[6277] in s-news-athena
re:documentation and ease-of-use
daemon@ATHENA.MIT.EDU (Bill Shipley)
Fri Jan 27 10:08:31 1995
Date: Fri, 27 Jan 1995 09:11:15 -0500
To: S-news@utstat.toronto.edu
From: bshipley@courrier.usherb.ca (Bill Shipley)
My off-hand comment on the difficulty of using the documentation to find
functions in S-PLUS seems to have generated a lot of discussion. I must
have touched a raw nerve...
I will summarize what I see as the problem and then let everyone decide for
themselves.
The data that I was analyzing involved plant weights harvested at various
times. However, varying numbers of replicate plants were taken at the
harvest dates. So, I had a vector of harvest dates (eg 1,2,3,2,4,1,5...)
and wanted simply to know the number of REPLICATE or REPEATED values in this
vector. To my mind a contingency table is a TABLE with rows and columns and
so it did not occur to me to even look under this heading in the help file.
I was looking in the index for things related to REPLICATION or REPEAT.
Yes, I know that (strictly speaking) I was looking for a degenerate
contingency table with only one row but I don't think that this would be
obvious to everyone.
The very flexibility of the functions in S-PLUS means that these functions
can be applied to problems beyond their "standard" uses. This flexibility
should not be compromised. The solution (I think) is to develop a good
system of cross-linking in the index and help file. This can only be an
ongoing and constantly updating proccess. Thus, in the future, one might
find the entry REPLICATES cross-linked to TABLE and RLE. This can only
occur if there is a continuous dialog between users and STATSCI. Perhaps
STATSCI should actively invite such cross-links.
Since we have all paid a lot of money for this program, it is in all of our
interests that it survive in the marketplace. This will be aided if the
program can be made more user-friendly to those of use who are not
professional statiticians without sacrificing the power and flexibility that
drew me to this program in the first place.
Bill Shipley
Departement de Biologie
Universite de Sherbrooke
Sherbrooke (Quebec)
CANADA J1K 2R1
bshipley@courrier.USherb.ca
