can do that.

Gillmor: I mean, who the hell needs that?

Gerrish: But all I’m saying is the technology is that it’s pulling those things in dynamically. It’s writing them into the browsers. There’s no need to refresh. So if Summize could do something similar to that.

Gillmor: Well, Gabe is basically saying the same thing.

It’s Gabe Rivera, Jerry, and Cliff Gerrish are talking right now.

[man speaks off microphone]

Gillmor: I mean, Gabe’s point is just that Summize is already doing that. They’re refreshing the screen automatically and telling you that there’s something you need to click on so that you can go ahead and find it.

Rivera: Right. But you have to do the manual refresh.

Gillmor: I understand. But the point is that the technology has been demonstrated with the announcement. It’s like the Steve Sinofsky interview, where he refused to say anything for half an hour. [...] at some point in the future, and how they were going to orchestrate the process of telling some people but not others.

Rivera: Right.

Gillmor: It’s just horse shit. I mean, I’m rapidly progressing from having good thoughts about what’s going on at Twitter to starting to believe that they just — it’s the cover-up that kills you. It always is.

And the other thing that’s going to happen is that there are going to be attacks. I mean, if you take a look at the Plan B post on TechCrunch, you’ll see that there are a number of attacks. I mean, there’s the “I’m an idiot.” And then there’s the “Why does Mike let this moron do this?” And then there’s the “You need to have an editor,” and all that kind of crap.

And then there are some people who are actually arguing the point, and one particular guy, who really engaged on the subject of whether or not, at the end of the day, that the Twitter folks should really be paying attention to this elitist group of Track assholes.

Rivera: [laughs]

Gillmor: No, seriously. There’s very intelligent comments to that effect.

Rivera: Mm-hmm.

Gillmor: It’s wrapped in some other stuff about how Notes failed, which is, of course, ridiculous. And there’s a little bit of open web standards religion, of the type that Chris Saad sells. And that’s fine. But it doesn’t mean that it’s real.

Rivera: Yeah. I think the thing that probably needs to be articulated is the usefulness and what Track actually makes possible, because it’s sort of a bootstrap at this point. But it only really became visible when there was enough volume at Twitter to make Track useful. And then it becomes something that just didn’t exist before. It becomes a way to get visibility into a stream of user gestures that just no one has enough ongoing volume and flow to really make that kind of a thing possible.

And that’s the thing that’s really unique about it. Go ahead.

Gillmor: No, you go ahead.

Rivera: Well, the problem about Plan B is that you need volume to make Track viable. It’s like the New York Stock Exchange. You’ve got to have a liquid market. Well, if you only do one trade a day, it’s not so interesting.

Gillmor: Right. I mean, you can hack Track right now with the “@” sign. The only problem is that everybody who’s on the network who’s trying to use this workaround, they have to run Summize in the background, which is what I do, or they need to constantly be bouncing back and around, use Twirl, put up with the gating of the API that they’ve imposed. Between those three things, you can get close to a real-time conversation with people that you don’t know are trying to talk to you.

And that’s fine. I mean, none of this is being saved. If you shut down Twirl, then it erases all the Tweets that it’s recorded.

Rivera: Right.

Scoble: This is Robert Scoble. The real problem is there’s only a small percentage of people who follow enough people and have seen this new, real-time-conversation world. This was the push-back from the FriendFeed guys, too. They asked their customers, “How important is IM when only 7% of Twitter users use it?”

Gillmor: Now, let’s not muddy it with the FriendFeed stuff. We had a good conversation with those guys, and you can judge for yourself. Just go download the show and listen to it. You can judge for yourself whether or not they got the message or are going to do anything about the message. I think they got the message. I don’t think that they necessarily will do anything about it.

But talking about FriendFeed as a band-aid for something that’s not happening on Twitter, it’s part of Plan B, but it’s more like Plan C or D, because hopefully we’re never going to have to get to the point where we have to wait for the FriendFeed developers to be able to be clueful about what’s going on here.

