Dear all-
Posting from Tom Bruce, Director, Legal Information Institute,
Cornell Law School.
--Sally Wise
________________________________________________________________
Folks:
I'd post this to LAWLIB, but I'm not a subscriber, so I'll rely on one
of you to cross-post. I figured I could reach a couple hundred of my
favorite law librarians this way, for starters.
As most of you know -- and as we don't talk about much here, especially
right after eating -- the currency of our US Code collection is a
disappointment. The official version that we get from the House Law
Revision Counsel's Office can drift away from current legislation by as
much as 18 months. And we're stuck with that, having neither a better
free version to draw on as raw material nor an editorial staff of 50 to
do "pre-codification" that would keep things fully up to date. Real
seismic shifts in the Code -- the recent changes in bankruptcy law, eg.,
or the establishment of the Department of Homeland Security -- catch us
out and cause a flood of snippy e-mail from individuals who don't
understand why a service to which they make no financial contribution
wouldn't be using its 50-person editorial staff to monitor these things
more closely.
And we should, or at least offer an update feature that works decently.
Most of you also know that we've taken a number of cracks at that and
while they've steadily improved they're still fragile and less capable
than we would like. Hence this note, which contains a brace of
questions:
a) First, let's distinguish between "updating" a section of the Code and
"providing an alert" with respect to a particular section of the Code.
For my purposes here "updating" means finding not only that a section
has changed but also viewing/obtaining the text of the changed section
(or perhaps a textual laundry list of changes that would allow me to
mentally transform the old to the new). "Providing an alert" means
determining that there is some greater or lesser likelihood that a
particular section has recently changed or is about to change -- and
can, perhaps, err in favor of overinclusiveness. There might be such a
thing as an acceptable level of false alarms, who knows. In the case of
an alert, we need not know what the change will be, exactly -- just that
one is likely to occur or to have recently occurred.
b) The questions, then, are two (or maybe four): What's the best way of
doing each of these, using exclusively public-domain, free-access
electronic sources? "Best" might mean "most accurate/authoritative" or
it might mean "most current". If these are two different methods,
please outline each. We'd also like to look at pending legislation that
has a reasonable possibility of passage (how one assigns probability to
a bill prior to enrollment is no doubt an interesting question, and
solutions that involve an "apply Windex to crystal ball" step will be
greeted with an appreciative chuckle).
I'm looking here for a series of cookbook steps that a human researcher
might use. We'll think about machines later.
Many thanks for your help,
Tb.
Just say, And hi to all you guys at WEXIS. Yeah, it's gonna be a
feature.
-- _________________________________________ Thomas R. Bruce (trb2@cornell.edu) Director,Legal Information Institute Cornell Law School http://www.law.cornell.edu_______________________________________________ You are currently subscribed to teknoids as: swise@law.miami.edu. To unsubscribe send a blank email to teknoids-leave@ruckus.law.cornell.edu -- See the web interface at http://ruckus.law.cornell.edu/mailman/listinfo/teknoids to get your list password, unsubscribe, and view your list settings.
This archive was generated by hypermail 2b29 : Wed Nov 14 2007 - 20:46:38 PST