Every day we take in a lot of information from a variety of sources. This information shapes our ideas, opinions and to some extent our personality. But where is it coming from?

Most of us don’t pay a lot of attention to this. We believe we make up our own opinions about the things that matter to us and leave the rest to professional and / or self proclaimed pundits.

Among the media that shape our opinions are movies, TV, radio, newspapers, magazines, books and of course (and largely for many of us) the things we read on the Web. If we would somehow keep track of all the content that we “consume” this way we could – at least to some extent – trace the origins of our ideas. Who it is that really shapes our opinions and views on the world. I think it would be amazing to see how uniform these origins are for most of us in the Western world. And talking about the Western world, it could be argued that it is the origins of our ideas that defines a culture and separates one from the other.

Tracing ideas
In the last couple of years tools have been emerging that make it possible to trace the origin of ideas more than ever before. This is largely due to the increase in “personal publishing” with blogs and blogging tools. At the same time, the tools are almost entirely limited to the blogosphere, but its a start.

One of the defining things for a blog is extensive linking to the sources and related information to the topic in question. The TrackBack functionality of some of the blogging systems such as MovableType make those links two-way so that not only do my readers know where my ideas are coming from they can also see who’s catching my train of thought and continuing the discussion, linking to me. This allows for some traceability of ideas from one weblog to the other.

But this is nowhere near sufficient. The TrackBack functionality is not that widely used and therefore covers only a miniscule part of all the content that appears online every day, let alone all the content that is already out there.

Many people have spotted this and you can now find several tools on the web that are dedicated to tracing links to and from web pages, especially weblogs. I think Blogdex was first in that line (coming out of MIT’s Media Lab), but others followed such as DayPop, Popdex and last but not least Technorati, probably the largest such index now tracking almost 3 million weblogs on a daily basis.

Theses tools allow you to enter a url or domain and see who have been linking to it. As an example I can see who’s been linking to Spurl lately. Interestingly enough none of these tools seems to allow you to see a list of the urls a given weblog has been linking TO only who has been linking TO IT. Such a list would be very interesting to do a quick “background check” on a blog you come across as ideas generally flow in the opposite direction of the link. When I link to a news article on BBC that almost certainly means I’ve read the article. Instead I suspect that the majority of searches on Technorati and co. are self-centered (ok self-conscious) bloggers trying to find out if anyone is linking to them. It should be very simple for the Pop-Day-Dex-Ratis to add this functionality.

Keeping track of the consumption
Obviously not everybody has a weblog and even those who do don’t write about everything the read. In fact it is interesting how bad the browsers we use every day are at helping users to keep track of the information we consume using them. Ever spent frustrating minutes or hours on Google or in the browser’s history list trying to find again that vital piece of information you read the other day?

Even the bookmarks / favorites that are designed to capture those “What a great page!” moments are very limited. They are only stored locally, so you don’t have access to your home bookmarks at work or in school and they usually get lost when the computer is upgraded or the hard drive crashes. In some browsers, bookmarks can not even be searched, let alone searched using a full-text search of the actual contents of the marked pages. The result is that few people use the bookmarks except maybe as a shortcut to a dozen or two of their most used web sites.

This problem has also been spotted and in the last few months, tools have been emerging that far exceed the traditional bookmarks in terms of functionality and usability, and even tap into some of the interesting social aspects emerging from the fact that thousands of other users are also using the same tools. Among these tools are Spurl.net (of which I am a founder), del.icio.us, Furl, Simpy and a few others.

To a varying degree, these tools offer functionality such as for example full text searching in all the pages users have marked, storing the entire contents of a marked page (addressing the “linkrot” problem), browser sidebars and toolbars for quick access, recommendations by matching user profiles, and of course access from any Internet connected computer (there is a lot more). Combined this makes the tools not only a replacement for traditional bookmarks, but rather a permanent record of the content a user has consumed during his or her browsing. The user can then use this record almost as a memory augmentation to later recall the information.

And there we have another piece in the puzzle that can help us seeing where our ideas are coming from.

Drawing the flow
Stephen VanDyke wrote a very interesting blog entry in the beginning of March, touching up on the same subject: “How News Travels on the Internet”. He even drew a flow chart of how he sees the flow of news online.