Scoble: Yeah.

Gillmor: This has never been about the number of people involved. That’s the other push-back that you get about this argument is “Oh, well. Who needs IM?” Or, “This is just IM, and we don’t need public IM.”

All of the arguments about this support the logic that there is something here — as Cliff put it very well in his post yesterday. There’s something here that is unique and powerful, and we know it, and we don’t have it anymore because this vendor can’t support the features. I mean, I didn’t ask for this. They put it in. I mean, Blaine Cook, according to him, wrote this feature into the software.

Scoble: Yeah.

Gillmor: And it was released, and now it’s been taken off the table.

Scoble: Yeah.

Gillmor: I continue to be more and more concerned, the fact that they don’t respond. They’re very careful about their use of words, as they should be. But I have never seen…

When you were standing there with the guy, if you had said to them, “OK, that’s great about IM. But what about Track, specifically? Are you going to bring it back, or are you going to somehow neuter it, or are you just not going to return it?” And it would have been interesting to see the expression on their faces, and we would know more than we do right now.

So, after the fact, all I can do is just be annoying about it, which I intend to be, and make it as impossible to not answer that question. Because when they don’t answer those kinds of questions, we can be pigeonholed as being elitists or a special fringe group of power users, that it’s not the network’s job, or, as Carol Lee or some other person puts it, this is something that we are willing to pay for. I’m not willing to pay for it.

Scoble: Yeah.

Gillmor: I mean, I’m willing to pay for it with the use of the service.

Scoble: Yeah.

Gillmor: So it’s not about FriendFeed, unless you have some sort of theory.

Scoble: No, no, no. I’m not saying it’s about FriendFeed at all. I’m saying that that’s the same attitude that I got from Twitter, that this is a feature only used by a small percentage of its user base, so therefore, they have a priority list of things to get working, and they’re going to put that down on the priority list down to number nine.

Gillmor: Yeah. They didn’t say that. They said, “IM…”

Scoble: Well..

Gillmor: On their website yesterday, on Monday, there’s a post, and it says something to the effect that, “Next on our list that’s most important to us is restoring IM.”

Scoble: Yeah.

Gillmor: That’s fantastic.

Scoble: And that means XMPP, though, doesn’t it? It just means the messaging interface. I think when you get IM back, you get Track back.

Gillmor: Well, we don’t know that.

Scoble: We don’t know that’s true, but…

Gillmor: If that’s true, all they have to do is just say, “Yes, when we get IM working, you’ll get Track back.” But they don’t say that. And I think there’s a reason for it.

Scoble: Yeah. I think, in between the lines, that they haven’t separated those two out. Those two are the same feature in their minds.

Gillmor: If that’s true, then they should answer that question.

Scoble: Yes, absolutely.

Gillmor: The question is, in your mind, are they the same thing?

Scoble: In my mind, they’re the same thing.

Gillmor: No, not your mind, their mind.

Scoble: [laughs] Right.

Gillmor: I mean, if you have some secret power to be able to get Track back, then I’ll talk to you. I don’t care what they think. If you’re in charge, fantastic.

Scoble: If they turn IM on and they don’t turn back on Track, then we know they’re playing games, right?

Gillmor: I think they’re playing games already, by not answering the question.

Scoble: That’s true.

Gillmor: So, what’s Plan B? How do we move the service? I mean, we need to inflict some pain here.

Scoble: I think they’re already inflicting pain. I mean, hell, a horrible copycat of Twitter, called Plurk, is already — I have 400 friends over there. It’s like I’m not even asking people to sign up because somebody stole my user name already.

Gillmor: So that’s not Plan B. That’s Plan Z.

Scoble: Yeah.

Gillmor: And Jaiku — so what? Exactly the right question. Doesn’t Jaiku have all the same features as Twitter? I don’t know. Does it?

Scoble: I don’t know.

Gillmor: OK.

