Please help me save the old PubMed. On March 1, 2000, it will no
longer be available and we will have to use the new version, which appears
to have been designed to discourage riff-raff like me from using Medline
data to embarrass my doctors.
The format of the old and new PubMed forms are compared below for a
sample citation/abstract saved as a test file.
Old New
format format
-----------------------------------
Offset from left 0 8
Words 322 402
characters 2,308 3,517
Lines 43 98
Pages to print 1 2
-----------------------------------
It is outrageous that in this era, when we are constantly exhorted
to recycle and save resources, the bureaucrats at the National Library of
Medicine (NLM)have designed a form which wastes a sheet of paper each time
an abstract is printed. The layout of the new PubMed is so bad that when you
print an abstract you not only have to print an extra page, you must keep
both pages and staple them together. This is necessary because the
institutional advertising and useless features at the top of the form cause
the abstracts to have a page break in the middle of the text. If they had
put the junk at the end of the form, you could at least just throw away the
second sheet.
There is a lot of junk on a sidebar that prints out on every
citation and abstract. Thousands of other websites have non-printing
sidebars, why not PubMed?
I wrote Word macros to reformat PubMed abstracts into a convenient
form. Abstracts from the new PubMed, with the junk at the beginning and end,
and with 8 blank spaces before every line, will be much more difficult to
reformat.
There are other problems with the new PubMed, and few improvements.
On the old PubMed, for example, you would see how your search terms were
grouped with each search. This would often let you know if you had invented
a new term that the MeSH server couldn't handle. On the new PubMed, you see
only your search terms as you have typed them. There is consequently a
greater chance of deluding one's self about the results of a search.
The new PubMed does not fix any of the annoying features of the old
version. Citation data is still in JOURNAL-TITLE-AUTHOR order. I have looked
at hundreds of journals over the years and I have never seen citations
listed in that order.
When "Saving as" abstracts from the old PubMed, the title returned
was always "PubMed medline query." When saving abstracts from the new
PubMed, the title is always "Entrez-PubMed." Neither are very useful-why not
return the first author's last name and the year?
The nicest feature of the new PubMed is the "Go to publisher
website button." But as nice as this feature is, there is no reason for the
button to be printed on every citation.
A very nice bureaucrat (VNB)at NLM suggested that I look into
taking a class on the new PubMed. I didn't need a class for the old one, and
taking a class will not fix the problems with the new one. If NLM had decent
designers and programmers they wouldn't need to waste millions of dollars
giving free training at sites all over the country. The VNB at NLM also said
that "NLM/NCBI has limited staff resources and cannot maintain and support
two versions of PubMed." If that is true, why not keep the good one?
I suspect the real reason they are shutting off the old PubMed is
because no one is using their new one. PubMed is not a free service--you pay
for it with your tax dollars and you have a right to expect that the new
version will be at worst no worse than the old one. I don't know whether the
new PubMed was done by a contractor, or in house by NLM personnel, but it
must have cost a fortune. We might as well have thrown the money down a rat
hole.
If you would like to see the old PubMed kept in operation until NLM
fixes the new one, please write to them at custq at customerq.nlm.nih.gov and
tell give them your opinion. Thanks.
Russ Farris
