Octavianworld

500 Mirrors: Channeling Light Into The Business of Virtual Worlds

I connected recently in NYC with my ex ArsDigita colleague and good friend Kevin Kelly.  Kevin is (among other roles) the CEO of 500 Mirrors, an enterprise-class virtual world platform provider, and was in town for the annual VirtualWorlds Conference at the Javits Center.

For the virtual worlds crowds, these are dark days, not unlike what things were like for the web generally in 2002.  After flying high in 2006 -- including making the cover of Business Week -- Linden Labs' Second Life, the poster child for the category, has hit rough patches on several fronts, including usability and, security, and scalability.  Electric Sheep, the leading interactive agency helping corporations build and pimp out their VW experiments, recently cut back its staff significantly.  And as most of us have seen, even mighty IBM, the corporate pied piper of the VW movement, has been advertising against its own efforts.  The remaining bright spots in the "classic" rich-client VW world these days seem to be in applications for kids:  Webkinz, Club Penguin, Habbo Hotel -- virtual babysitters for kids that parents are willing to pay for, perhaps partly as diversions to keep them away from first-person shooters,  MMORPGs, or perhaps as training for earning the big bucks playing WoW.  Though 1300 people registered for the conference, things seemed quiet even for the cavernous Javits, with few attendees from big-name companies/ brands in evidence.

My overwhelming impression remains that this is an industry whose technological reach and ambition greatly exceed its grasp of imaginative use cases for which this medium is uniquely suited.  Kevin and 500 Mirrors' CTO and founder Bob Flesch get this in spades.  Kevin tells a story of a recent conference in SL that illustrates the state of things:

• with a limited number of avatars each island (server instance) can support, the conference inevitably had an empty feeling;
• lots of power and flexibility for moving around means too much power for newbies and the less experienced;
• P-bombings have been an unfortunate reality;
• functionally, the experience's complexity exceeds its theoretical information and communication advantages -- for example, it's hard to read expressions when an avatar -- if human -- is programmed by default to look ironic and bored, if hip.

500 Mirrors' name reflects an approach to scalability it has developed that has effectively solved population limits for any practical enterprise and even consumer scenario.  (Proof comes from production instances that unfortunately Kevin can't disclose without unpleasant consequences.)  On the usability front, Kevin and Bob are focused on providing more by enabling less -- pre-scripting movement sequences to get someone into the right place in the right space, for example, or by turning off flying, or by simply making it impossible to get trapped in a corner.  Finally, since each instance they set up for a client is isolated (whether hosted by 500 Mirrors or installed behind a client's firewall), attendees can't jump out to inappropriate places, and intruders find it harder to get in.

But the biggest insight comes from conceiving of use cases that make sense, and Kevin's got a separate business going that's nailed one of these.  More about this in an upcoming post...

OpenACS Zeitgeist Next Stop: Brazil

Following the recent global user group meeting in Guatemala,  o resurgence de OpenACS/.LRN will next pop up in Brazil.  Here's Eduardo Santos'  summary:

I'm happy to announce the first (or the second, if you consider this) OpenACS Brazilian users group meeting. The event will take place at the 9th International Free Software Forum (FISL 9.0) in Porto Alegre, RS, Brasil, from April 17th to 19th.

The Brazilian community is growing and we already have 607 members from many different states in country. We also have a lot of translated documentation, which is an effort to make the OpenACS framework popular among Brazilian developers...

Here are some of the OpenACS/.LRN based services they will present:

1. OpenACS history in Brazil, by Rodrigo Proença
2. OpenACS projects in Embrapa by Orzenil Silva
3.  Citizens' Territories Portal, by Alessandro Landim
4. SPU Colaborative Portal, by Vitor Silva
5. Virtual Public Market, by Eduardo Santos

See his post with details here.  Please pass the word into your favorite e-learning/ online community/ web development forums/blogs/ listservs, etc. !

At Harvard KSG With Tim Berners-Lee

Jerry Mechling, who teaches at Harvard's Kennedy School, and runs the "Leadership for a Networked World" (formerly E-Government Executive Education) program there, invited me to a talk by Sir Tim Berners-Lee last Wednesday evening.  The audience included ~50 current and former senior public-sector information technology officials attending one of Jerry's sessions.

