CHEST-Papyrus Frequently Asked Questions File Version 15 Date: 3/April/2000 Author: Bernard Treves Brown Address: Department of Instrumentation and Analytical Science UMIST P.O. Box 88 Manchester M60 1QD United Kingdom Email: B.J.Treves.Brown@umist.ac.uk This is a compliation of information from the chest-papyrus list held on mailbase and the Papyrus-L list run by Research Software Design, authors of Papyrus. All messages sent to CHEST-Papyrus are automatically logged in monthly digests which can be retrieved. These are automatically deleted after two years. The present document is an attempt to save information from such files and as such is not exhaustive. Further, it may well be out of date. Anyone looking for information on Papyrus is strongly advised also to search the current log files for both CHEST-Papyrus and Papyrus-L. Probably the simplest way of searching both this file and the current CHEST-Papyrus logs is to use the WAIS interface at http://www.mailbase.ac.uk/lists-a-e/chest-papyrus/search.html Logs of Papyrus-L can be accessed from http://www.rsd.com/Lists.html ******************************************************************* Unless there is a note to the contrary, it should be assumed that comments apply to version 7 of Papyrus for Windows/DOS. Version 8 addresses many of the problems and limitations mentioned. I hope to have a note on applicability for each section for version 16. ******************************************************************* Comments on the file are welcome and should be sent to the above Email address. Thanks are due to Judy Adams , Duncan Branley , Gregory Brown, , Evan Cooch , Jenny Corbett , Beth Crutch , Sarah Davnall , Raisa Deber , Bill Foote , Heather Franklin , Dave Goldman , Donald Hadley, Fiona Hughes, Brendan P. Kelly , Dan Klein , Roel Kuijer , David Lindsey , Lis Parcell , Ed Richards, John Rodgers , Paul Schooling , Martin Schmid , Maggie Shapland , Graham Smith , and Robin Wilson for articles excerpted. CONTENTS Wordprocessors and HTML (Text Extract and bibliography generation) 1) WordPerfect 6.0 (DOS) Troubles 2) WordPerfect (Windows) 3) Word for Windows a) Master Documents b) Word for Windows 6.0 4) Word97 5) HTML Output Platforms 6) DOS Shell and Pap_res 7) Papyrus with Windows '95/98 8) Papyrus with Windows NT 9) Mac formats for Papyrus 10) Diskless Workstations and Read-Only Disks 11) User Directories 12) Running Papyrus on Novell 13) "Disk Full" Errors 14) Uninstalling Papyrus Databases and Import Formats 15) Importing from Web of Science (wos.mimas.ac.uk) 16) Importing from BIDS 17) Sources of Formats 18) Format Length 19) Printing Formats 20) Ovid (formerly CD-Plus Medline) 21) Win Spirs 22) Embase 23) Very Long References (Common with BIDS) 24) BIDS IBSS 25) Importing Names Correctly 26) Importing from EndNote 27) Importing from WordPerfect files Export Formats 28) IBM Research Reports (Four Letter Code) 29) Capitalization of Surnames 30) Harvard Format 31) References to Different Pages of the Same Book Using and Upgrading Papyrus 32) Overlong Notecards 33) Journal Abbreviations 34) Papyrus Locking Up, Repairing Damaged Databases 35) Upgrading Old Versions of Papyrus 36) Welsh Alphabetization 37) Keyword Search Problems 38) Browsing Through References 39) Useful alternatives to the %% delimiter 40) Listing Cited References by Author or Keyword (e.g. for Glossaries) 41) Merging Databases 42) Backing Up Databases 43) Large Groups 44) Printing from Papyrus 45) Reversing Text Extract Licensing 46) CHEST Issues - What to do if your site has not renewed its licence, Distribution of updates, Usage restrictions, Licences for Leavers &c. Other Bibliographic Databases 47) Alternatives to Papyrus 48) Importing and Exporting to Other Databases 49) Importing and Exporting to EndNote Further Help 50) Sources of electronic help ------------------------------------------------------------ Wordprocessors and HTML (Text Extract and bibliography generation) 1) WordPerfect 6.0 (DOS) Troubles When using papyrus (7.0.8) and wordperfect 6.0 together via upap, references are frequently brought in with the initial set of '%%' delimiters cut to a single '%', i.e. %reference%%. Problems occur as well when setting pap-res to use other markers. This causes problems with text extract. The default "trigger keys" for PAP_RES are Ctrl-Shift-Alt-A. (See pages W72-W74 of the Workbook.) Unfortunately, WordPerfect 6.0 seems to have a little problem with this combination; specifically, it loses the first of the two incoming percent signs, yielding things like this: %123(Smith, 1928)%% rather than: %%123(Smith, 1928)%% To get around this problem, you must change the PAP_RES trigger keys to a combination that does not include the Alt key. For example, to change the trigger keys to just Ctrl-Shift-A, you would modify the PAP_RES line in your AUTOEXEC.BAT file to look like this: c:\pap\pap_res cs-a See Chapter 37 of the Concepts/Reference manual for full details on loading and configuring PAP_RES. 2) WordPerfect (Windows) For WordPerfect for Windows 5.2 you can use any version of PAPYRUS as of 7.0.1. For WordPerfect for Windows 6.0 you need PAPYRUS Version 7.0.10c or greater. In "Text Extract" pick "WP greater than or equal to 5.0, +/- Windows", and PAPYRUS will automatically determine which version you are using and deal with it accordingly. The "List" function outputs a WP 5.0 file. That is just fine, as all newer versions of WordPerfect know how to read that format. 3) Word for Windows 3a) Word for Windows Master Documents [Information from Dave Goldman] PAPYRUS does not know how to Text Extract a Microsoft Word master document in order to work its way through each chapter document. The correct approach is to Text Extract each chapter document in turn, _without_ clearing the Group between chapters. For a few more details see the small print on page 160 of the Concepts/Reference manual. 3b) Word for Windows 6.0 With perhaps 15% of WinWord 6.0 documents Pap Text Extract reads the document, does everything it should, appears to create a valid .new document, but the .new document is not readable by Word for Windows. This problem was last reported with Papyrus 7.0.11b. There are two possible solutions: a) Save a copy of your document as a WinWord 2.0 file, and run Text Extract on that. With any luck you won't lose much in the translation to WinWord 2.0 and back to WinWord 6.0. b) Have WinWord save a copy of your document as an "RTF" file. (This is a plain-text encoding which is supposed to retain all information from the original.) and use that for Text Extract. [Information from Dave Goldman and Jenny Corbett] Some greek symbols (such as alpha and pi) can be handled by Papyrus and put them into titles etc; when these references are listed to a Word for Windows file they do not get translated over correctly. The problem is that when Papyrus creates a Microsoft Word output file it makes a file with the format used by Word 4.0 for DOS. When you then tell Word for Windows to open this file, one would expect it to be converted correctly. Regrettably, while Word for DOS uses the "extended ASCII" character set, Word for Windows uses the "ANSI" character set. For anything beyond unaccented letters and basic punctuation marks, these character sets are different. WinWord does not bother about this when reading a Word for DOS file. In addition, as of WinWord 6.0, WinWord has become very inconsistent in correctly converting character formating such as italics and boldface from Word for DOS files. Fortunately, there is a solution. Although WinWord cannot reliably convert a Word for DOS file, it does a really superb job with WordPerfect files! So instead of telling PAPYRUS to create a "Microsoft Word" file for your reference lists, tell it to make a "WordPerfect" file. This works very well; alpha, beta and pie are converted correctly to characters in the Symbol font, but one must be very careful not to dabble with the font created by the WinWord6 converter (CG TImes) as otherwise alpha, beta and pi convert back to their English equivalents. 