Slashdot Mirror

This is nothing new: the FOAF (friend of a friend) protocol with its cousin, the Semantic Web, was bandied about in the '90s

The FOAF project began in 2000, not "the '90s".

I kind of wish that someone would just create a desktop client that could read in FOAF files, look across the people who are in your network, download their status updates directly from them (or from mutual friends) BitTorrent style, and not allow any one entity (corporate or otherwise) to own you or your friends' data by default. Semantic Web forever- viva la revolucion?

The pieces are here- google to find software solutions that already automatize exposing this data in the linked data web (won't shamelessly plug the one I'm working for:)

FOAF http://www.foaf-project.org/
SIOC http://sioc-project.org/

If someone hosting your data space (if it's not yourself) proves to be a jerk, you just take your data and shove it elsewhere.

No walled gardens. No silos.

---
TELINT constitution undercover Perl-RSA Ermes SAPO nitrate Yukon AK-47
CDC BLU-97 A/B enigma Rubin INSCOM pink noise

Obligatory marketing blurb follow: OSN is a shiny new open source open protocol distributed social network. From a user perspective all the individual sites in the OSN federation appear as one. Users can search, browse profiles, send messages, and link to each other without regard to which sites other users are using. S/MIME public key cryptography is used to unambiguously identify senders and is combined with the social network to make the system resilient to spam. Spammers get voted off the island. User profiles are based on the FOAF XML file format and users can migrate their profile from one site to another. OsnLive.com is the first site running OSN.

Integrating data from a large number of sites is the vision of W3C's Semantic Web activity.

A lot of the issues discussed in the document have been solved already. For example a standard data format (RDF), fusing identifiers (OWL's inverseFunctionalProperty mechanis), API issues (SPARQL).

The FOAF project http://www.foaf-project.org/ tries to address the problem of personal data exchange for a few years now.

So why invent a new set of standards and not re-use the standards that are already there?

Too bad FOAF still hasn't taken off, then you could have the benefit of an address book you could reach at any time, but without the disadvantage of keeping it on a single site owned by a corporation that might just disappear down the line.

I believe that this is what semantic web projects like FOAF are designed to do.

There are ideas around.. Check:
http://www.foaf-project.org/
http://openid.net/

That is easily possible... the challenging part is who do we trust to manage such a large database of information?

Sure, we could tie everything in - from SSNs, to birth records, to marriage records, driver records, passports, real estate, phone directories, mailing addresses, employment...

Since I first read about the semantic web a few years back I've been hoping that some of it's ideas would take off. I've created my own FOAF file, but it's only recently that a couple of geek friends have created theirs that I can link to. I know a few social network sites generate these, but they may not be able to link to people outside that network. The project site does not seem to have been updated in a couple of years. There's the usual problem of nobody daring to publish an email address for fear of being spammed, although FOAF caters for SHA-1 checksums to reduce the risks.

I've also played with http://geourl.org/. I see a few sites using it, but, again, the homepage is not being updated.

Cory Doctorow wrote about Metacrap in 2001. Has much changed since then? There is the risk/certainty that companies/spammers will create fake metadata to attract the clicks, but perhaps someone can come up with a trust system to avoid that.

Some social networking sites (e.g. livejournal.com and d.hatena.ne.jp) already provide basic data export in FOAF (Friend-of-a-Friend) vocabulary. Search engines such as Swoogle and SWSE aggregate some of the content published in RDF. The problem is that crawling large database-driven sites with millions of files takes years when adhering to the Robots Exclusion Protocol. On the other hand, an API can provide on-demand integration, but with every site building their own API, a lot of schema wrapping (e.g. via XSLT's) is needed to aggregate data. Vocabularies such as SIOC could provide a standardised API and data format for all sorts of community sites, which would facilitate the integration of data from multiple places.

There's a decentralised RDF-based "Semantic Web"-type version in the form of FOAF. You can already browse it with software like FOAFnaut etc, and generate your own FOAF file with FOAF-a-matic. There was a crawler called Plink, but that seems to be dead now.

The hard work would appear to be done already:
OpenID http://openid.net/
FOAF http://www.foaf-project.org/

There are a bunch of efforts at linking sites and identities together using either HTML or RDF, including XFN and FOAF.

