Re: [patch] Accept # as Blank Indicator

2003-11-19 Thread Colin Campbell 
On Tue, Nov 18, 2003 at 10:50:22PM -0600, Chuck Bearden [EMAIL PROTECTED] wrote:
 On Tue, Nov 18, 2003 at 07:50:39PM -0600, Ed Summers wrote:
  On Tue, Nov 18, 2003 at 08:11:39PM -0500, Morbus Iff wrote:
MARC::Field-new('100','1','', a='Logan, Robert K.', d='1939-'),
MARC::Field-new('100','1','#', a='Logan, Robert K.', d='1939-'),
  
  I don't like this. The # is used simply as a typographical convention in LC's
  online docs. It has nothing to do with the actual content found in MARC
  records.
 
 I think Ed is right.  As I recall, OCLC used to use an underscore for
 blank indicator positions, but now they seem to be using the doodad
 represented in this image:
 
I'd second that. I've seem software that generated serious problems
through confusing the the typographical conventions with the data. Also
the fact the space is enclosed in quotes is doing the same thing for the
Human reader that LOC's hash is doing (showing it is a space not a null
string)
Colin

-- 
  Colin Campbell 
  Technical Services Consultant
  Sirsi Ltd
  [EMAIL PROTECTED]

Re: [patch] Accept # as Blank Indicator

2003-11-19 Thread Ed Summers 
On Wed, Nov 19, 2003 at 07:43:52AM -0500, Morbus Iff wrote:
 The LC also uses $ to represent sub-tags (I think that's what
 they're called; just woke up... the $a/$b things). But, I
 seem to see _a and _b more often. Which is more prevalent?

LC's MARCMaker/MARCBreaker utilities use $ if I remember right. 
It's mainly a typographical convention that should have little bearing
on how MARC::Record works.

//Ed

Re: [patch] Accept # as Blank Indicator

2003-11-19 Thread Chuck Bearden 
On Wed, Nov 19, 2003 at 07:43:52AM -0500, Morbus Iff wrote:
 The OCLC conventions are probably much more widely known than the LC
 ones simply because most libraries doing copy cataloging use OCLC as
 their utility.
 
 The LC also uses $ to represent sub-tags (I think that's what
 they're called; just woke up... the $a/$b things). But, I
 seem to see _a and _b more often. Which is more prevalent?

They are called subfields.  A variable data field (01X-8XX in USMARC
Bibliographic) comprises a 3-character tag, two indicators, and one 
or more subfields.

Well, if I were representing a MARC field in longhand, I would use the
dollar-sign with the alpha-num character identifying it.  I think the
underscore is used in MARCBreaker pretty-print.  I've also seen the
vertical bar used in other applications.  My feeling is that the 
dollar-sign is the most common, unless you are trying to emulate the 
output of a particular app (e.g. MARCBreaker) that follows a 
different convention.

Chuck