4) Word97 * This section applies to Papyrus 7 only * [Information from pbonate@QKAN.Quintiles.com and Graham Smith] Many users report red boxes in Word documents after text extract. The simplest solution is to upgrade to version 7.0.16c or later using the publicly available patches. In the interim until you persuade your sysadmin to upgrade, note that there seem to be a number of causes, one of which is memory problems, (eg with long docs with many graphics). This was supposed to have been sorted with Service Release 1. Check that you are using the latest patch for Word 97, which at the time of writing was SR-2b, available from: http://officeupdate.microsoft.com/downloadDetails/sr2off97detail.htm Another solution which works in some cases is to insert a pictures into the Word document using Paste Special rather than Paste, clicking off the "float over text" box. 5) HTML Output [Information from Evan Cooch and Maggie Shapland] Papyrus can be made to produce HTML output. Maggie Shapland has some details on her web site (http://www.cse.bris.ac.uk/~ccmjs/pap_html.htm). It can also be set to add HTML markup and appropriate tags for building into a database searchable with other programs (e.g. a perl CGI script). There is an example of what can be done at http://www.goose.org/gooseref/ ------------------------------------------------------------ Platforms 6) DOS Shell and Pap_res Problems been reported when using the DOS Shell to swap between Papyrus and a word processor, and thus paste formatted citations into Wordstar footnotes via pap_res. The most likely cause is starting dosshell before pap_res in autoexec.bat, because pap_res will not be started until the DOS shell is closed down. Editing autoexec.bat to move the Pap_res command above the dosshell command fixed the problem. Dave Goldman adds: pap_res does not include an unload feature. Since it only uses a couple of Kb we decided to keep it simple. However, it is a well-behaved TSR, so any of the freeware unloaders _should_ work all right. 7) Papyrus with Windows '95/98 * This section applies to Papyrus 7 only * [Information from Bernard Treves Brown, Mark Pack and Dave Goldman] In general, Papyrus for DOS works well under Windows95 and behaves identically under Windows98. Sharing a database across the Microsoft network system works fine as well. However, Papyrus 7.0.11 and 7.0.12 do not like long filenames. For example, if you have a Papyrus database in a directory with a long filename, Papyrus is not able to find it. However, if you stick to directory names that meet the old 8.3 format everything seems to work fine. RSD have a document (win95.txt) covering installation of Papyrus under Windows 95 and 98 on their ftp site. This covers the process of creating a shortcut in detail. It is also available as http://www.rsd.com/ftp/win95.txt A few people have encountered the following bizarre behavior. They go to the Input option to add a new reference into their database. As it is supposed to, the "Type" menu drops down to let them pick Article, Book, Chapter, etc. At this point, no matter what they do they cannot get the menu to accept their choice and allow them to proceed. This has affected some people as soon as they installed PAPYRUS. Others were doing just fine until they later installed some other piece of software. Simply changing the FILES= line in the CONFIG.SYS file normally eliminates the problem. It makes no difference *how* the FILES= line is changed; the value can be either increased or decreased, by any amount. A second approach is to: Right-click on your Papyrus icon. From the pop-up menu, choose Properties. In the Properties dialog, select the "Program" tab. Click "Advanced." Select "Prevent MS-DOS Programs from detecting Windows. Apparently this problem is due to some very esoteric DOS memory interaction. 8) Papyrus with Windows NT * This section applies to Papyrus 7 only * [Information from Duncan Branley, Gregory Brown, Ed Richards, Robin Wilson, Brendan P. Kelly, Dan Klein, Lis Parcell, John Rodgers, and Dave Goldman] RSD have a document (winNT.txt) covering installation of Papyrus under Windows NT and 2000 on their ftp site. This covers the process of creating a shortcut in detail. It is also available as http://www.rsd.com/ftp/winNT.txt Some users have noticed that whenever Papyrus is the foreground application it takes up 100% of CPU time. No ill effects (apart from slowness of other applications when Papyrus is running) have been reported. When it is closed Papyrus it goes back to under 10% or so utilization. In addition, as described in the README.TXT file, Papyrus cannot get to the clipboard used by Windows NT. As of Version 7.0.11d you can go to the Preferences option and pick the "OS/2, NT clipboard" option. This lets you designate a disk file for PAPYRUS to use in place of the Windows clipboard. Once you have entered a filename under "OS/2, NT clipboard", all references to the Windows clipboard will instead be directed to this file. This applies to the F3 and F4 keys ("copy" and "paste", respectively) and also to the Cite option. It should then be an easy matter for you to set up a macro in your word processor to put text into or take text out of the disk file that PAPYRUS is using for its clipboard. Be aware that this may alter the font of the line in word and change the style to Plain Text; it appears that because clip.txt is a text-only file, Word gives it the document's default style, rather than the style of the current line. A suitable macro might be: Application (A1; "WordPerfect"; Default; "US") FileChangeDir ("c:\") FileInsert ("C:\CLIP.TXT"; No!) Another approach is to use MJPM Thissen's application in the users' contributions section of http://rsd.www.com which integrates the NT clipboard with Papyrus. To use it, this program must be started before Papyrus is run. The mouse behaves oddly for a DOS program such as Papyrus when running under Windows NT. For example, on searching Papyrus pulls up the references ok, but the box that gives a choice of reference, notecard, or both has each selection instantly highlighted with the next rapidly replacing it. The sequence continues, randomly choosing an option if enter is pressed. Here are some of the things that WindowsNT/PAPYRUS users have found successful in dealing with this problem: (1) Run in full screen mode: [Gregory Brown:] This is easily solved. Right click the Pap.exe