Scoble: It doesn’t have the simplistic user interface because it’s an aggregator. This is the problem that we would hit with FriendFeed if they added all the features that we wanted. FriendFeed is not Twitter; it’s something else. And I think FriendFeed…

Gillmor: No. I disagree, Robert. Now you’re getting muddy again here. You have to really focus on this. OK?

Scoble: [laughs]

Gillmor: I know that you like to be on Techmeme every day.

Scoble: [laughs] No, just every other day. [laughs]

Gillmor: Well, OK.

Scoble: By the way, if Mike gets jealous of me, he starts grumbling. [laughs]

Gillmor: Well, Gabe’s on the call. And Gabe’s dealing with some guy who’s saying that, basically, the fact that stories don’t survive across the midnight barrier means something is not as good as it should be. Is that a fair characterization, Gabe?

Rivera: The problem is I haven’t actually read the whole post, even though I posted twice to it in the comments. But I read the conclusion and…

Gillmor: We will get to that in a minute, but my point with Robert is that we don’t need to change ourselves to FriendFeed in order to fix the problem with Twitter. I think he wants to boil it down to the simple thing which is what is the simplest question that we need the answer to.

Right now, I think the simplest question is that there is no Track available and that there is no IM available, and they say that they are going to bring IM back. Are they going to bring Track back with IM or not?

Rivera: And I don’t think there is anybody on this call who can answer that question. I don’t know if you can put it directly to them and get satisfaction, or one of those customer service types that they use, but I put that question on their blog and on Twitter, and really haven’t seen any..

Gillmor: No, they are not responding to it. I’ve posted a couple of Tweets about two and half hours ago, and I immediately asked them that question. They didn’t respond. Several other people did too. They are not answering the questions.

Rivera: Knew who I have had answer my questions about Twitter more than Twitter. Fred Wilson?

Gillmor: Well, do you think that he would be the appropriate person to ask that question?

Rivera: Somewhere in between. He may see the right pressure point.

Gillmor: I think the right pressure point is to ask them in public, and let them build up lack of credibility by not answering direct questions. I think it is good idea to ask Fred as well. Is it @FredWilson?

Wilson: Yeah.

Gillmor: OK. I will add that to the nomenclature.

Wilson: All right.

Gillmor: Now, Jim Posner says, “The only solution to an ugly baby is time.” In other words, getting Track back takes patience. There is no magic wand for Twitter. It is a long hard slog through a quagmire. I completely disagree with that statement. I think that is Clinton-spin of the highest order.

You know we need Clinton-spin because it’s about to end, like the political campaign. You want to come on the show and argue about this, Jim?

Canter: I don’t want to argue. I just want kumbaya.

Gillmor: Well, there you are.

Canter: Sorry I’m late.

Gillmor: That’s all right. I very much appreciated your article and Cliff Gerrish’s article and Erin McClusky’s article. Over the last 24 hours they all were to some degree or another, supportive of the idea that we may have something that we need to be concerned about. That we should push forward more aggressively than perhaps others would like us to.

Marc, would you summarize what you wrote?

Canter: I was immediately struck by your Plan B post and put a reminder to myself that I wanted to send some time really drilling down into the subtleness of your message. After re-reading it — and by the way, I re-wrote that three times. What I realized was that Microsoft is really a two-headed beast. There is the old school and the new school. And the new school is ultimately our strategy.

It is our strategy because that Mesh design cannot work in the old school way. It has to be an open world for the Live Mesh to work. So the Live Mesh benefits us all. That was really my gut when I heard that Ray Ozzie keynote as well.

Putting that in the context of Twitter, Twitter has a choice. They can continue to be simple, that their solutions where they can embrace the world of the Mesh and of redundancy. That’s why I had to create a little graph of both Fred and Evans, and said, it is up to them to decide whether they are going to remain old-school lock-in or participate in an open Mesh looking forward.

That’s what I said.

