[Papyrus-L] Saga of a Papyrus data conversion

[Jim Mallet]

I sympathize with both these queries (Bill Lindsay and Florin Curta).  I 
thought some of you might find my own experiences helpful.

After many years of using Papyrus, I too felt I ought to make the change.

First, until very recently I used Papyrus on the most modern versions of 
Windows XP, and it should work fine, providing you do a few tweaks to the 
"Command Prompt".  Bill Lindsay's *.BIB file must have got corrupted 
somehow.  Are you sure it still works on your Windows 98?  If you have a 
network card, perhaps you can make a local network with your two computers 
-- we used to run Win98 and Win 2000/XP on the same network OK; then you 
would just copy between the two computers -- get a geek to help if you 
don't know how to do that.

For my new software, I ended up using another Thompson-ISI programme, Ref 
Manager, rather than EndNote, because I heard it was more customizable.  I 
still don't know whether this is actually true.  They were all bought up by 
Thompson-ISI, so it's a monopoly now, which is a pity.

Theoretically, it is simple to transfer the data, you output the file using 
a RIS format from Papyrus, and then import it into the new database using 
Ref Manager's RIS filter.

But in practice, you lose a lot of formatting if you import from Papyrus 
using the standard filter.  Particularly italics (which are commonly found 
in titles of papers, and are important in biology for scientific names), 
and special characters such as accented letters in foreign languages, which 
Papyrus was quite good at handling using Extended ASCII characters.  The 
italics highlighting is specific to Ref Manager, but the special characters 
are a problem because whereas Papyrus and old fashioned pre-Windows 
programmes used extended ASCII text codes, the newer Windows programmes use 
different ANSI text codes for special characters.

To solve most of the problems, I needed to edit the RIS text file using a 
text editor (n.b. beware use of MS Word and other Windows programmes, which 
will often create line breaks where you don't want them). For a text 
editor, I strongly recommend TextPad, available as a fully functional trial 
version on the internet.

Eventually, I found there was a way that I could tell Reference Manager 
where the italics were. I was simply able to replace the italics on and off 
codes, which Papyrus helpfully put into the output file, into the Ref 
Manager preferred import codes, which you can set up.

I had to scan through the large text file of the resultant literature data 
to find examples of the special characters which showed up as strange 
characters in the text file.  Although I often couldn't see these codes 
properly in the text editor, I could highlight them and paste them into the 
"Find" part of a "Find and Replace" dialog of TextPad, and replace them 
with the correct special character (again placed in the "Replace" box by 
pasting from a text file).  After a few tentative attempts, I found it 
worked, and so I replaced all examples of weird characters I found 
globally.  Since I didn't know all the special characters I had used, I 
merely searched through replacing most things, i.e. most of the French, 
Spanish and German accents.  I think I got most of them.

Then I had a problem that Ref Manager thought that every single one of my 
journal names were the abbreviations, and not the full journal names that 
Papyrus had given it.  In fact, I had lost all the abbreviations!

To sort this out, I talked with the people at Ref Manager who helped me a 
bit.  Ref Manager, like Papyrus, contains a databased list of periodicals 
with several abbreviations allowed for each, which it treats as synonyms if 
there is an exact match.  So what I did in the end was export a list of 
journals with abbreviations from Papyrus, edit them into the appropriate 
format as a text file (I had to make sure all the special characters were 
correctly replaced in this new file as well).

I then started a new Ref Manager database with no data in it, and imported 
the journal list according to the instructions given me by the Ref Manager 
help personnel.

Then, into this new file, I imported the literature records from the 
existing database that I had converted from the RIS file (i.e. this was now 
in another Reference Manager database).  Then since the new Ref Manager in 
every case recognized the full journal title as a synonym of the journals 
it already knew about (both the literature records and the periodical 
titles were of course from the same original Papyrus database), the synonym 
was correct.  Now it knows all the journal abbreviations too.

As you can see, this was all quite tricky and took a certain amount of 
knowledge about how to manipulate text files, plus a lot of manual search 
and replace work.  But I think it was quicker to do this than to all the 
retyping it would have required if I hadn't done the automated editing of 
the RIS file.

Finally, I have all of my data, journals and abbreviations, and keywords 
etc.  When I say all my data, this is all the approx 11000 major papers I 
have ever made notes on, ever since I was an undergraduate (I am 50 
now).  I started with a card index, and when a Professor in the USA managed 
to get a work-study student to type it all into Papyrus.  It's kind of nice 
to have everything I have ever read as an academic!  And it's in a useful 
format for literature citation.  I have now successfully written articles 
with complex, numerical references, and which allow automatic citations 
using Ref Manager's MS-Word add-in.  Updating numerically cited reference 
lists (as in Science or Nature) was very hard to do in Papyrus, so it 
probably is labour saving to make the change.

But I still miss Papyrus!  I found it a lot easier to use than Ref Manager, 
even though Papyrus wasn't windowed, and I can no longer cut and paste a 
simple abbreviated reference, which was possible in Papyrus, of course. I 
could then pop the pasted reference into an email and send it to a friend, 
or print it out for a student.  Papyrus had an amazing number of clever, 
labour-saving attributes which today's literature database software still 
don't match.

Hope this helps

Sincerely, Jim Mallet

At 15:24 28/10/2006, you wrote:

>I have been seeking a solution to a similar problem: moving a large 
>database from an older computer with Windows 98 to a newer one with 
>Windows XP. I tried e-mailing the large INDX.BIB file from the old to the 
>new computer, and it was apparently received OK with the full file size. 
>But when I try to access it with the recent Papyrus download on the new 
>computer, Papyrus does not seem to recognize that it is there in the PAP 
>folder. (All the other, smaller files were moved earlier via floppy disks 
>to the PAP  folder in the new machine.)
>
>Question: Does the e-mail process somehow alter the INDX.BIB file so that 
>Papyrus doesn't recognize it?  If so, how can this be corrected or prevented?
>
>I know that I could put that file on a CD, but I cannot find an external 
>CD burner that will work with both the early Windows 98 (not 98  SE) and 
>XP. If anyone can help me with my problem, maybe they can help you too.
>
>Bill Lindsay   <mailto:w.t.l.jr at juno.com>w.t.l.jr at juno.com
>_______________________________________________
>Papyrus-L mailing list
>Papyrus-L at ResearchSoftwareDesign.com
>http://five.pairlist.net/mailman/listinfo/papyrus-l

James Mallet
http://www.ucl.ac.uk/taxome/jim/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://five.pairlist.net/pipermail/papyrus-l/attachments/20061028/f98275fe/attachment.html