file (the same procedure can be used on any shortcut to the Pap.exe file you have) and then on the menu item Properties. This will bring up a tabbed dialogue box which enables you to set options for an MS-Dos program. Click on the Screen tab and you will see where you can choose to run Papyrus with a full-screen view [Brendan P. Kelly:] To display Papyrus to full screen size under Windows NT (3.51), I had to set both the screen size AND the buffer size to 80 X 25. In NT, those settings are made under MAIN, CONTROL PANEL, CONSOLE, SCREEN SIZE AND PROPERTIES (2) Mouse "exclusive mode": [Robin Wilson:] On my Win NT 4 system, the mouse behaves OK now that I have checked the mouse "exclusive mode" box under the "misc" tab of the Papyrus shortcut icon properties. Even then, the highlighted menu choice still dances around a little when the cursor is outside the menu box, but it stops the dance when the mouse cursor moves onto a selectable menu choice. (3) Disable "fast ROM emulation: [Ed Richards:] Just found that you can cure some of the display problems with Papyrus (when using the F2 list function) using NT settings. Click on the program's icon, go to properties (Alt-enter) go to the Screen tab, deselect Fast ROM Emulation, then save the settings. Makes a big difference! Without this, you can get into scrolling locks in the F2 display that require shutting down the program. 9) Mac formats for Papyrus [Information from Dave Goldman] There are plenty of formats for Papyrus 7 (DOS) but fewer for the Mac version. This is how to go about doing the conversion: (1) In Version 7.0, make a fresh database. Bring into this database whatever formats you wish to convert. (2) Run Create Back-up Files to create a set of *.BB files. (3) Bring those *.BB files over to a Mac. (4) In Version 8.0, create a fresh database (or you can use an existing one). (5) From the File menu pick Transfer..., and tell PAPYRUS to bring in the Version 7.0 data from your *.BB files. This will convert the formats. (6) In Version 8.0, make a fresh format library, and drag or copy the relevant formats into it. Usually formats convert okay when you move a Papyrus Version 7 database over to Version 8. However, import formats are sometimes a bit tricky, and might need a little tweaking after they arrive in Version 8. 10) Diskless Workstations * This section applies to Papyrus 7 only * Version 7.0.10 of Papyrus gives errors which look like a memory or FILES= problem when used on diskless workstations. Use version 7.0.8 or 7.0.11 instead. With 7.0.11b, Papyrus is happy if there is no C: drive. If there is a C: drive, it insists on trying to write to it in order to test the FILES= bit from config.sys. If C: exists and is not writeable, it will throw you out. This problem was fixed by 7.0.12 11) User Directories [Information from Maggie Shapland] The default drive and directory for Groups created by the Search and Group options can be contolled using the PAPGROUP environment variable. If you add a line of this form to your users' AUTOEXEC.BAT files: SET PAPGROUP=C:\whatever then when a user uses the Search option, the resulting SEARCH.GRP file will be placed in the indicated directory. (Without such a PAPGROUP setting, the SEARCH.GRP file goes to the user's C: drive, in the root directory.) 12) Running Papyrus on Novell. The only technical limitation is that if one person is accessing a particular PAPYRUS database, no one else can access that same database at the same time. However, you can have an unlimited number of people simultaneously accessing one shared database if they _all_ launch PAPYRUS in its "read-only" mode (described in the READ.ME file). Any user, of course, is bound by the CHEST agreement with respect to copying the software, using the software for non-institutional-related work, etc. 13) "Disk Full" Errors [Information from Dave Goldman] The error message which appears is: "While trying to extend file #12 - Either your disk is full or contact Research Software Design." File #12 is the INDX.BIB file. Here Papyrus needed to enlarge INDX.BIB, but the operating system (e.g. a file server) didn't allow it to do so. The most likely reason being that the directory holding your *.BIB files is as big as it is allowed to get -- in which case, you need to talk to your network administrator. If this happens Papyrus will have to do an Index File Regeneration, since it was in the middle of writing to the index file when the error occurred. But you'll first need to resolve whatever problem prevented PAPYRUS from enlarging INDX.BIB in the first place. 14) Uninstalling Papyrus [Information from Martin Schmid, Roel Kuijer and Bernard Treves Brown] This approach relates to Papyrus 7 (DOS). Locate the Papyrus directory, probably called PAP (location: C:\PAP). Normally, it would be sufficient to delete this directory and all its contents (take care of keeping a copy of all files with the extension .bib or .bb on a floppy disk or elsewhere, if you intend to use your database in future). If you have created several databases you must find their directories and delete the contents as well (again backing up .bib and .bb files if they may be needed again). Next, remove the line with PAP_RES from autoexec.bat (if this feature was installed). Finally, delete the Papyrus shortcuts on the desktop and Start Menu. ------------------------------------------------------------ Formats and Databases 15) Importing from Web of Science (wos.mimas.ac.uk) [Information from Bernard Treves Brown and Dave Goldman] The easiest way of doing this is to upgrade to Papyrus 7.0.16c or later and get the latest version of import.flb Copy the format WEB SCI 2000 from import.flb using Formats:Library:Copy from Library When you have found the references of interest in Web of Science and selected them remember that there are a number of additional fields which can be saved, these are listed at the bottom of the page. Only use addresses, keywords and abstract: at 15th March 2000 none of the others worked - although it looks straightforward to add them if you need to. References can be downloaded using any of these buttons: SAVE TO FILE EXPORT E-MAIL Do not use FORMAT FOR PRINT The document you save on your hard disk from the browser or save from your email client should import correctly. If you don't have Payrus 7.0.16c or later then this import is rather harder. You will need to obtain an old version of IMPORT.FLB format library with a format named "WEB SCI". This can read the output from the ISI Web of Science site. However, you must use a text editor or word processor to make one global change to your downloaded file prior to importing it into PAPYRUS. The Web of Science output format separates beginning and ending page numbers into two fields: BP 31 EP 37 You must combine these into a single field: BP 31-37 This is easily accomplished with a single search-and-replace: Look for: EP Replace with: - 16) Importing from BIDS If having trouble importing references from BIDS ensure that you are using a downloading format. This applies whether you are using the web interface or the references are being sent by email. If you are saving the contents of a web page, be aware that some web browsers have a bug, and by default save the page in HTML format even when save as plain text has been selected - if the suggested file extension is changed from .htm to .txt however, the page is saved as plain text (which is necessary for most import formats). 17) Sources of Formats * This section applies to Papyrus 7 and 8 * If you feel that an output format for a particular journal should be included in Papyrus, RSD are happy to receive lists of suggestions. 18) Format Length [Information from Maggie Shapland and Dave Goldman] Version 7.0 Format Libraries are limited to 657 characters per format. If it looks as if this will be a problem, try to eliminate enough characters by removing spaces and other characters that may not be absolutely necessary in your template. For example, change this: [|PT - 0][|LG - 7][|PT - 0][|EM - 0] to this: [|PT -0][|LG -7][|PT -0][|EM -0] If you need to save a few more characters, then you can further shorten the fields to be discarded: [|PT0][|LG -7][|PT0][|EM0] 19) Printing Formats [Information from Dave Goldman and Duncan Branley] For a variety of reasons, particularly editing, it can be useful to print formats. At present there is no way to simply tell Papyrus to print out a format for you, in either version. In a Windows/DOS environment this can be done by copying and pasting from a DOS window. The best approach on the Mac is to do a screen capture - ever since System 7 Shift-Cmd-3 will capture your whole screen - in Mac OS 8 there are additional key-combinations for capturing, e.g., just the frontmost window. This results in a PICT file, on your System disk, which you can then open in SimpleText or most any graphics application. 20) Ovid (formerly CD-Plus Medline). CD-Plus, and now OVID, has always offered two different styles for downloaded references. The first is the style which they call "Standard" or "CDP" or somesuch. The second is a "Medlars" style, which has two-letter tags in front of each field ("AU" for authors, "TI" for titles, etc.) There is an import format for "Standard" in the CHEST-Papyrus archive at: http://www.in.umist.ac.uk/papyrus/files/manbids/quilty/ and ftp://ftp.in.umist.ac.uk/papyrus/manbids/quilty/ Chris Graves and Maggie Shapland have co-written a format that can cope with Medlars format from both Ovid Medline on CD and BIDS EMBASE. This is available from: http://www.uwcm.ac.uk/uwcm/dm/stafinfo/papyrus.html 21) Win Spirs One of the output formats is "Formatted ASCII". This does not put in a blank line between records, and will therefore not import into Papyrus without fixing using silcvt.exe. A better approach is: Download search history Download record numbers This puts in a blank line before the record number most, if not all, records will import (may need to check the first and last records) 22) Embase Multiple pagination, in the form: 33-34+37-38 can give trouble. References with such pagination will be rejected at level "Tolerant", the default for the format definition. However, if you then import the Rejects (.REJ) file specifying level "Oblivious" as you set the import running, they will be accepted. But check the references afterwards, because "Oblivious" may not have stored them as accurately as "Tolerant" would have done. 23) Very Long References (Common with BIDS) When importing very long references, on occasion Papyrus will exit partway through the procedure with the following error message: ************************************************************************** OUT OF SPACE IN CHARACTER STRINGS! Write down everything on the screen, and notify Research Software Design!! ************************************************************************** This is commonest with ISI references, since they frequently have an extremely long Cited Reference section. In such cases this section should be removed with a word processor prior to importing into PAPYRUS. 24) BIDS IBSS Even using the "Manchester" formats, IBSS Review references may be rejected because of ragged indentation of the Review keyword. Nothing can be done to the format definitions to prevent this. However, the reject file (.REJ) can be run through the CLEANUP utility program (in your Papyrus software directory) to fix it. 25) Importing Names Correctly [Information from Sarah Davnall] As the manual suggests, "full-names" permits names which have either spelt out forenames or initials only. The prompts in the format definition process are the way they are to enable you to tell Papyrus what names look like when they are a combination of full names and initials - "Jones, Ivor L." for example. The initials in an initials-only name should have the same style as you chose for full-names-and-initials at the format definition prompt. Usually, you can tell whether this will be the case when you see the example names displayed by Papyrus at the end of the author-name prompt sequence. 26) Importing from EndNote [Information from Dave Goldman] The procedure for importing from EndNote to PAPYRUS is as follows: (1) Have EndNote create an ASCII file of the references in its built-in "Refer" style. (2) In PAPYRUS, use the Format liBrary option to copy the predefined REFER format from our IMPORT.FLB format library. (3) Use the iMport option to read the file from step 1, using the REFER format at a Fussiness Level of "Tolerant." Note that each EndNote user has the ability to change the fields and reference types of his/her EndNote database, and has complete freedom in the way author/editor names are entered (full first names vs. initials, surnames first vs. initials first, etc). So the REFER import format will usually need to be customized a little to match the "REFER" output from any particular EndNote database. 27) Importing from WordPerfect files [Infomation from Paul Schooling and Dave Goldman] Some people already have extensive bibliographies written up in WordPerfect. The major problem with this is that bold and italics are generally used to define fields (eg title in italics). When you plain-text this (by whatever means provided in WP) there is no trace of this delimiting. The following importation method should work with all versions: The best method is to write a program to hack the WP files to produce a plain text file with character sequences to represent the onset and finish of bold, underline and some others, eg "abc emboldened-text def" in WP becomes "abc def" in the plain-text file. Then it is possible to explain to Papyrus where things are. At least, it is if the originals have been written with perfect consistency. Generally a clean-up program, is needed to move markers around, e.g. end-bold immediately followed by start-bold is eliminated; spaces and newlines at the extreme ends of an emboldened section are moved outside it. Such a program has been written at least once, but it is not publicly available at present. ------------------------------------------------------------ Export Formats 28) IBM Research Reports (Four Letter Code) [Information from Dave Goldman and Fiona Hughes] IBM research reports generally use the following format: BGHH 95 Mihir Bellare, Juan A. Garay, Ralf Hauser, Amir Herzberg, Hugo Krawczyk, Michael Steiner, Gene Tsudik, Michael Waidner: iKP - A Family of Secure Electronic Payment Protocols; Usenix 1995. The first