"I AM continually shocked and appalled at the details people voluntarily post online about themselves." So says Jon Callas, chief security officer at PGP, a Silicon Valley-based maker of encryption software.

Sounds like imeem

Or like FOAF, the XML (RDF) dialect for describing social networks that really shows the power of Semantic Web concepts. FOAFNaut is a good example of how by combining these simple RDF descriptions with visualization technologies, one can easily create a easily and pleasantly navigable source of information. (There's some other cool examples in Springer-Verlag's Visualising the Semantic Web ). FOAF files are a cinch to create--there's already a couple of user-friendly generators--and I have no idea why the concept hasn't caught on. Well, Orkut, MySpace, and Friendster obviously won't export FOAF files so they can lock users in.

I am actually sympathetic to the basic idea here: New platform.

I'm newly skeptical of the approach of endlessly creating side-systems on the web browser.

There are amazing things that are possible when you make a new platform for integrating ideas.

For example, we can envision a world where you can watch people writing blog posts as they write them. We can imagine working on documents together with others in real-time. We can imagine social networks, we can imagine shared web browsing. We can imagine going to a web page, and seeing other people who happen to be browsing the web page at the same time as well. We can imagine looking at them, seeing what their affiliations are; There are all these things. We have seen voice communication. Within 10 years, good voice synthesis will be coupled, and we'll be able to look and sound like anybody.

Now, what we haven't seen, even in our imaginations, is all this stuff working together. Integrated into one platform.

Doing this stuff piece-meal, a little bit at a time, on the edge of the network, isn't going to work. It's just not. It'd take forever. Building new standards into the existing network just takes forever. There is no design team. Nadah. Nothing.

Where we see the cool stuff happening, really, is in these large behemeouth new platform.

Now, sure, we can get some milage out of AJAX. We can do sophisticated things with that.

But are we really going to make a 3D world with live document editing, voice & synthesis, presence, infinite versioning on everything, avatars, the whole thing, yadda yadda yadda, using just AJAX? Within 10-15 years? Hell no! It takes at least at least 5 years to make a new specification pretty much standard amongst users. Even RSS aggregators have only 10% penetration amongst blog readers.

What does this mean? It means that a new platform is in the works. Whether you know it or not, a new platform is in the works. Which of the new upstarts is going to be it, remains to be seen.

Sure, sure, sure-- there will be gateways between the world of Vanilla HTML + AJAX into these new worlds.

At some point, you can make a computer render pictures of the new world, and ship them off in AJAX. You can even play Lemmings in the browser now. (Well, you could have...) But the new world is going to be built in the new world. It's not going to be built piecemeal out here in weblandia. When we use browsers to access it, it will be a window into that world, but it will not be that world.

This seems like just another company offering to be the engine for lots of dating sites.

What we really want, if our goal is interoperability, is something similar to the FoaF project's RDF description framework for describing people, then using technology to match them up.

In fact, using something like FoaF, we can describe people in more than the "29 dimensions of compatibility"- we can look at things like interests, where they blog, geography, etc.

WRONG ! Semantic Web expects minimal agreement within communities and domains, for example all camera companies agree on a 'camera ontology' and TV companies create a 'TV ontology', such domain specific ontologies may or may not be linked to a 'master ontology'.
SW is very much out there.. and is already weaved in to the Web of today..

Well again these may sound just 'specifications' and less of an 'ontology'.. then look in to the rapidly growing billion dollar industry.. bio-chem-pharmaco informatics.. ontologies are becoming backbone of their entire computing, data collection and analysis infrastructure..

Whats more..
Flip through last month's Nature Biotech and you ll find articles talking about ontologies, RDF & Semantic Web.. Yes, its already here
Remember, these Biologist are those people who finished the Genome project 2-3yrs earlier than it was orignally planned.. They are very good at collaboration, strong proponents of open-source and very hard workers.. Semantic Web is the right platform for them that gives them tools and a standard to share data seamlessly.. Lets just wait and watch what these people do with it...

