Tom Morris has a nice thought about using SPARQL as a universal API.
Makes great sense, but I followed the link and it was another long and seemingly academic RDF article. Well, he did warn me.
The problem is, I like this stuff and will probably get back to it later, but a lot of web developers might not, and that’s why we are seeing slow adoption for SPARQL, I think.
Then again, maybe adoption isn’t really slow, but I am.
Danny Ayers has some wisdom about not reinventing the wheel for the semantic web.
Apparantly there seems to be a disconnect between the MySQL community (or CEO at least) and the Semantic Web community.
It’s hard to disagree with Danny most of the time, but I think the RDF web communty can take one thing away from this whole debate. (and I think they do)
The Semantic Web community needs to make this stuff more easily accessible to the average mortal web developer.
There was a time when there was bit of talk about using OPML and RSS as a springboard to getting kids to adopt some more sophisticated RDF stuff.
We do want to make this stuff relatively easy to experiment with, and then developers will see it’s power and want to delve deeper.
When someone makes something like a GrazrScript layer on top of a SPARQL query (hint hint), I think we’ll see some more mainstream adoption.
But ultimately, Danny and others are right. The foundations are in place.
But, as Tim Berners Lee pointed out recently, we need to have incremental adoption, and that’s all I’m driving at.
I’m hoping part two of the latest Gillmor Gang will prove more interesting.
If you remember the Jason and the Argonauts tale, you might know how Jason succeeded in conquest over the Seed men by casting a stone at one, who thought it was his neighbor, and letting them all kill each other.
That’s what Steve Gillmor seems to do by letting the fellas discuss the importance of Google algorithms and whether site owners can get a cut by having search engines bid for their site search.
If Steve would have put the “knockoff” Cheerios down for a sec I know what he would have said.
It’s not whether Google’s algorithms hold up, it’s whether they can garner more stock in the conversation with all their attention data.
The winners of the future are not the best technologies. We’ll all be able to plug into those the same way we plug into an electrical outlet.
The winners are the services which add value to the conversations happening throughout distributed web networks.
These networks and conversations are fluid and changing constantly in response to our gestures.
Those who don’t get this are either thinking too hard or just not enough.
In a similar way that facial and hand gestures are a meaningful supplement to spoken conversations, the gestures which we talk about with attention are the metadata of the conversations happening on the web.
That equates to economic power because markets are conversations.
I agree with Jason Calacanis that many in the SEO business are trying to game this system, but I disagree when he says the system works. People are trying to game the system because it does not work. It just works better than the previous systems.
I can prove it Jason. I’ll write a better piece on a new cell phone than Engadget and see which shows up higher on Google.
No. Those dynamics are only part of the game.
The richer system envelops us with answers using our data and our network’s data in a chameleon like fashion, never static like Google. That’s child’s play.
Jason(Argonaut) succeeded in getting the Golden Fleece but was fickle and left Medea for another Princess.
Likewise, in the shorter term companies may succeed by amassing link attention.
The true winners won’t be seeking the Golden Fleece at all. They will be removing the barriers and letting the crystal waters flow in, filtered and clean, Pure Conversation.
So, Im back from RSS Alley and OPML Camp today.
I did not blog from the conference simply because the conversations were so rich and thought provoking that I didn’t want to only give them partial attention.
Attention itself was in fact a big topic and there is some big news coming our way on that topic. Other topics included namespaces, the spec, RDF and tags. Oh and Second Life seemed to permeate the breaks. I even talked a little on SSE, and I think we might see some progess on SSE used with OPML soon.
Special thanks must go to Adam Green, who did a great job organizing and Berkman for hosting, and Halley Suitt and the Top Ten Sources folks for hosting the party.
And if Dave Winer is listening, we all thanked you for RSS and OPML in general. Great work.
The whole weekend was truly inspirational.
But speaking of conversations and attention, this one is nearly over for me.
Perhaps a couple transitional posts and I’ll be moving on.
A couple things are for sure. OPML and Reading Lists (Glists if you ask me, Beeds if you ask James Corbett) have a huge future. Attention and gestures have a huge future. Grazr has a huge future and Second Life-like environments have a huge future. I’d like to be in the middle of it all.
Or I could just hang out on the podcast.com platform(search for podcast in second life) and listen to streaming music.
Either way.
Alex Barnett gives his take on the current status of the semantic web.
I can only add one observation.
We often say about site users that to successfully gather metadata, we need them acting in self interest.
For example, the del.icio.us users are tagging for themselves, and everyone benefits from the collective wisdom that can be imparted.
The same should be applied to site developers and designers and even plain old bloggers.
They will only add the extra effort of making their data available in a semantic web friendly format if they can see a return on that investment.
They only cleaned up their designs when they needed to rank high in search engines.
They only linked and permalinked when it was obvious what the benefits were.
In other words, if Google started heavily favoring semantic friendly content, there would be a rush to incorporate all the great ideas brewing out there.
It reminds me of Heiman Roth in Godfather II.
I didn’t ask who gave the order, because it had nothing to do with business
Until business folks can see an ROI, it’s tough to get them to act altruistically. However, I will leave open that the citizens can be the leaders here and create the market, as is happening with so many other fronts.
Then the business folks will follow.
If you don’t feel like clicking ,
Stowe Boyd
Alex Barnett
Robert Scoble
Ken Yarmosh
I’ll let you know that all these posts are about dissatisfaction with memetrackers. That’s just a sample of a growing feeling.
You can just as easily put together a list about the attention problem caused by generic feedreaders.