four letters are the first letter of the surnames of the first 4 authors and each author is cited first name followed by surname. If there are fewer than 4 authors then the second letter from an authors name maybe included in lower case to make a 4 letter code. This is easy in Papyrus Version 8.0, which has a new "Reference ID" field for this sort of format. In Version 7.0, one of the user-defined fields could be used for this purpose. Change the name of (say) Field A to "Reference ID", and enter the appropriate "BGHH 95" or whatever there. Then set up an output format that includes that field at the start of each reference. To cite references in this way, in Version 8.0 the equivalent of the Text Extract function can be told to insert these ID's into the manuscript in lieu of sequential number or name and year. With Version 7.0, it is probably easiest to type the "BGHH 95" citations directly into the manuscript in her word processor. Then, using PAPYRUS, create a Group containing the appropriate references, telling PAPYRUS to sort the Group by Field A. 29) Capitalization of Surnames [Information from Dave Goldman] When Papyrus sees that you have chosen all-uppercase surnames, it assumes that what you really want to appear in your word processor are capitals and small capitals. Papyrus marks the entire Authors field - including any "and" - as being in small caps. If the typeface used does not have small capitals then everything appears capitalized. Now an italicized word in the middle of a title will be un-italicized if your output format indicates that the entire title should be italicized. In the same way the "and" can be set as being in small caps to force it to end up un-smallcaps when the entire Authors field is smallcapped. You might want to do this if you want the "and" in lower case, which some formats expect. So in the Names/Year section of your format, do this: Format for 2 Authors: 1 {sand} 2 Format for N Authors: 1, 2, {sand} 3 (There's a mention somewhere in the READ.ME file of using "{s...}" to indicate small caps.) 30) Harvard Format [Information from Maggie Shapland] Maggie Shapland has prepared a Harvard format based on the the British Standards document. It is available from her web page (http://www.cse.bris.ac.uk/~ccmjs/) in the Bristol library. 31) References to Different Pages of the Same Book * This section applies to Papyrus 7 only * [Information from Roel Kuijer, Donald Hadley and Bernard Treves Brown] Often, references to books are most useful when a page number is given, but any book should only appear once in a bibliography. Where the references have page numbers set the page numbers can be supressed by adjusting the format, but the reference set shows multiple entries and the text loses the page reference. For example: This is a difficult reference%%226p27(Atlas, 1991); 4257p56(Gean, 1994)%%. When do you cite these%%226p99(Atlas, 1991)%% and when are these ones cited%%226p1021(Atlas, 1991)%%. When is it correct do do this%%19738p6(Osborn, 1994)%% and what will happen if I reference a series of them%%269p56(McDonald & Silberberg, 1986); 226p23(Atlas, 1991)%%. This text extracts to: This is a difficult reference (1, 2). When do you cite these (3) and when are these ones cited (4). When is it correct do do this (5) and what will happen if I reference a series of them (6, 7). References (citation order) 1. Atlas SW, ed. Magnetic Resonance Imaging of the Brain and Spine. New York: Raven, 1991. 2. Gean AD. Imaging of Head Trauma. New York: Raven Press, 1994. 3. Atlas SW, ed. Magnetic Resonance Imaging of the Brain and Spine. New York: Raven, 1991. 4. Atlas SW, ed. Magnetic Resonance Imaging of the Brain and Spine. New York: Raven, 1991. 5. Osborn AG. Diagnostic Neuroradiology. London: Mosby, 1994. 6. McDonald WI, Silberberg DH. Multiple Sclerosis. London: Butterworths, 1986. 7. Atlas SW, ed. Magnetic Resonance Imaging of the Brain and Spine. New York: Raven, 1991. The only way of dealing with this problem is to use a generic reference for the book and put the page numbers into the text. This is a difficult reference %%226p27; 4257 (Gean, 1994)p56%%. When do you cite these %%226 (Atlas, 1991)p99%%, and when are these ones cited %%226(Atlas, 1991)p1021%%. When is it correct do do this %%19738(Osborn, 1994)p6%% and what will happen if I reference a series of them %%269(McDonald & Silberberg, 1986)p56%%; %%226 (Atlas, 1991)p23%%. ------------------------------------------------------------ Using and Upgrading Papyrus 32) Overlong Notecards Be careful about entering data into notecards as there is no warning that the 8000 char limit is exceeded (unlike entering abstract info). The result of this is that all notecard info for that reference is deleted. This problem was remedied by version 7.0.10 33) Journal Abbreviations Papyrus needs to be fed the correct journal abbreviations in order to produce bibliographies correctly. Some users have made their lists publicly available on RSD's ftp site (this section is mirrored at ftp://ftp.in.umist.ac.uk/pub/papyrus/users/). For instructions on loading them see (3) later in this section. The Chemical Abstracts style is given in CASSI (Chemical Abstracts Service Source Index). The Index Medicus form can be obtained on-line by telnet to the National Library of Medicine's catalog. The collection is divided into books, journals, and audio-visuals. If you choose "journals," you can search by journal title and it will show you the standardized abbreviation as well. The Internet address is: locator.nlm.nih.gov At the login, type locator and follow the directions from there. Dave Goldman writes: If any of you have done a lot of work in creating your PAPYRUS journal dictionary (i.e., correct names and abbreviations), and would like to make that work available to others in your field, here is how we could proceed: (1) In PAPYRUS, go to Journals, and then choose List. Have PAPYRUS list out all of your journals (or whichever ones you wish to share), to a file. When asked: Suppress header and frequencies, for subsequent Journal Load? answer "Y". (2a) If you would like us to make this file available on our Web site, just e-mail a copy of it to me (dave@rsd.com), along with any explanatory information you'd like us to provide. We will mention the name of the person who created the file, and if you wish we will also include your e-mail and/or physical address for feedback. (2b) If you would rather take responsibility for providing the file to interested PAPYRUS users, then just e-mail me whatever information you'd like to provide. We will post that information on our Web site, so that people can contact you directly to obtain a copy of your file. (3) Any PAPYRUS user who obtains one of these files can merge its journals with his/her existing journal dictionary via the Journal "lOad" function. 