AND...yes there's more.. 5 days ago NIH approved a 20million grant to group at Stanford to create a NATIONAL CENTER for BIOMEDICAL ONTOLOGY. Its the same group which developed the only OWL editor (Protege) available out there !
I just hope that those guys at NIH are not fools to give away hard earned tax payers money on something thats not gonna work

Semantic web is not something you can thing of as a concrete application nor we can consider it mature. As you surely read, semantic web is an extention of the current web. So I can link you to firefox or some HTML editor. Joke aside, it is more complicated than that and if you want to embrass semantic web you should get to know XML, RDF and OWL (in this order). In fact, if you are not working to build sw, you should consider another approach. I suggest you to look at RSS there and foaf which are IMHO concrete, but limited, examples of semantic web working examples.

As a web developper... try to generate web pages from RDF (mindswap as some tools) or XML (ala gentoo) source.

People who don't have a clue about semantic web tend to refer about it as semantic horse shit. It's a petty that those who don't believe in things try to demolish them rather than let it go... or let it perish if they are so sure about it's doom.

Everyone checks the gpg signatures right?

Well, sure, but... Can I trust what I'm seeing?

Often, distros rely on that the keyring has been distributed by trusted means. That the keyring hasn't been compromised. But since for PGP to be useful in checking e-mail and stuff, people generally import lots of keys, so this is not the best thing to rely on.

Anybody can generate a keypair "Red Hat Security " or whatever, and sign their trojan with that. Sure, gpg will report that the "update" is unmodified and signed by security@redhat.com. But since you do not know who security@redhat.com is, you can still be duped.

We should be building a huge social network, a PGP-based web of trust (WOT), so that you can actually check if the guy who signed a package is trusted by you or someone else you trust. People need to generate keypairs, go to keysigning parties, take the opportunity to exchange signatures whenever one is out travelling, etc. Go get yourself registered as willing to sign.

The next problem is to decide who you "ownertrust". To extend your WOT to people you haven't met, you assign a "ownertrust" to people, which says something about to what extent you trust them to correctly verify the identity of others.

I think this is rather hard to do these days. I don't know enough people personally to know their key signing habits, if they keep their private key safe, and stuff like that. Such things are important to know if you are to ownertrust.

I have thought about it a bit, and I think it would be nice if one could declare important aspects of one's policy in such a way people could easily find the policy when going through their keyring, to set ownertrusts.

Say, one could for example use FOAF to say things like "I only sign people I meet face to face after carefully checking their photo IDs and having them respond to an encrypted e-mail" and "I keep my private key on a networked computer that I control", "my passphrase is a mangled 20+ letter string" etc.

gpg --update-trustdb

The problem with this approach is of course that people can state a policy they don't follow, but the non-personal WOT is really built on signatures, so I don't think that is a problem.

What do people thing of this idea?

Go see the Friend of a Friend (FOAF) project it is a live implementation of what we can think of the first layer of the Semantic Web.

You have to consider that we will not see a "final" version soon. And the Trust layer is on top of pyramid. So wait for a couple of years... :)

I predict Semantic Web will have a greater impact than the Internet has now. From individuals to business to software and AI, let the revolution begin...

The same can be said about any semantic web technology - whether it's FOAF (an RDF vocab for describing people and their interests) or a vocabulary for reviews. As soon as major authoring tools (i.e. both web editors and content management systems) start integrating these technologies, people will use them if they are useful. Do not expect web designers or bloggers to have a clue about all the great things that the semantic web can do - give them one useful thing which they understand, package it in a pretty UI, and they will start using it.

Marking up Hilton as <motel> or <celebrity> is all very well. This is what XML is for.

One of the key points behind the semantic web is to define meanings to your meta tags. My system has a <partnumber> tag and so does yours, but that doesn't mean they're the same. I can publish my definition of <partnumber> so that other apps can know how to interpret my partnumbers. Complex definitions can be provided in computer-readable format, which can then be looked up, referenced, shared etc. with other systems.

Take Dublin Core, for example. A standard set of tags to describe document attributes, such as title and author. Why should I write my own <author> tag when I can simply pull-in part of Dublin Core's vocabulary. Not only does that save me (the developer) time, but it means any app that knows about Dublin Core will know what I mean when I say "author". Or, if an app doesn't know about a particular term it can simply go look it up.

