Month: July 2008

Not long ago, Podtech was a video company with a bright future — or at least so it appeared. Run by John Furrier, with some high-profile social media types like Jeremiah Owyang and Robert Scoble on board, the company had high hopes of being a new video-content provider. And then the train left the track at some point, and now the company’s assets (whatever’s left) have been sold for just $500,000. Owyang left to join Forrester and Scoble left to go to FastCompany, and John Furrier was effectively forced to resign. So what happened? How did $7.5-million worth of VC money get vaporized so quickly?

We may never know the complete answer to those questions, until someone like John chooses to talk about it (Update: He has posted a comment on FriendFeed), but we can draw some tentative conclusions from what Robert Scoble has said on FriendFeed. Among other things, he says:

“Podtech was screwed up by a number of decisions. Everyone played a part, but I sure learned a lot about how a company can screw up big time. Major learnings for me? 1. Have a story. 2. Have everyone on board with that story. 3. If anyone goes off of that story, make sure they get on board immediately or fire them. PodTech did none of the three and I’m sorry for my part in not making the three happen.”

And then, apparently unable to resist adding more details, he says:

Continue reading “Podtech failure: Scoble’s lessons”

Apparently, congratulations are due for Henry “I used to be a famous Wall Street analyst” Blodget and the rest of the team over at Silicon Alley Insider: the site has apparently raised about $1-million or so in venture capital. PaidContent has the news — although Rafat and the gang at PC (who just got their own windfall of $30-million or so from Guardian Media Group) seem a little miffed that their scoop was broken by SIA itself, after Henry blogged the funding news. Alley Insider is part of a Web 2.0 holding company of sorts called Alley Corp., which was created by former Doubleclick execs Kevin Ryan and Dwight Merriman, and recently launched two new properties: Clusterstock.com and The Business Sheet.

I can’t remember where I came across this. For some reason I think it was on FriendFeed, but now I can’t recall who posted it (if it was you, let me know and I will give you a shout out — Update: It was FriendFeed intern Dan Hsaio, who got it from Philipp Lenssen). In any case, it is right off the geek-o-meter, but I felt compelled to take note of it anyway, in part because it illustrates just how far things have come in a relatively short space of time — just 15 years or so. Now, we’re worried about how things will render on our iPhones with their multi-touch interface, or how many video clips we can pack into the gigabytes of memory we have, or if the Ajax and Flash on a website is annoying or useful.

In 1993, Marc Andreessen — who was then working on a little piece of software called Mosaic while working at the NCSA (National Center for Supercomputing Applications) at the University of Illinois in Urbana-Champaign — suggested a new tag that could be used in HTML: the IMG tag, which could be used to (wow!) display an image on a page. Bonus points if you follow the thread: Tim Berners-Lee, who has since been knighted by the Queen for his contributions to technology by effectively inventing what we now know as the World Wide Web, doesn’t seem to like Andreessen’s suggestion much. He says he’d much rather use the existing HREF tag and put image attributes in it.

My friend Om Malik has an interesting post about Yahoo — and not about the interminable Yahoo vs. Microsoft-plus-Carl-Icahn takeover, which has become the beast that refuses to die, but about the kind of thing Yahoo should arguably have been focusing on instead of trying to compete with Google on search. As he describes it, one of the blog posts at Web Worker Daily got a spot on Yahoo’s home page, and then got voted up by users of Yahoo Buzz, a Digg-style feature. As Om says:

In a few hours, the story … was viewed over 200,000 times and attracted over 350 comments. Now that’s a lot of traffic — but more importantly, a gigantic amount of engagement displayed by Yahoo visitors. The traffic sent our way by Yahoo was many times the traffic we get from, say, Digg or StumbleUpon.

As Om notes, it’s not so much the sheer volume of traffic that is impressive, but the engagement of the audience. Even during the biggest Digg storm or Stumbleupon flood I’ve ever experienced, I’ve never gotten more than a handful of comments. As beaten-down as it is, Yahoo still gets a ton of traffic — but Om is right that it needs to find better ways of taking advantage of and leveraging that traffic, instead of just trying to go head-to-head with Google. Should it sell off its search arm to Microsoft? Perhaps. At least then it could concentrate on what makes it different from Google, instead of trying to duplicate it.