Gillmor: OK. Most people who criticize what I am doing where I talk about Mesh, they criticize it as being either naive about the fact that there is no possible way that Microsoft can get involved in anything that is truly open because they are all convicted monopolists. That’s one argument.

The other argument is that Twitter is a trivial application, has no enterprise capability or importance, and is not going to be sufficient to up-end the powers that be inside Microsoft. That’s another attack on that.

Then there is the idea that, well, you don’t understand what Mesh is. It is just a synchronization technology, and why are you talking about Twitter anyway. It was sort of a roll-up of the first two arguments.

Do you have any response to any of these?

Canter: Yeah. Obviously, people only see technology at its first blush. They see Live Mesh is a synchronization technology. Obviously, these people don’t understand the ramifications of what it means for Bill Gates to retire, and for Ballmer to be pretty much over the hill, and all of what you and I know is going on at Microsoft.

This is a tremendous change for Microsoft. If there is one thing that old hardened users like us know is that Microsoft is a company that can morph and change its DNA. It had done it before and it will do it again. They will adjust.

Knowing that that’s possible, having seen that happen before - I mean, how many of us would have believed that Microsoft would end up with the predominant browser? That was absurd in 1995.

Knowing what Microsoft can do to adapt itself to the marketplace, the significance of Live Mesh, what Ray Ozzie was doing before he came to Microsoft. Now, it is a lot more than just synchronization technology.

Scoble: Yeah. In Mesh, there is a little server on everybody’s desktop. Look at what Twitter’s architecture is. If I ran a Twitter, I don’t push to the client. They pull when somebody wants to see my Tweets.

Because I got that answer out of them, that it doesn’t… I don’t send 26,000 copies out when I Tweet. What happens is when somebody like Marc Canter looks at a Twitter page, it actually pulls the 10 Tweets on that page and pulls them over and [...] at once. Something like that, some sort of architecture like it.

Look at Mesh. Mesh has a little server. You hit the little local server, it goes over and says, “Hey, who has new Tweets for me, for this server,” and it pulls them in and re-mashes them up and displays them to me.

That’s sort of a synchronization engine, but it is not what people think of.

Canter: What you can do with the desktop server — we have been learning from Day 1, or since the 1990s about that. There is tremendous amount of things you can do with a desktop server. That’s just the beginnings of Mesh.

Gillmor: Just for the sake of explaining some of this to the audience. What you are referring to is that the original blog reader generator that Dave wrote, called Radio. I think it was Radio 8, right, Rob?

Scoble: Yeah.

Gillmor: When it came out, it supported an HTTP server on the client. It used that as a way of being able to essentially use the web nomenclature to create data and store it on the local client, and then replicate it to the server, which is how it got on to the network.

Scoble: That was also included in an aggregator that went and swept RSS from everybody’s blog, which could be either a desktop client or it could be a server/cloud-based clients, or cloud-based services, and built a little feed right there, which was local to your machine. It was not up on the cloud. And so now you could have a PHP-style Twitter thing.

Gillmor: Essentially what it did was to decouple client services from server services. And that allowed for a certain kind of bootstrapping to begin to occur that has continued to be followed.

The major vendor that essentially mined or copied that strategy was Lotus, with Notes, when they imported the Domino server, which was a bunch of IBM server code that they had for an HTTP server that IBM owned. And after they were bought by IBM, some engineers decided that they would just co-opt this HTTP server, and they turned the access to Notes from being a proprietary protocol to being one based on a bunch of open standards over the web.

Canter: This stuff has been with us for a while.

Gillmor: Right. I just want to stitch together the reasons why, from a historical perspective. Again, I’m a little sensitive to being criticized for not knowing what I’m talking about when people don’t do their homework about what it is that I’m talking about, so I’m trying to provide a little bit of extra value-add here by suggesting that these technologies, as Marc just said, have been around for a while.

And not only that, but the major vendors have been exploiting these technologies — none more so than Google today. But a lot of this stuff, the whole Ajax stack was the result of Microsoft being unwilling to unburden the browser of write capability. In other words, you can’t really use a browser to write to the architecture, except as a workaround, which was first identified as Outlook Web Access.

