hort.net Seasonal photo, (c) 2006 Christopher P. Lindsey, All Rights Reserved: do not copy
articles | gallery of plants | blog | tech blog | plant profiles | patents | mailing lists | top stories | links | shorturl service | tom clothier's archive0
Gallery of Plants
Tech Blog
Plant Profiles
Mailing Lists
    Search ALL lists
    Search help
    Subscription info
Top Stories
sHORTurl service
Tom Clothier's Archive
 Top Stories
New Trillium species discovered

Disease could hit Britain's trees hard

Ten of the best snowdrop cultivars

Plant protein database helps identify plant gene functions

Dendroclimatologists record history through trees

Potato beetle could be thwarted through gene manipulation

Hawaii expands coffee farm quarantine

Study explains flower petal loss

RSS story archive

Iris Checklist versus Iris R&I dB

  • To: "Iris-L" <Iris-L@rt66.com>
  • Subject: Iris Checklist versus Iris R&I dB
  • From: "Hall_Gigi" <gigi.hall@mtv.gtegsc.com>
  • Date: 8 Nov 1996 17:32:42 -0800
  • Return-Receipt-To: "Hall_Gigi" <gigi.hall@mtv.gtegsc.com>


A long winded message, the theme of which
is - just get the data all in one place. Then
worry about how to make it consistent, or
whether or not enough people care about
making it consistent to "go-the-extra mile".

Different people use the R&I for different
reasons.  My most common use is to look
up the description of an Iris for a friend
who wants to keep their Iris properly named
and labeled, but does not own a complete
set of the R&Is and checklists.  My second
most common use is to look up the
descriptions of Iris to make a list of Iris
available at our club meetings and sales.

For both of my most common reasons to use
R & I data, just having the text that
presently is in 13 separate volumes (1939,
1949, 1959, 1969, 1979, 1989, 1990, 1991,
1992, 1993, 1994, 1995, [and soon to add
1996]) in one master list would be a big,
big, plus.

The third reason that I use the checklists
is to verify whether or not a name has
already been used.  This is in support of
suggesting possible names to hybridizer
Vern Wood for the PCI seedlings I guest
and believe are worthy of introduction. Not
only do I have to check thirteen books, I
need to check them twice.  I often get halfway
through a pass and realize that I have forgotten
AIS's peculiar alphabetizing scheme (which
ignores spaces, but not punctuation in the
middle of an Iris name), and have to start over
or repeat the search to have confidence that
I have not missed a listing.  There is no need
for database capability to completely eliminate
this frustration - just a complete list of the
names with hybridizer and either year of
registration or year of introduction would be
wonderful.  And scanning the data in would
serve my purpose just fine.

I can see the argument for a database - it
would make checking for all Japanese Iris
or all Arilbred Iris or all the direct
descendants of Iris cultivar 'xyzzy' a snap.
But most of us are not using the R&I for
research into ancestry most of the time.
The principal concern is 'are our Iris
correctly identified'.  I can understand where
translating the color classification codes
of the original checklists will be tedious,
time consuming, dull and prone to argument
(because there are many instances of two
different older Iris which are similar, but
not identical, in existence using the same
name).  But get them all in one place first -
then worry about making them consistent.

I understand the the classification scheme for
Iris has changed over the years. For example
"medians" as currently defined, are the result
of a committee to study Iris classifications
formed by AIS in the late 50's.  They "corrected"
classifications for the decade of the 50's
when the 10-year compilation was published,
but did not go back to the 1929, 1939, 1949
"other bearded" and revise the classification.
But again, I can deal with different classification
symbols at different time spans as long as the
abbreviations, and the dates for which they
are valid, are included with the list.  Just get
the list all in one place.

I understand that different data has been
included in the descriptions during different
periods of time (and believe me when I say
that I consider a R&I entry for a Japanese
Iris that doesn't even list whether the flower
is supposed to be single or double just about
worthless), but again, I don't care about every
entry being consistent near as much as I care
about all the entries being available in one
volume rather than 13.

Scanning technology has improved greatly in
the last couple of years.  Many, many "desktop
publishing" houses now have the capability to
do multiple scans of the same text and compare
the scans, highlighting differences. Where
differences occur, an operator checks against
the original and either makes corrections or
"accepts" the correct version.  This is still
somewhat tedious, but much, much faster than
manual reentry and Quality Control and check of
the reentry.  Local companies that used dual
manual entry with comparison of the two sets
of manual entries only a couple of years ago
have almost without exception converted to
scanning, then comparing the scans within the
last two years.

A couple of years ago, I was involved with a
project that compiled a database from hand-
written paper records - there were approximately
190,000 records, of about the same complexity
(15 fields) than an Iris database would entail.
Entry was manual.  Over seventy different people
worked on the project before it was over (not
all of them full time or for the number of
months it took to complete).  Total cost of the project
was several hundred thousand dollars (more than
1/2 million).  Divide this by 1/3 (about 70,000 entries
required in the AIS database) and you have an
estimated cost of the manual entry approach of
over $200,000.  AIS does not have this kind of
money to spend on an full compilation.  You can
state that you will use volunteers for the entry
and therefore it is free, but you have no clue what
issues of computer incompatibility and translation
you will run into trying to compile the volunteers

In a different vein, another friend of mine worked
on a task under a program to bring the 26 paper
maintenance manuals for the program "on-line"
(total text larger than the 13 R&I and checklist
books).  This was a few years back when scanner
technology was not nearly as good as it is today.
The total task took under $50,000 (still more
than AIS wants to spend, I'm sure) for the text,
with figures and graphics scanned separately from
the text and then "pasted" back in.

But you say - the twenty six manuals did not have
to be interleaved - with all the "A" entries placed
before the "B" entries. - this is not a valid
comparison.  I feel it is  because as long as the list
is all in one place in softcopy, I can search for the
information I need and don't care about the order
being alphabetical nearly as much as I do in a hard-
copy version.

AIS definitely owns the copy-right to the latest
checklists.  It is less clear that the 1929 and 1939
lists are still under AIS copyright.  However, as
long as AIS owns any of the copyrights, approval to
incorporate the information in the lists into some
other document, must be approved by AIS.  Our
speculation on the best method is just that until
AIS shows a will and desire to do something to
change the current state of affairs.

And in fairness to the very real people involved,
AIS is an all volunteer organization.  The token
payments made to the AIS membership secretary,
the Bulletin Editor, and the Registrar are not
a living wage.  The registrar must have a
vote in how to proceed and must take
responsibility for updates - and this may be a
greater issue than the technology (or lack of it)
for initial compilation of the "master R&I".


 © 1995-2017 Mallorn Computing, Inc.All Rights Reserved.
Our Privacy Statement
Other Mailing lists | Author Index | Date Index | Subject Index | Thread Index