 |
 |
 |
Recently in foundational Category
I just came across an amazingly brilliant post by Glenn McDonald at ITA software (it's been written a while ago, last summer, but just read it entirely now). Do me a favor and read it entirely here: Never Mind the Semantic Web (or, 13 Reasons Not to Let a Computer Scientist Choose a Name (or a Problem)).
I must say, I truly wish for 2009 (yes, I am a bit aggressive in expectations) that we finally get some sense into our community and spend some time working on the positioning of the semantic web/web3.0, or however else you'd like to call it (this in itself a semantic problem, sadly). Glenn has 13 reasons, and hence 13 great points that point us in the wrong direction we've taken. I am going to analyze and ad lib on them in the next posts one by one, because I think this is the juice that we've been missing all along.
For right now, I will pick his 11th point:
" 11. "Metadata". There is no such thing as "metadata". Everything is relative. Everything is data. Every bit of data is meta to everything else, and thus to nothing. It doesn't matter whether the map "is" the terrain, it just matters that you know you're talking about maps when you're talking about maps. (And it usually doesn't matter if the computer knows the difference, regardless...) "
This is brilliant. By obsessive-compulsively talking about metadata as core to the semantic web, we made it sound like an Aristotelian/Platonic concept of "substance", or "essence", as if metadata is an onion's "core", so that when we want to discover/generate metadata, all we have to do is go to metadata's outlets and just get it from there. This very absolutist data modeling philosophy assumes there is a consistent and stable metadata core. The problem is that the onion of data generated on the Internet has no core: it's layers and layers of data. As Glenn eloquently puts it, there is no metadata, or rather, every data is a metadata to another, depending on what you want to build, generate, etc.
Funny how upholders of semantic orthodoxism (there is only one right semantic way, right? as they say) simply created a semantic problem.
What do you think?
I must say, I truly wish for 2009 (yes, I am a bit aggressive in expectations) that we finally get some sense into our community and spend some time working on the positioning of the semantic web/web3.0, or however else you'd like to call it (this in itself a semantic problem, sadly). Glenn has 13 reasons, and hence 13 great points that point us in the wrong direction we've taken. I am going to analyze and ad lib on them in the next posts one by one, because I think this is the juice that we've been missing all along.
For right now, I will pick his 11th point:
" 11. "Metadata". There is no such thing as "metadata". Everything is relative. Everything is data. Every bit of data is meta to everything else, and thus to nothing. It doesn't matter whether the map "is" the terrain, it just matters that you know you're talking about maps when you're talking about maps. (And it usually doesn't matter if the computer knows the difference, regardless...) "
This is brilliant. By obsessive-compulsively talking about metadata as core to the semantic web, we made it sound like an Aristotelian/Platonic concept of "substance", or "essence", as if metadata is an onion's "core", so that when we want to discover/generate metadata, all we have to do is go to metadata's outlets and just get it from there. This very absolutist data modeling philosophy assumes there is a consistent and stable metadata core. The problem is that the onion of data generated on the Internet has no core: it's layers and layers of data. As Glenn eloquently puts it, there is no metadata, or rather, every data is a metadata to another, depending on what you want to build, generate, etc.
Funny how upholders of semantic orthodoxism (there is only one right semantic way, right? as they say) simply created a semantic problem.
What do you think?
Just read Doc Searls' post on VRM is personal. You know what? He's right! I hear so much "social, social" this, social that, etc. etc. that I am about to regurgitate it all. I hope I'm not going to walk you into a philosophical (individualist v. collectivist cultures a la Hofstede) debate because it's not my intention, but: I believe the future of the web is individual (with the social component still being subsumed under "individual" because if I enjoy sociality, it's ME that enjoys and therefore allows it, not the other way around). I don't believe in social per se, just as much as I don't believe in individual per se. I guess you can call me a true relativist or practical guy, in that I don't really believe in any regimenting new golden rules. If I let my friends (and invite them to my thousand LinkedIn, Twitter, Facebook, etc. accounts) is because I am the one in control of who am I exposing my sociality to, with whom, and when. I sometimes feel like being social, I sometimes don't.
And another point where Doc makes complete sense: all the conversation and Power Point decks I hear around "social network marketing" from agencies is still about "owning" the user/customer/consumer albeit in a different channel, with different methods. But the mindset is the same. So to me keeping the "social" rhetoric ongoing is a fad. Unless of course we come up with a truly consumer-friendly model that also happens to make money for the biz without "owning" the social consumer.
This is why I believe the next web will be personal, not social (but pray include "social" in the "personal" denotation). At least that's until someone comes up with a better explanation of "social" than the standard PPT as seen on projectors. I almost feel like the "Che" of data, lol.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=3522a576-91f3-46c0-903f-793094b38f61)
And another point where Doc makes complete sense: all the conversation and Power Point decks I hear around "social network marketing" from agencies is still about "owning" the user/customer/consumer albeit in a different channel, with different methods. But the mindset is the same. So to me keeping the "social" rhetoric ongoing is a fad. Unless of course we come up with a truly consumer-friendly model that also happens to make money for the biz without "owning" the social consumer.
