Interactive keys vs. conventional keys

[Mike Dallwitz]

                                                                  19 June 1995

> From: "MICHAEL A. IVIE" <ueymi at MSU.OSCS.MONTANA.EDU>
> To: Taxacom
>
> I just cannot let pass Mike D's statement that no one should be doing keys
> any way except his way because his are so much better! This patently
> self-serving statement is absolute bullshit.
>
> Mike, I use and love keys that use your software, and I wish that all keys
> were available that way, but this is the real world. To fully utilize your
> system in an isolated situation takes years of work, years that could
> produce good, usable keys along the way. I have worked with one of the
> users at your institution that has published an intkey CD, and he admits
> that without you right there, and the huge storage capacity in the back-up
> computers on your local network, he would never have been able to do it.
> It is ridiculous to think that only institutions with the wealth to have a
> Mike D and such a network should be producing keys.
>
> Your system is still 'way too expensive to be practical. I expect it to
> become cheaper as time goes by, and younger workers will eventually come
> on line in the profession with INTKEY type expertise as a matter of course.
> That is in the future, not today.
>
> I hope that no 80 year old systematist with a bad heart stops working on
> the last few couplets of the genus they have worked on for 40 years to
> learn a stops doing so in response to your view, and begins all over to
> learn a new system similar to yours. The same goes for many other
> excellent taxonomists without the means, skills, back-up, or inclination
> to do it your way.

What I said was

    No one should be contemplating constructing conventional keys as
    identification aids, as interactive identification is so much better.

This statement _was_ too sweeping; no doubt circumstances make it difficult
for some people to use computer methods. The point I was trying to make was
only that interactive keys are (potentially) much better than conventional
keys (and it seems you agree with this). I was referring to interactive
identification systems in general, not just our own (I posted a list of
interactive identification programs a few months ago).

You have overstated the difficulty of using our programs to create keys
(interactive or conventional). Most people use them with little or no help
from us. You can be doing productive work after a learning period of a few
days. This time will be repaid many times over in increased productivity. To
become really skilled in the use of the programs _will_ take years of
experience (as with most other activities), but you can certainly `produce
good, usable keys along the way', and benefit in other ways as well.

The hardware requirements are modest. The MS-DOS versions of the programs can
even be run on a computer without a hard disk. The CD-ROM package referred to
requires about 20MB of disk space, most of which is occupied by image files.
The network is used mainly as a convenient way of transferring the image files
from the illustrator (who uses a Mac) to the scientist.

> From: Robin Wilson <rwilson at ARIEL.UCS.UNIMELB.EDU.AU>
> To: Taxacom
>
> I would love to see an 80 year young systematist put a view forward ...

More than 20 years ago, a colleague started my `taxonomic' career by
suggesting that I write a program for generating (conventional) identification
keys. He is now 82, and uses DELTA regularly for revisions of genera
containing several hundred species. He does not, as far as I know, intend to
publish INTKEY packages, but he uses INTKEY to assist him with producing
conventional publications. For example, he uses it to try to identify new
material. If the identification fails for some reason (often because the
specimen is from a species not yet in the database) he amends the database and
continues.

> From: Wayne Maddison <wmaddisn at CCIT.ARIZONA.EDU>
> To: Taxacom
>
> Let's not forget that most of the world's biodiversity is in countries
> that have an awfully tough time affording computing resources. We must
> consider the needs and resources of third-world countries if we are to set
> priorities in the techniques used in systematic work. It's great to set
> our sights high and dream of an ideal world, but we are not there yet, and
> keys printed on paper will still be used by those sweating over
> microscopes lit by flashlights in tropical forests.

We tend to take for granted such costly infrastructure as libraries. Several
years ago, I attended a conference at which Arturo Gomez-Pompa claimed that
the cheapest and most effective way to help research in third-world countries
was by providing computing infrastructure. Judging by Deb Kama's enquiry about
his recent CD-ROM, I presume he still holds this belief.

In any case, printed keys and interactive keys can be produced with little
extra effort from the same database. Both can easily be updated when
corrections or additions (e.g. new species) are made to the database.

> From: David Maddison <beetle at AG.ARIZONA.EDU>
> To: Taxacom
>
> ... it is unrealistic (and self-centered) to expect the rest of the world
> to follow the same path, for whatever reasons.

I don't _expect_ the rest of the world to follow the same path, but I would
like to _pursuade_ them to do so. I gave reasons why interactive keys work
better, and no one has challenged these reasons except to say that other
considerations may outweigh them. I doubt whether these other considerations
(mainly the lack of computing facilities) are the real reasons why people,
particularly those on this mailing list, are not using these methods.

> And as far as expense goes, DELTA is very expensive if one has to
> buy an IBM compatible system along with it, for those of us who love
> Macs. :-)

As Nicholas Lander pointed out, our programs will run under DOS/Windows
emulation on Macs. But I reiterate that our programs are not the only ones
available for these purposes. There are several that are written specifically
for Macs.

Mike Dallwitz                                  Email md at ento.csiro.au
CSIRO Division of Entomology                   Fax +61 6 246 4000
GPO Box 1700, Canberra ACT 2601, Australia     Phone +61 6 246 4075