Sir Tim's comments included:

• a discussion of how the WWW came to be
• an examination of some of the risks that could have killed it early on, and how those were overcome
• an exploration of some of the possibilities of the Semantic Web
• an exhortation to members of the audience to "set their data free"

An Unevenly Distributed Future: MITX Internet Video Panel

Last night I went to a great MITX panel discussion at MIT on Internet video.  Scott Kirsner moderated a great and diverse group representing the media, CDNs, VC, and agency worlds.  Scott organized the discussion into three buckets:  how are users interactions with web video changing; how is the production side of the business evolving to enable new things; and, the "monetization" of all this.  Here are my notes, and a few ideas:

Just When I Thought I Was Getting Clever

Since publishing your email address on the web invites spam, I recently used a free external service to add an "email me" form to this blog (see link at left).  The service uses captcha, the system that asks you to prove that you're human when you register for something online by copying a set of squiggly letters and/or numbers into a form.  I was feeling very smug about this until I read my former ArsDigita colleague Carsten Clasohm's post on how some spammers now get around this.

Memo to MIHOO: A Modest Proposal for The Next Step

There are three ways an ad network can compete:  better reach, better data, better analytics.  Know more people, know more about them, know what to do with that information.  Jeffrey Rayport noted yesterday in MarketWatch that recent consolidations in the ad network business have made it a game that "only the mighty can play."  Meanwhile, Danny Sullivan wonders in Ad Age whether the Microsoft-Yahoo combination will be enough to catch Google, even if they can pull the integration off.

So here's an idea, borrowing from a time-honored software business strategy most recently exploited by Facebook when it opened up an API to its "social graph" (also inspired by recent fiddling I've done with the IBM Many Eyes project).  MIHOO should provide a way for "third-party algorithm developers" to query its combined data hoard of users and their clickstreams.  That way, they can "unleash the power of open-source algorithm development" by letting the unwashed mathematicians of the world back-test their ideas for getting consumer X to respond to offer Y in context Z.  Of course, this would come with a price: when you go to run your algorithm on the MIHOO network to "monetize" it, MIHOO takes the gross cut to the network and pays you, the algorithm developer a royalty.

Symbiotic way for MIHOO and the little guy to take on big GOOG.  Lots of problems with this to be sure, starting with privacy (even though it already "is", can data be further anonymized to make it tough to infer my personal clickstream?).  And, as a practical matter, can MIHOO expose a data warehouse that could take mass pounding from the world without slowing to a useless crawl?  Further, would it be worth it?  There is a diminishing return to finer and finer targeting ideas, after all.  As for the competitive risk -- that people will get insights on MIHOO's data set that they can deploy on some other network (Google's or otherwise) perhaps there is a way to abstract the data being queried until a promising algorithm can be "optioned" by MIHOO.

Granted, it's a bit out there... but work with me people, or help me find the fatal flaw.

Sunny Days in Uppsala

Sun's bought MySQL AB for a cool $1 billion.  Why?  Maybe Amazon's recently adding SimpleDB to AWS made Sun, which is also offering utility-based computing, a bit nervous.  Some folks wonder if a billion is too much to pay for open-source software.  I say they get the full attention of the core of the MySQL project, and once they figure out how to run distributed MySQL clusters on their version of EC2, they can rent (SaaS) what they can't sell.

This open-source thing is going to be big one of these days...

Related: Tim O'Reilly's recent Radar talk offered some cautionary notes about the whole cloud-based SaaS trend that are worth considering:

http://itc.conversationsnetwork.org/shows/detail3364.html

Not Dead Yet

Went to the Web Innovators Group meeting last night in Cambridge.  The ballroom at the Royal Sonesta was full, with a healthy mix of entrepreneurs, engineers, students, VC's, corporate types, etc.   Rumors of Boston's demise as a tech hub are overstated.

Memorable snatches of conversations/presentations:

• "I hear there are assets in Germany available..."
• "We started with a subscription model, and morphed it from there..."

and my favorite:

• "We're a real-time mixer service, halfway between Meetup and Dodgeball.  It's 3 o'clock, you want to meet people, you sign in, and at 4 we've got you in a bar with five people you don't know... but you have something in common with!

I caught up with Paige Arnof-Fenn.  We speculated on Whether It's A Bubble.  "No, there's twenty times the money (advertising anyway) there was in 2001, and maybe 75% of the VC money invested back then," was our consensus.  What about a recession? Money flows to the most measurable media, which these days is online, so maybe this time a recession might kill off the old (golf-and-a-rate-card analog formats) instead of the new.

The River of Time(s)

The New York Times has graciously allowed Dave Winer to innovate with their content.  The result is http://nytimesriver.com, an alternative presentation which is halfway toward a branded reader.  I like the outline view especially, since it dynamically sorts out topics, in a way that subscribing to fixed tags or categories in an RSS reader wouldn't.  Judging from the traffic, it will be a bit before we see ads here, but I'm impressed it registers since it's so new!  I think the bigger point is the happy result of this open-minded partnership.

