DELTA. Comments and opinions continue to trickle in.

Andrew Taylor andrewt at CSE.UNSW.EDU.AU
Fri Mar 18 12:24:11 CST 1994


In a long discussion about Delta someone writes:
>8.  As far as the data basing features of DELTA, I don't see
>anybody having the time or money to go back and enter in all the
>data on each species that they have already done.

There is the possibility of automatically extracting the information
from textual descriptions.

I'm currently engaged in doing this with the 4 volume Flora of NSW
which contains description of roughly 5,000 species.  I've constructed
software which reads the text and extracts characters and states.  This
information is then used by an identification program which lets a user
describe a specimen in simple English phrases.  At any time the user
can ask to see descriptions of species that match their specimen.  This
program is aimed at non-expert users.

I intend at some time to try to convert the extracted information to
Mike Dallwitz's Delta format.

This is very much work in progress - but the results are already good
enough to convince me that automatic extraction will be sufficiently
accurate and complete for the needs of my identification program.  How
useful it will be for other purposes I don't know yet.

I've looked at a few other Floras and they seem equally amenable to
automatic extraction of the information.  Descriptions in other fields
may present a much more difficult problem.

I'd be very interested in hearing from people attempting anything
similar.  I only know of Renaud Fortuner et al's work with nematode
descriptions as part of the Nemisys system.