I could pontificate today about possible solutions, but instead I think I’ll take some action. I’ve been looking for a project. This seems to be the one.
I’m launching an open-source memetracker at http://glistn.com , but it’s going to heavily integrate reading lists (glists) and just approach the whole problem differently. In fact, I mean it to be something more than that. I want it to be a metacommunity. More on that later.
Since it will be open source, you’ll be able to know exactly how the filtering takes place and scream if you don’t like it.
If you want to be notified when it goes live, you can sign up at the site and also subscribe to the blog.
If you’d like to take a larger role (perhaps advisory or hacker), just contact me at matt {at glistn dot com}
Dan MacTough channeled through Danny Ayers, or was it Ayers channeled through MacTough . . .anyway
More important than the fact that OPML may not be a good format for a particular use is that the end use always seems to be to render the information in HTML, a la Grazr, Bitty, OPod, Optimal, etc. I just don’t get why anyone would want to transform their information from a format designed for it INTO a format that’s not designed for it only to then transform it AGAIN INTO HTML. Haven’t they ever played with Babelfish? English-to-Japanese-to-French may be fun, but it’s not a very accurate translation.
Great quote, and I hear where y’all are coming from.
But couldn’t the same argument have been made for RSS. In the majority many cases, the reason for putting any data into OPML is because you want a common format to share it with others.
I know there are other options, better formats, and that you can argue that RSS lists or even HTML lists are just as good, but buzz isn’t always a negative thing. Sometimes it can signify a growing consensus.
In such cases, for good or bad, to allow the greatest number of people and services to share your data, you need to put it into the container they are prepared to accept.
So I think these renderers are just ahead of their time, and in order to show them off, the creators may have to force the data into OPML. The developers are looking forward to a day when when OPML is a highly common exchange format, and then these conversion of OPML to HTML will make more sense.
A new concept is sprouting in the OPML landscape.
The allusion to flora is not accidental, even if banal.
Consider these two unrelated posts, the first from Lisa Williams,
OPML’s biggest impact will be in making it as simple to add a record to a self-assembling worldwide directory as it is today to write a blog post. (Did that make any sense at all? I hope so.)
Yes, that makes sense Lisa.
It sounds like an “organically” created web directory, seeded and fed by the natural actions of an ecological-like community.
Next we move on to James Corbett, commenting on one of my posts,
I’ve been wondering if we should label these multimedia Reading List as…. SEEDs = Sensory Feeds. Seeing as SEEDs are the fruit at the leaf nodes in a tree I think this will tie in nicely with the direction some feed grazers are going. And as a SEED meme accumulates momementum it can actually spawn a whole other OPML tree, just like a real SEED.
Before we get lost in this placid garden imagery, we must also note one of Corbett’s posts that indicates there is also a food chain, or feed chain, if you will, that is in intense competition for our ravenous attention.
He concludes,
Of course the fleet footed Feed Aggregators won’t die out, they’ll just evolve Feed Grazing capabilities.
Our current crop of aggregators are likened to reptilian eating machines. The next generation of consumers, the mammals, will use adaptability to flourish where the reptiles could not.
Man, however, is the only creature in history to have conquered agriculture. Thus, the information consumption tool that wins will not only hunt and forage, but harvest.
This, I think, is a key conceptual transition that must be made to address the growing attention inundation issue.
To consume what is available naturally will not be enough. Social structures must be built to enhance the bounty which abounds.
Adam Green’s River of Feeds is certainly pointing us in the right direction. Annotated lists turn that river into a mill. Lisa Williams hints that we are at the dawn of a new type of information economy, one built upon the small actions of the masses. And so we stand at the launch of a new era, similar in many respects to the industrial revolution.
Large economies of scale, mediation and complex societal structures were produced by the historical industrial revolution.
In this metaphorical one, we will produce some of the same, but moreso, an ecosystem. Both economy and ecosystem, stem from latin for household or habitat.
It seems to be largely held that these social communities can be sown out of the metadata that exists like tagging, linking and subscribing.
I’m going to conclude this post by contending that a more definitive gesture will arise that will create smaller communities among the larger ones that we conceptually know of today.
In fact, I’m going to borrow a concept from the science of ecology called the metacommunity [PDF].
{End of Part I}
John Milton and others call Man’s fall from Eden into sin a “felix culpa”, or beneficial fall.
The same idea can be applied to inferior technologies like HTML which have a greater ultimate good than their superior counter-parts.
The web wouldn’t have exploded so quickly if it had been built on XML.
Some will put down OPML and RSS 2.0 as inferior to other formats like RDF and Atom.
But there seems to be a felix culpa happening here as well. The inferior formats are being adopted at a faster rate than the richer ones.
If HTML begat XML, will OPML begat RDF?
That’s what Danny Ayers suggests in one of the most interesting pieces I’ve read in a while,
Say you click on a link in the content view pane, there’s no reason why this couldn’t trigger the on-the-fly generation of a new “feed” from the triplestore. When you bring in the notion of subscribing to “Reading Lists“, i.e. dynamic feed source lists, you’ve got a great way of connecting to new sources of info, again by bridging through the formats the agg understands.
This is an incredible idea. To use reading lists as a conduit to the sematic web. I love it.
Small incremental developments which offer immediate rewards stand a better chance of adoption than big leaps forwards, not matter what the reward.
I’d like to point out the irony that there are footnotes in his blog entry that suggest he’s not done with the post. He’ll come back later and finish his thoughts. Let’s get it out there and then we'’ll expand and improve upon it. He’s living by the words he suggests we adopt as an industry.
Hear. Hear.