Open Social: The Empire (and Friends) Strike Back

I recently wrote a post reacting to Brad Fitzpatrick's August 17 essay, "Thoughts On The Social Graph".  Brad recently left Six Apart to go to Google.  Now, as many predicted, the other shoe has dropped, with Google and others announcing the new "Open Social" standard. 
Here's today's NYT article on the subject.

The concept of the social graph -- a digital map of who knows whom -- represents an increasingly critical element of many web applications, especially public ones.   Early social networking services like Friendster, MySpace, LinkedIn, Facebook (and behind-the-firewall versions like Contact Networks, where I used to work) became valuable because they offered advertisers and other business users  "viral  pathways" to communicate through that more closely matched how people really filter/ absorb information -- word of mouth from trusted, or at least known, sources -- in an age of information overload.

Facebook recently pushed the utility of the "social graph" one step further by providing an API to allow others to build applications that take advantage of it.  This strategy has been a runaway success.  Third-party developers get more efficient access to an audience than would be possible through the conventional "email to a friend and then have each one register separately" MO we've been using for ten years.  Facebook gets inventory to advertise against, and maybe a cut of commerce at some point.  Users get more utility out of a platform they already like -- which means free stickiness for Facebook.  All this goodness is why Facebook is worth 100x revenues to Microsoft (and whatever other investors get announced).

Now, as is the way of all flesh in the software business, a group of Facebook competitors, who individually and collectively find themselves behind in the race to build their own social graphs, have banded together to see if they can re-capture the initiative.  The vehicle for this is a new standard called "Open Social".

Domain Farming: This Is Clever

From Seth Godin:

http://sethgodin.typepad.com/seths_blog/2007/10/farming-domains.html

Neat.  What other domains have this potential?  Sure beats simply parking them.

Online Video: That Sound You Hear Is The Dam Bursting

By various measures there are some $500 billion spent globally on advertising.  About 40% of this goes to TV, which is why Oxygen is worth a billion dollars to folks like NBC.

Yesterday I saw a press release in which Sony announced more content partnerships, with folks like CondeNet and Sports Illustrated, for the "Internet Video Link" service it packages with its BRAVIA high-def television sets.  Sony gets unique content it can sell ads against.  The online publishers presumably get a rev share on video assets they've already produced, and extra traffic going back to their sites.  If I understand how the service works, this completely bypasses the cable TV infrastructure by using whatever broadband you have available.   Memo to cable company, re: $100/month digital cable subscription:  "Ask not for whom the bell tolls..."

(Ok, you say, but why buy an HDTV if all you can watch is low-res Flash clips?  Not so fast.  See this interesting announcement from Brightcove and BitTorrent yesterday?  How long will it take Sony to add BitTorrent nodes into the   Internet Video Link modules on its sets?)

So, how do we know if Sony's video content bet is ultimately better than NBC's?

NBC paid $12 per Oxygen channel subscriber.  According to news reports, there are about 300k Oxygen viewers in prime time.  Let's assume of course that different people watch different shows each week, and that prime time is as popular as all other times combined.  So I figure that 300k prime time viewers translates into 1.2 million unique viewers per month (cross-check: Oxygen.com has 1.2m uniques).  OK, so $900 million (Oxygen's acquisition price), divided by 1.2m = $750 per unique viewer.  (NBC also has to continue producing programming and pay carriage fees to cable networks to continue to reach those viewers.) Against this up front and ongoing cost, Oxygen gets about $7/unique (per month of course, ~$100M in sales/ 1.2m uniques/12 months). Figure a 15% margin, and NBC's netting $1/unique out of that top line (Oxygen Media LLC didn't disclose income AFAIK).

I don't know the terms of Sony's rev share, or how they are placing those ads with advertisers.  Let's assume the overall split is 1/3 to each party (Sony, content partner, video ad network).   Let's assume a $30 gross CPM for the premium content.  So  Sony's getting $10 CPM.  If  I assume each unique is good for 25 impressions per week, this translates into $1/unique, but that's essentially cost-free (no programming, no carriage fees, no amortization of acquisition costs) if you ignore the cost of the Internet Video Link network infrastructure.  So, Sony's getting the same $/unique.

Now for the sound of the dam bursting:  if I'm an advertiser and I can pay an ad network $3/unique (split three ways between publisher, ad network, and Sony) to reach my audience of women watching TV via, say CondeNet content, vs. paying NBC $7/unique, and get better targeting and measurability (remember, this is *Internet* video), which do I prefer?

Seems to me Sony's got the better growth/ scalability story, though reach is more limited in the short term.