Andrew Taylor






>From TAXACOM at HARVARDA.HARVARD.EDU Fri Mar 18 06:25:56 1994
>Return-Path: <TAXACOM at HARVARDA.HARVARD.EDU>
>Received: From HARVARDA.HARVARD.EDU By jarrah With Smtp ; Fri, 18 Mar 94
 06:25:46 +1000
>Received: from HARVARDA.HARVARD.EDU by HARVARDA.HARVARD.EDU (IBM VM SMTP V2R2)
>   with BSMTP id 4383; Thu, 17 Mar 94 15:23:19 EST
>Received: from HARVARDA.HARVARD.EDU by HARVARDA.HARVARD.EDU (Mailer R2.08
> R208004) with BSMTP id 9149; Thu, 17 Mar 94 15:20:02 EST
>From:         Ingolf Askevold <IASKEVOL at VM.CC.FAMU.EDU>
>Sender:       Biological Systematics Discussion List
 <TAXACOM at HARVARDA.HARVARD.EDU>
>To:           Multiple recipients of list TAXACOM
 <TAXACOM at HARVARDA.HARVARD.EDU>
>Date:         Thu, 17 Mar 1994 15:06:29 EST
>Message-Id:  <940317202543.9021 at cse.unsw.edu.au>
>Reply-To:     IASKEVOL at VM.CC.FAMU.EDU
>Subject:      DELTA.  Comments and opinions continue to trickle in.
>X-To:         ENTOMO-L at UOGUELPH.CA, TAXACOM at HARVARDA.HARVARD.EDU
>
>      Well, a few pretty strong comment about DELTA have come in
>since I posted the last bunch of opinions I'd received, last week.
>Perhaps these will generate some further discussion, for I think
>it's informative and illuminating.
>      Anyone, as have those who contribute below, who wants to
>contribute to the commentary but wants to do so anonymously, may do
>so through me.  Sure, I'll be the fall guy and take all the flack,
>why not?
>      Ingolf Askevold  iaskevol at vm.cc.famu.edu
>MORE DISCUSSIONS ABOUT DELTA...
>
>[these are not presented in any particular order.  I'm deleting the
>names of those who sent me them, which some sort of even
>requested... ]
>
>[statements in [brackets] are those I've added to inject my opinion
>or for clarification]
>
>Date:         Tue, 15 Mar 94  16:56:06 EST
>
>   So maybe I should confess to being a secret DELTA user!  A
>little bit like one of the "groups" people attend these days.  I
>have used it to create tables, keys, write texts (which I edited
>rather heavily since I do not like the consequences of DELTA for
>the written language).  I am in fact, using it all the time.  The
>first DELTA files are very nearly always disastrous:  It takes time
>to learn how to write a decent character-list, and certainly to get
>it corrected etc.  However, once one gets into the swing of things,
>it gets to be relatively routine.  THE ONLY MAJOR PROVISO TO
>REMEMBER IS THAT THE DELTA CHARACTER LISTS ARE NOT IMMEDIATELY
>USABLE IN CLADISTICS.  Especially when using DELTA for writing
>descriptions one ends up focussing on  what might be called
>"differentiae" rather than on shared similarities.  One can modify
>a DELTA list to use in cladistics, but using it directly leads to
>some rather entertaining phylogenies.  Oh, by the way, I work on
>polychaetes and I know that several ichthyologists are using the
>techniques:  It is more widely used than suggested by the
>comments so far.  Keep on scoring!!
>
>                     *****************************
>
>[now, the above letter contrasts with next, somewhat, don't you
>think?]
>
>                     *****************************
>
>Dear Ingolf:
>
>I saw your earlier message about DELTA and Entomologists, but did
>not want to respond because my comments were basically negative.
>Now that I've seen what others have said, I can tell you what I
>think about DELTA.
>
>I am currently developing computerized keys for the identification
>of **********.  I am using DELTA primarily as an example of how not
>to make interactive keys.  Because I am only making keys for
>non-entomologists to use, I am only interested in INTKEY, and not
>in any other features of DELTA.  The number of other programs out
>there that do essentially the same thing as INTKEY indicates to me
>that this is what everyone else is interested in also.
>
>I also have a problem with the idea of making all my hard earned
>knowledge available in an easily manipulated form that basically
>obviates the need for ME to work any further on the organisms.  I
>get the feeling that I'm computerizing myself right out of a
>career!
>
>There are several problems with DELTA.
>
>1.  It is not an expert system, which is really what I want in an
>interactive key.
>      [DELTA itself does not profess to be the expert system; INTKEY
>      is the expert system software; other aspects of DELTA are used
>      to generate and manage the data INTKEY runs on]
>
>2.  I feel that the search algorithm is the opposite of what it
>should be.  I would prefer a routine that separates the taxa with
>unique characters FIRST. Instead, the program looks for characters
>that have poor consistency and tend to show up in all taxonomic
>groups.  This is really frustrating for a phylogeneticist.
>      [Good taxonomic keys split sets of taxa into large, potentiall
>      equal sets on each half of the dichotomous keys.  Keys that
>      clip off taxa one at a time based on unique features are
>      lousy, and get damned long, and generally don't work.  The key
>      algorithm selects taxa as you weight them or predefine that
>      they should be used, so you have a great deal of control over
>      how the key is generated.  I don't see why this writer says
>      that the program seeks characters of poor consistencey and
>      which occur in all taxonomic groups.  That makes no sense to
>      me, and isn't what occurs anyway, in my experience.]
>
>3.  It is a DOS program.  Most academic types are more comfortable
>with Macintosh.  Had the programs been originally written for the
>Mac, which of course, could not have been done by low-powered
>programmers, the software would now be wildly popular.
>Furthermore, many of the interface problems would not have existed.
>Perhaps a Windows version will make the programs work better.
>      [I don't know about this "most academic types" idea - that's
>      perhaps more like people around him are Mac users and
>      therefore that's the majority; around me, most everyone is an
>      IBM user, so it seems to me that's the norm...  PAUP 2.4.1 was
>      originally written for IBM, by the way; HENNIG86! shall I go
>      on?  Seems to me that there are a heck of a lot of IBM people
>      out there.]
>      [A windows version of INTKEY is due to be available soon, and
>      I hope that this will improve the interface problems that this
>      writer is talking about!]
>
>4.  The interface really sucks.  It is not intuitive.  This is the
>single most important aspect of any software today.  The command