Outlook Web Access was a service that did the same thing, what Domino has done for Lotus Notes, and provided web page-based access to the proprietary protocols of the Exchange Server.

And the problem there was that in order to get a rich enough client that it would be usable, they invented, effectively, the Ajax libraries, which was mostly designed and implemented by Adam Bosworth when he was at Microsoft. And once that started to happen, there became this collision between the Office and Windows groups at Microsoft and the so-called Ajax groups, which were perpetually underfunded and pretty much despised by the people inside Microsoft who were in control of the crown jewels, namely the revenue models of Office and Windows.

So it’s kind of migrated, first to the whole XML architecture, and then from there to the use by Netscape and other players.

Canter: Amen.

Gillmor: What?

Canter: Amen, brother.

Gillmor: So, what happened after that is very close to where we are now, which is that Ajax became something that could be exploited, along with the basics of an on-demand architecture, by Google. And now we have forces at Microsoft who are gaining the ascendancy, beginning with Scott Guthrie, who — oh, god, it was eight, nine years ago. I went up to Microsoft with about three or four reporters — I think Steve Lohr from “The New York Times” and others — and we were invited to this off-site by Eric — what was that guy’s name, Robert? Eric…

Scoble: Eric Rudder?

Gillmor: Yeah, Eric Rudder. And there were a bunch of people around the table, and they were showing the next generation of Microsoft technologies. This was about a year and a half before they became public. And one guy named Scott Guthrie was showing this ASP.NET application development environment, which was, at that point, using some of the Visual Studio capabilities, but it was so early that it wasn’t integrated. Today, Scott Guthrie is running Mesh and all of this stuff, in terms of ASP.NET is now basically the visual creation technology of Microsoft’s strat.

So, Microsoft has come full circle, from inventing, or being forced to invent as a workaround, Ajax, to coming back and developing access t.NET and, through Silverlight, access to some of the richer services, the capability of being able to put together, re-orchestrate their architecture around these web technologies. So, to think that Microsoft is somehow clueless, a dinosaur, out of the game, yadda yadda yadda, is just myopic bullshit.

Scoble: Well said. I’m sitting on a couch in Microsoft’s building in Silicon Valley. [laughs]

Canter: And I’m watching wild turkeys run around my backyard in Walnut Creek.

Scoble: [laughs]

Gillmor: Is that what those sounds are?

Canter: [laughs]

Gillmor: So, how do we tie this together with what the importance of Track is? Do we really, seriously consider? I mean, I know that most people think that we’re wasting our time here. Gabe Rivera, do you think that we’re wasting our time talking about this?

Rivera: I think odds are more than 50% that you are.

Gillmor: Really? OK. Explain how that’s true.

Rivera: Because I think they could change. When they finally get their architecture person who actually fixes it, then they’ll put all the other stuff back in place. Well, you haven’t wasted the time. There have been other good things that have come out of it, but…

Gillmor: No, that’s a legitimate criticism of it — assuming that you’re right, for one.

Rivera: Yeah.

Gillmor: I mean, the same criticism was made of the Obama campaign. There was no way that he was going to become the nominee because he was arrayed against much larger forces. That turned out to be untrue.

So let’s reduce that argument to its essentials, which is what period of time do you think that they could take in order to not respond to these kinds of questions and then get it together that would be acceptable to the audience?

Rivera: Oh, I don’t see acceptable as the issue here. I mean, a few people will always be upset.

Gillmor: No, I’m not talking about that. I’m talking about, what would it take for them to have people move away from their architecture, to stop contributing? Scoble’s already doing this. He’s going off and siloing his posts inside FriendFeed comments, where they are absolutely invisible to the rest of the world.

Scoble: [laughs]

Rivera: Yeah. It’s a tricky puzzle that involves the whole competitive landscape and what other services do and the new services that pop up — and maybe more importantly, what the Scobles of the world do, because if they organized for a mass migration..