I'm not criticizing the Oxygen deal;  there are sure to be cross-promotional opportunities with iVillage that add to Oxygen's value to NBC (if a small fraction of iVillage's 15M uniques go to Oxygen, that's a huge leg up).  And of course, there are near-term limitations of Sony's Internet Video Link value proposition to advertisers: low reach (the number of people with Internet-ready HDTV sets), and low availability of content from its partners.   Plus, there are a number of apples-and-oranges problems with my analysis (like, are these ways of reaching viewers really the same?  Do they ultimately reach the same viewers?  Are viewers in the same frame of mind?). 

The broader point, though, is about the significant arbitrage that is appearing for advertisers between reaching TV viewers the old-fashioned way and the new-fangled way, and about the dramatic shifts in fortunes that will occur over the next couple of years as dollars flow accordingly.  It's a good time to be long high-quality video content, and to be aggressive about getting online with it.

LinkedIn: Facing Reality

Many of us have been wondering for a while when LinkedIn would get around to opening up its API so others could take advantage of its networks of registered users to build applications  that could be spread virally through those networks.  Facebook has stolen a march on LinkedIn, first by allowing anyone to create a group, and most recently by exposing an API to allow others to develop applications that use its registered user base and networks -- or what they call the "social graph". 

Facebook applications have taken off like wildfire, and with this initiative Facebook has raised the "platform ante" beyond where Google, Amazon, and Yahoo had it (i.e., rich APIs and data sets to query through them, but limited networks of users) for anyone aspiring to build a large-scale web presence.  My favorite Facebook app, which I think best (most simply) demonstrates what Facebook has made possible, is the "Friend Wheel".  When I look at mine, I realize how many friends I still have to introduce to each other!  (Maybe what I need is a Facebook app that implements "graphic friendship" ideas...)

Now there's speculation about LinkedIn getting with the program.  Folks fret about whether the LinkedIn UI could handle the weight of a bunch of apps.  That's a red herring, IMHO.  Based on the relative rate of connection requests I've been getting in the last few days (4 or 5 to one in FB's favor), it would seem LinkedIn has no choice. 

But what's even more compelling about what Facebook's done, and why the imperative for LinkedIn is even more urgent,  is the economic opportunity it creates.  Everyone in the network can now make informed choices of apps to place on their profile pages, and a smart platform player will ultimately do a three-way rev share -- some for the member, some for the app developer, and some (ok most) for itself.  It will be interesting to see how soon Facebook gets around to this.

How Sad For Us All

Via my log files, this result when you Google "web application ideas"...

The Great Click Rush of 2007

Ad networks are hot, and in many cases highly-profitable businesses right now.  On the heels of Google's snapping up DoubleClick for $3 billion, Yahoo! bought (the 80% it didn't already own of) 4 year old  online display ad exchange Right Media last week for $680 million.  So, a logical question to ask is, who buys whom next?

Video Search Part II: ScanScout

A week ago I had lunch with Waikit Lau, president and co-founder of ScanScout (http://scanscout.com).  Waikit, whom I got to know when I was at ArsDigita and he was part of the http://photo.net team a few years ago, describes ScanScout's video analysis technology like this (this is the shareable version): 

Media as Software: A Conversation With Doug Turner

Kiki Mills at MITX introduced me recently to Doug Turner, whose past includes eight years as a member of the 3D graphics research team at Apple's Advanced Technology Group.  Doug and I met for breakfast and talked shop about digital media.  One of Doug's ideas, which I found particularly interesting, is (his words) the concept of "media as software".  Right now rich media streams are largely analog audio and video once they are "published".  (If you've composed or edited a digital video "project" and then converted it into its final form, you know what I mean.)  Doug describes this  as publishing digital media as platforms on which other people can add/edit their own stuff. 
 

Syndicate This

In the post right before this one, you will see a widget for this blog I created using Goowy Media's yourminis service.  You can help make me famous by embedding it in your blog.  Or, you can install it on your desktop (you need to have Adobe's Apollo player first -- that's their cross-platform desktop equivalent of the Flash player for web browsers) , and wait patiently for it to alert you of my latest big idea.  Or, you can ignore my widget but check out many more useful ones people have built using Flash 8 and the yourminis API.

Kaboodle's "Help Me Choose": Another Clever Structured Collaboration Example

The social shopping service Kaboodle recently announced the availability of its "Help Me Choose" Widget.  Essentially, this allows Kaboodle users like me (my wife and I use it to maintain a shopping wish list for home and kids) to publish a poll on external properties like this blog, so friends can offer input into certain purchase decisions.   

