"I was thinking about you all day today and what a great person you are."

"I wanted to be #1...After 2 months I reached the top position for my most popular keywords."

Read More...

The SEO Book

PPC Information

Need Links?

Need Traffic?

Use these Free PPC Coupons

SEO Tools

WordTracker - keyword suggeston

Interview of Spurl Creator Hjalmar Gislason

28 July 2004

Recently I interviewed Hjalmar Gislason, who is the owner of a social searching bookmark project by the name of Spurl. Here is what he said...

What is Spurl?

Spurl helps users remember the interesting web pages they come across. Collectively, Spurl.net users are collecting the interesting bits from the entire web.

Spurl.net is an advanced bookmarking service. It stores the useful and interesting things users come across while browsing. Once stored, these quality information sources are easily accessible again through full-text search from any Internet connected computer.

As thousands of users are using the same tool, Spurl can recommend things that you will be interested in and enable you to find related information that other Spurl users have come across. You can also easily share your findings with others or even work together in groups, collecting information on a given topic.

I have played with a few other tools built around similar concepts or themes. What makes you different or better than say del.icio.us or Furl or Simpy?

I'm obviously not the best person to answer if or how Spurl.net is better than these excellent services, but I can mention a few of the main differences.

1. Spurl's users are anonymous. All a user has to do is choose a username and password. The links are then shared anonymously in various ways, without involving the user more than he or she wants. The ability to publish your links (on a dedicated page or included in a webpage or blog) is certainly there, but the user is in full control of how much he reveals his identity. Spurl.net is an attempt at creating social software without involving specific users too much. It may look on the surface like interacting with a computer program, but behind every interaction is the collective intelligence of thousands of users - and somehow you feel that. Some people will certainly see the whole anonymity thing as a drawback, but for serious research and business use it may very well be exactly what is needed to be able to make full use of the social aspects without compromising your own data.

2. To an increasing degree the group functionality, that allows users to selectively work together in groups, collecting resources on a given topic will become a key element of the system. Imagine a programming team that can search and share their collectively gathered information on technologies or subjects related to the projects they are working on. Or a class at school that can share their online findings on a subject. Or just a group of friends that wants to share the funny and interesting pages they come across browsing each others blogs or the more casual parts of the Web.

3. The Spurl Bar gives quick access to all the main functions of the system through a compact interface. Quick access to recently spurled pages, recently visited pages and most used pages through the Spurl Bar becomes centric to many user's experience.

Recently blog software and blog tracking systems have be working harder to look at social connections. What is missing from their systems?

That obviously depends on what they want to achieve. But they have a lot of interesting possibilities given the mass of information and - even more importantly - links that are flowing through their systems.

I'm a huge fan of social software of all kinds. We are just beginning to see these systems' potential, not only as social networks but also as tools to collectively make decisions, filter content, and personalize information and interfaces. Social aspects will make their way into much of the everyday tools we use, even the ones that we do not normally see as social today - like word processors, spreadsheets and anti-virus programs.

The general feeling I get from many of the social systems today - including the blog tracking systems - is that their main use (and seemingly main expected use) is a bit Narcissistic. Take Technorati as an example. The main use for the tool is almost certainly to look up "who is linking to me?", but it would be at least as interesting to see "what is this blogger linking to?". The data is already there in Technorati's database - so it shouldn't be much of a problem to add. The really interesting things start to happen when you combine the two and do two-way analysis of the flow of ideas and links from one blog to the other through the entire network of blogs. A lot of people have talked about this kind of ideas but I've not seen a proper implementation of it. In any case Technorati is probably in the best position to do it as they seem have the most data.

As I understand it your service is completely free. What is your business model?

You're right. Spurl.net is free of charge to users. There aren't any ads either - just the pure and simple user interface.

Our business is built around the idea of The Interesting Web. Our thousands of users are out there, constantly searching for interesting and useful content, marking it, categorizing, labeling, writing descriptions and indicating the most interesting parts of the pages by highlighting snips from the pages before "spurling" it.

One could say that where Google has its spiders, crawling the web and automatically indexing pages, Spurl.net users are doing the same - storing the interesting things they find and benefiting from the findings of others. A humanly written description of a page will beat the best indexing robots any day of the week. Let alone the "consensus" that can be determined when the information about a page from hundreds of users is compared, reflecting the different views and reasons-for-interest in the page.