Sharing vocabularies is time-saving, but also helps computers process information automatically. Mr Berners-Lee and some colleagues had a good article published in Scientific American a while ago which explains their vision of intelligent software agents doing the sorts of things computers should be doing with the information the web has to offer. Such as automatically adjusting your schedule if your gym's online timetable has changed and your squash game needs to be moved. OK, that's a very basic example, but the point is that although the information needed to do this sort of stuff is already on the web, it is currently only readable by humans.

If anyone is interested in learning more about this stuff then have a look at the Resource Description Framework (RDF) which is a foundation technology of the Semantic Web (There's more to it than HTML META tags!). There's a lot of activity involving RDF-based technologies such as OWL, FOAF and the popular RSS.

The WWW also has Annotea, to allow for people to submit annotations. Now, you can imagine lots of people having a simple way to rate pages, a rating option could for example be "Supplied metadata are bad/fraudulent", or something like that.

You would first and foremost make decisions based on ratings from people you trust. That is, people who are close to you in your FOAF-based social network.

When every Internet user becomes a reviewer, and people are well connected in a social network, so that there is a review available of most pages, there is going to be a very strong incentive for authors to supply accurate metadata. Think of it as moderation.

Face it, allthough it happens that you stumble upon pr0n involuntarily, the vast majority of pr0n surfers do it on purpose. Pr0n0graphers (this is getting a bit too leet for me...) then will have strong incentive to refrain from such tactics, they will be modded into oblivion anyway, and accurate metadata is going to bring them traffic, since they are modded up by those who actually surf pr0n.

So, unless the goatse guy is a friend of yours, I don't think it is a big worry.

Provided SW becomes a reality that is.

FOAF is a really good start, though, go create it now!

I'd like to see something like FOAF used for whitelisting. I posted a SA bug about it, and then there's things like Trust and Reputation in Web Based Social Networks . I think this looks like a workable approach.

The reason, I would guess, that Google is interested in social network based systems is because they are looking at the Semantic Web idea (or at least some of the technologies).

FOAF is an effort by Dan Brickley/Libby Miller/Many others to create an open way to describe people - the world wide FOAF as opposed to the WWW. It shows the power of RDF in a practical (and fun!) way.

Now, Lets say I'm Google. I spend millions on maintaining my marketshare with R and D; I'm going to realise sooner or later that gossip is a powerful communication tool. Friends talk to friends. If your friend gives you something, you are far more likely to accept it. If you make it so you can only get it from your friends, its a commodity.
How do I leverage cutting edge ideas and attract users?
and *bam* there we have it.

Furthermore, SPF enables domain reputation systems such as GOSSiP (currently under design) which enable domain's to be given a "spaminess" score based on their previous behaviour.

That's interesting! I'd like to plug two bug reports of mine (I wish I had time to hack, but I haven't). Friend-of-a-friend makes great start for a reputation system, at least for whitelisting people you know well.

So, I there's a Spamassassin bug on this, and it has generated some interest.

Now, the problem is to generate FOAF-records easily and reliably, and for that, I suggest for example enabling KAddressbook to export them.

FOAF:

The Friend of a Friend (FOAF) project is about creating a Web of machine-readable homepages describing people, the links between them and the things they create and do.

20:21 Central
While we scramble behind the scenes to put things back together, we'll share the slashdot love and link to other sites where you can get more information about FOAF.

FOAF Info:

• rdfweb.org
• foaf-project.org
• The FOAF FAQ (from rdfweb)

FOAF Tools:

• FOAF-A-Matic
• FOAFnaut (needs SVG plugin)
• FOAF Explorer

Looks like I picked the wrong week to stop sniffing glue.

19:14 Central
Oh hell. We got slashdotted. And the main site wasn't even running the current code revision. Back in a bit.

FOAF (which PeopleAggregator uses, as do many other sites) and it's relation to XFN are discussed here, and in depth by Leigh Dodds. It would be pretty hard to make something like PeopleAggregator using XFN since it's concerned with typing relations, not describing people. FOAF and XFN don't really compete. (btw, FOAF came first)

FOAF, for the enlightened ones among us who appreciate RDF's directed graph model

In the Wiki world, we've been thinking about ideas such as having Local Names.

