Posted by: rolfsky | October 11, 2007

viva la semantic revolution – get started by rolling your own

If we do a quick search for why web 2.0 sucks, there are plenty of resources and even attempts at correcting those issues. But we’re not here to talk about that. We are here to talk about why we need the semantic web to suck, desperately. Tomorow.

The sneaky secret here is that the sooner we all recognize it’s broken, the sooner we’ll all start pitching in to fix it. Tom Sawyer doesn’t need help painting this fence, he needs help building it. A simple plan might be to pile some milk crates and pallets in a rough line and walk by for someone to tell him “really” how to build a fence. In this vein, let’s start making a terrible semantic web.

I say we want a terrible semantic web because understanding that something is terrible is to acknowledge that you:

  1. know that it can be better
  2. more importantly: you know that it is a thing

You can complain that “these are doughnuts are terrible,” but that implies you know what a good doughnut is like. Complaints are the by-products of failed expectations; they are the release of the difference between what you think you should get and what you did get.

Expectation is the stored energy which inventors, entrepreneurs and other people who stay up all night seek to tap into. They understand that giving people what they want puts them in a place to profit. If we hear no complainers, then there’s no apparent reason to “fix” it and business can merrily ignore the issue while the entire community doesn’t know what it is missing. As a business person, the loud wail of complaining signals a market opportunity, “obviously they need good doughnuts, over there.” Viola, someone invests in a doughnut shop.

In this situation, it’s up to us, the thinkers, engineers, fanatics, product managers and other tech gurus to show the rest of the world what they’re missing. Can we do it all at once? Can we serve the feast without forks, plates, fruit and a table? No. So let’s get them started with some bread and cheese on a picnic blanket and stare at them with lovey eyes and raise our eyebrows as if to imply, “oh, darlings, you haven’t seen anything yet.

So, if we’re all going to start, where should we begin? Because we are lazy, let’s start by borrowing some thinking smart people have already done. If we care to, there are plenty of microformats which are the amoebas in our proto-soup of the semantic web.

Before I scare anybody off, let me disclaim that word a little first. The word “microformat” is another ridiculous word invented by someone clever, probably not unlike myself. While being technically correct and essentially descriptive they defy any attempts in decryption by those not in the “in-crowd”. This is doubly unfortunate because, of course, the creators are giving microformats to help us and would want everyone to embrace them.

Without jumping too far in, microformats are just agreements about how data is structured, tagged, and represented in regular’ol web pages. There is no voodoo magic, it’s just a set of understandings that when you write something like this: <a rel="friend met co-worker" href=...>, we all know that this link is referring to a resource owned by someone who is a friend whom I’ve met and also happens to be a co-worker. Now, what you do with that information is entirely up to you, or your browser or whatever appliance you use to interact with internet content. Perhaps, it could put a small icon for each aspect next to the link. There are more complex microformats like hCalendar, but let’s start small.

The particular microformat that I’ve described above is called XFN which stands for “XHTML Friends Network”. The idea is that you can essentially “tag” links with contextual data to state that this link points toward a friend, while this other link actually points to another page that I own. I know that this is light-years behind what we want the semantic web to be, but that’s kind of the point, isn’t it? It also has applications for building some type of distributed “social network”, but don’t hold your breath on that quite yet.

The good news is that if you’re a WordPress user, they’ve got you covered. Embracing a whole bunch of mircoformats, setting up your own little corner of machine-readable is easy as pie. Jump into your blog-roll management section, click “edit” on a link you already have and then expand that little section labeled “Link Relationship XFN” you probably didn’t notice before or were too scared to fill out. Well folks, that’s what it’s here for. Fill one out, label and link and post a comment to let me know you’ve participated.

To prove your success for yourself, drag this bookmarklet to your favorites toolbar to install the XFN Dumper v0.21 found on the XFN Tools page, visit the page including the links you just made, and click the Dumper.

Now, with your new-found knowledge of XFN, what else can you do? Click on over to for a nice glimpse, it may be more than you think. With your help and understanding, I’m positive that we can start showing the world how much a pain doing this all by hand will be and also the promise that it shows to help out. Stay tuned for more explorations. 🙂

Don’t worry, we’ll all get the point eventually.



  1. Rolf, I am a bit sceptical about where this semantic web is going to. I agree that it sucks in its current form. But the examples you provide hardly get me enthusiastic about the whole concept. I understand that there are 2 different issues to be solved. The “I have already 1 identity filled out”, why do I have to do it all over again issue. And the “What are the relations between people in my network issue. XFN does provide tools to tackle them both, but the question is do I really want to have these solved?
    It would be nice to set up 1 profile and re use it in different services. Dick Hardt with his Identity 2.0 is working on that. But I am afraid that managing my profiles during usage might me more work than simply entering it a few times. I would like for friends I have in one network to also be available in another. But personally I don’t want to take the effort of setting the relationships of friends in a network (social graph). It seems a way to enrich the social graph, but does it also enrich my interaction with others? The thing is, I know these relationships already. That is why I interact with them in the first place. No need for me to describe them on-line. I’m curious what problem you would like to solve with this. Where is this going in the future? Am I missing the point ?

  2. Ah, hah! 🙂 You’ve wandered into my trap.

    The viewpoint that you are expressing is exactly where we need to be. Your dissatisfaction is proof to the designers of tomorrow that you’ll be willing to support something like that.

    You’re absolutely right about not wanting to transport/re-input your profile, friends and contacts. The XFN that I described is like writing your own HTML. Or sure, you CAN, but why?

    These tools are laying the groundwork for the ideas of what come next. 🙂

  3. Aha, another cliff hanger, can’t wait to see where this goes.

    I see you’re traveling over here in Amsterdam, wanna meet for a cup of coffee and redesign the web?

  4. Coffee it is. Hey, I’ll be in the Netherlands at the end of March… coffee or tea?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: