Topicle: Is That All There Is?

You can build your own vertical search engine using Topicle. Topicle is a platform for vertical search engines, rather as Ning is a platform for social networks and BricaBox is a platform for social content.

That’s about as exciting as I can make Topicle sound. It’s less of a platform than a veneer on top of the Google Custom Search platform. I set up a Custom Search engine about a year and a half ago. There’s a link to said engine in the sidebar of this blog, since it provides Directions Along the Changing Ways: it enables search of, not just this blog, but the whole happy Changing Way family of sites.

I did create a search engine for Boston at Topicle. I remain underwhelmed. It’s essentially a list of site URIs. Note that you (yes you) can edit the list.

The most interesting part of the exercise came when I tried to make the engine understand the Boston-ness of this blog. Posts I put in the Boston category should be searched. Posts I put in other categories (e.g., WordPress) should not be searched by the Boston engine.

So I included the Boston category by specifying the URI Although Topicle/Google accepted the URI, it doesn’t appear to do the corresponding search. I got similar results, or lack of results, when I included the feed for the Boston category.

I know that eclectic blogs like this one present problems for vertical search, but I think it’s a problem that vertical search engines need to solve. Lest you think that this blog is alone in its eclecticism, I give you the example of Fred Wilson.

There’s more enthusiastic coverage of Topicle at RWW and at Mashable. It notes that “former Google Product Manager Steffen Mueller” is behind Topicle.

Google Sites

So I set up a site on PTO (Putting Things Off), mainly so I could try out Google Sites. As Glyn points out, Sites:

  • is to some extent a repackaging of JotSpot, the wiki outift that Google acquired over a year ago.
  • isn’t packaged as a wiki, but as “a kind of super-duper easy-peasy Web site creation tool that even idiots like us can use.”
  • thus is rather similar to Google Pages.

In fact, an idiot like me could find himself using Google Pages when setting up his site, even though he meant to use Google Sites. It’s not the first time that I have found “easy-peasy” software rather confusing, and I suspect it won’t be the last., Google, Yahoo Making Music Together

For those of us hosted at, there are multiple ways to include music in a post. the simplest is to point the WordPress audio player at an MP3 file.

This raises the question of where to stash the MP3 files. In a recent support forum thread, DZonson suggested the use of Google Pages. You use Google Page Creator to set up a site, upload the MP3s, put them on a page, publish the page, and you can then use the WordPress MP3 player at the MP3s.

You can take one more step in order to make such GPC pages available as playlists. You can edit each page’s html to add a line of javascript invoking the Yahoo Media Player. I like this simple, lightweight player, and I like the way it turns a storage bin for MP3s into a page that can be made interesting in its own right.

I should note that GPC is part of Google labs, which is a place for projects “that aren’t quite ready for prime time.” GPC imposes space limits, currently 10MB on a file and 100MB on a site (but you can have multiple sites).

I should also note that Yahoo Media Player can’t be used directly at itself, since it is javascript. I wish it could be made available, in much the same way as services such as Sonific are available.

BigCo Banter

Just a few comments on goings-on at the web BigCos in the last week.

OpenID: BigCos on Board

This morning the OpenID Foundation announced that Google, IBM, Microsoft, VeriSign, and Yahoo! have joined the board. This is good news, since OpenID is good.

However, there are limits to the goodness of the news. As Michael Arrington points out:

OpenID looks like it’s going to be a winner, so big companies making their user accounts OpenID compatible is a good hedge. Everyone, of course, wants to be an ID issuer, since they get to “own” the user. Less attractive is allowing users from other sites to log into your services, so don’t expect that functionality to come for some time.

Google Online Marketing Challenge

or, Google Goes to School: Hey Kids, Want Some Free AdWords? That does sound, as Marshall Kirkpatrick puts it, “a little bit creepy.”

But the people in question are university students taking marketing classes. They are not, for example, high school “kids.”

How does the program work? Student teams will receive US$200 of free online advertising with Google AdWords and then work with local businesses to devise effective online marketing campaigns.

I don’t have a problem with this, but then, I teach in a business school (although I don’t teach marketing). I hope that the profs involved will challenge their students to recognize that the Google Online Marketing Challenge is itself a marketing program – for AdWords.

Utility of Data and Openness of Code

Consider the following interesting assertion.

Data has this really weird quality. In economic terms data has an increasing marginal utility… Each incremental point of data adds value to the ones you all ready have.

