Renkoo released a public beta last week. My first impressions were "this is a pretty crude effort." But I had deeper misgivings about it, Skobee, and Planyp.us and other new entrants in group social planning that I couldn't quite articulate. Today those misgivings crystalized when emails flew back and forth about a dinner invitation tonight that was originated last Friday. What was my friend thinking when she made that invitation by email instead of using Evite? We know Skobee failed but why? What are the real challenges facing similar projects?

TO EVITE OR NOT TO EVITE?

That is the question. NOT TO EVITE means sms, email, or phone. Skobee, Renkoo, Planypus wanted or want to be a new alternative to SMS, email or phone in the NOT TO EVITE alternative. Problem is, those alternatives are too similar to evite and hence represent TO EVITE. You could argue how the voting mechanisms and the communications features represent novel features that really improve upon all the existing choices, but to evite using hipsters, they are more similar than different when the deep proposition is really

FORMAL OR INFORMAL

TO EVITE means FORMAL and NOT TO EVITE means INFORMAL. Informal is meant to be bumpy and imperfect. People willingly make that binary choice. For all the shiny different features, making them create and manage the invite on a site means FORMAL. However you want to spin it, Skobee et al represents merely another kind of FORMAL and Evite owns FORMAL. If you try your new social planning service on five hundred people and it doesn't take off virally from there, with a few nudges, you know you're slamming against the central proposition: you're too FORMAL to be INFORMAL and too similar to king evite for people to understand the difference in a glance.

[added later] I went to dinner with the friend and she told me that she wanted the dinner to be "intimate." I didn't probe deeply as it was a dinner but clearly she associated "intimate" with sending out personal email vs letting a social planning web intermediary handle it. I write more in the comments below, but I'm really fascinated to see the applicability of marketing and consumer survey tools in this domain.

Wallop started as the secretive MyWallop back in late 2003 after Google bought Pyra labs and Friendster was launched earlier that year. It was initially announced as a blogging service. Meanwhile a whole lot of activity transpired in the social networking space, including Microsoft's own launch of MSN Spaces, as Wallop remained under wraps. Wallop finally launched as a social networking service in invite-only beta September 2006.

Technology Over User Experience

Microsoft lost its way, Mr. Berkowitz says, because it became too enamored with software wizardry, like its new three-dimensional map service, and failed to make a search engine people liked to use.

"A lot of decisions were driven by technology; they were not driven by the consumer," he said. "It isn't always the best technology that wins. It is the best experience."

The cultural passions -- and tactical weaknesses -- Berkowitz attributes to Microsoft's online services unit could easily be applied to Wallop, Microsoft's social networking spin-off. Though Wallop is an entity with independent management and outside funding, it seems to have inherited the parental worldview of technology-uber-alles.

In social networking, as in search, technology must be subservient to the mission of delivering a positive user experience. Google has arguably the most advanced search infrastructure but most people submitting queries through the simple search page are not aware of the grid beneath even as revealing tech articles trickle out. Users just know it works and they return.

Friendster had the user experience down pat until it tried to scale when the much celebrated open source platform failed to deliver pages without interminable waits. According to Markus Frind,

Friendster on the other hand did a search of friends of friends. This increased the over all computing power needed to keep the site operational expotentially and killed them off. Myspace on the other hand stripped out everything that would require ranged searches and it became really easy for them to win.

So if you want to make a business that scales really well, try and make one that avoids searching and ranged searches.

I would add that the connector feature on Friendster that shows the chain of friends connecting any two people for every profile view must also have been a computational drag. (I remember being wowed by a six or seven degree chain connecting me to a Russian woman living in Beijing.)

After users started suffering waits of countless seconds for page requests that may or may not come back correctly, they stopped returning, especially when a more compelling, creative, and functional MySpace arrived on the scene.

MySpace wasn't perfect -- page requests failed frequently under heavy loads, but they always failed gracefully. You would hit refresh until the page got served, which took less than a second or two. The social experience was compelling enough for users to tolerate these little glitches. Tom and crew found continually tweaked the recipe that members consumed in pageviews numbering the billions or trillions. Technology worked just well enough, and, aside from the hiccups, transparent.
Read the rest of this entry »

I got interviewed by a writer for a business magazine on Friday about Twitter. Fittingly, Twitter's the only service I use regularly of the dozens of new Web2.0 social or communications sites, but paradoxically, it is also the one that I've yet to dedicate even a whole paragraph of writing. I had dismissed Twitter as a flash mob-type fad and resisted attempting to understand this multi-headed hydra.