Scoble: I don’t know. I mean, Leo Laporte already talked about Plurk, and that kicked off 400 people to join Plurk that were in my friend network. So, 400 people out of 26,000 have migrated. Which is probably a bigger percentage than it actually seems, because I think only about 30% or 40% of Twitterers were actually active. But still, if it continues, the problem is Plurk just doesn’t have a good UI. And it doesn’t have all the features of Twitter, so…

Gillmor: But what does it do? Why is that a significant thing?

Scoble: It is not a significant thing. It is just Leo cracks up any competitor that Twitter can… He is always hated Twitter’s name, because he has a trademark for Twit, “This Week in Tech,” that’s his radio show, or podcast.

What’s going to get me the move? Right now, nothing. It’s been up and down for two weeks. I am still getting a message every second when it is up. Hell, I can even look at it on IM, I’m just looking at it on the web browser. I haven’t noticed a massive migration. People have come over to FriendFeed and checked it out. And they stick around for one or two things that — it’s not a Twitter.

It doesn’t feel like Twitter to them. So they leave and they go back to Twitter when Twitter is back up. It is a good place for them to come over and get their… It’s like methadone for Gillmor’s heroin. When the heroin is taken off the streets and Twitter is completely down, they come over to FriendFeed to get a little hit of methadone.

Gillmor: Let’s just talk about, be more specific about this, because Plurk, I have no idea what it is, so I don’t really care. FriendFeed, at least we understand that there are some serious developers behind this who are willing to have conversations in public about what their plans are. They may be vague about it, but at least they’re willing to stand up and be asked questions and respond, so that the people out there can figure out, what do we think?

Do we think that these guys have a clue? Are they so embedded in the value proposition of what they created their product for that they can’t absorb or divert some resources to be able to absorb some of the energy that Twitter has created, not necessarily through design, but rather by serendipity.

Scoble: What I heard on Friday was they want to be the body of the starfish. They don’t want to compete with the arms of the starfish. They see Twitter as one of the arms that’s bringing food to them. They don’t want to become an arm, because it messes with who they think they are, what their values that they have. I think both of us will ask him, “Hey, will you please build out that arm because it is not working for us? We would like to have that flexibility.”

Gillmor: I think that we are asking something different. Maybe that’s what you were asking them. What I was asking them was, do you have a reason to believe that building out XMPP is going to be valuable for your service? That was a simple question, not the question that I think you are asking which is, are they going to go out and compete against Twitter.

I don’t think this is about competing against Twitter. I think this is about recognizing that something exists that wasn’t necessarily conceived of, but is definitely something that people on this call and others who are listening believe is something important.

Canter: Yeah, I totally agree.

Scoble: But the way it came across was we were asking them to compete with Twitter.

Gillmor: I disagree. First of all there is a transcript of that call. So we can all go read it and we can make up our mind about whether our perceptions of what they had to say and what was asked and how they took it, and how they took it throughout the call, was monolithic or not.

I don’t believe it was. I think that they had a nuanced response to it after some repeated questioning, basically pointing them in the direction not of competing against Twitter, but of answering the question, which is do they have any sense that this transformative capability of Track is something that would…

It comes down to this. If I had the feeling that they or a combination of them and Surmise and Twirl, or any combination including Twitter, was able to come to the table and establish some sort of rationale for supporting this kind of a service, then I would be willing to move my data over there.

It is not just me, but if a bunch of us started to essentially boycott Twitter and move that type of conversation in real time to another service or services, I think that would be extremely useful in encouraging Twitter to come off of this kind of stonewall that they are doing right now.

Scoble: I think that they know that we’re going to move if they don’t get their service back up and running. It’s not running right now. It was down again today. Forget the IM and the Track feature, the core functionality was down. Until they fix that it’s hard to…

Gillmor: The core functionality, the latency is a very tricky thing. Most people today are — in ten minutes the networks are going to start their broadcast about the end of the campaign tonight, Democratic side. Kara Lee Kuhns has already suggested that she was looking at Plurk as a possible workaround in case Twitter goes down tonight, in terms of the communications of an important event.

