name server coverage

Doug Yanega dyanega at UCR.EDU
Mon Jun 6 11:46:35 CDT 2005


>As usual, Doug and others do not do their homework.
>
>Doug wrote:
>
>>>and no one seems to be up-front about what percentage of names in
>their group they actually have
>>>gotten into their database.
>
>Norm will undoubtedly respond on behalf of his Hymenoptera Name Server,
>but I want to point out that we have since we first went online 2000
>have tried to provide users with FULL DOCUMENTATION as well as status
>information on the Diptera names we served.
>
>For status, see http://www.sel.barc.usda.gov/Diptera/names/BDWDss.htm

My apologies, Chris, for omitting the word "virtually" from the
clause above - I had actually intended to write it that way
("virtually no one"). Your site is exceptional in that regard, and I
did know that. As for the HNS, I didn't mean it as an attack on
Norm's work there - it's just that I had hit that particular wall
searching for that name last week, so it was fresh in my mind when
Roderic's post appeared (again, no offense was intended to you
either, Norm). However, your comment below points out something that
upsets *me* (i.e., both sides can complain about the other):

>What upset me is while Doug is right about how difficult, building, and
>then maintaining Name servers, people like him continue to complain but
>never contribute to helping get the job done.
>
>In the years, we have had our Diptera site online, I have noted that
>there are three kind of people.
>
>Those who find errors, omissions, etc., and write you about them so
>that they get fixed, as well as those periodically send along reprints,
>data files, act as reviewers, etc. They are few but we are truly
>dependent on them to get the job done.
>
>Then there are those who find errors and write to various media
>complaining how bad, useless, ["induces an
>undesirably false confidence"], etc., are the current products. And do
>nothing else. Yes, we are not yet perfect but we are a long way along
>the track to all names online, far more than we were just a dozen years
>ago. So, stop kicking us and help us along.

For my partial rebuttal, I will point out the following: as you,
Norm, and others are well aware, my database-related gripes over the
years have been focused mostly on the *interface* - not on content,
not on your integrity, motivation, scholarliness, or other things -
I'm mostly worried about the INTERFACE. I think what you and Norm and
others are doing is great, and necessary, but why do people get
locked into a particular interface, and never re-vamp it to make it
work better?; *that* is and has been the core of my complaints, and
this latest matter about issuing caveats is a small thing compared to
that, and not one I expected to generate such steam. Your present
comment, though, actually can be linked to two of these
interface-related issues:
(1) It isn't exactly easy to detect errors in a database when you
can't generate reports (such as catalog-style output) *from* it, even
on a small scale; the lack of customized reporting formats DOES make
error-spotting harder. Why? In part, because a labor-intensive
interface (e.g., where one is forced to go to an individual species'
page in order to access any of the data for that species) greatly
reduces one's motivation to actually examine the fine-scale data, and
spot errors therein; also, it's a lot easier for a person to spot
errors when they have two *similar* data sets they can compare
side-by-side (e.g., if I can't make your data set look like my data
set, or the catalog I have sitting next to me, I won't find
discrepancies as easily). For example, in neither the BDWD nor HNS
can one generate a list of synonyms for any given set of taxon names
(in the BDWD it can't be done at all, in the HNS, it can only be done
one taxon name at a time, and only for those few that have literature
citations entered).
(2) It isn't exactly easy to *report* errors to any of you folks. The
BDWD, for example, has a "questions and comments" link on each record
page, but that is a fairly clumsy interface for error reporting. A
better design would have an "error correction request" button on an
individual record's page, so that if someone spots an error, they
click that button, and a message is composed that contains a copy of
the data on that page (and references the URL), allowing a person to
simply edit the appropriate portion and submit it directly, rather
than typing out an entire message from scratch. When you have a few
dozen to a few hundred errors to report, that kind of time-saving
step can make a big difference. Other databases don't have any
feedback links on record pages at all, so it's even tougher to send
in corrections. While you have some justification in chastising folks
like me for not sending in corrections, also think about whether
you've set things up so we can do so efficiently.
Ultimately, of course, all this would be rendered meaningless if we
had one big list that everyone used; we probably never will, but I
still believe we could have, and benefit from, clearer standards
regarding explicit limits of each data resouce, and - more
importantly - of data *output*. Things like the Taxonomic Search
Engine are the future, but they - and we - work more efficiently when
data sources are well-defined and compatible.

Peace,
--

Doug Yanega        Dept. of Entomology         Entomology Research Museum
Univ. of California - Riverside, Riverside, CA 92521-0314
phone: (951) 827-4315 (standard disclaimer: opinions are mine, not UCR's)
              http://cache.ucr.edu/~heraty/yanega.html
   "There are some enterprises in which a careful disorderliness
         is the true method" - Herman Melville, Moby Dick, Chap. 82




More information about the Taxacom mailing list