Graphic Friendships, Part III: The OpenACS "Collaboration Graph"

Gustaf Neumann, who teaches computer science at Vienna University,  has been a very important contributor and innovator in the OpenACS/ .LRN community in the past several years.  Recently he  has authored a very powerful wiki module, built with the OpenACS toolkit.  More recently he's integrated some other tools to deploy a  "Collaboration Graph" page he's published on the xoWiki instance running at openacs.org.  The tool graphs co-authoring relationships for wiki pages among OpenACS community members.  Here is an example that plots co-authoring relationships for my friend Caroline Meeks, and here is Gustaf's introduction on the OpenACS forums.

Twying out Twitter

With encouragement from my friend Perry Hewitt, I've been fooling around with Twitter to get a better sense for what all the fuss is about.  Conclusion:  it's a service that provides an asynchronous chat room for group interaction via SMS, and this has its place and time.

Grokking Yahoo Answers 2.0

"Yahoo! Answers Adds Social Networking", writes Loren Baker at Search Engine Journal (nice demo video from the Yahoo! Answers team included).  With everyone trying to build an online community to grow these days, what to make of Yahoo's latest gambit?

Clouded Vision

My new colleague Steven Forth, who is CTO of eMonitor (the content technology arm of Monitor Group) referred me last night to Many Eyes (http://services.alphaworks.ibm.com/manyeyes/home), which is a social data visualization and interpretation service developed by the Collaborative User Experience (CUE) Research Group at IBM's Watson Research Center.   As the intersection of social software and content analysis is currently a high-priority professional interest, I decided to try it out. 

Among other visualization approaches to structured data sets, Many Eyes generates tag clouds from free text files.  Steven noted that in particular, the two-word view seems like a very powerful 80-20 cut at inferring predominant meaning in a body of text. 

I experimented by exporting the contents of this blog as a text file, progressively scrubbing useless Typepad artifact words and html tags that appear frequently (like "title", "breaks", "comments", and my name) out of the source file -- to do this I simply ran "edit/replace/'word', '[]'" in Windows Notepad  -- and then publishing the file on Many Eyes.  Here's the result (click on the image to manipulate the cloud on Many Eyes):

The two-word view does a pretty decent job of communicating the themes I write about, I think.  Unintended side benefit:  highlights recurring cliches and verbal tics I need to purge from my writing, like "drive higher" (argh).

This whole effort took about 30 minutes, from registration to pasting the syndication html into this post.  Two-thirds of that time was spent scrubbing the data iteratively.  This could have gone faster in one of two ways.  First, Many Eyes could provide a custom scrubbing interface where I could register multiple words to be eliminated or replaced from a text file.  Second, and better, they could allow users to share not only comments, but scrubbing filters that would be applicable to data sets coming from common sources with common problems, such as Typepad exports, or government information.

Beyond this, I can imagine a thematic matching capability -- "based on two-word 'keyphrase' frequencies, this data set seems to have lots in common with these other ones..."  Such a capability could be further enhanced by ex-post user rating,  so people could confirm whether, for any given algorithmically-suggested match, the result was actually good, a la "was this useful to you?"  This, like the "Graphic Friendships" idea I wrote about a while back, could help to make the web browsing experience more productive.

Nice job guys! 

"In Their Tribes", Or, "How Do You Handle 10,000 Tech Maniacs' Votes?"

Dell  got a lot of press recently for launching its new Digg-knockoff "ideastorm" site (http://dellideastorm.com) for receiving and prioritizing customer feedback on its products and plans.  Right around the same time, The Wall Street Journal pointed out that on Digg itself, there's an extreme power-law distribution among contributors:  one guy, "Stoner", accounts for 13% of posts that got voted onto the Digg home page recently, and only 30 of the 900,000 registered users account for a third of Digg home page posts.  The bias that such extreme concentration can create is immediately apparent on the ideastorm site:  Linux (or more-generally, open-source)-oriented suggestions account for 18 of the 30 highest-rated suggestions on the first two pages of the site's list.

Set up this way, where so many "precincts" vote in the same general pool, voting sites like ideastorm can quickly alienate folks that don't have time (or money) to organize support for their ideas. Erick Schonfeld's post, here on Rojo, suggests what sites like these need is a "trustworthy reputation system" where people can disclose more fully who they are and who, if anyone is paying them to place stories there.

I come to a different conclusion.  I believe voting sites, like other forms of "structured collaboration", are best managed in subgroups, defined by shared interests, and coordinated through active moderation by volunteer "editors" -- much as Wikipedia has evolved.  From direct experience in managing a social bookmarking application, I can foresee (in fact we've specified) several specific administrator-defined (subgroups, categories, recognition and networking opportunities for volunteers) and user-defined (e.g., tags, profiles) mechanisms to make this work. 

