Palm pilots - possibly a better usage & interactive keys

[Stinger Guala]

The issue of community adoption of interactive key technology is a complex one
and several have mentioned many of the main points.  In my own experience,
having authored at least three different "key programs" and countless data
sets, and carefully watched people use them there are some points that I would
like to give my own perspective on here.

Wide adoption in this arena requires several things I think, and they haven't
really been available enough in the past to force the kind of tipping behavior
that makes oddities into basic tools.

- Comprehensive data sets. For example, a key to Erigeron is no good to a
person who doesn't know that they have an Erigeron. USDA PLANTS
(plants.usda.gov) is going to release keys to all wetland monocots and a few
other taxa in the summer (initially in SLIKS, www.stingersplace.com/SLIKS/)
and plans to do the entire US flora over the next few years. All of the data
will be freely available for download and use. The difference between these
data sets and others that have been developed for US plant taxa so far is
that, by design, there is a large specified amount of overlap in characters
recorded across all taxa. Thus, the data can be sorted independantly across
all taxa and distributions, giving comprehensive data sets for any
taxon/geography combination, down to the resolution of species and county
(with a buffer). The general ideal is envisioned on pgs 13-15 of
(http://www.neoninc.org/documents/RIC_Boston_Report.pdf).


- Portable software. Interactive keys have have come a long way with the WWW
and advancements in storage and processor speed. Larry Morse's IDENT in 1969
had to run on a mainframe to key out 10 species of maples using only brief
text characters. Sunday I was in the field with often well illustrated
interactive keys to nearly 6000 species and backup reference of more than
10,000 photographs on my PDA. I've run SLIKS on my cell phone and Bob Morris
is running his Electronic Field Guide
(http://www.cs.umb.edu/~alyons/efg_web/index.html) on an iPOD. There is good
data in the commercial computing community that (all else being equal) web
sites that require a download (a plugin etc.) to use their content, can expect
only one ninth of the usage of those that don't. If you have to pay for the
plugin, it goes down to the range of 1 in 1000 if you are lucky and have a
very competetive product.  As exceptionally good as many interactive key
programs are right now, I don't think that the community will truly embrace
them until they run ubiquitously and without plugins. Handheld, full windows
machines will make windows dedicated software more portable in the physical
sense but they won't help in the platform sense where a great deal of
diversity will continue to exist.

- Easy to use software. As has been mentioned by several participants, the
software has to be easy to use. There is the traditional look (e.g. LucID &
Intkey) and an exceptional number of other neat prototypes out there. Mike
Dallwitz in particular has written a lot of excellent stuff about what makes
an interactive key program easy to use - see his post for refs.  Other have
their own ideas. The past, in my opinion, has been the MSOffice try-to-be-
everything-to-everyone-approach but the future will look more like Google. I
think in the very near future we'll have lightweight portable keys that go
everywhere and really really smart server based keys for either online ID or
for downloading those lightweight portable keys. For example, there are
no "Best" characters. There are a lot of optimization algorithms (divide the
pile evenly, what the experts say is best, what other people use most, what
the data are most complete in, the most easily observable etc.). In some web
environments (e.g. AJAX - see Amazon.com) you can have the key suggest what is
REALLY the best character state to use based on all of those - and the key's
own experience with thousands of people in millions of keying sessions and
give you help when you actually need it, not just when you think you do.  It
isn't as scary as it sounds and it isn't far off.


- Standards. The third thing that I think will put an interactive key in
everyone's hand is the broad use and integration of standards. It's really
hard, I didn't use a standard data format in SLIKS, but it has to be done.
DeLTa was the standard for interactive keys for a long time. Nexus is a more
broadly accepted standard for this kind of matrix (See Kevin Nixon's WinCLADA
keys), but people are generally moving on to the SDD
(http://wiki.cs.umb.edu/twiki/bin/view/SDD/WebHome) which will - I think -
eventually migrate into an ontology and integrate with the OBOs
(http://obo.sourceforge.net/cgi-bin/table.cgi) and all of the other
descriptive ontologies out there. That will show the real power of these kinds
of data sets and foster very broad adoption. For our part in the Systematics
community, I think the best thing that we can do now is to start integrating,
amending, and building data reference standards that fit into this framework
for characters within and across taxa.


Cheers,

Stinger Guala

www.stingersplace.com
plants.usda.gov