Those are the words of Brad Burnham. I saw Brad’s post linked to by Fred Wilson, Brad’s partner at Union Square ventures. Brad’s post is mainly about Google; his perspective is that of a VC who would like to see Google’s dominance disrupted, preferably by USV-backed firms.

Matt Asay picked up on Brad’s point about marginal utility of data (MUD?), and tied it to his own thinking about abundance and free/open source software. Thinking about Google, MUD, open source, and abundance didn’t make me think that Google’s dominance is due to end any time soon.

Google takes some of the abundance of open source software and uses it to power web services that are mostly free to the user. These services benefit from, and add to, Google’s private abundance of data. Google cannot realistically be called upon to open up this abundance, since it is data from and about users. The implicit contract between Google and the user is: we’ll give you free web services, you give us your data, and we won’t do anything evil with it or to you.

So Google’s private abundance of data is part of its positive feedback loop. At first glance, that doesn’t bode well for Brad’s wish for his firms, or for Matt’s wishes for open stuff. But I do wish each of them well in 2008.

Google Profiles

Yesterday, I realized that I had something called a Google Profile, and that it was linked to my Google Reader account. Today, several of the blogs to which I subscribe using Reader included posts about Google Profiles. One of these, GigaOm, directed me to its colleague Web Worker Daily.

The post at WWD, by Mike Gunderloy, is pretty good, raising some interesting questions. A Google Profile “is simply how you represent yourself on Google Products” – but what are the future plans?

I’m not sure why I’d want a profile that was limited to Google products. I do want a profile that spans products, but I also want a profile that spans vendors.

I’d like to know how Google Profiles will be related to standards such as OpenID. Without a good story on that, Google Profile will begin to look like Microsoft Passport.

Prison to Hotel: In Telecoms and in Boston

If Google’s bid for 700Mhz spectrum materializes in January, it will bring the trillion-dollar infotech and telecom industries into direct competition for the first time in 50 years. That’s how Daniel Berninger starts a rather gracefully-written post on infocom vs telecom. I particularly liked this passage.

A survey of companies will reveal a range of approaches, but the prevalence of competition leaves telecom and infotech populating opposite ends of a spectrum analogous to that of prisons and hotels. Hotels and prisons offer similar functionality, but the range of available choices produces opposing experiences for their respective inhabitants.

The typical telco pursues a lock-in strategy in which the resources consumed to keep customers captive exceed those applied to refine service offers. The absence of lock-in options, on the other hand, makes innovation imperative for the survival of infotech companies.

We have right here in Boston a prison turned into a hotel. According to Robert Campbell, the Globe’s architecture critic, the Liberty Hotel doesn’t live up to hopes.

Email is Social!

Stop the presses at the NY Times! Or at least read the rather good post by Saul Hansell at the NYT’s BITS blog.

Google and Yahoo have come up with new and very similar plans to respond to the challenge from MySpace and Facebook: They hope to turn their e-mail systems and personalized home page services (iGoogle and MyYahoo) into social networks.

Web-based e-mail systems already contain much of what Facebook calls the social graph — the connections between people.

Let’s talk about Yahoo first. Fred Wilson estimates that Yahoo Mail, with 250 million users, is the largest social graph on the planet. But Yahoo’s plans to use this graph make Michael Arrington sad.

Yahoo’s Brad Garlinghouse is talking about creating yet-another-social-network around Yahoo mail… He says the project is called “Inbox 2.0″ internally… It makes me sad because it is absurd for Yahoo to keep launching new social networking products, almost monthly, without what appears to be any sort of high level strategic vision…

I mean, I follow these products for a living, and I can’t keep their strategies straight. Or even figure out if there is a strategy. If Inbox 2.0 is part of Yahoo’s big vision for the future, then tell us more than the bits about the news feed and profile pages. Tell us how it can change the entire company, as OpenSocial appears poised to do with Google.

I have to agree with Michael that a more coherent message is coming from Google. Recognizing Gmail as a social graph fits very well into OpenSocial. For much the same reason, I disagree with Saul when he opens his post with the advice that we ignore OpenSocial.

Other remarks from “usual suspect” blogs include: it’s about time that Yahoo and Google unlocked the social potential of their email user bases; what’s really at issue here are two concepts that Hansell… didn’t name explicitly… RSS and Attention Data; it would make more sense to focus on start pages than on email.

Then there are many good contributions to the conversation in the comments to Saul’s post, and to the other posts referenced above… But this post has gone on long enough already.