Regardless of whether Plurk will do this for her, or whether there is some capability that will, by compositing together all of these hybrid services — which is what I am doing right now. I don’t really care whether or not Twitter goes down for half an hour if we’ve already got a chat room, for example, going on this call.

Now I am trying to reboot so that I can get the video back up to see. It doesn’t apparently seem to be working. So I’m not looking to the chat room, but I will get back to it in a minute. We have infinite elasticity in terms of being able to create these services and bootstrap existing ones to be able to get the kind of functionality we need.

So it is not about is Twitter going down. Twitter is not going down. Even if it goes down for half an hour, if I’m not looking at it for half an hour I don’t know that it went down.

The latency issue helps them mask the direness of their service situation. We are not going to abandon Twitter because they are having a hard time supporting what they have invented. There is a lot of goodwill that’s been built up there. There is no good reason to abandon them for that.

There is a reason to abandon a service: if they start either not responding, or not telling the truth, or being arrogant or any number of things that people just shy away from. Then they will only be abandoned if there is an alternative.

Scoble: Yes. I think the whole problem is there is no real alternative out there today. Plan B to me was Track could become an alternative tomorrow down the road, but it is not an alternative tonight. If Twitter goes down tonight during the Obama campaign news, there is not a really good alternative.

There is Jaiku, there is FriendFeed, there is Plurk, there is Pownce. Those are totally not satisfying alternatives.

Gillmor: None of them have Track.

Scoble: None of them have Track and none of them have users.

Gillmor: If I was to right now go over to FriendFeed and just absolutely stop, turn off Twitter, not post anything on Twitter at all, period, and move over to FriendFeed, and somehow use their conversational model in such a way that it was easily available.

In other words, don’t put any content whatsoever in — use Twitter as the signaling methodology for FriendFeed. So that the only thing that happens on Twitter is a word which says “Plan B,” and then a link. If you want to read it you got to go to FriendFeed. All that data is built up inside FriendFeed.

After a few weeks, there is going to be an interesting group of people that are only communicating over there because if they want to do what we are doing they may well vote with their feet and do so. Or, they may do it on a part-time basis.

Scoble: And they are. There is Michael Gardenbreak showed up today, a congressman showed up today, too.

Gillmor: Well, I wouldn’t go over there if you paid me.

Scoble: I know that.

Gillmor: The reason I wouldn’t go over there is not because I think that those guys aren’t deserving of this. I have an account over there. The only time I ever go over there is to take a look at somebody who is following me, or whatever the nomenclature is, over there. And I go over their stuff and if there is something interesting that jumps up then I subscribe to them too.

I don’t use any of it. I don’t use the friend environment on Facebook either. But over time, it builds value. So the question is, Plan B is not necessarily about what are we going to do tonight. I mean there is a Plan B for tonight, and I just described it. There is a combination of a chat room — and Weiner has used an IRC, which I won’t touch because I hate IRC. But there are people who like it and they will use that. There will be workarounds and maybe even Twitter will be up.

But if we start to develop a syntax and a strategy for proliferating our data on multiple services, then, as those services respond to what the Track model is, with their own support for it, even forget about whether they support using XMPP to talk to Twitter directly, then it may make a difference. It may make a difference a lot quicker with a high value audience than it might otherwise.

Rivera: I think there is room for an independent Track service, just like Summize just seems to be something that searches Twitter. You can have something that just does tracking and I think you can basically get it to be ultimately as real-time immediate as Twitter tracking was.

For two reasons. First — if anyone has confirmations, that would be good to know — I think Twitter does give feeds of all their Tweets to other companies, just streaming, so that they don’t have to scrape.

Gillmor: That’s right. That’s what they were talking about when they put IM back up. That’s XMPP, and they have committed to doing that. At some point in the future, that will be available and a third party could, in fact, do what you are saying.

