Trillian Possibilities

Enough people have mentioned Trillian to me lately that I should probably consider its importance. There may be others, but Trillian is the only instant messenger I'm aware of that works with all other messengers-- Yahoo, AIM, MSN, ICQ and soon Jabber's Exodus, I believe. The suggestion, along the lines of what Michael Bernstein has argued, is that GlobeAlive use an interoperable chat proxy on the user side (something like Trillian) and let the experts/participants use their own favorite instant messenger, whatever it may be. This would dramatically increase both sign-ups, and more important, quantity of participants online at any given time, as there would be no GlobeAlive Desktop to download or keep logged on.

We're already working on making GlobeAlive compliant with a Jabber client to eliminate GA Desktop, but this will still require experts/participants to download a new messenger. If indeed there's a way to do it, it's certainly preferable to let the experts/participants use their own favorite messenger (which they already use regularly) to interact with their visitors at GlobeAlive. But Trillian, unlike Jabber, is not open-source, and without a formal partnership with them, we could not use their code on the user side.

I wonder if there is a Jabber client that allows users to chat with any of the other major IMs? In other words, a Jabber client that does what Trillian does? Conceivably one could be created from hacking a current Jabber client, but our programmer mentioned that creating a Trillian-type chat from scratch is an enormous undertaking and not one he's prepared to even consider at this point, which is understandable.

Wiki Alive

Apparantly wikis, to some degree, turn blogs into conversations, which means they might have a place in the GlobeAlive model. Here's where I first learned about wikis and here's Joi Ito's wiki with a discussion about what it's all about.

The Virtues of Jabber

We're switching the current GlobeAlive Desktop IM to Jabber, which is a big and necessary move for us. Here's Doc on the The SIMPLE case for XMPP, the initial core of the Jabber protocol:
XMPP and Jabber are taking off because they are (truly) simple infrastructural building materials that are highly compliant with the NEA nature of the Net.

SIMPLE is being pushed by IBM and Microsoft for a variety of reasons; but that very variety may be the core of a problem.

...Meanwhile, any company, any customer, can develop its own Jabber-based IM system. Countless numbers have done exactly that, including IBM.

SIMPLE may be a fine protocol when it's done; but it's not. And from the sound of what Microsoft says above, it risks violating POGE: the Principle of Good Enough. Without POGE we would have no TCP/IP, no HTTP, no HTML, no SMTP. In other words, no Net, no Web, no Net-based email. POGE also accounts for the success of XML and Linux. It's why XML-RPC moved faster than SOAP.


The Word

Doc suggested I use Technorati to check out what people were saying about GlobeAlive. Found the following bloggers and here is what they had to say:

Scott Knowles:

Peoople. Globe Alive is Google for finding people: The World Live Web.

Not since I discovered blogging have I been as thrilled with a new idea on the internet as I am now with Globe Alive.

Daniel Von Fange

GlobeAlive - not yet primetime.

GlobeAlive is definitely still beta quality. Although there are great posts on the web about what it can become, at the moment it is not well implemented.

The Head Lemur
The Next Big Thing

Real time communication with folks who have information to share, expertise to offer, and are willing to do so on a one to one basis is going to revolutionize the way folks can use information to improve their lives, their business, and their society. Globe Alive is the next big thing.

Tom Maszerowski

Check out GlobeAlive, it's a search engine that finds live people you can help answer your questions. For instance, here's the results for homebrewing beer.

Christian Vidmar
Matt Mower: The expert web comes alive. Couldn't be more alive than this, after having read about GlobeAlive from Matt I went there and met...Matt! Live.
My thanks to those who think GlobeAlive has what it takes. I should add that Daniel Von Fange is right that we are not ready for prime time. Perhaps I need to put "beta" on the front page. We are certainly a grass-roots beta site without money trying to make it work with a small but growing following of people who just really like the idea. He's also right that the system defaults to "all experts online" when there are no matches, which means you get a lot of people asking you live questions you can't answer. This is a temperory measure that we plan to eliminate as soon as we have enough targeted traffic.

Hopefully, GlobeAlive will catch on and truly serve as a live alternative to the Googles of the world one day. Until then, please bear with us.

Bernstein's Critique

I got an email from Michael Bernstein today. He makes a strong point.

I was interested in your weblog posting "The Difference". Eliminating
the barriers for users is obviously important, but you've still not achieved the simplicity of Google.

Here's the main difference: Google doesn't require you to register before they index your site.

Increasing the number of experts indexed by GA would make the service *much* more attractive to users.

The way to do this is to get webloggers to include an 'expert profile' in some XML format or other and link to it from their weblog, in much the same way that FOAF (Friend Of A Friend) files are right now. Then, you could crawl the internet looking for experts to index, or changes to profiles, much as Google crawls the net looking for pages to index.

Said profile could include information on IM accounts, and then, in order to continue to provide annonymity, GA would have to act as an IM proxy for users (and proxy to multiple IM services).

