counts of scores, scores of counts

[bruce.simonson]

Hello folks,

Apologies if this is the wrong forum for this question, or if it has been dealt with before.

I am wondering if IMSLP can help produce an estimate of the number of compositions there are in the world, classified, kinda, by date. OK, sounds a little weird, but subject to the constraint that "in order for a composition to exist, someone posted a copy of it on IMSLP", the following questions might be answerable:

1) How many different (unique) composers are represented in the IMSLP database?

2) For each of the above unique composers, how many different (unique) compositions are in the database?

3) (harder), For each of the above unique composers and compositions, how many in each "time period"?

4) (not so hard, if you can get to (3)), How many unique compositions did each composer compose in any given year?

5) (okay, this query sounds like a "bot' wants it), Is it possible to get a list of unique compositions in the IMSLP database, with the composer, and (best guess at the) date of composition?

Why do I want to know the answers to these? Fair question. I am trying to get some kind of handle on the volume of musical composition across time, for a music appreciation class. Given (5), I could figure out (1)-(4). Given (1)-(5), and the assumed proxy (IMSLP "knows all" (compositions, anyway)), I could get an estimate of this. Incidentally, the year of composition is an important part of this problem, for me.

Anyway, this is my first post into IMSLP-forum land. If this question is more appropriate elsewhere, please help me out.

I would be the first to admit, if it were easy to get the above questions answered, and if I were to get the basic answers, that IMSLP is seriously cool. Which I think it already is, but heck, I'd even go to very seriously cool.

Cheers,
-Bruce

PS: Since I'm such a newbie, I'll have to figure out where this got posted, and if anyone has read it. All in good fun, in the world we call the internet.

[KGill]

The first three questions are extremely easy to answer, because IMSLP has categorized stuff that way since its inception. You can look in the left sidebar for some of these links.

1) How many different (unique) composers are represented in the IMSLP database?

See the main page: http://imslp.org/wiki/Main_Page IMSLP currently has 49160 works by 7162 composers (160352 scores total).

2) For each of the above unique composers, how many different (unique) compositions are in the database?

You mean a separate count for each composer? I guess some kind of SQL could be used for this - not sure - but it does at least give the count in each individual category. Otherwise, you can find the total number of works on the main page (as I indicated above).

3) (harder), For each of the above unique composers and compositions, how many in each "time period"?

See http://imslp.org/wiki/Browse_people_by_time_period - you can obtain individual composer counts of each period by clicking on the respective URL (and then clicking on 'Restrict to composers'). For works, simply go to one of the work pages under a composer from the time period you want, scroll to the very bottom of the page, and click on the category link that tells the work's time period (for instance, Romantic).

4) (not so hard, if you can get to (3)), How many unique compositions did each composer compose in any given year?

Actually this is much harder than 3) because it is not an automated function. There is not a composition date entered on every work page, so any attempt to automatically determine this would be pretty inaccurate in general.

5) (okay, this query sounds like a "bot' wants it), Is it possible to get a list of unique compositions in the IMSLP database, with the composer, and (best guess at the) date of composition?

Not really, but the closest thing can be found at http://imslp.org/wiki/Special:CategoryWalker/Scores/, probably. Scroll down to the 'Periods' section and click on the [focus] button for that portion of the list; unfortunately the 'Composers' section cannot be expanded in this way because it would overload the server.