34) Papyrus Locking Up, Repairing Damaged Databases Papyrus can lock up when the database is damaged in some way and a user tries to list some records. Incidents where Papyrus announces that it has a negative number of records have also been reported. The following is based on a message from Dave Goldman of RSD describing their usual advice when a PAPYRUS database is acting strangely, e.g. records appearing twice when listing in numeric order, accessing certain records causing the computer to hang. The first thing to do is make a back-up copy of all the *.BIB files. Just use the normal DOS "copy" command to do this. If anything goes wrong in the following steps, this back-up copy might prove important. Next, go to the Utilities menu and run Index File Regeneration. N.B. This is not the same as going to the List menu and choosing Keyword Index. If Index File Regeneration seems to run all the way through without complaints, then see if your problem has gone away. This normally fixes references appearing twice in lists. If not, then return to the Utilities menu and run Create Back-up Files, followed immediately by Restore From Back-up Files. If this process runs to completion without errors, then your problem should be gone. If either the Create or the Restore fails to complete successfully, then there is a serious problem in at least one of the *.BIB files. This sort of problem is usually caused by a computer crashing while PAPYRUS is saving new information to the *.BIB files. RSD have also seen cases where SMARTDRV or another disk caching program has not written all of its caches back to the disk by the time the computer is shut off. In any of these circumstances, you can send RSD a copy of all the *.BIB files (the set copied in the very first step above); they can more often than not repair the files and return them to you. 35) Upgrading Old Versions of Papyrus [Information from Sarah Davnall] * Deleted * 36) Welsh Alphabetization Until this is implemented as an option in Papyrus, Dave Goldman suggests: tell PAPYRUS to use "Spanish" alphabetizing. Spanish also treats "ch" and "ll" as separate letters for alphabetizing, so at least those two cases would come out right. Of course, this won't help with any of the other double letters. 37) Keyword Search Problems [Information from Dave Goldman] You can find all references that have no keywords (so that you can change this condition) by running this search: Search for: NOT KEYWORD="*" In some cases keywords will contain punctuation such as parentheses, i.e. "(" or ")". It is not normally possible to use this punctuation in search terms however, under Preferences, there is an option named "keyword pUnctuation". If (and only if) you use this option to tell PAPYRUS to consider punctuation within keywords to be significant, then you can do a search such as: Search for: key="*(*" This will locate all references with any keyword containing a left parenthesis. As explained in the Preference option, though, most PAPYRUS users will want to stick with the default setting, where keyword punctuation is ignored for indexing and searching. If you are a typical Papyrus user in this regard, but have a special need right now to find your parentheses-containing keywords, you could list out all your keywords to a text file, open that with your word processor and do some quick searches in there to find the parentheses. If there aren't too many of these keywords, you could then build your search specification in the word processor, and copy and paste it over to Papyrus. 38) Browsing Through References [Information from Sarah Davnall, Dave Goldman and Bernard Treves Brown] Papyrus does not provide a simple facility to browse through references, e.g. for checking accuracy or looking through with the intention of creating a group to order from a library, but the following at least avoids having to type in refnos or author names: Do a search of your entire database, eg by au=* in the Search option. Choose View/Edit Highlight the first entry in the list and select it If it is obviously in need of correction, press E to edit it. If you are not sure, select it, in which case it is displayed, along with a set of letter-indicated options. Again the letter which invokes edit is E. Either route takes you to the edit screen. Edit the record, then press Alt-S to save it, and you will be returned to the View/Edit list ready for you to highlight and select the next. If you want to drop a reference from your hit list press R in the View/Edit list. Sarah Davnall notes: I've never tried this with a database of any size. You may find that it is more easily done by taking your refs in smaller groups, eg by alphabetical groups of authors (au=a*, au=b*, etc) or by ranges of refnos (ref>0 and ref<100, followed by ref>99 and ref<200, etc). 39) Useful alternatives to the %% delimiter [Information from Brendan P. Kelly] Rather than using the default delimiters (%%), use the ` symbol (single left quote). It has no other use in most documents (percent signs do) and is often less distracting to the eye. 40) Listing Cited References by Author or Keyword (e.g. for Glossaries) The idea here is to create a keyword or author index with the references that are relevant, i.e.: Eskimoes 9,12,45,28 Flamingoes 3,22 Gorillas 8 (The digits are the citation numbers.) Famulus can produce output like this. The layout cannot be achieved directly by Papyrus, but you can get close and then finish off with some global editing in your word-processor. A. create a format definition which specifies just the one field which contains the citation number. (If you plan to use the Papyrus reference number as the citation number, you specify this in the "indentation/numbering" option, and leave the format strings in the "Article", "Book", etc options blank.) B. Use the List option "Co-Author" specifying `*' as the co-author, (for an author index) or "Keyword Index" (for a keyword index). List your reference to file using the format definition which you have created. This should give you a file whose contents look something like: Aardvaark 50 62 71 Ape 7 55 79 and so on. C. Use your word-processor's find/replace facilities to get rid of all the new-line characters and replace them with commas. Additional markers in the format definition might help with the editing. Whether or not this is necessary depends on what you are wanting to achieve, and the capabilities of your word-processor. Papyrus can produce output like the following in a similar way. Eskimoes Smith and Jones 1994: Some study of something... Thornton 1992: Igloos, summertime, and property values... Flamingoes Vegas 1987: Yard ornamentation among the... 41) Merging Databases [Information from Maggie Shapland and Gabriel Egan] First decide which set of reference numbers you intend to use in future; call this 'A' and the other 'B'. The references from 'B' will be imported into 'A'. Start database 'B' copy the PAPX format from the library and use it to list the data to file. Now open the database 'A', copy papx from the library again and import the data from the file you have just created. You can specify what to do about possibly duplicate entries when you start the import. If you suspect that there are duplicates, run a `winnow' process afterwards to show you records in the new database which Papyrus thinks are suspiciously similar to one another. 42) Backing Up Databases [Information from Raisa Deber] "Create backup files" in the utility menu, breaks the backup files into small enough chunks to fit on even ancient floppy disks (the files are, at their largest, about 300K each). It is then very simple to copy them onto multiple diskettes for backup purposes. 