Obviously the flowchart only shows a fraction of the entire news-flow and the graph is rather centric to the blogging world, leaving out some of the more traditional ways news travel, such as press releases from the Source directly to the Traditional “Big” Online Media or even all the way to the Offline Media. A very thought provoking graph none the less.

The material that travels in the “Dark Matter” is especially interesting, i.e. emails, chat and instant messages where people are sending simple messages saying “Hey check this out.”. In the end, what captures the attention of the Big Media is largely coincidence, but as they reach a far larger and broader group, their coverage feeds into the blogosphere again and the circle can even repeat itself.

Another interesting attempt at drawing the flow of ideas is the Blog Epidemic Analyzer from HP. The project has indexed the flow of some 20,000 URLs as they go through the blogosphere from one blog to the next, spawning more blogging until they gradually fade from the discussion (and a new “hottest thing” emerges).

All the way to “The Source”
…sounds a bit like the plot for Matrix 4 – doesn’t it?

What I dream of is that I can one day – using tools similar to those described above – find out where the news and other information I’m consuming is really originated. That way I can see if my sources are colored by the influence of a single media company, religious group or government; the Bush administration or Michael Moore; the WWF for Nature or Texaco; the fans of the Pistons or the Lakers.

And if my sources turn out to be uniform I can at least say that it is on purpose or spend a little time studying the arguments of the opposite party. Then I can say in good faith that my opinions are formed after taking the arguments of both sides into account and making up my own mind, but were not deliberately or accidentally forced upon me by a like-minded group of people with a single view on things.

There is still a long way to go, but these tools and other similar attempts hint that this may very well be an achievable goal.

While I have no intentions to use Spurl.net to tell you who you are, the proverb highlights how important the information we consume is. Every day we take in a lot of information from a variety of sources. This information shapes our ideas, opinions and to some extent our personality. Given this fact, it is amazing that people don’t try to keep a better track of their information consumption.

I guess there are two reasons:

1. Suitable tools have been missing.
2. Because of the lack of tools, people haven’t really given its importance a thought.

Spurl.net is aimed at solving this problem. I’ve tried to implement it so that it becomes a seamless part of users’ browsing activity. See something you find interesting or want to store for later reference – a single click of a button marks it, indexes it for you to search later on and even stores a copy of the page so that it will always be available even though the original page changes or becomes unavailable (so called linkrot). If the user wants, he or she can add more metadata about the page such as a comment, a custom title and a suitable category in a personalized hierarchy of categories, all efforts to help the user locate the information again later.

The conscious effort to mark something is important, as only a part of the abundance of information we consume is worth being able to track later on. The marked information thereby becomes a record of the consumption that the user rates as important or somehow significant.

While I’m not totally convinced that “personalized search results” are the holy grail some people claim it to be, I’m sure that most users will find it hard to imagine how they managed without being able to search the interesting part their own browsing history, once they’ve become accustomed to it.

With this record in hand, Spurl.net can also offer the user a lot of interesting value adds on top of the core “store & search” functionality mentioned above. The recommendations, syndication possibilities and the “spurl beam” (see: Spurl.net page, all mentioned features require login) only scratch the surface of a wide range of functionality that can and will be offered utilizing users’ spurl history. Some of these are already being built into Spurl.net and will be introduced in the coming weeks.

While these are all nice frills, I will try to keep focused on the main vision as detailed above. As before, all comments and suggestions are welcomed.

– - –

As an end note, nicely portraying the importance of the information we consume, here are a few of the sources that inspired my Spurl.net implementation in the first place:

• “As We May Think” by Vannevar Bush
• Grand challenge Memories for life led by Ehud Reiter and Andrew Fitzgibbon.
• “Tapping into the Beam” by David Gelernter (published in The Next Fifty Years).
• Microsoft’s MyLifeBits project
• Out of Control by Kevin Kelly
• Smart Mobs (book and blog) by Howard Rheingold
• Emergence and a range of articles by Steven Johnson
• Linked by Albert-László Barabási

There are two main reasons for the switch:

• The old URL wasn’t all too easy to tell your friends about – was it?
• The dedicated URL will make any forthcoming upgrades a lot easier.

