Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
branch: master
Fetching contributors…

Cannot retrieve contributors at this time

123 lines (92 sloc) 5.736 kB
Thoughts:
1. Move to using proper MARC records?
2. Use the new XML MARC format? Then there could be a stylesheet for
each of the different views and all you'd have to do is apply the XML
for a book to a stylesheet and bingo, there's your web page. Of
course the XML would have to generated by database fields, but maybe
there's some way to tie MySQL and XML together so it's transparent.
Desired:
Taking a course on classification and cataloguing certainly showed me what
was missing. It would be good to use MARC records, use subject headings,
and have authority control. Any good library catalogue needs those.
Still, this thing works well for one person (especially if that person is
me).
1. Patron/librarian modes! Patrons can only search and browse, but
librarians have full editing privileges.
2. Right now, only one author per name is allowed (there's no way to
distinguish writers with the same name) but this should be changed
and, when an author is renamed, if the name matches with an existing
name there should be the option to shift all the books under the
incorrect name over, or to keep them under a unique version of the new
name. If you see what I mean. This would be easier to manage if the
author table were expanded to include date of birth, notes, etc.
3. Browse FIC titles, and browse all titles together.
4. When browsing, page through 25 or so titles at a time.
5. The one thing (aside from the initial database set-up) that keeps
Mrs. Abbott tied to MySQL is the insertid DBI::mysql function that's
called after a new title/author/series is inserted. If there was some
way to abstract this so one function could be called that worked for
all databases, this would be perfectly portable.
6. More ways to browse the titles, say by ISBN, publisher, purchase
date, first printing, etc. Also maybe a by-the-numbers list, where
authors are ordered by the number of titles in the catalogue written
by them. When looking at an author, there could be different ways to
view the titles, such as in alphabetical order (the default) or by
purchase date (showing the date, which would show how interesst in the
author progresses).
8. Generate some stats about the catalogue: number of titles, number
of authors, most popular author, etc.
9. Should probably add some information to the author table, like date
of birth, URL, etc. This is separate from pretty much everything else
in the database, though, so can go in whenever.
10. Change the way purchase dates are entered so that it's three
pop-ups, for year, month and day. Same for the edit screen.
12. Check on characters like <, > and & when displaying a book. All
stuff pulled from the database should be escaped with CGI.pm's escape
function. Unfortunately FreeBSD's default CGI.pm is old and doesn't
have this function right now.
13. The hook is there, but use isn't made in browse of the way you
can pass a call letter to it so that that title appears BIG, thus
making it easier to see where it is and what's around it.
14. SERIES: It should be easier to create a series after adding the first
book in that series. Right now you have to shift around a few screens.
When a book is deleted, it is now also deleted from any series it's in,
but the series could end up empty. It should be possible, when editing a
series, to edit the name and also completely delete it. Series should
have notes, too. It's not possible to have two different editions of the
same title in the same place in a series (like two different editions of
THE TOWER TREASURE, Hardy Boys #1). This is because getSeriesInfo in
MrsAbbott uses placeInSeries as the hash key.
17. Logging. Whenever a major change is made, log it. The log
should be browsable by the tool. Perhaps offer a menu of choices when
a book is deleted, e.g., "Donated," "Given away," "Sold," "Taken to
cottage." Could keep all books, old and new, in the database, but with a
flag indicating whether or not they're here. After a while, deaccessioned
titles could be moved into a data warehouse!
18. When browsing by first letters of last names, report how many
authors there are under that letter (and perhaps how many books per author)?
20. Anthologies. Not only of short stories, but of novels. It would
be nice if there were a way that listings of anthologies could be
included in such a way that they were visible in title listings.
21. Error-checking. Right now any errors when connecting to the
database just make it all die (like when a " is entered in a field).
It should use eval or something to catch errors, and report on them
sensibly.
23. There's a field for edition. Should there be one for volume? Or
should I just use a standard format like V 1 in the title field?
24. No way to manage pseudonyms--e.g., deleting one if you find X
didn't really write as Y, it was actually Z. Generally: authority control
over names.
26. Games! Pick six titles and first lines, mix them up, and see if
the patron can match them up.
Problems:
1. You can't add a title with a " character in a field. They should be
escaped somehow. Are there are special characters like this?
2. The byCallNum sort routine in browse, used when browsing by call
letter, doesn't handle call numbers that don't have a dot, like "PN 6081".
If you add a dot, it works out, but if there isn't one they appear at the
very top of the listing, ahead of (for example) just plain P.
3. When I tried to change Philip José Farmer's middle name from Jose to
José, the entry somehow got deleted! I had to insert it into the database
by hand after going back to find his author.id number. Never any problem
putting in new book titles with accents, though.
Jump to Line
Something went wrong with that request. Please try again.