258 weeks agoWeb is Atomic
What makes web different?What puts Internet apart from other computing? It may sound dumb, but Internet is different because it is online. What does that mean? We can request as much information as we want at the moment, no need to buy multi-CD Encarta Encyclopedia. If something is on the web then it is within our reach when we want it. Imagine if we had to download website at it’s entirety to have a glance, that would suck. Or let’s imagine an alternative-reality-web where all images are embedded in HTML. That’s stupid, inefficient and inflexible. Web is smart, efficient and flexible, and the reason for that is that it is atomic. CSS is useful, it allows content and presentation to be separated, how nice. CSS makes web pages more atomic. That is good, period. However web is atomic in more ways! If you know anything about HTML you should remember how you’ve learned that, that wasn’t too complicated, was it? You can look at the source of a web page and see what makes things bold, italic and such (at least that was true back in the day, I guess it’s not as obvious with present layouts). Or you can find a tutorial and understand some core concepts behind hypertext and markup pretty easily. Later one would realize that what he learned was not entirely true, but even incomplete knowledge worked pretty well. This is exciting! It ’works’ that way because mechanics of the web are atomic, smaller self-sufficient parts form bigger structures. Is Web2.0 atomic?Now let’s see how this applies to semantic web. Let me quote Tim Bray: ’Maybe I’m just a geek, but I think it’s the data and the hyperlinks that are at the centre of everything; if you focus on keeping that as good, clean, and open as possible, the right software and services fall out. Like Sam Ruby says, ’It’s just data.’’ Absolutely right! Hyperlinks (and related concept of URI) are the magic that allows web to be all interconnected, while staying atomic. If the technology does not submit to this, then will fail on the web, simply because it will not fit in. I’ll give you just one example. How much more Google Maps are useful because they allow user to make permalinks? A lot! Is RDF atomic? (It’s not)Let’s take a look at RDF now. It’s a set of triples, can you link or address specific triple? I don’t know, and I think it’s too hard to find out. RDF makes too big a jump from what data on the internet is to something some highly intelligent architects think it should be. I think that’s not going to happen. RDF is not atomic enough. One single worst thing that kills entire RDF business for me is that it puts too many things in one place.
Solution to what problem?I see a few areas where separating things some more would be nice:
One last quote, from Paul Miller this time (I mixed different paragraphs and changed emphasis. Web is remixing – so true.): Web 2.0 presages a freeing of data, allowing it to be exposed, discovered and manipulated in a variety of ways’(And so on, read the original, it’s good.) See? It’s all about being atomic from different perspectives. Web is atomic! |
|