OpenTopics
So Mary Hodder and I have been discussing “how could you do Technorati tags” = open.
Apparently it is NOT required to use the Technorati domain in your tags - but that’s something people haven’t caught onto and since the first generation of tagging ‘plug-ins’ went out without that feature - needless to say almost everyone IS now directly contributing to Technorati’s link flow.
Now I really don’t care about link flow, traffic, or any of that shit - but I can see why others might. I myself really APPRECIATE Technorati’s contriobution to teh community - since I’m a folksonomies nut - and it was nice to see Flickr glued in there - but “where’s BuzzNet? or Fotolog.net?”
And before I go on about what OpenTopics could be - here are a few comments from some Technorati peeps on my post on ‘Why Tags matter’ (responding to the David Weinberger post of the same name):
From from someone named Seth Russel:
I totally agree. We should find some way to eliminate the domain name from our tags. A tag is a tag is a tag, is the same tag, regardless if it is tagged at technocarti or del.icio.us or fliker or whatever. There is a simple way to do this actually … i keep trying to write a white paper about it. But in a nutshell all it requires is that the tagging services share RSS feeds. Thanks for bringing this up !
Posted by: Seth Russell at March 3, 2005 06:01 AM
Technorati is a collector of tags, but you are not required to use the Technorati domain in your links to participate.
You can link to the Apple iPod page, the Wikipedia iPod page, or the Technorati page for the iPod tag. Each link is meant to provide a relevant destination for your categorization link. You choose the destination, Technorati aggregates the conversations, and the topics are open.
Technorati also treats the category and dc:subject elements within your RSS and Atom feeds as a post tag.
The rel value of “tag” is a W3C and IETF proposal that predates Technorati’s evangelism of the usage in January.
Placing tags in your posts is decentralized tagging, open to everyone to share on their own servers and their own weblogs, wikis, etc. and open to personal definition. I believe OpenTopics is already here and content publishers are just beginning to find interesting ways to use the feature and create new ways for others to discover related content.
Posted by: Niall Kennedy at March 3, 2005 08:45 AM
Thanks, Niall, i learn something every day. But i still think that context servers like Technorati should share tags and items such that items intentionally tagged at one server would also be also tagged at other servers.
OK so first - thank you Seth and Niall - its nice to hear from Technorati peeps what’s up with all that. But as I stated above - it’s really too late to tell folks about that - if they’ve already plugged in their plug-ins and are using the system that way.
Meanwhile - so I was a Matt Mower LiveTopics user, loved what Matt and Paolo did with ENT and also LOVED eVector’s k-collector product.
So how can we take those ideas - get folks using the ENT namespace (instead of some RESTful technique like Technorati nows persues) and create a truely open topic exchange. Well maybe not INSTEAD of a RESTful technique - how ’bout IN ADDITION to the RESTful technique.
That’s one coolio thing about Flickr’s APIs - they support XML-RPC, SOAP and REST techniques. So that’s TRUELY the open, agnostic approach - support all three.
Oh yah - let’s not forget Mr. Phil “I hacked it up over the weekend’ Pearson. Not many people give credit for the ‘link’ ranking lick - which Phil’s blogging ecosystem did six months BEFORE Technorati. But Phil also did something called the Internet Topic Exchange - which was the first blogging aggregator channel around topics.
So anyway all ego and credit aside - Mary and I have been trying to figure out how to get this to work. If any of you wanna participate in such an effort contact me or Mary.
Thanks.
We already have category and dc:subject in RSS feeds. Why do we need ENT? I guess I’m missing something.
“Each cloud has a URI which may point at a resource containing a complete description of the topics used in the cloud.” A folksonomy probably doesn’t have this.
We already have category and dc:subject in RSS feeds. Why do we need ENT? I guess I’m missing something.
“Each cloud has a URI which may point at a resource containing a complete description of the topics used in the cloud.” A folksonomy probably doesn’t have this.
We already have category and dc:subject in RSS feeds. Why do we need ENT? I guess I’m missing something.
“Each cloud has a URI which may point at a resource containing a complete description of the topics used in the cloud.” A folksonomy probably doesn’t have this.
In a way i agree with Julian that the topic string in the RSS (catagory, dc:subject, whatever) should give us all we need. But I am still studying the ENT standard, maybe i am missing something. For me the whole point of Folksonomy is that when we try to nail the metadata to the wall of exact precision, we end up loosing more than we gain. One solution is that when we want a more exact catagory, we just make a longer more unique string, example: FolksonomyContextProject. Check out my posting about this yesterday http://can-you-hear-me.org/?p=18. Now it just might be that your tag OpenTopics refers to the same thing as my tag FolksonomyContextProject …. err time will tell. Anyway we dont’t need no stinking domain names in the topics, and that is gonna be my point in this white paper that i wish i could finish.
In a way i agree with Julian that the topic string in the RSS (catagory, dc:subject, whatever) should give us all we need. But I am still studying the ENT standard, maybe i am missing something. For me the whole point of Folksonomy is that when we try to nail the metadata to the wall of exact precision, we end up loosing more than we gain. One solution is that when we want a more exact catagory, we just make a longer more unique string, example: FolksonomyContextProject. Check out my posting about this yesterday http://can-you-hear-me.org/?p=18. Now it just might be that your tag OpenTopics refers to the same thing as my tag FolksonomyContextProject …. err time will tell. Anyway we dont’t need no stinking domain names in the topics, and that is gonna be my point in this white paper that i wish i could finish.
In a way i agree with Julian that the topic string in the RSS (catagory, dc:subject, whatever) should give us all we need. But I am still studying the ENT standard, maybe i am missing something. For me the whole point of Folksonomy is that when we try to nail the metadata to the wall of exact precision, we end up loosing more than we gain. One solution is that when we want a more exact catagory, we just make a longer more unique string, example: FolksonomyContextProject. Check out my posting about this yesterday http://can-you-hear-me.org/?p=18. Now it just might be that your tag OpenTopics refers to the same thing as my tag FolksonomyContextProject …. err time will tell. Anyway we dont’t need no stinking domain names in the topics, and that is gonna be my point in this white paper that i wish i could finish.
Julian, the URI of the “cloud” about this from my point of view is http://del.icio.us/canyouhearme/FolksonomyContextProject and the golbal uri is http://del.icio.us/tag/folksonomycontextproject . If the meme spreads, then I wont be the only one posting to this context. Now how do we factor the stupid domain name out of the identifying string?
Julian, the URI of the “cloud” about this from my point of view is http://del.icio.us/canyouhearme/FolksonomyContextProject and the golbal uri is http://del.icio.us/tag/folksonomycontextproject . If the meme spreads, then I wont be the only one posting to this context. Now how do we factor the stupid domain name out of the identifying string?
Julian, the URI of the “cloud” about this from my point of view is http://del.icio.us/canyouhearme/FolksonomyContextProject and the golbal uri is http://del.icio.us/tag/folksonomycontextproject . If the meme spreads, then I wont be the only one posting to this context. Now how do we factor the stupid domain name out of the identifying string?