kristoff@NET.BIO.NET (Dave Kristofferson) (05/27/89)
It appears that agreeing on a standard for the Table of Contents
formats on the BIO-JOURNALS newsgroup is going to require another
query to the users. Below are examples of three "flavors" of Medline
format in widespread use. One is from the NIH and the others are from
Dialog and BRS. Comments about the strong and weak points are
included with both and are set off by >'s in the left hand column. In
a nutshell my current opinion is to go with the BRS version, but I
would like to know if people's software could swallow a modification
of the author line (suggested by A. Coulson) which would solve several
problems. If I hear that this would cause too many problems then we
will probably stick with the straight BRS format (as in the recent J.
Biol. Chem. postings).
Dave Kristofferson
----------------------------------------------------------------------
NIH Medline
AU - Goldstein A
AU - Brutlag DL
TI - Is there a relationship between DNA sequences encoding peptide ligands
and their receptors?
SO - Proc Natl Acad Sci U S A 1989 Jan;86(1):42-5
> My main problem with this format is that a single line is used for
> each author. For those of you who peruse this info on-line and do
> not use it in a computerized database, this would stretch out the
> information over more screens than would be comfortable for reading.
> A more minor problem is the waste of space between the line
> identifier on the left and the start of the information field.
Dialog
TI- A biophysical model of cochlear processing: intensity dependence of
pure tone responses.|
AU- Shamma SA; Chadwick RS; Wilbur WJ; Morrish KA; Rinzel J|
JN- J Acoust Soc Am; 80 (1) p133-45|
> This is the most diverse of the three formats presented. The order
> of the lines and the JN identifier differs from the other two. The
> authors are fortunately presented on a single line, but see further
> comments below about author line format. Finally, the use of the |
> symbol offends my sense of aesthetics 8-)!
BRS Medline
AU Santibanez-M. Rohde-K.
TI A multiple alignment program for protein sequences.
SO Comput-Appl-Biosci. 1987 Jun. 3(2). P 111-4.
> The line identifiers here agree with NIH Medline and waste the least
> space. This seems to be the least objectionable although the use of -'s
> seems a bit artificial (more natural delimiters can be chosen as shown
> below). The use of hyphens on the author line in particular can
> cause problems with hyphenated names but this can be overcome as
> long as each author initial is separated by -'s, e.g., Smith-J-W-F.
> However the scheme proposed by Andrew Coulson below strikes me as
> being superior. The problem is that it is non-standard and BRS and
> the others above have already become default standards. My question
> then is "DOES THE FOLLOWING FORMAT MAKE YOUR REFERENCE DATABASE
> SOFTWARE GAG?" If so, I would like to stick with just plain BRS
> Medline format above since that is what I am currently receiving
> from some of our BIO-JOURNALS contributors. Professor Coulson's
> description of the proposed AU line format follows the example below.
AU Santibanez,M., Rohde,K.
TI A multiple alignment program for protein sequences.
SO Comput-Appl-Biosci. 1987 Jun. 3(2). P 111-4.
> The following is from Dr. Coulson:
> 'The "surname" part is always present, and it may contain anything except
> ",", including particularly "-" and " ", which stand for themselves.
> It is terminated by "," or end-of-field
> In the latter case, or if the "," is followed by " " or end-of-line, the
> name is complete. Otherwise there is an "initials" part, which usually
> consists
> of a series of single characters separated by ".", but which may contain
> anything but ",", and which is terminated in the same way as the "surname"
> part.'
-------
(end of Coulson text)
Thus a hyphenated name together with Jr. (another parser problem in
some formats) would look like this according to the above:
Doe-Smith Jr.,J.K.
A final question to all of you. Some of the journals like to put in
subject grouping title lines, e.g., ENZYMOLOGY, to set off a group of
references in their TOC's. These lines to date have not had
identifiers in the left hand column. Would it be acceptable to make
the following rule:
Text can be placed without using a line identifier before or after any
reference (i.e., before or after any grouping of AU, TI, and SO lines)
as long as it starts in the left hand column (to distinguish itself
from continuation lines inside a reference).
Or should a line identifier be used (which?) for headings or other
text, e.g., errata info?
I would appreciate feedback as soon as possible as I would like to
resolve this issue within the next couple of weeks if at all possible.
When agreement is reached I will ask our contributors to make one last
set of changes to conform to our standard.
Sincerely,
Dave Kristofferson
BIONET Resource Manager
kristoff@net.bio.net
or kristofferson@bionet-20.bio.net