Danny Sullivan totally misses tagging

In a recent post, Danny Sullivan goes ballistic about social tagging (such as del.icio.us) and its usefulness for improving search engines:

…we’ve had tagging on the web for going on 10 years, and the experience on the search side is that it can’t be trusted.

The meta keywords tag has been around for nearly a decade. The idea behind it in part was that people could use the tag to classify what their pages are about, as well as provide copy that search engines could index if the relevant text for some reason couldn’t fit on the page itself.

…In addition, none — NONE! — of these search engines now or ever has made use of the tag in a way to let you perhaps see all the pages "tagged" to be on a particular subject. Why not? The data is largely useless.

Mark me extremely dubious that tagging will make major inroads in improving search. And if I’m wrong, I’ll happily mea cupla. But after 10 years of tagging, the experience so far gives me good reason to be dubious.

Danny misses the whole point. Meta-tags and ‘bookmark tags’[1] like del.icio.us are totally different beasts. The difference is not, as Danny points out, the fact that meta-tags are self-provided tags and del.icio.us tags are not – that hardly matters. The huge difference is the basic underlying value proposition that each of these tagging types offers:

Meta-tags were designed primarily for exposing certain keywords to a search engine as a means for improving ranking on those keywords. That is their sole purpose and sole value. So I, as a creator of meta-tags for my site, am not ‘awarded’ for creating good meta-tags, or useful or relevant ones. I am awarded only by getting improved search rankings and increased traffic on those keywords I choose. There’s no personal value gained by writing good meta-tags… on the contrary – if everyone else is doing keyword stuffing, I may drop in ranking if I write proper meta-tags.

This underlying design of meta-tags almost incentivizes misuse and is therefore fundamentally flawed. No wonder none of the search engines use a flawed system.

Bookmark tagging solutions like del.icio.us on the other hand provide a fundamentally different value proposition. As a del.icio.us tagger, I am organizing interesting pages for myself so that I can go back to them in the future and easily find them. I am doing it for pure egoistic reasons, and I have all the reasons in the world to do a good job of it. Sure – whatever tags I put on del.icio.us may have some grander value for someone else (meme trackers, search engines, etc). But frankly I don’t really give a shit about those other uses when I’m tagging something. Therefore the underlying nature of social tagging apps (or ‘bookmark tagging’) incentivizes proper tagging[2].

Of course there could be attempts to abuse bookmark tags like del.icio.us if a significant financial incentive gets tied to their tags (in the form of using tags for search engine rankings).  But:

  1. The more people use it for their own selfish reasons, the more difficult it is to abuse it. Think of this as sort of a Wikipedia… sure – some folks may try to abuse it once in a while and change entries, but the more people you have using it legitimately, the more difficult it is for abusers to misuse the system.
  2. Abuse attempts are 100x easier to identify on a bookmark tagging service than they are on meta-tags or links because the usage patterns of a non-abuser (big variety of URL’s, variety of tags, etc) are so different than an abusing pattern (stuffing of specific tags on specific sites). In the meta-tag world there is hardly any such differentiation, making it extremely difficult to sort the valid from the abusive.

Saying del.icio.us tags are necessarily crap because meta-tags are, is like saying all self-published sites (aka ‘blogs’) are crap just because self-published sites on Geocities were crap 10 years ago. What’s the connection? Only the word ‘tag’ I guess… not much else.

I applaud Yahoo for foreseeing this and making one of the smartest moves in search in recent years (and I believe this was primarily a search move).

[1] The proper name for del.icio.us class applications is social bookmarking. However, for the purpose of this discussion, I think that bookmark tagging is probably a better name for these apps. While they do have a significant social aspect, the social piece has very little to do with the motivation people have in tagging on del.icio.us. The primary motivation is much more of a ‘bookmark’ motivation – ‘how can I save something so that I can easily find it later?’.

[2] I’ve quoted Joshua Porter in the past, but this applies here as well:

From now on I’m going to call this idea the “Del.icio.us Lesson”. This is the lesson that personal value precedes network value: that selfish use comes before shared use. We’re seeing it more and more everyday in services like Del.icio.us, Flickr, and is an interesting aspect of networked applications. Even though we’re definitely benefitting from the value of networked software, we’re still not doing so unless the software is valuable to us on a personal level first.

Share this post!

Jim Lavoie on Jimi Hendrix

I recently posted about Rite-Solutions and how it harvests ideas from its own employees. Chris Flanagan of the Business Innovation Factory (BIF) was kind enough to post a comment on this blog shortly after and point to the video of Jim Lavoie (Rite-Solutions CEO) speaking about these ideas at a BIF conference.

I saw the video and highly recommend it (though the first half is sort of skippable). Go see it here.

I wanted to post the transcript of the video but couldn’t find one, so I’ll contribute my own little transcript for one Jim Lavoie quote which I particularly liked:

Imagine Jimi Hendrix comes into a room and he’s got a new idea for a song. He’s standing in front of 6 fat white guys with their arms folded asking him if he looked at all the other music ever published in the world to see if his song is really unique. Or if he’s done a cost analysis. Or if he’s looked at doing it offshore because ‘maybe we don’t have to do it here’. And there’s always the boss who doesn’t even talk to Jimi. He asks the lawyer “Is there any risk?”.

I call it ‘when the inventor meets the preventers’. And that’s what’s really stopping innovation: white guys that haven’t written a song in a lot of years that stop the innovator with stupid questions.

Every word set in stone… Design (or management for that matter) by committee is a bad bad thing.