I'm a big fan of Ries and Trout market positioning, and I keep their lessons in mind when I look at any service. They advocate the power of simple, clear positioning in the mind of the consumer. In apparent violation of those prescriptions, Twitter strives to be so many things. I realized that Twitter is so flexible and entertaining that users find their own combination of useful features. There is no incoherence because they focus on what works for them. Twitter's feature rich like a buffet or Swiss Army knife and there's definitely a market for those things.

Whether Twitter will turn out to be a mere fad or find legs and break out is something for the market to decide. But whatever its future, Twitter deserves to be studied like a movement or school in stream of art history. Here the stream is internet communications services. Online chat hasn't changed fundamentally since the inception of the currently dominant forms. Meanwhile mobile communication and social networking has risen and the web has spawned a new effort to reinvent itself (web2.0). Convergence and synthesis is happening all over the place and Twitter is certainly one of the shiniest new examples. Anybody starting a new social web service should take a hard look at Twitter and try to understand why it works. (The best tools of analysis probably come from the field game engineering, eg WoW or GTA, but that's another subject.)

So what is Twitter? Let's look at it through the prism of conventional categorization and maybe we'll get the gestalt by the end. The first group comprises functional features and the second abstract qualities:

• Social Networking System
• Chatroom
• Microblogging
• Multiplexer
• Group Communicator
• RSS Feed
• Salon
• Meme
• MLM

Salon

Perhaps by virtue of its origins - SF based with somewhat prominent founders -- and social networking and chatroom features, Twitter draws a slice of internet cognoscenti early adopters. Examples of solid but not flashy webizens I've seen are Christine Davis, Steve Smith, Frances Berriman, Lynn Wallenstein, and Stowe Boyd. Yesterday I saw Yan of Planyplus and Emily Chang. Just now, I see the artistic Johanna McDonald. This is just an arbitrary bunch of people I picked out to give you a flavor of the subscribers. Think tribe.net. (I hope these people don't mind being "outed" this way. I'll put the genie back in the bottle if you ask nicely.) Sam Sethi calls it a backchannel, I call it a salon.

Social Networking System

Works well. Twitter follows Flickr in deviating from the standard add/approve model used by the likes of MySpace. In Twitter, you can add somebody without their approval and they can add you independently. What you are really doing is subscribing to their message streams. It's like social RSS for microblogging. The profiles are simple, succinct and colorful. The personal URL in the profiles have a high emphasis befitting active webizens. People tend to find each other in the public timeline, ie the chatroom.

Microblogging, Portability, Multiplexing, RSS, Persistence

The unit of currency on Twitter is the message. In the typical chat context, the message is sent and published in the receiving chat client or in the chat room. Then it disappears as it scrolls off unless someone archives the log. In Twitter, messages are broadcast to cellphones, IM clients, web accounts and the public timeline, RSS feed, any or all at once depending on the sender and subscriber settings. For posterity, the messages are archived in time order with a permalink to each entry.

Twitter multiplexes the mode of messaging. Not only can messages be received through a mix of channels, they can be sent through those channels too: {IM, SMS, Web} -> { IM, SMS, Web}

The microblog entries - or lifestream - can be republished intact via Flash or other widgets that tap the API. Aside from media and blogosphere buzz, this viral mechanism could be a key marketing tool for the platform. A quick search using Google shows ~2,000 twitter badges on the blogs and ~160 on MySpace. Small numbers but supposedly growing fast. We'll look back in a few weeks to see how far Twitter runs.

Chatroom, Public Timeline

Twitter's public timeline and the people it draws reminds me of an IRC chatroom. In fact, some people from Twitter have formed a parallel #twitter room on the server blitzed.org. One way of spinning Twitter is seeing it as an elaborate marketing scheme for one chatroom. Whether by design or necessity, having only one public chatroom is huge way of concentrating attention and mixing people up in Twitter's bowling alley stage of adoption.

Clotaire Rapaille says we go to the mall to connect. Webizens go to Craiglist, the blogosphere, and now also Twitter to connect.

An irony is that people don't really connect much of the time on the Twitter public timeline. Some participants actively do, but many are not even actively aware of the public space. Their message entries get posted there automatically. The public space is really both a chatroom and an exhibition space: a grab bag of conversations and nonversations that provide easter eggs for what Rapaille would call a treasure hunt and I refine as a social treasure hunt. You see what you like and subscribe to the feed. Advertise your personality and someone might subscribe back. If it works out, you make new friends.