Rivera: Yeah, and just how the best clients are… Well, Twitter doesn’t have a desktop client. They don’t have search, and third parties have come along and done that better. Steve, you and I both have said that we have long awaited — we find other features that Track doesn’t have really essential to what Track should be. For instance, “Track these terms but not from these users.”

Gillmor: Yes, [...] is what we call that. And IM is definitely being built out on top of that infrastructure.

Rivera: I am certain that some third party will do that better than Twitter, just because Twitter is — as well funded though as they are now, they are not going to do better than the rest of the world. There are people who are interested in all the nuances of tracking.

So I think we can expect, as long as Twitter is sharing their Tweets in real-time or at least scrapable very quickly, we can expect a world-class tracking thing as soon as someone builds it.

Gerrish: Doesn’t it sound like we are trying to build out an XMPP server architecture that is going to sit next to an HTTP server architecture?

Rivera: Well, Twitter may give the streaming feed.

Gerrish: Right, but that’s via XMPP.

Gillmor: Well, I think that this is kind of what Marc is hinting at when he talks about an instant messengar, correct?

Canter: Yes.

Gillmor: Right, so why don’t you flesh that out a little bit for us?

Canter: XMPP is an accepted standard that can get used in a lot of different ways. They could jump on the bandwagon of the usage of it as IM. We’ve got a client that is moving weather data over XMPP. Other people could update profiles and social graphs, and others could do interactive conversations.

That’s the whole idea, that you use an open standard in a lot of different ways. At least, then we could have an even playing field. I also really like the idea of this third-party tracking service. As you look at the different features of Twitter, isolate out the things that really matter, make that open and share it with the community and let Twitter put the value-added on top of that.

Gillmor: Well, both on top and also as partners. Right now they have the largest cloud. But one of the things that I like about this idea is that if we establish a nomenclature for how we use other services that will work across both Twitter and this Track service, then we are somewhat insulated from the vagaries that any of these individual services are doing.

In other words, if we can establish, for example, a FriendFeed model which routes around the private siloed comments structure of that service, and translates that into a form that would work on Twitter. My suggestion would be to wrap that up in TinyUrl or Snarl or whatever the url is, to wrap those comments that are beyond 140 characters or whatever the number would be, and be able to encapsulate those conversations in a way that would work across Twitter, and therefore work across Track of that, as well.

Gerrish: I’ve got to take off, thanks.

Scoble: Me too. I’ve got to go back to the end of the Radar conference and judging a bunch of startups in about 10 minutes.

Gillmor: All right.

Gerrish: Bye.

Gillmor: I think we have got Gabe. Are you still there?

Rivera: Yes, I am still here.

Gillmor: OK. I think we’ve got a plan now. I don’t share your optimism about how this is just going to emerge for no reason.

Rivera: It seems over-optimistic for me too, but in the world of Twitter much weirder things have emerged. I put it more than 50% and some number of months. That’s my intuition on the third-party tracking issue.

Gillmor: All right. I think that what we have got here is a significant addition to the Plan B theory.

Canter: All right!

Gillmor: I think that we will just leave this group of people who are interested in this, as the board of directors of this, and we will continue to meet on this as much as possible, over the next few weeks.

Canter: And we will get Fred Wilson to fund it.

Gillmor: Well, he already is funding it, and will just be able to continue to fund this.

Canter: All right, talk to you later.

Gillmor: All right, thanks so much, Marc.

Canter: Bye-bye.

Gillmor: I want to thank you and we will be in touch. This is Steve Gillmor. This has been the Gillmor Gang for Tuesday, election day, last day of the Hillary Clinton campaign, and thanks to everybody who showed up and especially to those who didn’t. We will see you again tomorrow. Bye-bye.

[music]

This entry was posted on Tuesday, June 3rd, 2008 at 3:56 pm and is filed under Gillmor Gang. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.