Spurl.net gathers information about web pages from the database and sells this unique source of information to search engines, filtering companies and web researchers. As clearly stated in our Terms of Use (http://www.spurl.net/termsofuse.php) and Privacy Policy (http://www.spurl.net/privacypolicy.php), the data that is sold is only aggregated information about web pages, not data about individual users or their usage.

The buyers of this information use it to improve their web page indexes, interest relations and usage pattern information.

To emphasize the importance of human information in search engines, two examples stand out:

1. Yahoo's humanly edited directory and index in their early days.

2. The importance of human information (namely links in humanly created content) in Google's PageRank algorithms.

Both became the places to go for searching the web. The search industry is constantly looking for new sources for such information. A recent example is Ask Jeeves' "subject specific popularity" that rates links from certain (humanly edited) sources as authoritative on a specific subjects and ranks links from there higher than links from other sources. Collective intelligence at its best.

Off the start your data will be of extremely high quality, but it will lack the distribution necissary to facilitate selling it to large partners. How do you intend to capture enough participation to make the data extremely valuable?

In only six months we've gone from zero to 100,000 links flowing through the system a month. There is already a lot of value in there and some of the experiments we've been doing with the data are highly promising. Sure enough we need to attract more usage, and there is a lot of room for improvement in that regard. So far it's only been word of mouth - no active marketing or PR. We're starting that now.

The key thing though is the user experience itself. We're constantly improving the product, but try to keep a delicate balance between introducing new features and making what's already there as user-friendly and straight forward as possible.

These kind of services are not as viral as communication software - as it does not matter to you whether you are receiving links from your friend or someone half the way around the globe, as long as it is useful. The group functions - or Spurl directories as we call them - changes that and we're certain that when we upgrade that part of the system from the experimental phase it is currently in to the production version in the next 2 weeks or so, it will dramatically affect the viral distribution of the product. At the same time it is highly useful for the users.

So, it's a mixture of active marketing and product improvements that will take us to the next level in terms of distribution.

After the data gains value and is used you will likely run into the problem of attempts at artificial manipulation. How will you keep the data completely anonymous and yet prevent your system from being heavily manipulated?

This will inevitably happen at some point and we anticipate it. The solution to this lies in a "karma" system of some sort, where the importance of data in the "consensus" will depend on the user's usage history. It is a lot easier to deal with manipulation within a system like Spurl.net than - say - comment spam in blogs. A manipulating user will have to fake a normal usage pattern to really be taken into account and that may be too much of a hassle for the "spammers" to be worth it. Bots should be relatively easy to identify as it will be hard for them to build their karma within the system. In blog comments the expected behavior is dropping in, one comment and out again, making it hard to identify "real" users. In this sort of a system one builds their collection over time, making it much harder to fake.

This does not mean that new or inexperienced user's data will be somehow blocked or overlooked in the system, just that when gathering material for the consensus information, clever tricks will be used to identify the good information.

The concept of your product will likely be integrated into many search engines and portals. Google buying Picasa caused their distribution to nearly immediately increase 6,000%. Why do you believe that large search providers and other sites will purchase your information versus building similar technology in house?

I'm pretty sure that we will see the "big guys" offering Spurl-like functionality within the next 12-18 months. It makes sense for them, both as a service to their users and to be able to utilize the data for the purposes explained above.

With their strong presence in the market, these players could probably get as many users in a week as we've signed up in six months. But as soon as one of them starts such a service, others will have to follow suit. If a partnership with an established service can save them the 4-6 months it takes for a big organization to make a decision, implement and roll out a new service - that can be crucial in the race.

There is a lot of turmoil in the search industry right now. New players are emerging and some of the IT industry's heavy-weights are joining the race. Everybody is looking for innovative solutions that can help them "beat Google". I firmly believe that new and innovative sources of human web page intelligence will play a central role in that, and not everybody will be building everything themselves from scratch.

In the meantime we continue to build our user base, gather data and build experience in turning that data into useful information. We will also be rolling out additional services that demonstrate the power of the concept. There is a variety of ways that this information can be monetized right away - by selling it and by rolling out commercial services that make use of it ourselves.

-----

Thanks Hjalmar.

If you would like help Hjalmar gow his network by giving Spurl a try today.

- by Aaron Wall, author of The Search Engine Optimization Book.

We Value Privacy

This article may be syndicated in whole are part. Simply provide a link back to the original article or http://www.search-marketing.info. Please note that I do not usually update articles over time and the date last modified on article pages is usually referring to a navigational change.