This is why I believe the next web will be personal, not social (but pray include "social" in the "personal" denotation). At least that's until someone comes up with a better explanation of "social" than the standard PPT as seen on projectors. I almost feel like the "Che" of data, lol.
I had a brief email conversation with Jamie Taylor, Minister of Information at Metaweb, and he framed the issue of open data from within a behavioral economics perspective: expected utility theory. I never thought of this but had been on the look for a food (or at least better than the pundit blogs that are pro or con on S(s)emantic application methodology or another). I think I have found it, via Jamie, whom I thank for opening my eyes. I will write more on this tonight and attempt to establish the value function and its parameters, because I think being rigorous at this will set us straight on being able to "sell" the 3.0 story in the marketplace. Here I go:
There is a fundamental problem/conflict between user needs (e.g., I want all my data to be mine, portable and across all data providers/sources/sites, etc.) and data owners (big Internet companies, sites, networks, Walmarts, Amazon's, your grocery store, your medical insurance, anyone holding, processing, doing something to the data you leave behind in your behavioral trail of interacting with them). The data owners must be convinced (in a granular fashion, by answering every parameter in the "expected utility function" equation, e.g. size of payout (to data onwers, of opening up their data to developers/users), probability of occurence (of this payout, which is essentially our business models of Web 3.0 app providers), risk aversion (check the panel Business Risks of Web3.0: What Risks? at Web 3.0 Conference & Expo next week), and account for differential utility of the same payout to different companies/verticals with different assets and/or needs.
That is the question that needs answered. If we continue to talk tech mambo-jambo that the market doesn't understand, but don't have pretty damn good answers for the above, we're screwed! The Web 3.0 Conference is an attempt to get us all straight.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=e4060310-4307-4c23-ab9a-bae648417e14)
There is a fundamental problem/conflict between user needs (e.g., I want all my data to be mine, portable and across all data providers/sources/sites, etc.) and data owners (big Internet companies, sites, networks, Walmarts, Amazon's, your grocery store, your medical insurance, anyone holding, processing, doing something to the data you leave behind in your behavioral trail of interacting with them). The data owners must be convinced (in a granular fashion, by answering every parameter in the "expected utility function" equation, e.g. size of payout (to data onwers, of opening up their data to developers/users), probability of occurence (of this payout, which is essentially our business models of Web 3.0 app providers), risk aversion (check the panel Business Risks of Web3.0: What Risks? at Web 3.0 Conference & Expo next week), and account for differential utility of the same payout to different companies/verticals with different assets and/or needs.
That is the question that needs answered. If we continue to talk tech mambo-jambo that the market doesn't understand, but don't have pretty damn good answers for the above, we're screwed! The Web 3.0 Conference is an attempt to get us all straight.
Finally here! This is the blog that will track everything relevant in the semantic web/Web 3.0 world, from both a technology and business utilization perspective. I have been a "student" of this area for quite some time, and am working on a startup project of my own related to this.
What I hope to achieve with this blog is to help create a community around "mainstreaming" Web 3.0. I believe that after 7 years of promises, the marketplace (press, investment community, startups) are just about at an inflection point of larger (than academic projects) adoption. To get a little bit of background context about where I am coming from (Web 3.0 is all about context, right?) read my articles here: Where is the Semantic Web Killer App (Part 1), and here: Where is the Semantic Web Killer App (Part 2), and take a stand whether pro or cons.
Here's how it will all work out: every week I will post 3-4 comments/thoughts about the area, some technological, some "philosophical", others very business application-driven, and would like to hear from you.
Did I also mention that we are "dogfooding" the blog by using Zemanta semantic blog capabilities, that automatically generates metadata from text and is able to hook into open data? This is all about moving from the "newspaper" web to the database web. So, here we go, dive in at your own pace.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=7021df77-e260-4b1f-bfe6-02748344c02b)
What I hope to achieve with this blog is to help create a community around "mainstreaming" Web 3.0. I believe that after 7 years of promises, the marketplace (press, investment community, startups) are just about at an inflection point of larger (than academic projects) adoption. To get a little bit of background context about where I am coming from (Web 3.0 is all about context, right?) read my articles here: Where is the Semantic Web Killer App (Part 1), and here: Where is the Semantic Web Killer App (Part 2), and take a stand whether pro or cons.
Here's how it will all work out: every week I will post 3-4 comments/thoughts about the area, some technological, some "philosophical", others very business application-driven, and would like to hear from you.
Did I also mention that we are "dogfooding" the blog by using Zemanta semantic blog capabilities, that automatically generates metadata from text and is able to hook into open data? This is all about moving from the "newspaper" web to the database web. So, here we go, dive in at your own pace.
Search the site
|
Archives
Internet.com / Blogs
|
Other Links
|
Web3beat marketplace
Jupitermedia Corporation has two divisions: Jupiterimages and JupiterOnlineMedia
Legal Notices, Licensing, Reprints, Permissions, Privacy Policy.
Advertise | Newsletters | Tech Jobs | Shopping | E-mail Offers