(this is page 107 of 121)





Latest

Shovel it

I've always thought that Bob Metcalfe, who I've known and considered a mentor for the past six years, would make a fantastic addition to the blogosphere. He used to write a column for InfoWorld which was an absolute gas and often very insightful as well. However no matter how many times I pitched him on the notion of his starting a blog, his "old economy" notions of "getting paid" for work always got in the way.

Which is why it was such a pleasure to see him making his debut with a guest post on Mike Hirshland's blog to weigh in on the hoopla surrounding whether Metcalfe's law has overstated the value of a network in its formulation. Needless to say, it is too bad that his inaugural post had to be on such a fundamentally lame issue (but read on to see how he still pulls out a great post). I don't know what is more ridiculous: the fact that this debate was kicked off from an absurd article in IEEE Spectrum (a supposedly respectable magazine) or the fact that it has gotten such attention from the blogosphere. Bloggers everywhere have jumped on both sides of the debate and even some of Metcalfe's co-workers have piped in to defend "The Law" despite the fact that... well let's just say it is no F=MxA.

The most impressive thing about Metcalfe's law is that George Gilder (who coined it) and Metcalfe himself (who promoted it) were able to embed it so deeply in the hearts and minds of analysts, MBAs, and engineers everywhere. At its core, all that Metcalfe's law states is that when the benefit of something grows non-linearly while its cost grows linearly, good things result. That is basically it (and you can see it from the original 35mm slide Bob presents). Economists have a name for this effect– it's called economies of scale, and while networking computers comes with the hard problem of understanding how to quantify the benefits, the underlying math is really straight forward. So let me have a go at it:

Because it's summer, I've spent some time on the beach and noticed that every drugstore within 50 miles has got these new space-age plastic shovels that are easier for little hands to use. The normal sized one (pictured here) is maybe 3 inches by 4 inches (12 square inches of shovel area) and costs $3.99. Recently I noticed that there is a bigger one that is maybe 2 inches bigger on each side but costs only $4.99. Can we call the added power of the shovel relative to the cost "Rodriguez's Law?" Now networked devices, unlike shovels, benefit from each other, but if we look at the entire network as the object that is benefiting from economies of scale, we're talking about pretty much the same thing.

Bob and George were geniuses not for their insight nor for their quantification of the observation, but for their uncanny ability to capture the zeitgeist of the era in it (explosive PC growth and an insatiable appetite to believe in the power that all of these networked computers were going to bring us). This, and the fact that Metcalfe is a genius salesman (just watch how he hawks his 5 year old book in his inaugural post), is why Metcalfe's law is up there with Moore's law and not in the dustbin of other such "laws" where my Rodriguez law would most likely end up.

Because he is a bright guy with a biting sense of humor, Bob gets right to this in his post and manages to lampoon everything from Al "Mr. Internet" Gore, to the hundreds of bloggers who tracked back without realizing that there is a lot more humor his defense than serious meaty analysis. The title of the post, "Metcalfe’s Law Recurses Down the Long Tail of Social Networks," pretty much says it all right there but you can read into to post for even more fun.

(Incidentally, I am sure that those IEEE guys are distant relatives of the dudes who spent their last few minutes rearranging the deck chairs on the Titanic to make access to the bar easier)

If you really want to understand something more valuable about the nature of social networks, there are a couple of wonderful books that treat the subject fairly rigorously (the value of social networks). In the meantime, here's to the possibility that Bob may enter the blogosphere on a more permanent basis!

Comments Permalink

Sometimes a blogosphere meme is so good and on-point with what you've been thinking that it's hard to do anything other than agree vehemently with it. Such is what is going on today with the volley between Caterina and Fred about "Business Development 2.0." Basically, they argue that we're past the point of needing the big biz dev dances that we all did during web 1.0 with suits at long meetings and contracts and meetings and sushi bars and meetings, etc., etc.

Around the office I've been calling it "lightweight business development" because it feels much more agile than any of the other stuff I've ever done. Fred outlines a bunch of good examples on his blog so instead I'll try to give the geek's perspective: doing biz dev between unprofitable sites, or in the context of unexplored types of relationships (say Tabblo and a major photo retailer for example) the old way is the business people's equivalent of premature optimization and is therefore almost likely to lead to a bad outcome.

Now there are cases when partnerships need more than the lightweight approach– usually because there are CAPEX or marketing resources that need to be applied. But, as I was telling a prospective printing partner today, if he's going to have to make a significant investment and venture into unexplored waters because of us, it's probably not the best fit at this point in time. Sometimes it may be (though I am hard-pressed to think of a good example), but more often than not, this probably leads up to a situation where escalation of commitment becomes an obfuscating factor in figuring out that it probably wasn't meant to be in the first place.

Comments Permalink

Personal homepage redux

People have been telling me to try Netvibes for a while now and I've been reticent to do so because it just seemed so 1998 all over again to think of the personal home page as being the new black. Then I read this morning on TechCrunch that this little Parisian outfit had raised a $15MM round from real VCs and I decided that it was time to go and check it out for real.

Note that I went there full of skepticism for what the company could possibly do with AJAX that would make it worth cracking open a category that was so tired 8 years ago already. And the answer is a lot, not because of the AJAX per se (though this is the way we were meant to configure any sort of webpage) but because of the richness that comes from their module-based approach to developing content widgets. Want Flickr pictures? No problem, just stick in the RSS feed and you are off to the cool races. For that matter, anything that emits RSS is just a click away from being a permanent fixture in one of an infinite set of pages that you can create as are a whole host of other non-obvious data sources (Meebo for IM for example). The whole experience is done so well that before you realize it, you've sunk 15 minutes into setting up a few pages without having every registered or gone through any kind of cumbersome signup process. It is really slick.