43) Large Groups [Information from Dave Goldman] Papyrus version 7.0 sometimes has a problem with a group containing more than approximately 800 references. This is particularly true for groups sorted alphabetically by author. If you have problems, try sorting your group by Reference# (and only by Reference#). Then see if you can perform your searches and loads. If so, you can re-sort the group alphabetically as a final step prior to exporting. 44) Printing from Papyrus [Information from Duncan Branley and Bernard Treves Brown] Many people have found printing from Papyrus 7 (DOS) unreliable, especially with networked printers. The easiest thing to do is to select "File" when you're prompted to choose an output device (Screen|Printer|File) and then open it up in a Word processor. You get the added advantage that you can format a lot better too. Many sites instruct users not to attempt to print to networked printers from Papyrus 7. 45) Reversing Text Extract [Information from Chris Martin and B E Davies] This is most likely to be important when working with others on a long article or book. There are two approaches: the manual way and a rather complicated method which needs a wordprocessor that can use wildcards in search and replace operations. The simple way: Edit in a new keyword such as the file name of the document (paper) being written. A subsequent search for this keyword gives a group with the citations. Listing these in PAPYRUS native format allows the text to be to crosschecked with the references. A second search and list then allows the references to be output in the format closest that required by the editors. The keyword might be deleted once the document is published. The complicated way: Before running Text Extract to convert your '%% refs %%' to formatted citations, run this Search and Replace operation: Find: %%*%% Replace with: ^& ++^&++ Note the space after the first '^&', it could be any character. The asterisk means "find any sequence of characters", so '%%*%%' finds all your unprocessed Pap cites. I am using the sequence '++' here simply to make it clear in unformatted email; you could use any characters you like. The sequence '^&' (caret-ampersand) is special to Word, it means "insert the Find string here"; in other wordprocessors use the equivalent. So, if your original document said: ...furthermore, prior studies %% 4, 9 %% have shown... you would get: ...furthermore, prior studies %% 4, 9 %% ++%% 4, 9 %%++ have shown... Next perform 2 more Replace operations: Replace: ++%% with ++ Replace: %%++ with ++ (this is why you needed the space above) Now the only occurences of '%%' left are the ones Pap should convert to citations; run your text extract in the usual way. You would have to ask your collaborators to please ignore the ++4,5++ sequences in the document and not disturb them. As mentioned, you could use anything in place of the '++', i.e., {} braces, which you could also superscript. When you get back your heavily edited document, you need to do 2 things to get return to square one: 1. Delete all the actual citations in the document. To assist with this, I recommend using superscripted square-bracketed format when you do your original Text Extract (format #6). If you do, you can easily Replace all the citations by finding all occurences of: superscript\[*\] Hopefully, it is unlikely that you have other non-citation superscripted square-bracketed text in your document. BTW, you tell Word to find superscripted text by using the Format-Font buttons on the Advanced Find dialog. The '\'s are necessary since the square brackets have special meaning to Word's search feature; \ tells Word the []s are literals. 2. Replace the '++' with '%%', and you are back to your original, un-text-extracted document. This could be stored in macros to improve the efficiency of the operation. One other thought. If you can get your co-authors to ignore the '++2,4++' stuff, can't you just send them the original, non-text-extracted document, replete with '%%5,6%%' Pap cites, and just have them ignore that stuff, thereby avoiding all the conversion? ------------------------------------------------------------ Licensing 46) CHEST Issues - What to do if your site has not renewed its licence, Distribution of updates, Usage restrictions, Licences for Leavers &c. [Information from Heather Franklin and Judy Adams] The latest version of Papyrus available to CHEST can be downloaded from the CHEST web site at http://www.chest.ac.uk/software/papyrus/contents.html There will be one person at any licensed Institution who is authorised to download the software. He/she may apply to Judy Adams@chest.ac.uk for an appropriate username and password. Sites that are already licenced for for the DOS version will have to complete an additional form (Addendum B to Schedule 4) before they are given access to download the Mac version. Sites that have not previously had a Papyrus licence must complete a different form, (Schedule 4). New releases are announced on the chest-papyrus mailbase list by CHEST once the software is available to download from their ftp site. (There have been problems with downloaded files in certain cases, notably disk.id files have contained blank lines between the text lines - in other words they should contain single spaced text, rather than the double spaced text). Under the site-licence:- 1. usage should be "for the purposes of the normal business of the University". Specifically, the normal CHEST exclusions apply. That is: (a) work of significant benefit to the employer of students on industrial placement or part-time courses; (b) consultancy or services where the product is commercially exploited. 2. the user may not copy or transfer a registered version of the software to another person, although copies of the demo version may be distributed freely. 3. Technical support, when needed, must be obtained from the local computing support services, not directly from Research Software Design. (The computing support people may, of course, contact RSD if they fail to solve the problem themselves.) 4. When a user leaves the University, they must either purchase a "personal licence" (see below) BEFORE they leave or remove all Papyrus software and databases from their computer. 