Share this post!

Quigo’s 6th anniversary

Quigo_logo
We (Oded Itzhak & myself) founded Quigo 6 years ago this week. It’s been a wild roller coaster, especially during the years 2000-2003 when investors thought we were either insane, or martians, or most probably both. If we were to hire a panel of experts to identify the worst ever moment time to start a company like Quigo, I don’t think they’d achieve the amazing precision that we did in selecting that worst possible moment in history (excluding the big bang, possibly). The bubble popped and the market crashed. Investors, who only weeks earlier were throwing insane cash at insane companies, were not touching folks like us with a stick. An internet company?? Dealing with this dead-in-the-water search engine stuff?!?!

Founding_quigo_1

David Beisel of Masthead Venture Partners wrote a while back:

…one of my favorite questions to ask is, "What is the founding story?"
How did the founders meet and eventually come together? How did the idea for the current incarnation of the business emerge? How has the winding road of the initial days given rise to the current perspective?

So for whatever it’s worth, I’ll take the opportunity to talk a bit about the founding story of Quigo (from my perspective at least):

At the end of ’99 I was moonlighting[1] on an idea I had re search or discovery of pages on the web. The idea was essentially this – the back button was a fully functional piece of the browser, but there was no ‘forward button’ (even if only metaphorically). So my idea was to look at the content of the page a user was browsing, and based on the analysis of that content – provide links to other pages (or – the next pages) on the web that may be of interest to that user. In many ways, I guess, this idea had the seeds of the contextual matching that’s at the core of Quigo’s AdSonar product and Google’s AdSense (though I’ll admit that in many ways it was also quite different… ;-). I hired an engineer to help me hack a prototype of the idea. We never reached a fully working application as this evolved into Quigo shortly after.

At about the same time, Oded was fresh out of a company he was involved with, which was developing geographical streaming technologies (sort of a very early Google Earth  in 3D would be a good description I guess). He was tinkering with a search related idea he had – a downloadable application that would provide direct access to form-based search applications on a variety of different websites. In many ways, Oded’s app had the seeds of what later became Quigo’s FeedPoint product.

At that point Oded and I have never heard of each other. My brother, a classic Galdwell Connector, made the connection through a friend of his wife who happened to be married to Oded. He thought we were working on similar stuff and hooked us up (thanks Eytan, Sharon & Shiri!). We met a few times, opened our kimonos (which is, btw, extremely difficult for 1st time entrepreneurs who are confident the ideas will be stolen…how stupid…), brainstormed both ideas, and decided to take the best of both and start a single company.

Now, before I start bragging about how amazingly smart Oded and I were and all that BS, I want to point out the following points which put everyting below in its due perspective:

  1. Starting a successful company is as much about luck, determination and persistence, as it is about smart ideas. Coming to think of it – the idea is probably one of the least important factors in successful entrepreneurship of any kind.
  2. If it weren’t for our amazing employees today and along the way (especially those who joined us in the early days and stuck with us through some pretty insane periods….), the company would not have lived to celebrate it’s 1st anniversary, not to mention it’s 6th…. So thanks all you Quigo’ers!

Now I can brag freely…;-)

The combined idea with which we started Quigo was that search engines aren’t properly exposed to some of the most valuable content and product pages on the web (now commonly referred to as the Deep Web), and therefore we’d develop a gateway for the search engines into those pages. In essence, we identified the need to expose product pages from ecommerce sites (and not only sites’ home pages) to users searching for those products on search engines. This now is obvious to any 2nd grader, but at the time there were very few who foresaw this, and even fewer (I dare say – 2?…) who developed technologies to solve this.

While we got many things right (much of the foundation we laid back in 2000 still serves the company today and generates much of its revenues), one area we did get totally wrong was the original business model. We saw ourselves more as a platform for improving search engine technologies rather than a platform for exposure of retailers (aka ‘advertising’…). So our original plan was to crawl high quality sites (both content and ecommerce), and try to license our technologies to the search engines. While that model may have worked spectacularly well just a few months earlier ("we’ll pay any amount of $$’s for a better service/more eyeballs"), it was a total dud in mid-2000 ("eyeballs? yuck! Anyway – we’re running out of cash and can’t spend a dime on anything"). So we ended up flipping the business model: instead of asking the search engines to pay us (for the technology), we’d pay them! We’d charge ecommerce sites for the exposure on the search engines, and share the revenue with the engines.

Bear in mind that this was all a couple of years before search engines conceived of paid inclusion (PI) or pay-for-placement (P4P) programs[2], so the notion of paying search engines for exposure of pages was at the time fairly exotic, to say the least.   

Back to April 19th, 2000 – After working for a couple of months on things necessary for starting a company (seed funding, lawyer, etc, etc) – Quigo was born. Happy Birthday, and many more to come!

[1] I may be off on the dates and details here, but I think that at the time I was doing the following:
a) Trying to finish my senior year studying Product Design at the Holon Technological Institute.
b) Doing client design work under the design shop I owned – NetWorks.
c) Working on the product for another company I co-founded earlier in ’99 – Ad4ever (now part of Atlas).
d) Moonlighting on the product that later evolved to become Quigo.

Busy year…

[2] To be perfectly accurate, GoTo (aka Overture) had introduced it’s P4P program a little while earlier, but it had been available at the time only on GoTo.com which was a pretty obscure engine. Non of the major engines had any significant means of getting pages exposed and promoted on their SERP‘s.

Share this post!