The name, spurl.net doesn’t mean anything, i.e. the “url” is url for sure, but the meaning of “sp” is up to you – ideas welcomed. On the other hand the name is short, easy to remember and not too many ways you can screw up the spelling of it when you hear it pronounced – all very good qualities. Thanks to Frosti at Sonos for the name.

The logo was made by Egill. I think it is pretty nice. He’s however not to blame for the rest of the spurl.net UI – I made that mess all by myself.

There are several new features introduced with the move:

• You can categorize your spurls and add personal comments to them (use the “Advanced” tab in the “Add” window).
• Quick lists: These lists show the activity at spurl.net in a brief overview, but leave out some of the details found in the older lists.
• I have separated “My spurls”, i.e. what each individual user has spurled, from the other lists, as the usage of that list is obviously quite different.
• You can now click the “Related” icon next to any URL to see other URLs that have also been marked by the same people.
• …and more. I will add details here as the new features are sufficiently tested and documented for your use.

Well, I hope you enjoy the new and improved service. As before, any ideas and comments are welcomed and if you find something you think is a bug, please let me know.

For those of you that were using “I like!” before, I ask you to delete the “I like!” link and replace it with the new one (you will find it on your “My profile” page in spurl.net). Same goes for all RSS feeds and Javascript links. To update them, simply replace “http://mercury.hjalli.com/ilike/” with “http://spurl.net/” in the URLs. Simple – right?

The old links will however work for some time to come, so don’t worry.

Spurl.net is still in development, so expect to see even more new and improved stuff coming soon.

The new Mercury blog will from now on be THE source of information on what is happening with spurl.net, and my Wetware blog will go back to the normal talk on non-traditional computing, engineering and biology.

Thanks once again for using “I like!”. Enjoy spurl.net!

By the time I post this, it will be almost exactly a week since I introduced the service and I would not have believed the amount of traffic the site has got since.

As a result, the server is sometimes a little slow in response. Usually it works ok, but it is a rising problem. I am in the process of upgrading my hosting server, so bear with me for a few days.

From the numerous emails and blog entries I’ve seen from users (see bottom of post), most people seem to like, I like! (pun very much intended).

Here are answers to a few of the most common questions and comments I’ve seen:

• There is a problem with international characters: Yes I know, it’s being fixed. As a native speaker of Icelandic I surely understand your frustration. Give me a couple of days…
• I want to be able to manage the pages I’ve marked as my bookmarks: That kind of features will be added. Expect the first version of that late next week. When the time comes you will be able to categorize and sort the pages you’ve already marked, so keep “Liking” as you please.
• My top recommendations have no relevance to what I have marked: Actually I’ve been surprised to see how many users have been happy with their recommendations, given that the current recommendation algorithm is only my second version. Even the ones that have given the recommendations bad reviews have still admitted that they found pages that amused them in the recommendations. Keep in mind that the recommendations will improve:
• …as the overall database grows.
• …as you mark more pages you like.
• …as I will further improve the recommendation system.
• I want to be able to add details about the page as I mark it: As it is now, you will have to go the I like! page to add details about a page you have marked. I’m changing this so that you can do this right away when you mark the page. This will however be done so that it will not change the functionality for those that just want to quickly “like” a page and leave it at that. This is actually the next feature update. Hopefully ready by tomorrow.

I have a long list of feature and improvement ideas that I’m going to implement, but at the same time I will implement your good ideas and requirements as well and prioritize in line with the feedback I get, so please keep the comments and suggestions coming. I will evaluate the ideas thoroughly and try to do the best I can (use the comments here, or drop me a line at ilike.hjalli.com (replace first dot with @)). And if you find a bug, please-please let me know.

Thanks again for all the feedback and for using my service.

Here’s what people have been writing about I like!:
- I Like It! – Chew’s News
- I like, un outil de bookmarking avec recommandations – Outils Froids (French)
- I like – Dr. Web Magazin (German)
- I like – Cynical-C Blog
- I like – She=Marmod (Dutch)
- I like – Couchblog
- I like. – The Presurfer (no permalink)
- I Like… – Blog-Fu
- Do You Like? – J-Walk blog
- I like … – NurSo ¦ OderDochSo
- I like – *.*
- I don’t like – Green Gabbro
- It’s called I like – HubLog
- I like! – Library Stuff weblog