Notice that this also prevents lock-in, as anyone can index the same info.

If Michael is right and there is such a way to search live people/bloggers without having them sign up with GA first, it would make GA a true search engine, and make all the difference in terms of sheer quantity of live matches. There are complications, however, which may lead to a compromise solution, as Michael points out:

1) you need to get the IM account info from the participant in some way. Ideally without them having to do anything. The closest thing I can think of to 'nothing' is having them include contact info in an autodiscoverable way on their site/blog (like rss autodiscovery). I realize that this isn't *actually* less work at this point, because no weblog tools currently support this automatically, but it could certainly become part of the standard weblog setup/service over time.

Which leads us to:

2) Spam. How do users prevent their autodiscoverable contact info from being abused? I don't have a good answer to this, unfortunately. The same problem exists for email, and people have been becoming more reluctant to include an email on their websites and weblogs.

So, an easier solution for now might be to still require self-registration, but replace the GA Desktop download with a form field
that lets the participant register an existing IM account, to enable the proxying I outlined above.


Ratcliffe Writes Again

Mitch Ratcliffe, an author who I met on the phone last week and the man behind the Strip Mall Infomediary concept, says that blogging made him want to write again. After years of writing short 600 to 1200 word columns, he found himself writing an 8,300 word article recently The Invisible Dogma (a sobering analysis of misconceptions surrounding the application of technology as a fix for broken human processes.) He is prolific again and feels blogging is largely responsible for his reinvigoration:

Something happened to my brain last week when I finished the draft of an 8,300 article -- you see, I'd been writing 600 to 1,200 columns for years, many of them each week and usually as a sideline to what I did for "work." It got so that I was unable to write beyond that limit with any real success. By the middle of 2001, I just stopped writing, because it had become formulaic and not fun anymore.

Blogging made writing fun again, largely because there was nothing to worry about: length, typos (though I am careful about this), or fees (I'd gotten to the point where if I didn't make a certain amount per word I didn't want to write, like a lazy star outfielder who has several years left on their contract). That article on Invisible Dogmas broke a barrier in my head and I am thinking about writing not just one but several books over the next couple years. I haven't written a book in almost 10 years, so this is a big thing for me. And I've been thinking how to use this renewed energy.

Another GlobeAlive?

Just got an email from Alf Eaton, the creator of the alpha site Thinkbot. This is about as close as I've seen so far to another "chat engine." Alf says that we beat him to it, but his idea still sounds very distinct to me.


Anyone know how to fix permalinks at Blogger? Just found the following so maybe I'm not the only one:

Now if they can just get around to fixing the Blogger permalinks which have been broken since the templates blew up on April 1st...

Kind Words from Strangers

Thanks to Doc's reference to GlobeAlive last night, GlobeAlive has been the recipient of many kinds words from strangers who hopefully will soon become friends. Perhaps the kindest words came from Head Lemur who had at the very least the kindness, and at the very best the insight and foresight, to call GlobeAlive "the next best thing." I sure hope I can prove him right. Here's some of what Head had to say:

Information without education is the old way of communication. Blind faith in the acceptance of news by organizations whose principles are being bent by ratings, marketshare, and advertising, are being examined and pointed out with increasing regularity by simple folks whose eyes are being opened to the potential of communication and discourse in near real time.

Doc's announcement of Allen's new enterprise, Globe Alive, points to the next stage in this revolution in communication. Globe Alive is the next big thing.

Real time communication with folks who have information to share, expertise to offer, and are willing to do so on a one to one basis is going to revolutionize the way folks can use information to improve their lives, their business, and their society. Globe Alive is the next big thing.


Just got off the phone with Scott Johnson, founder of Feedster, a search engine that specializes primarily in finding bloggers. We're exploring possible parallels between the Feedster blog search model and the GlobeAlive live search model.

Scott has modeled Feedster on the simplicity of Google. You can see from the Feedster front page that they're all about simplicity. We discussed the importance of being a one-trick-pony. Here's Scott on Feedster:

Feedster is a search engine for what is called an "RSS Feed". An RSS Feed is an XML tagged file which allows a website, news site or blog (actually any site) to provide to the world a list of its current contents. RSS feeds can contain all kinds of information from news to blog / weblog posts to stock quotes and more.

The Difference

Britt Blaser, founder of XpertWeb, sent me an email today regarding Google Answers. I have actually never used that site, but I have been watching it since its inception and I'm glad to hear Britt is watching it too. He has precise stats on their traffic, their growth and their revenue.

Basically Google Answers is a first class version of the standard expert site model.

  • Non-live, blind-email based interaction.
  • Members only.
  • Pay per answer.
  • Choose category

To a degree, Google Answers is a success, but I think sticking to the restrictive standard expert site model is the main reason why Google Answers had only 500 users last week while the Google search engine had over 1 billion. Except in special cases, it seems people want the answer now, they don't want to be a member first, they don't want to pay for the answer before they have a relationship and they'd much rather use a search box than a directory.