On ideastorm, which takes the trouble to tell you how many people have endorsed a particular person's (in this example, Gautam's) idea, it would be especially useful to publish a page that ranks submitters by the number of endorsements from unique others.  These listings might include links to these submitters' profiles (including addresses for blogs if they have them, or perhaps at least bios).  Dell could then draw its volunteer moderators via further screening of this pool, maybe offering a piece of equipment or some public recognition as a thank you for assiduous moderation. (Reddit provides a stats page that's sort of like this, though its "karma"-based rating -- explained here -- isn't particularly intuitive, to me at least.)

More generally, I think there's lots of room for "shades of gray" models between "fully-edited" and "fully-user-generated" content that would provide more transparency about how members of a group promote submissions.   Think in terms of "concentric circles of delegated moderation", where trusted insiders recruit trusted outsiders from a pool of folks who have demonstrated passionate, constructive participation.

The "Edutainment" Future Is Now

With the help of several friends, I've written a wiki page on OpenACS.org that explains the "what" and the "why" surrounding the recent announcement that OpenACS/.LRN is the first to support the IMS LD specification for designing open-ended, collaborative learning experiences online.  It may seem arcane, and the examples may not be much to look at today, but this is a very big deal if you think that online games, communities, and learning have any synergistic future at all.

Yet another reminder (among others) about why I continue to believe in this project and the community that contributes to it.

Beyond Peanut Butter

I recently posted a contrary perspective on Brad Garlinghouse's recent "Peanut Butter Memo",  which presaged and maybe influenced the recent reorganization at Yahoo.  My post suggested that placing lots of early-stage bets ("peanut butter") isn't a bad idea at all, as the Yahoo memo suggested.  Rather, it's a question of how you manage them -- "drowning kittens", as some graphically call it.

Scanning my prolific friend Bill Ives' blog recently, I came across this post on "Collective Intelligence Networks", and especially John Maloney's excellent comment on GE's internal "Imagination Markets".  This, on the same day that the NYT gave Intrade a huge plug (my colleagues and I have been tracking prediction markets for a while, as yet another really interesting form of "Structured Collaboration").

It occurred to me that running a prediction market internally, on the likelihood of success of early stage ventures in companies like Yahoo and Google among others, might be a really interesting complement to formal management reviews.  In such services, like in other media properties, the apps are frequently visible to highly-informed end users who can be excellent proxies for early-adopter, power-user consumers -- but verbal feedback from such users can often be unreliable.

The tools for doing this appear to be available, though we haven't installed and tried one yet.

Perhaps another idea illustrating a principle that if you want to  "organize for digital", you've got  to (with apologies to Nicholas Negroponte) "be digital".

Gotuit: Video Search, and its Implications

A couple of weeks ago I had breakfast with Patrick Donovan, and old colleague from when we worked together in the early 90's at a Lexington, MA consulting and software development firm called Symmetrix.  These days Patrick is VP of Product Development at Gotuit Media.  Gotuit provides technology that allows users to "deep-tag" a slice of a video they look at on sites like YouTube and Metacafe. 

Here's a slice I made of a four-minute video on YouTube about the BT around the world sailing race.  I wanted to highlight what I thought was one of the more "that's gotta hurt!" moments in a video otherwise full of them:

 

Here's the url for this slice on Gotuit: http://www.gotuit.com/player/index.html?c=SM_Entertainment&t=8503&s=59028

Gotuit doesn't actually rip and store video itself; rather, it's an interface through which you create a data layer (on Gotuit) that identifies and describes (with title, tags, and a free-form text field) slices of videos hosted elsewhere.  Gotuit provides a browser toolbar with buttons that make it easy to quickly deep-tag a slice of a video and then share it via embedding or a hyperlink in a blog or other web page, or by simply emailing it to one or more friends.

So what?  Search is already a killer app on the web.  Video is exploding on the web.  Ergo, search for video will be huge.  Since video (and audio) is consumed linearly, meaning you can't browse it the way you can browse text, simply tagging and otherwise describing a source file on YouTube or somewhere else is only partially helpful to getting you to what you're looking for and helping you consume it efficiently.  Being able to deep-tag slices becomes really useful particularly for form factors and contexts -- like mobile -- where efficient use of limited resources (time, bandwidth, and memory) are more important. 

It doesn't stop there.  Of course, if you can tag inside a video, and you can sell ads against tags, you can advertise inside a video, in a highly targeted way.  And if you can tag inside a video, you can also string together slices to create what we used to call "highlight reels", but now would call "tag-dimensional mashups" I guess, allowing "omni-directional re-purposing" of content.  Videos of NFL games could be sliced and mashed up to create "best touchdowns", "hardest hits", "ugliest players" series.  Again, particularly interesting for mobile applications and contexts.

