#794 Add 'Printing number' and "Printing # Details" fields to pub records

[Ahasuerus]

Add 'Printing number' and "Printing # Details" fields to publication records. The first field should be strictly numeric, allowing values like "1", "3" or "27". It will be displayed as a new column in the standard Publication table. It will be used for sorting publications that have the same publication date and publisher within it.

The second field will allow arbitrary values like "stated fourth Ace printing but actually at least the 6th printing because Ace reset its printing numbers at some point". The value, if present, will be displayed in a mouse-over bubble next to the numeric Printing Number value. (See http://www.isfdb.org/wiki/index.php/ISFDB:Community_Portal/Archive/Archive48#.22Printing_Number.22_field_--_an_alternative_approach for further discussion.)

[Anonymous]

Suggested use in sorting publications - by date, by publisher, by printing.
Would be useful in the publication lists as a separate column.

[Uzume]

Perhaps this is not within the defined boundaries of this feature request, however, I can see this as more of an "issue designation" where ISBNs usually define the boundaries of an edition and this adds to that with an issue where one can specify a print ranking but it would also be useful for other items that are printed with essentially the same title. For example, I was thinking perhaps we could capture issue or volume and issues information (since we usually label magazine pubs with "Title, month/season year" as per the entry instructions/rules) here for magazines, etc. Then if there is a magazine run where the issues are just labelled with seasons (and have unknown month of publication) we can still order by this issue designation. Anyway, something to keep in mind.

[Ahasuerus]

From a 2019-12-13 Rules and Standards post:

The main problem that I have run into is that the definition of "printing" can be vague. A popular book which has been published by multiple publishers using multiple formats can easily have half a dozen "first" printings: "first printing of the Dutton edition", "first printing of the Ace paperback edition", "first printing of the Italian edition", etc. If a title is associated with 6 "first printing" publications published over the course of 20 years, sorting them by their "printing" number would create a mess. In addition, some publishers (like Ace in the 1980s) occasionally reset printing numbers, which creates a difference between "stated printing" and "actual printing".

From a purely technical perspective, adding a new field to publication records would be straightforward, but first we need to decide how we are going to use it.

[Uzume]

Also from a 2019-12-13 Rules and Standards post:

Printing rank is only important in conjunction with a year and publisher though - to allow us to sort the 20 0000-00-00 or the first and second printing in 1979 by Ace. Sort by "date, publisher, rank" is how I think we should use it. Which will sort the editions from the same publisher. Still need to figure out the dates :) The other case we need the printing for is "I hold undated 23rd printing of a book we have 17 printings of. How do I know that we do not have 23rd?" Most of the times, new editors just add it again and a moderator need to open all 17 and read through notes because the statement can be anywhere in the notes... Older editors do their own research but it is tedious.) So yes - that won't solve the date problem completely - but it will help a bit. Visually if nothing else (and if we are adding it, it needs to bubble to the pub table on the titles or we are back to square one) And off on a tangent I went again. :)

[Uzume]

I personally am of the persuasion that we want such a field and it need not be a numeric field and instead a general text string is probably the best (with some useful Help/advice about how it should be used also necessary).

We could also add some sort of issue rank typing field to go with this allowing one to select "via statement", "via number line", "assumed; see pub note for evidence", etc. I recommend the issue rank type field be limited in value much like format and pub type are.

I believe this could be useful for simplifiying reprints of magazines as well (instead of creating new editor title records and new series).

[Anonymous]

I am not sure it makes sense to attempt to create only a linear printing history. In some cases that might be alright but not in general. For example we have have data on and want to capture 1st, 2nd, 3rd Canadian printings as well as 1st, 2nd, 3rd US printings amongst a set of publications from a single publisher with the same title and content and without other dates (i.e., dated 0000-00-00 unknown). This is why I think it better to avoid a purely numeric "printing number". We might also have other types of printings without dates or true orderings that could still be used for groupings, etc.

This is why I endorse a sortable texual printing field that can be used differently based on the publication history (often determined by publisher).