New UDRP Search Tool
Date: Wednesday April 23 2003, @05:08PM
Topic: Uniform Dispute Resolution Policy (UDRP)

Intellectual Property Notes reports that, "Four foundations, including The Cornell Legal Information Institute and The Markle Foundation, have created a new search engine that can locate decisions from all domain name dispute resolution providers. Currently, the search engine database holds 4,090 cases, but the developers plan to have all decided disputes accessible within a few weeks. Searches may be conducted by domain name, complainant, respondent, trademark, or panelist. Users may also specify further information, such as whether the panel determined that respondent had no rights in the disputed domain name, or whether the panel found that the purpose of registration was to disrupt the business of a competitor. Users also have the option of performing a quick search, which finds near matches if the user provides a misspelled domain name. "

This looks like it will be a really useful tool....

This discussion has been archived. No new comments can be posted.
New UDRP Search Tool | Log in/Create an Account | Top | 4 comments | Search Discussion
Click this button to post a comment to this story
The options below will change how the comments display
Threshold:
Check box to change your default comment view
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Errors
by vbertola on Thursday April 24 2003, @12:29AM (#11525)
User #3435 Info | http://bertola.eu.org/
Great... though the homepage of the site currently shows:

Current Decisions
in Database:

ERROR /usr/local/udrp/phplib/udrp.inc at line 46 SELECT COUNT(CaseID) FROM Cases: Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (2)

I would suggest these people to consider publishing the home page statically every hour or so, to avoid a database hit each time the home page is loaded... and to avoid to have nice PHP errors on the home if the database goes down. (And perhaps you could disable display_errors in your php.ini)

Apart from technicalities... I think such a database is a great idea - it will help in monitoring the general trend of UDRP decisions, and in allowing their public review.

--vb. (Vittorio Bertola)

[ Reply to This | Parent ]
Not bad, but incomplete
by Mueller ({mueller} {at} {syr.edu}) on Thursday April 24 2003, @02:06PM (#11528)
User #2901 Info | http://istweb.syr.edu/~mueller/
Hey, it works! I just spent half an hour playing with it.

As someone who put together a prototype for this effort I had fun checking this out, comparing it
with the one we did.
[syr.edu] This one has strengths and weaknesses relative to the prototype.

Strengths:

I thought our user interface was much clearer, simpler and easier to navigate, but this one is much more powerful. It enables Boolean searches and wildcards under most fields (under "Advanced Search").

It has a more extensive and law-focused way of classifying Respondent interest and bad faith

It tells you whether the domain name was used or not (this data element seems to be incomplete, however).

It is online rather than downloadable. This improves accessibility. but has weaknesses: slower performance and you can't get into the database and crunch numbers (mainly a concern of social scientists, I guess).

Weaknesses

The real key to the usefulness of any of these databases, as Ethan and his colleagues know well, is current, accurate data. New cases need to be put in quickly, and all cases should be in. But the UDRP-DB data is not complete - dates and some categorizations are missing from many cases. More important, it is not clear how their selection of cases is bounded - what cases are in there and which ones are missing? Nowhere on the web site does it tell you how many cases are in there and which ones they are (e.g., from dates xx through yy).

What I like least is that after a search records are listed by domain names and not by case. Thus, a case that has 10 domain names in it will be listed 10 times in a hit list. We learned that users want to see case lists, not domain name lists.

From a graphic/navigation standpoint, the hit list is ugly and confusing. For God's sake, even the online NAF database doesn't use all those silly zeros in its case numbers, why should we?
Some case IDs are capitalized, some are not; the colors are hard to read. The hit list order is not really based on decision date, but some other date, and we don't know where it comes from.

You can't search by defaulted cases. The summary sheet does not show the names of all the panelists.

Regarding trademark categories, I cannot figure out how these cases were categorized. If you search for "geographic indicators," you get only 5 hits. The famous newzealand.com case doesn't pop up (but it's in there). There are nearly 50 of such cases at least. So either the data is incomplete (many cases have not been categorized yet) or worse, wrongly categorized.
 
All of the cavils above can be fixed over time, but the most serious one is completeness of data. You can't use this as a research tool at all unless one of two conditions applies:

1) All the cases are in, or 2) A known, bounded set of the cases are in. Until that happens, this must be considered Beta.

Anyway, the people we really need to hear from are the lawyers who do UDRP cases.
[ Reply to This | Parent ]
a little patience
by Ethan on Sunday April 27 2003, @09:08AM (#11533)
User #3418 Info
I invite you to try this new tool but would ask that you wait until Wednesday April 30th), which was the date we were intending to make it public. Some of what Milton doesn't like now will still be an issue then but there are also some bugs and inaccuracies will have been removed by then.

Whether you like this or not or find it useful or not, I would encourage you to think of how primitively the UDRP process uses informational tools which we now have readily available. The UDRP requires that decisions be published simply as whole files. I applaud Milton and others who have done empirical work on UDRP cases but the amount of research that has been done has been hampered by the lack of good tools to analyze cases. If there is ever version 2 of the UDRP, I would hope that the people who know something about information retrieval would have some input.
[ Reply to This | Parent ]
  • 1 reply beneath your current threshold.




  • This article comes from ICANNWatch
    http://www.icannwatch.org/

    The URL for this story is:
    http://www.icannwatch.org/article.pl?sid=03/04/24/0115210