While I'm not sure that Gotuit in its current business model incarnation (today's separate service vs. a licensed, embedded capability in major video sites) or UI expression has reached its final stage, this is a really interesting company, and Patrick is a very smart guy.  Check them out.

Yahoo Pipes: The Custom-Filtered Web

One of the biggest challenges for many of us these days is *filtering* the fire hose of information that the query-able or "subscribe-able" web (nice ring to it!) can provide.

Michael Yoon recently pointed me to Yahoo Pipes, a new service that allows non-programmers to create highly-customized RSS feeds, starting with a variety of source feeds and allowing a broad range of pre-determined and/ or user-defined intermediate processing.

I'm a big Miami Dolphins fan.  To keep my habit under control, I've decided I'm only interested for now in what their new coach Cam Cameron is going to do to fix the offense.  I built a pipe to filter several major Dolphins news sources for "Cameron" and then serve me the results in a pubdate-sorted RSS feed.  Took 5 minutes to do, and I started from scratch without cloning any prior pipes as models (a really nice feature, like what Ning provides for social media sites, here's an example we built).

What's even more interesting is to create mashups across different kinds of services.  One popular example is the NYTimes/ Flickr mashup that serves you an RSS feed of Flickr photos based on an analysis of the content (such as looking for the most frequent keywords) in the NYTimes feed.

A very interesting service to track, still a bit techy for the mainstream, so we'll probably see some Pipes-based intermediaries emerge, folks who will build a collection of specialized pipes for niche audiences.  As for monetization:  PPC keyword advertising in the feeds?

Here's a good post by Anil Dash on this new service.

Postscript: Perry beat me to it.

Recursive E-Commerce

Just fooling around, with one-click ordering Javascript popup.

 

Missed Connections

Last night I checked out TripConnect.com   In theory, a useful service: travel recommendations are a popular thing to share, as other folks like Tripadvisor have demonstrated.  Presumably, they would be even more popular to share within groups of folks you know and whose opinions you trust, which is TripConnect's premise.

Problem is, as far as I can tell, you have to manually enter / invite the network of folks you'd like to share trip reviews with.  No thanks.  Within the circle of folks whose names I'd have the patience to  put into the system manually, I'm better off calling or emailing them.

TripConnect's UI looked eerily like LinkedIn's (the usual Web-two-oh pastels and rounded buttons).  So I thought, "TripConnect is a useful focus for collaboration in search of pre-existing networks of folks to use it.  LinkedIn is a network of folks looking for useful things people could share.  Why don't they partner?"  I'm sure someone has thought of this of course. 

I'm more interested in the general case -- what other networks and other service ideas, existing or envisioned, should get together like this?  Where's the LinkedIn API so folks like TripConnect could build mutually beneficial services onto it?

Referrals to stuff like this appreciated!

Carmun.com: Social Search Goes Vertical

Jeffrey Rayport and I had breakfast last Friday with his former colleague Lori Cohen and Jonathan Edson, a former AOL business development executive who is founder and CEO of Carmun.com.  Carmun is (my words) "social citation search", principally for academics, but also for anyone trying to find (good) books or journal articles on a topic (like school kids and college students).

Carmun helps you find good materials on your subject by crawling the Library of Congress index, and parsing out citations from footnotes and bibliographies into a structured data format (is there an RSS extension that makes sense here?).  Search is "social" in that it relies on ratings by users to help filter results.  Next, Carmun allows you to create folders that contains the citations associated with your projects.  Once you've added all relevant citations to a project, you can download a complete, properly formatted bibliography to include in your scholarly work.  Beyond an individual's project folders, Carmun also supports setting up groups within which people can collaborate around projects with conventional tools like a calendar and a blog, to which members can subscribe via email alerts.

I like this service, which is currently in beta.  It seems to me like a really good example of an application that follows structured collaboration principles: citations are valuable things to share, especially within properly-defined project-oriented groups.  My only suggestion is that to improve usability, it might reorder the major boxes on the home page so that research is leftmost and community is rightmost.  This would help folks like me get the natural flow of using the service a lot better.  Past this point, I found it all pretty intuitive.

Good luck Lori and Jonathan!  (And thanks to J.B. Lyon for introducing me originally.)

A Web 2.0 Travel Service That Gets Web 2.0

My friend Perry Hewitt, who runs the online marketing consulting firm Colechurch Consulting (so named because it helps bridge marketing-IT divides for its clients), pointed me to FareCompare this weekend.  FareCompare, an airfare comparison engine, uniquely albeit gratuitously  does the Google maps mashup thing with airfares.  Gratuitous, since the map doesn't add much to your processing of the information provided -- most people know where the cities listed are.  But what I like are the multiple approaches to syndicating its service:  Firefox plug-ins, Google and Yahoo desktop widgets.  (I checked Expedia to see if they offer the same thing -- they do, but only one, and it's the buggy, out-of-date version of what FareCompare distributes.)