What I was hoping to do with GlobeAlive was to not only have all the advantages of human response, but too keep all the advantages of the search engine format (at least for first-time users of the site):

  • Answers right now. How many people would use Google if Google got back to them with website matches within 24 hours?
  • Anonymous. How many people would use Google if they had to be a member first?
  • Free. How many people would use Google if it cost per search?
  • Search box based. How many people would use Google if it didn't have the simple search box format?

I would venture to guess that the answer to the above is about 500. About 500 users per week would still use Google under the conditions above, which is the traffic that Google Answers has right now. I think this is a very clear signal that the simplicity of the search engine model is the superior system, but what that search engine produces, websites, may not be what everyone wants.

The most important question for GlobeAlive is: how many people would still use Google if it listed live people, rather than websites, as search results? Live people that matched your keyword and could answer your question right now, free, anonymously, by chat. I would guess the number would be less than the current 1 billion per week, but still in the tens or perhaps hundreds of millions. At the very least, it would be a whole lot more than 500.

I hope Britt agrees with me. I do believe that people need to be paid for their time, but not before the first "hello." And I do believe that memberships are useful, but not before one has had a chance to use a site. I also believe that blind email-interaction is a good medium, but not as an exclusive. Live interaction, answers right now, is very often preferrable. A fast nickel beats a slow dime. Finally, I do believe that directories are important, but a database of people, like websites, also needs to be effectively searchable. The search box is the way people are now used to finding things.

Like Father, Like Doc

Doc's "markets are conversations" thesis specifically, and Cluetrain in general, served as the initial inspiration behind the formulation of the World Live Web. I knew in the abstract that I wanted something along the lines of a live one-stop-world of people, not websites, but I didn't have any idea how to do it, or whether it could be done. My father's idea crystallized it into a formulation for me and the conversation-based, live search model (the world live web) came into being soon thereafter. The World of Ends is Doc's latest conceptual breakthrough and one which in many ways takes off running where Cluetrain left off. Apparently I'm not the first person to create a new model from Doc's ideas, and I'm sure the World of Ends will do for many people what Cluetrain did for me.

Doc Speaks

My father, Doc Searls, known to many people as the co-author of The ClueTrain Manifesto , has been kind enough to offer his analysis of the GlobeAlive concept:

It's basically a 'live' search engine: one that finds human beings who might be available to answer questions in real time. There's a lot of synergy with what Britt is doing with Xpertweb, and what Mitch has been saying about the Strip Mall Infomediary, both of which also, like GlobeAlive, could stand to benefit from the kind of identity infrastructure I wrote about in Making Mydentity, and expect to see coming out of SourceID and similar efforts. There are also natural synergies with smart mobbery, social software, moblogging, and most of the stuff in Marc's blogrolling column. Even ActiveWords, to name another potentially interested party. And, of course, instant messaging with presence detection, which is why Allen and friends are currently developing a new Jabber-based client.

Blaser's "Commander Data"

Britt Blaser very succinctly sums up the value of both XpertWeb and The World Live Web with the notion of "Commander Data." Blaser's own words:

You know what I want? I want Commander Data. I want him in my coat closet, using no resources until I have a question and then he activates, solves my problem and goes back into stasis (he might be expletive-activated and then expletive-deleted). Because he's Commander Data, he does everything almost immediately, so I'm willing to pay him a lot per minute.

I'll bet that's what you want too: an expert on the software you've got, not more software for you to be inept with.



Had a conference call this morning with the always interesting Britt Blaser and his close friend Mitch Ratcliffe, who I met for the first time today. Both are entirely out of my league in terms of Internet mileage and insight, but they were kind enough to listen to the idea behind GlobeAlive and offer suggestions on how to make it work.

XpertWeb Model

Britt and Mitch are exploring some new ideas for XpertWeb that may bring a new dimension to the money-making side of the XpertWeb model. The idea centers around a third-party trust account that works as a unique form of insurance for both parties involved. Beyond that I neither know enough nor have license enough, probably, without their permission, to go into further detail.


Looks like Britt Blaser likes the idea of the and has taken it upon himself to purchase it for us. I'll be thinking of ways to give bloggers their own live niche in "the world live web." Maybe Britt will have some ideas in connection with his upcoming XpertWeb as it rapidly approaches completion this year.


Emphasizing the World Live Web

We made some substantive changes to GA today. We're attempting to emphasize the "world live web" idea over the GlobeAlive name. Richard Moose built us a nice "world live web" logo on our front page with a lunar look to the GA logo in the background. He's also experimenting with some chrome looks and some new icons.

Seeable Search Results

We've also made changes to our search results page which now has a clean white background.

The Importance of Being Generic

Maybe the most significant change is in the names of the participants. We've stopped calling all the participants experts and have opted for "participants," the most generic term available at the moment. We don't want to ciphen off any interested group, given that GA works for almost any real-time communicative interaction, not just asking and answering questions. All the front page links and sub-page references were changed.