One of the interesting things to me about Mike Arrington’s interview with Twitter founder Evan Williams isn’t so much the discussion of business models (although that’s obviously something the company will have to deal with eventually) but the debate that seems to be going on inside the company about how it handles API access to Twitter’s data. As Mike notes, the service only gives four outside companies full access to the entire Twitter data feed, and one of those is Summize — which of course is now part of Twitter. The others are FriendFeed, Twittervision (which overlays Twitter posts on a map) and Zappos, which is an online shoe retailer that makes extensive use of Twitter.

The ability to get access to that “firehose” of XMPP data was what allowed Summize to produce @ replies from Twitter when users couldn’t get access to those replies through Twitter itself, one of the features that I think drove interest in Summize over the past few months — although it would arguably still have been a useful service even without Twitter’s repeated downtime issues. But while Summize used it to build something that made sense as an adjunct to Twitter, a service like FriendFeed.com is using the data firehose to build something that is closer to being a very real competitor. Like some other people, when Twitter was having its issues, I effectively duplicated the most important part of my Twitter follow list by following or creating friends in FriendFeed.

Continue reading “The Twitter API: Giving away the store?”

Full-text search of video has been the Holy Grail of search — or at least one of them — for years now, pursued by startups like Blinkx and others. The vast majority of video searching involves the text that surrounds a video (the description, tags, keywords, etc.) but that only scratches the surface of what is possible. Now Google has given us a glimpse of what full-text search might look like, although for now it is restricted to a handful of political clips and is only available as a gadget for embedding on your Google home page. When you search for a word, the video progress bar is marked with yellow strips where the word appears, and hovering over the strip shows you the word in context. Will this be added to all of YouTube someday? We can only hope.

http://www.gmodules.com/ig/ifr?url=http://www.google.com/ig/modules/elections_video_search.xml&synd=open&w=336&h=440&title=Elections+Video+Search&border=%23ffffff%7C3px%2C1px+solid+%23999999&output=js

Netflix and Microsoft announced a deal earlier today that allows Xbox 360 owners to download movies through their game consoles, which appears to be another step in the software company’s Trojan Horse approach to winning the living room/media-server wars. Will this give it an edge over Apple TV? MG Siegler seems to think so — he’s dropped his earlier championing of Roku’s media server box and is now calling Microsoft the winner. I’m not so sure just yet, however. Apple TV has its flaws, but so does the Xbox, as more than one person has pointed out (Chris Albrecht of NewTeeVee still likes the Roku server, and so does Webomatica).

I’ve been looking for a media-server solution for some time now. I’ve used a repurposed desktop PC that I hooked up to the TV, and I bought a MediaGate audio/video server (essentially an external hard drive with audio and video output jacks) which unfortunately isn’t working at the moment, and I’ve thought about buying an Apple TV or an Xbox or a Mac Mini — the latter being what my friend Rob Hyndman uses as a media player for all his movies, music, photos and so on.

Continue reading “Is the Xbox the living-room winner?”

Umair Haque, the new-media theorist who writes the Bubblegeneration blog and recently launched the Havas Media Lab, has a great take on the phenomenon of “user-generated content” and how it hasn’t really measured up to what many proponents — both outside the traditional media and inside it — were hoping for and/or expecting. In the first research paper released by the media lab (available in PDF only at this point), Umair argues that this is because user-generated content is mostly a misnomer. What users generate is *context,* he says.

As many others have, Umair notes that the Techmeme leaderboard is composed primarily of mainstream media sources — but says this makes perfect sense when you think about user-generated context instead of user-generated content. Most of those commenting on media stories or blog posts aren’t generating content per se, but are taking content that is created by others (whether the traditional media or new media outlets) and contextualizing it in some way that is meaningful for them and others who share their interests. As Umair says:

Continue reading “Umair: It’s user-generated “context””

Maybe Radiohead should change their name to something a little more futuristic — Laserhead, perhaps. For their new video “House of Cards,” the band decided to experiment with laser-scanning technology and make a video without using any cameras at all. Instead, they used 360-degree scanners that created what amount to 3D maps of the band and various actors in the video, as well as a much larger scanner that created a 3D image of an entire street. The parts where the data forming the image seems to disintegrate are very cool. Not crazy about the song though. Google has a whole page devoted to the creation of the video, and TechCrunch has some more info here.

[youtube https://www.youtube.com/watch?v=8nTFjVm9sTQ&hl=en&fs=1&w=425&h=344]