In Wiki, you can name a page just by putting "[[ ]]" marks around it, and it links to the page. Recent advances such as the NearLink have made it so that you can refer to pages on "nearby" wiki, even without naming the wiki. If the word you are linking to isn't defined on the immediate wiki, but it is defined on a near wiki, then the word links to it's definition on that nearby wiki.

But we're carrying the concept even further. With Local Names, we want to be able to link not just to wiki pages, but any sort of page. For example, you could bind [[Slashdot]] to http://slashdot.org/ .

But wait! There's more! We want to store these bindings in a "Local Names Server", which you could then tell people about, or store in your person preferences server, or a FOAF file. Then, when you post to a website, or slashdot, or whatever, and refer to something that it doesn't know about, it can look it up in your personal local names server. Of course, Slashdot would have to know what local name servers are, and would have to know to look at them.

At the end of the day, what you effectively have, is a world without URL's- just lots of local names. You'd have a mechanism for "picking up" and "giving away" local names. So, for example, if someone refers to something by a name, and you like it, you can "pick it up" into your own local names server. There are all sorts of possibilities here.

FOAF is an open XML/RDF standard for describing these social networks, it seems like that would be a good way to implement this. Plus, since it uses SHA1 sums of email addresses it would be possible to check addresses without giving them up to spammers.

A lot of sites like Tribe.net and my own project SongBuddy are working on integrating FOAF into the site, so that you won't have to worry about the mechanics of it unless you want to. Seems like an easy way to build these kind of white lists.

If you haven't already, you should take a look at the Friend of a Friend project.

While it does define a simple "knows" relationship for bootstrapping, it's all about implicit relationships like coauthorship, codepictions and colocations.

The w3c also has a list of projects that use RDF. Some of them seem a bit academic, but one that looks particularly cool is eventSherpa - a semantic calendaring application that lets you publish and subscribe to RDF calendars. The FOAF project has also been gaining steam as Typepad and others join the movement.

Hey, we should be over of this 'i run this service' mania. FOAF project is way more interesting and it's open - everybody can specify their friends. without invitations. viva la open web!

I have gone on to create a basic FOAF-file. That's Semantic Web vaporware project. Well, it's not vaporware, since it works, and it is pretty cool. There's even SlashFoaf, so create your FOAF RDF today!

Tribe
No download, runs anywhere. More nerdy, uemphasis on freedom of use, discussion groups. Supports lots of pictures.

MySpace
No download, runs anywhere. Supports restricted blogs, popularity contests, 10 pics. Does not emphasize actual RL friendship dynamics.

Friend of a Friend
Open standard for creting friendster-like network apps. Used by PeepAgg to build OSS system.

There are more, and I'd love to see replies with links to this rapidly growing class of services/apps, with brief descriptions attached. Thanks

Picture sending is definitely possible (I've received/sent pics to/from others plenty of times), most Bluetooth phones tend to be higher end devices with cameras anyway, so a picture of the current environment is trivial to send.

As for social networking, that's already being looked at, the Friend-of-a-Friend project people (social networking based on open format XML/RDF files) are already looking at FoafMobile with Bluetooth being a major component.

Furthermore, using PGP, trust values could be assigned to the information.

- Serge Wroclawski

Perhaps it's just me, but the graphics on the homepage of The Friend of a Friend (FOAF) project(linked to from the review), actually makes me wonder if these so called friends have other reasons for describing us and our webpages. ;)

• Foaf integration
  
• Trackbacks
• Books, Movies, Music, Games
  
• Blogshares shortcut for adding your blog into a fantasy stock market
  
• GeoURL for adding lat/long and adding your blog into world database
  
• Creative Commons licensing on content so you can retain rights while promoting the CC movement
  
• Yahoo Groups features: Msg Boards, Calendar, Chat, Address Books
  
• The ability to create contests:
  • Football Pools
    
  • Baby Pools
    
  • Square Pools
    
  • Dead Pools
    
  • Bracket Pools