Here's the dilemma. These are all great ideas for how to promote consensus-based decision-making but we need ways to tie discussion to specific research tasks. The PTO wants to do this project because we (the community) are going to supply them with research and analysis of how that research relates to the claims of the pending application. Hence - they don't simply want comments or discussion, especially not an H20 discussion where people are not directly responding to one another and the specific research. Comments that do not directly relate to the assessment of the "novelty" and "non- obviousness" of the claims, in fact, will be redacted (not by us but by the PTO). So what we want to do is to get people focused on uploading prior art and assessing that art's relevance to the claims. The uploaded prior art and annotations on that art (as distinct from the general discussion) is what gets forwarded. It is very reasonable for the agency to want that information that helps it make a better decision and this is an experiment in figuring out if we can do that. Best, Beth On Apr 9, 2007, at 8:26 PM, Luis Villa wrote: > On 4/9/07, Beth Noveck <noveck@xxxxxxxxx> wrote: >> Luis, >> >> While we're distracting you from studying, could you describe the >> collaborative comment submission process in a bit more detail? The >> whole point of the project is to push towards collaborative research >> and we're on the look out for good, specific ideas for how to do >> this. > > It is just a fuzzy idea. What you want to have, I think, is a > mechanism by which people can have a short, rational discussion about > something (prior art, interpretation of the language, etc.), and > indicate to the PTO that the resulting comment is the agreed-upon work > of multiple people, instead of just existing in isolation. > > For the conversation, I might draw inspiration from Berkman's H2O[1] > and have discussion that is perhaps more structured than a regular > discussion board thread. In H2O, comments are held so that each > commenter gets a set period of time (usually a day to a week) to make > one and only one comment, and then a second 'round' of comments can be > made a set period of time later. This encourages longer, more > thoughtful responses, since you only get one of them per 'round', and > since you can't post them in anger (since they don't go public until > everyone else's comments do, and you can edit them until then.) > > Alternatively, if the statements to be submitted to the PTO are very > short and fact-heavy, you might look into using a wiki-like structure > for creating the comments, with a comment/discussion thread > per-comment. Wiki generally works poorly for opinion, but it might > work well for something like creating a neutral-ish statement of fact > about prior art- wiki is pretty solid at creating descriptions of old > technology, for example. Again here, the more contributors to the > final statement, perhaps the more weight given to the comment when > presented to the PTO. > > I might also draw inspiration from slashdot's moderation, or apache's > community discussion standards. Instead of requiring a textual > comment, allow commenters to simply attach a '+1- Agree' or '+1- > Extremely Relevant' (something like that) to comments which they find > interesting. Comments with more +1s are more relevant. In apache, you > can +1 as often as you want; in slashdot you have only a limited > number. Either way, you've provided a way for people to very concisely > provide more information. At least in theory, something with a large > number of +1s is likely to be more fruitful for the PTO to spend time > on than something with no +1s. This could work together or > independently of the other two ideas; e.g., people who did not > participate in a wiki or H2O-like discussion might still +1 or -1 the > results of those discussions. > > I feel like I've missed out on too much of the design discussion for > these to get too much more concrete, but I hope they help stimulate > some discussion. I'd be happy to answer more questions about the H2O > stuff in particular, since I was briefly involved in that project, and > I realize it is a bit hard to wrap your head around. I do encourage > folks to create accounts on H2O and experiment a bit so that they can > see what it is like- I really do think the 'staggered' conversation > model it uses might really be of use in this context. > > HTH- > Luis > > [1] http://h2o.law.harvard.edu/index.jsp _______________________________________________ p2patent-developer mailing list p2patent-developer@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/p2patent-developer