That said I'm not sure that there is a real business here. Most of the content does not belong to Netvibes and most of the interesting stuff is likely to take place elsewhere (after you click away to gmail or flickr). However, I have to give it to these guys for putting together a really cool framework for building these little widgets and encouraging what I think is already a fairly rich third-party ecosystem of widgets.

But that is not all; I realized after playing with the site for a while what is the one meaningful thing that has changed since the personal home pages of 1998. Content now comes in micro-chunks and most smart sites can emit it through RSS in a way that makes it both easy and powerful to repurpose for a Netvibes-like experience. And Javascript lets us glue it both on the consumption side, and perhaps more importantly, on the collection/authoring side in a way that creates a much more compelling user experience.

I guess some seemingly bad ideas are just good ideas waiting for the right pieces to fall into place.

Comments Permalink

I didn't get a chance to read about Apple's World Wide Developers Conference as this was a crazy week but lots of people did nice write-ups of the whole event, sounding mostly disappointed that Apple didn't introduce an iPhone (which by the way is sorely needed as I've discovered that the Treo 700p, "smartest of the smartphones," is very much at the Windows 3.11 level though that is a different story). My favorite part of WWWDC had little to do with snazzy hardware though, and was instead around a feature of the OS called "Time Machine" that let's you restore your entire hard disk to some previous timestamp.

This is a really good idea. In fact, it's so good that over a decade ago David Gelernter and a bunch of Yale computer scientists came up with a whole system for managing information on a computer called "Lifestreams" that basically worked as a time-indexed set of events (emails, IMs, files written, files edited, files deleted). Lifestreams was cool because it argued for something other than the stamdard WIMP desktop metaphor, and because it seemed to come at a time when the Internet was giving computing a whole new vector which most regular users had not anticipated (email? IM? web search history?)

But Lifestreama was a commercial failure and a meme failure because it was just too different. The brilliant thing that Apple has done with Time Machine is to fit the whole paradigm into the standard desktop metaphor: you go backwards and forwards on a timeline in order to restore the state of the entire desktop at a certain moment in time– with a snazzy interface. This is something that regular non-propellerheads can get their heads around.

Another way to look at this may be to think of it as source control for the masses. Imagine that every app you ran on your desktop was wired through Subversion (in fact I think in some O'Reilly Hacks book this was exactly what is recommended). Then with a simple svn command you could easily get back to any previous state in your workspace. The challenge would be to make sure things like app-specific metadata got into the svn repository– but here again, the Mac shines since it does such a good job of avoiding the PST hell that most Windows apps have (one mega file with a proprietary binary format that makes it impossible to diff). In the meanwhile, I've got to give it to Apple for bringing good (if old) ideas to the masses in a nice package.

Comments Permalink

Find me some kewl blogs!

While rich feedreaders like NetNewswire (and their fast-following web-based cousins like Bloglines and Google Reader) made reading lots of blogs possible, it still amazes me that no one has been able to make any kind of significant improvement in the discoverability of cool new content. To this day, with one notable exception (more on that below), the best way to find new blogs still remains the blogroll and direct link from respected source. From sites crafted around aggregating attention metadata (like Rojo) to user-generated directories, I've tried at least one of everything and left it for my homegrown list of 200-odd feeds after just days of experimentation.

I was excited for the big Technorati redesign mainly because so many bloggers swear by it, but after just a few days of looking at this page, I find myself dizzy and confused about which of these highlighted (and often duplicated) posts, I might want to click on to follow a particular thread in the blogosphere.

This weekend I came across Brad Feld and Fred Wilson's (two ex-abuzz investors) excited posts about FeedBurner's new content and advertising networks which are really aggregated feeds along topics with a "manager" deciding who is in and who is out (for example, they are both part of the VC network). I love Feedburner as a company, and am always surprised to see it left out of most industry conference discussions about "Web 2.0," particularly because of how much the company/concept/product seems to exemplify so many of the things that are exciting about Web 2.0, but I'm not sure I buy that Feedburner has got a winner in remixing feeds topically like this. Though they do build a "homepage" that shows some basic stats around who is in a network and what the posts are, you have only to look at the community section of any open source project page that aggregates bloggers (i.e., Rails bloggers or Django bloggers) to see what a mess this can quickly become without the careful attention of an editor. Basically, feed remixing is cool, but we need much finer and richer editorial control (sometimes at the sub-post level) if we're going to get a tool useful enough to keep using beyond the initial gee-wiz phase.

Now there is one tool which has become a part of my daily information consumption diet: Tech Memeorandum or Techmeme as it has now been rebranded. Essentially, Techmeme tries to be a front page for blogosphere using nothing but its algorithm for editorial control (think what Google News should be to news but for blogs). Techmeme is not without its flaws– the greatest one being that a lot of the most interesting parts of the discussion as often left out of the Techmeme rollup and has to be ferreted out of the blog posts that are mentioned, but overall I have been amazed at how often I've come back to Techmeme to try to get a "snapshot" view of what is going on, especially when I find myself backed up and looking at 8,000 unread posts in NetNewswire.

Can it be that this problem of blog discoverability is actually hard to crack? I don't believe it– but what I do think is that it is very hard to monetize, and as such, some of the people I think would be most likely to crack it aren't even thinking about it. Google AdWords are not enough, and it is not clear that a really amped up Techmeme can survive as a destination property on its own. In fact, the closest model I can think would work for something like this would be the Topix model where the core technology development is subsidized/owned by big media companies that are hoping to make it part of their overall offering.

In the meanwhile, I will keep drowning in posts while I try to poke my head about water for long enough to find the cool new stuff...

Comments Permalink