Joi Ito's Web

Joi Ito's conversation with the living web.

Karl-Friedrich Lenz and others are dumping the Creative Commons license because they don't like the fact that the original licensor guarantees to the licensees. If I understand this correctly, it means that if I snap a photo of something, someone copies it from my site and posts it and get sued, it comes back to bite me.

Me sense is that this is the way it should be. Tell me if I'm missing something.

Here's my view.

Copyright is not some clearly defined law. It is law and technology that is the codification of what society thinks is "fair use", "free speech" and many other things at the time. Hollywood and a lot of other people are pushing to limit things like free speech and fair use. Someone has to fight for it. There have to be suits, people have to get mad and we have to fight.

If you try to pass the risk to service providers and corporations, they'll just say, "we don't want to be the next Napster," and will most likely prohibit sharing and posting rather than take the risk. What the CC license does is allow the service provider to say, hey, all of the stuff here is tagged with copyright info and guaranteed by each of the people. Sue, them, not us.

For the individual who gets sued, yes, it's tought and we have to fight to help these folks (that's what the EFF is for), but having the individual who originally posts the stuff be responsible does a few things. It distributes the risk. It's much more likely for Hollywood to sue a big company with money than an individual. It will get people thinking about and fighting for their right to fair use and free speech.

I'm sure there will be a chilling effect and this chilling effect might be non-trivial, but my sense is that the chilling effect on service providers who have to "police content" will be so quick that we'll have restrictions on our free speech via technology faster than you can say, "Mickey Mouse."

So what does this mean to Joe CC user? When you are about to post a picture or a quote that is questionable from a free speech or fair use perspective, be prepared to defend yourself. But, as the Chicago Manual of Style says:

The Chicago Manual of Style
The right of fair use is a valuable one to scholarship, and it should not be allowed to decay through the failure of scholars to employ it boldly. Furthermore, excessive caution can be dangerous if the copyright owner proves uncooperative. Far from establishing good faith and protecting the author from suit or unreasonable demands, a permission request may have just the opposite effect. The act of seeking permission establishes that the author feels permission is needed, and the tacit admission may be damaging to the author's cause.

OK I figured out my next python project. A bot that polls technorati for inbound links to my blog and sends them to me as they come in via IM (Jabber). This is probably going to challenge my programming skills, especially since I know nothing about Jabber... yet. So, if someone is already working on this or has already done this. Stop me now!

Also, if you know of any relevant python modules I should look at, that would be greatly appreciated. I plan to try out Mark Pilgrim's PyTechnorati.

I added a new style sheet and a new style sheet class. On my sidebar, if you click on humble style, it will set your style sheet to a style sheet based on the polite fonts style sheet (big fonts) where I've added a new class called "boastful". This style sheet will render text that I think might seem boastful in a tiny white font so you don't see them. (I couldn't figure out a way to delete them all together in css.) So if you want to stop seeing my boastful disclaimers and generally cut down on boastful comments altogether, change your stylesheet please.

Also, I ripped off an idea I saw on Liz's site and made my archives in my sidebar pulldown forms.

sputnik.jpgI got my Sputnik. It rocks. It rocks in many ways. It rocks because it uses Jabber to talk to the Central Server (The Central Server is a machine on your network that manages the access points. If you don't want to set up your own Central Server, you can use the community server. You log into the Central Server to access the AP you're using. Similar to the web page that pops up when you first log into WiFi networks in Hotel, except this one is yours.). It rocks because the Central Server controls and manages access and security for all of the access points across the network (even WAN). It rocks because Sifry is giving me real time tech support via... Jabber. (It's amazing that Sifry could build Sputnik AND Technorati at the same time. He was giving me tech support for both at the same time...) It rocks because it's secure. It rocks because it lets me do port forwarding, firewalls and peering for only $185. Anyway, I'm supposed to get my second AP and my own Central Server soon. If it all works out, I'm going to buy a bunch of AP's and try setting up my own little network of Sputnik nodes.

Disclaimer: This is a totally un-scientific review of a product by a friend ;-). I have only tried a few AP's and don't have much empirical evidence to support my notion that Sputniks "rock". I've used MELCO, Linksys and Apple AirPorts. I do think the central server and the Jabber thing are quite unique and seem to be good ideas. We'll see how I feel after I finish building out a network of Spuniks.

Good entry in Web Dawn about Technorati as a reputation system. Via Marc Canter. So we need to get Sifry to catch permalinks of what people are linking to better and we need a little more metadata in the XML feeds. The addition metadata which would allow me to implement something like a reputation system would be:

1) The name of the person who posted the entry
2) The URL (permalink) of the entry on my blog the link points to
3) Filtering out of all blogroll "false positives" and entry imbedded links only or a separate feed for blogs rolls. (Can you work with Jason D on this?)
4) Implement pagerank. Take the inbound links/blogs assign the blogs with high inbound links/blogs with high page rank and then do another calculation assigning outbound links from those blogs higher value and give each page a technorati ranking. And do this every 5 minutes.
5) Make a contextual pagerank where the user can set their own rankings for blogs and ask technorati to calculate the final pagerank.

It would be cool if 4 & 5 could be done on your own site in some way, but not sure how that would work. Anyway, a bit off the top of my head, but I'm trying to figure out my NEXT super duper python project.

Lucky for me Aaron just wrote xmltramp and made my life a whole lot easier.

Aaron Swartz
In trying to write some code to use the new Technorati API, I noticed that all the tools for accessing XML documents sucked. So I wrote my own: xmltramp. It makes handling XML documents in Python a piece of cake:
Via Cory

By the way, Dave Sifry and Aaron Swartz rock.