Group Communicator

Having a friends list, multiplexing, especially to mobile devices, ability and an API creates a platform for group communication. The problem with this function is that at this early stage very few people on the system are real life friends and all you have is a list of virtual friends with who you may not really want to keep up with all the time. But that's something that will get worked out once you get the world onto the system. Again, the Salon and Social Networking are proving effective at drawing people to the system and forming a sense of community. Other more pure attempts at group communication such as 3Jam work perfect technically but have a huge adoption problem.

I would like to have later the ability to partition friends into different groups. Both 3Jam and Yahoo's Mixd have this feature. Again, it's understandable that they concentrate everybody in this stage of their growth.

Meme, MLM

Twitter has claimed recently to have doubled the user base. I think a lot of it has to do with word of mouth and widget viral buzz. No slam against the widgets, but I wonder how many people who see them really care about what their friends are doing? I think men may care about what a particular gal is up to in a stalkerish way but do healthy people really want to receive real time and continuous status updates? I suppose I am being too heavy.

Perhaps Twitters widgets should be seen as the blog within the blog. Plenty of blogs share quick short thoughts. MySpace has already demonstrated the notion of blog sub- or collateral communication with the optional "current mood" and "Tell us what you're reading, viewing, or listening to" that accompanies each blog entry. The Widget merely generalizes this notion and gives the user to flexbility to place it wherever she wants. Optional to place, optional to read.

I say MLM because I see the widgets as self-propogating empty memes whose real function is building community. It's like round about way of advertising a club of internet wonks. Again, this perception of mine could change as more mainstream people jump on board.
Read the rest of this entry »

Yahoo quietly launched Mixd yesterday. On the phone side, Mixd is equivalent to and even slightly better than 3Jam. Mixd’s web interface for creating and managing the SMS groups is much better than 3Jam’s though still raw and has the capability to receive and display videos and photos.

Like 3Jam, Mixd implements the IRC model over SMS. Mixd does make –- or adopts -- tiny innovations. Each invitee can be allowed or denied the privilege to invite others and each invitee can kick his child invitees from the group. Chats replies are sent to the web interface in the form of stacked talk bubbles. If the chat transcript were to be widgetized, you have the equivalent of a Twitter. There’s no API yet, but that should come later if Yahoo follows the strategy for its other services.

The phone interface was a bit confusing until I realized it was just like 3Jam. Each conversation, ie Mixer, is tracked by the initial phone number the SMS came from. Replying to that same number replies to that Mixer. For example, “Apple” Mixer is tracked by 445-556-000 and “Orange” Mixer by 445-556-001. I could not get Mixd to take the picture I sent from my Sprint phone. The receiver is a single email address pix@mixd.com so it’s not clear to me how Mixd would figure out which Mixer to sort the picture or video to.

Strategic Implications

Startups in this space like 3Jam and Twitter will lose sleep at night. Nobody has a technical or legal lock has on these simple ideas and or has a huge user base as we are still in the bowling alley phase of adoption. 3Jam is doomed unless it can find a buyer. No VC will pump more money into that venture after today. Google should pick up 3Jam’s assets for loose change to bolster their stagnating Dodgeball asset.

Dodgeball was afflicted overambition and miscalculation and now is locked into an unappealing model of operation. Whatever user base it has is growing at a snails pace and nobody is using the service. Though Google was lured by the Dodgeball’s shiny features, they didn’t realize that the “check in” location model doesn’t fly on the phone interface and with a sparsely connected early community of adopters. Simple communication ala 3Jam or Mixd is probably possible but the Dodgeball doesn’t position the service that way and non-subscribers can’t be invited into a convo which is basically an anti-viral killer. Google should pay $1 or $2 million for 3Jam and incorporate the assets into a remodeling of Dodgeball, if they still care at all. Dodgeball is a rounding error in their financials, so maybe we even bother to speculate.

Within the Yahoo house, the opportunities are rich. We won’t see them in the near term but shouldn’t be later surprised by ties between Mixd and Yahoo TV and Flickr. YouTube gets video from the web and resells to the likes of Verizon. With Mixd, Yahoo TV can reverse that flow and use phones to gather video. The UCLA student tasering incident was captured on phone video. If photos and videos prove popular on Mixd, expect YouTube to be soliciting your phone soon.

SMS group chat is pretty cool and thus far the big problem has been marketing. In my own tests, people get it quickly enough but no one really wants to use it. Yahoo applied its huge account base to Answers and got a lively community asking and answering questions. I think the same magic could work for group chat but with more modest success. Strategically it is important for Yahoo, Google, Microsoft, and anybody else to mark their territory in the customer mind as SMS is merely an early transitional battleground. The real battle comes when rich phone interfaces become prevalent.
Read the rest of this entry »