>line interface makes it a dinosaur.  There are too many things to
>do to get to an answer.  The software does not guide the user
>through each consecutive step.  It is not clear at any given point
>what to do next.  You cannot view text and graphics at the same
>time.  With all the really slick multimedia stuff out there these
>days, I don't think DELTA has a chance of being adopted as a
>standard for zoology.
>      [I agree, and so do others I've talked to, that the interface
>      sucks and that it is not intuitive.  No argument from me.  I
>      think Mike should post his fulsome reply to this statement,
>      for I'm still not sure that I buy it.  I'm assuming the
>      windows version of INTKEY will solve the multimedia problems
>      mentioned here.  The intuitive part is something that is a
>      whole other ball of wax.]
>
>5.  The people trying to write keys using DELTA are accustomed to
>writing and using dichotomous keys.  For better or worse, the
>program would be more popular if it could think like a taxonomist,
>i.e., it WAS an expert system, and could flow through the data the
>way I would.  The problem here, of course, is the search routine
>(see #2 above).
>
>6.  I do all my figures on a Mac (= super VGA).  Everything is fine
>with DELTA as long as the computer you're using has a super VGA
>card and driver.  If not, the figures, which are the only good part
>of the entire key, will not work.  So much for that idiotic
>argument that DOS and IBM XT/AT compatible PCs are some kind of
>universal "standard."
>      [Let's face it, Mac users think the IBM platform sucks the big
>      one.  IBM users think that Mac users are just a bunch of lazy
>      louts who haven't the aptitude to use a real computer.  I use
>      both and think they have differing assets and there really
>      ought not be this animosity.  Nor should we be hurling insults
>      here.]
>
>7.  The data files, and to some extent, the directives files, are
>too difficult to modify.  Let's face it.  The DELTA system is
>really a programming language. Not a very complex or powerful one,
>but a programming language nonetheless.  I think it is expecting a
>lot of a taxonomist, who may have just gotten used to the idea of
>typing a manuscript on his PC, to ask him to learn a programming
>language.  You should be able to write the data and other program
>files with the same ease that you should be able to use the
>program.  The data files should be organized more intuitively and
>input should be more automated.  Much like programming environments
>for the Macintosh are done.
>      [At this point things are degenerating. There is a pretty
>      marked difference between the expectations that our competing
>      platform camps have of software.  I should reiterate some
>      comments that Mike Dallwitz has made to me, paraphrasing:
>      basically, "if you want software that can do all the things
>      you expect of it, it's bound to become complicated and a user
>      has to be willing to spend a little time to learn it."  Let me
>      add to this also, that there's a difference between DEVELOPING
>      an expert system, and USING IT.  Those who develop a data base
>      that INTKEY runs on, are the ones who have to learn all the
>      ins and outs of DELTA, the intricacies of the data files and
>      such.  I'm not saying this couldn't be set up a little more
>      ituitively, just that those who make up data bases surely have
>      to be prepared to understand the workings more than those who
>      are going to use the system.  For example, why should those
>      who buy the beetle larvae system have to know everything there
>      is to know about DELTA?]
>
>8.  As far as the data basing features of DELTA, I don't see
>anybody having the time or money to go back and enter in all the
>data on each species that they have already done.  There is also
>the problem of cross-leveling all the data for each species which
>means going back and looking at them again.  It is just too much
>work to do whether it results in a pub or not.
>      [No-one's expected to go back and redo his life's work; that's
>      pretty much a given because it's not very time effective. Of
>      course, it might be another story if there were some special
>      value in it]
>
>Well, I could go on and on about DELTA.  Give me call sometime if
>you really want to get me started.  I hope that we can get some
>money here to develop the kind of interactive keys that I think
>most people would like to see.  I see the data basing of taxonomic
>information as a separate issue, but maybe that will eventually
>catch on also.
>
>Cheers,
>
>                 *************************************
>
>      I consider it an excellent tool for helping organize one's
>data gathering and for INTERACTIVE identification.  Manuscripts
>that I have seen that were written relying heavily on delta have
>convinced me that it is not the way to go to prepare floristic
>accounts - or only as step one one a long staircase.  Perhaps the
>criticism should be that people tend to treat as it as step - 2 or
>3.  The keys that I have seen have used relatively obscure
>characters; some did not allow for all the variation in the
>description (which surprised me).  After reading a number of
>strictly parallel descriptions, I have reached the conclusion that
>they can easily conceal the significant information in a deluge of
>unhelpful information.  The wording achieved with Delta is also
>often very awkward [but the most recent release has significant
>improvements in this area; besides, you can really develop your
>character set to adjust these oddities].  As a means of providing
>an initial draft, or a reference work, fine.  For a final product,
>not without more work than is generally used.  And I suspect less
>work is involved if one starts hand work from the first delta
>product.
>      Why do I not use it myself? Too impatient. Then again, it
>might improve the quality of my work... I think people are not
>always prepared to admit how much basic taxonomic work needs to be
>done before one starts using DELTA to write a revision.  I have no
>particular quarrel with using it as a teaching device.  I do not do
>so, it is difficult enough getting the material I consider
>essential across.  As you may gather, I reached the decision that
>DELTA would not be a more efficient way of achieving my goals for
>the course.
>




More information about the Taxacom mailing list