5. The current understanding is that usage under the site licence can be "at any location within the United Kingdom" thus it can be used off-site. (The implications for owners of portables who go conferencing abroad are unclear...) When Papyrus users leave their HE institution, unlike users of other kinds of site-licenced software, they may wish to hold on to their databases, and therefore the software which gave access to it. Fortunately an arrangement was included in the CHEST deal whereby individuals who had used Papyrus under a CHEST site-licence could have the right to buy a full personal licence at a special price, which should be around 15-20 pounds. (Notionally 14 pounds + VAT, but the site is allowed to impose a surcharge to cover administrative costs) Dave Goldman describes the procedure as follows: (a) CHEST can at any time order any number of Registration Cards from RSD, for 14 pounds each. (b) Individual sites can then purchase these Cards from CHEST, and re-sell them to end users who wish to convert their CHEST-licensed copies of PAPYRUS to perpetual personal licenses. The end-users will complete the Cards and return them to RSD. These "Personal Licences" are useful in two other circumstances: a. Your university has chosen not to renew the CHEST licence and has not bought a "perpetual licence" (this allows perpetual use of the release of the software available at 30th June 1997, [probably 7.0.14, but will someone please confirm this -- Bernard TB]) b. Your university has chosen not to not renew the CHEST license, and did buy a "perpetual licence" but you need to upgrade. In either case become an personally-licensed PAPYRUS user by buying a Personal Licence. Once RSD has received it you become eligible for personal updates and support and can buy update disks (Version 7.0.11 to 7.0.15 cost $15 including air mail shipping) and/or download patches from RSD's ftp site (see the section "Sources of electronic help" for details). On 30th October 1998 Judy Adams wrote: The following clarification of licensing conditions may be useful. Clause 3 contains the sentences: "3.2 Product may be used by any employee, student or other persons authorised by the Licensee for the purposes of the normal business of the Licensee's organisation, whether or not they are located on the Licensee's premises." This means that the student or staff member can have a copy for use at home in support of their work/studies whilst they are part of the university. When they leave they must remove the software from their machine, because they will no longer be using it for the purpose of the activities of the University. The paid "Personal" copies can be purchased for their own use, whether or not it is for University use. Because they have purchased it there is no requirement to remove it when they leave the University. This option is only available to people who have had use of the product whilst they were at the University; in other words, someone who had never used Papyrus at the University could not purchase a copy under this option for their use at home. They must also have purchased their Personal copy BEFORE they leave the University. Here are examples of how the two options work: - Site becomes licensed for use of Papyrus - Staff/student (a) uses Papyrus at University and wants a copy to use at home for University purposes; gets copy from Computer Centre. - Staff/student (a) wants personal copy for non-University work - Purchases "Personal" copy - Staff/student (a) leaves University. Returns University copy to Computer Centre. Retains Personal copy. - Staff/student (b) does not have copy for University purposes. Wants to use Papyrus for non-University purposes. CANNOT purchase Personal copy through CHEST* - Staff/student (c) Used Papyrus whilst at University, but did not have personal copy whilst at University. Leaves Uni. Wants Personal copy. CANNOT purchase Personal copy through CHEST.* * They can get their copy from: GeoMEM Consultants for 79.00 plus VAT (see CHEST Directory) ------------------------------------------------------------ Further Help 47) Alternatives to Papyrus [Information from Dave Goldman and David Lindsey] Research Software Design suggest that you lean toward EndNote, as opposed to Reference Manager or Pro-Cite, when recommending a Windows-based alternative to PAPYRUS. One reason is that they have always had a cooperative relationship with the makers of EndNote. In particular, each company provides the means for users to import data from the other's product. This message should not be taken to imply that RSD is in any way less than fully committed to producing a full-fledged Windows edition of PAPYRUS as soon as possible. They understand the level of desire for this! David Lindsey suggests that ProCite be considered since there are network versions of the software which allow shared databases as well as shared usage of the software. 48) Importing and Exporting to Other Databases [Information from Dave Goldman] There are some useful notes on this on RSD's website at: http://www.rsd.com/Exchange7.html 49) Importing and Exporting to EndNote [Information from Dave Goldman] EndNote->PAPYRUS ---------------- Have EndNote export the relevant references using its built-in "Refer" output style. PAPYRUS comes with a "REFER" import format that should work for importing the resulting file. In some cases the PAPYRUS format must be slightly modified to make it match the way that particular EndNote user entered his or her data into EndNote. Our technical support department can help with this process if needed. Reference Manager->PAPYRUS -------------------------- Have Reference Manager export the relevant references using its built-in "RIS" output style. PAPYRUS comes with a set of formats with names such as "RIS 6" for importing the resultant file. Again, our technical support department can help with any problems. 50) Sources of electronic help There is an anonymous ftp server for the list at ftp.in.umist.ac.uk in directory /pub/papyrus The latest copies of the "Manchester" formats are held here as well as other formats which have been submitted. For those without ftp access the server can be accessed by ftp-by-mail services such as those described in "Accessing The Internet By E-Mail -- Doctor Bob's Guide to Offline Internet Access". The latest edition of which is available by the following routes: (and yes, I have tried it!) Mail. To: mail-server@rtfm.mit.edu (for US, Canada & South America) Enter only this line in the BODY of the note: send usenet/news.answers/internet-services/access-via-email To: mailbase@mailbase.ac.uk (for Europe, Asia, etc.) Enter only this line in the BODY of the note: send lis-iis e-access-inet.txt Anonymous FTP. Site: rtfm.mit.edu get pub/usenet/news.answers/internet-services/access-via-email Site: mailbase.ac.uk get pub/lists/lis-iis/files/e-access-inet.txt Research Software Design, the authors of Papyrus, can be contacted at info@rsd.com for general information, or support@rsd.com for technical support questions. Please note that CHEST users should not use support@rsd.com address directly. Always contact your institution's support staff in the first instance. The latest version of formats provided by RSD is available at any time via anonymous ftp from the CHEST-Papyrus site at ftp.in.umist.ac.uk in directory /pub/papyrus/rsd. This is a mirror of RSD's site at ftp.rsd.com, directory /pub/vendors/rsd, which has some other interesting things. European users are advised always to try the CHEST-Papyrus site first. The transatlatic link gets very congested, especially in the afternoon. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% This article is provided as is without any express or implied warranties. While every effort has been taken to ensure the accuracy of the information contained in this article, the author assume(s) no responsibility for errors or omissions, or for damages resulting from the use of the information contained herein. The contents of this article reflect my opinions only and not necessarily those of my employer, CHEST, or Research Software Design. %%Administrivia Fill Column = 65