This syndication thing is going to be big one of these days.

(BTW, Perry, love your site's new design.)

The Case For CPM (And Other Digital Media Learnings)

From a July survey of 6,000 people by DoubleClick, via November 20th's NYT:

About a third of consumers sometimes click on banner advertisements on the Web. But twice as many consumers sometimes respond to such ads indirectly, avoiding clicking on them but later visiting the Web sites advertised...

DoubleClick makes its living serving ads, and obviously has an interest in making the case for all those ads that people don't click on.  Nevertheless, it's useful to have some numbers here on the relative impact of CPM-based online advertising.

Here's the study:

http://www.doubleclick.com/us/knowledge_central/documents/RESEARCH/dc_touchpointsIV_0611.pdf

Surprisingly, the Times omitted mention of the word "sometimes" that DoubleClick had used in the study to describe the behaviors above.  So, as I read these results, folks are twice as likely to sometimes visit an advertiser's site after seeing its web ad as they are to sometimes click on the ad itself.

The rest of the study has some very interesting findings about the relative importance of different media channels to different stages of purchase decision processes, across different industries.  It's worth a look.

Adify: A First Step Toward A Multi-Tier Online Ad Network Ecosystem

Yesterday I posted ideas about the likely emergence of multi-tier publisher aggregation mechanisms for the online ad market.  Things move fast.  Today, via TechCrunch, I learned about Adify, one of 13 firms selected from over 200 to present at today's Web 2.0 conference in SF. Adify allows publishers to set up their own virtual ad networks to market ad space they have to sell on their web properties.  Translated:  it's sort of a specialized eBay or Amazon storefront for online ad space.

So, now advertisers can search and browse these storefronts to buy space at retail in highly targeted ways.  But we can also predict that new intermediaries will emerge to package and broker value-added bundles to advertisers who lack the time or the sophistication to do the shopping themselves.  These intermediaries could be new services provided by the traditional ad networks, or they could be new players entirely. 

More broadly, what we're seeing is the predictable exploitation of the big arbitrage opportunity in online advertising.  Rough numbers:  a $20 billion online ad market, with (conservatively) a 15% margin*, featuring (my guess) a 50% arbitrage opportunity =  a >$1 billion market available to the quick and the clever.

(*Prior to its recent stumble, Aquantive's March 10Q reports that, excluding consulting by Razorfish and DNA,  revenues and pre-tax profits were $37 million and $15 million, respectively.  After a share of corporate G&A and taxes, I make that a 15% margin more or less.  Of course one company's profit margin is only a vague, flawed proxy for the industry profit pool, but it gives you a feel for how lucrative this business currently is.  I'd guess Hellman & Friedman have done very nicely after taking DoubleClick private in 2005, what with lower overhead and tax-deductible interest on buyout debt.)

Related:  Madhens

De-Optimized

Follow-up to my post on MegaKarma: traffic to my blog doubled (rough estimate), and a couple of people voted my post up on Reddit.  But then I got voted back down into the dark hole I came from.  Oh, the humiliation.

Moral of this story:  Just as is true in the real world, there's no substitute for personal networking.  Send your posts to others, ask them to comment constructively on their blogs and link to you if appropriate.  Services like Megakarma might be helpful, but only on the margin.  Anyone have a different experience?

MegaKarma: Social Media Optimization

Via an ad on Adverblog, I learned of MegaKarma.net, a service that helps you game submissions to social bookmarking / voting services like Digg and Reddit.  The basic idea is "you scratch my back, I scratch yours":  once you've submitted your link to the social bookmarking services, you submit it to MegaKarma, which then emails it to a list of other members of these services who vote it up (if they like it) after it's posted.  I'm just trying this service now (with this post!), so I can't say yet whether or not it seems to be effective.  But it does seem like a way around the "credibility power law" effect that emerges in systems like these, where "power posters" have disproportionate clout in the algorithms the services (properly) use for rankings.  Just as we had PR firms to spread the word and build buzz in Old Media, we now have services like these for the New.  Technology changes, but socioeconomic dynamics remain the same.

Postscript:  Mike-O-Matic's take

The Medium Is Not The Message

PHP may be the new Powerpoint, but my friend Ben Kline, a senior ad agency exec, says he's seen too many situations recently where people confuse having a [pick your new media channel:  blogs, podcasts, video, "community"] for having good content on them.  So what makes good content for these new chann