Don't forget that there are existing tools out there that cover at least part of this use case.
For example, Zotero does a nice job of scraping the most relevant bits off an OPAC item page, and you can later dump it out in just about any format you please. There's even some limited functionality for sharing selected libraries of citations online. [not intended as a commercial for Zotero; I'm sure that one could do similar things with Endnote Web, etc.]
Edwin V. Sperr, MLIS, AHIP
AU/UGA Medical Partnership
Office of Graduate Medical Education
Clinical Information Librarian
St. Mary’s Hospital
1230 Baxter Street
Athens, GA 30606
e: [log in to unmask] | [log in to unmask]
From: Code for Libraries [mailto:[log in to unmask]] On Behalf Of Karen Coyle
Sent: Tuesday, October 24, 2017 11:35 AM
To: [log in to unmask]
Subject: Re: [CODE4LIB] mini-catalogs
I like this idea.
We could add any reviews we can find, let people annotate the entries, and call them bibliographies. There is an obvious (to me) need for something other than a dump of bibliographic data based on a search. The hard part will be facilitating selection, the wheat/chaff problem.
Not all of the items in our libraries are of equal import, but we have no way to provide a ranking. (I think OCLC does this based on # of
holdings?) If a sub collection could be a SELECTION based on INTELLIGENCE that would be ideal.
On 10/24/17 7:56 AM, Eric Lease Morgan wrote:
> I think a “kewl” (as well as cool) idea is the creation of mini-catalogs.
> Our libraries have large collection. That’s nice. But often the student/scholar only wants to look at a smaller subset of the collection. For example, they might want to look at only the books about painting. Alternatively, they might want to only look at items in a particular sub collection — a “special” collection. Unfortunately, and to the best of my knowledge, our library catalogs are not really amenable to such things.
> In order to facilitate greater use & understanding of these sub collections, I think it would be fun to:
> 1. dump all the MARC records describing a sub collection
> 2. generate a set of text files intended for printing,
> and these text files would manifest a VERY traditional
> library catalog 
> 3. generate a computer-searchable index designed to be
> used by a hand-held device 
> 4. promote the use & availability of the outputs of
> Steps #2 & #3
> What’s kewl is that the text files can be given away, printed, and even (“Gasp!”) written in. They require zero technology, and can last a long, long time. Heck, they are even portable and copies can be placed at the head of the collection(s). In days of old, librarians paid hundreds of dollars for these sorts of “catalogs”. They can still be valuable today.
> What’s more, the computer-searchable indexes and can be carried into the stacks and used like a Star Trek tricorder to home in and browse the collection(s). A bar code reader on the “tricorder” would be a helpful interface.
> Fun with the blending of newer and older library techniques?
>  example set of printed catalogs/indexes -
>  example computer-searchable index -
> Eric Morgan
> University of Notre Dame
[log in to unmask] http://kcoyle.net