Regarding match detection, I agree that there is not much point in reinventing the wheel but, at the same time, I'm not ready to call search a solved problem. Any time you partner with a third party search, as I found out using other products, you share the limitations they have. There's good and bad to that approach though, usually the good does outweigh the bad.

Resolution Assistance is a tough art in general. This is one thing I'll be looking at closely. I have a pretty big virtual roledex of DMCA agents that I've compiled over the years. If this can be worked out and automated, it will be worth almost anyprice.

As far as speed goes, I think the main goal right now is to be faster than doing it by hand and, barring any major server issues, It think you will be that. However, I get nervous when I hear about people automating resolution efforts. That is how you get problems such as the YouTube debacles and the recent AP Drudge Retort controversy. I guess I'm just asking that you move with caution into that area.

Finally, I agree that links and licenses are more valuable. The only issue right now is that there is no legal system. With my personal resolution efforts, my link request efforts have averaged about 50% resolution, DMCA about 95%.

Hope that helps!

Regarding match detection, I agree that there is not much point in reinventing the wheel but, at the same time, I'm not ready to call search a solved problem. Any time you partner with a third party search, as I found out using other products, you share the limitations they have. There's good and bad to that approach though, usually the good does outweigh the bad.

Resolution Assistance is a tough art in general. This is one thing I'll be looking at closely. I have a pretty big virtual roledex of DMCA agents that I've compiled over the years. If this can be worked out and automated, it will be worth almost anyprice.

As far as speed goes, I think the main goal right now is to be faster than doing it by hand and, barring any major server issues, It think you will be that. However, I get nervous when I hear about people automating resolution efforts. That is how you get problems such as the YouTube debacles and the recent AP Drudge Retort controversy. I guess I'm just asking that you move with caution into that area.

Finally, I agree that links and licenses are more valuable. The only issue right now is that there is no legal system. With my personal resolution efforts, my link request efforts have averaged about 50% resolution, DMCA about 95%.

Hope that helps!

Regarding match detection, I agree that there is not much point in reinventing the wheel but, at the same time, I'm not ready to call search a solved problem. Any time you partner with a third party search, as I found out using other products, you share the limitations they have. There's good and bad to that approach though, usually the good does outweigh the bad.

Resolution Assistance is a tough art in general. This is one thing I'll be looking at closely. I have a pretty big virtual roledex of DMCA agents that I've compiled over the years. If this can be worked out and automated, it will be worth almost anyprice.

As far as speed goes, I think the main goal right now is to be faster than doing it by hand and, barring any major server issues, It think you will be that. However, I get nervous when I hear about people automating resolution efforts. That is how you get problems such as the YouTube debacles and the recent AP Drudge Retort controversy. I guess I'm just asking that you move with caution into that area.

Finally, I agree that links and licenses are more valuable. The only issue right now is that there is no legal system. With my personal resolution efforts, my link request efforts have averaged about 50% resolution, DMCA about 95%.

Hope that helps!

Match Detection — we do our own “fingerprinting” of the content. We do use a major search engine to find matches. No need to reinvent the wheel. The big search engines have indexed more pages and have better spiders than we could build.

Resolution Assistance — i think Discovery really shines here. It captures various points of contact for the site and allows notices to be sent to some or all of these contacts. Discovery will find the right people. At a minimum, it will find the host ISP and serve them.

Speed/Usability — the speed of identifying matches and sending redresses and following up to see if the site took the required action is very good. Where Discovery could use some improvement is doing this automatically so that the publisher does not have to review and act on each suspect individually. We are working on letting the publisher pre-define rules and policies for letting Discovery ID the sites, send redresses and tahe escalation action when appropriate, without human intervention.

The objective of Discovery is to verify legitimate users and to identify non-legit users so that they become legitimate users. It's not as much about getting sites to stop using content — although Discovery can do that. It's about enabling sites to use content in a way that compensates the publisher, gives them credit and brings them new traffic. A license or a link action is more valuable than a take-down action!