IMSLP and How It Is Relatively Unknown

[Leonard Vertighel]

This is great news! Thanks for posting.

[jfarrington]

I would say that most music librarians are aware of IMSLP, and many regularly search it. At the most recent Music Library Association conference, in Chicago in February, IMSLP played a large part in one of the panel discussions I was on that covered all kinds of digital score sites.

There are a couple of things working against more music libraries linking to IMSLP. One, not many libraries any more are creating lists of links--they are just too time-consuming to maintain. Another disadvantage that IMSLP has--especially from an academic library perspective--is the nature of the indexing of the site. For example, what we do is post scores to our digital repository because of its use of persistent URLs. Because we link to digital scores from our online catalog, we need some assurance that we are not going to have to go back and change links every time a server name changes. Plus, we link directly to the edition in question. In IMSLP, you have links to the work level, and even to the movement level, but not particularly strong links to the edition level.

I did put in another good word about IMSLP on my panel, and that is if library's do not have access to a digital repository like the UR maintains, they could nonetheless at least share their wealth by posting scores that they scan to IMSLP. I don't know if any will take me up on the idea, but at least it was floated as an alternative.

[Peter]

Does this mean that a parallel submission from your side to both your repository and to IMSLP will never be the case?
Then, what is your opinion about IMSLP users trying to mirror every submission to Sibley's online library? It is quite easy noticable that a lot of works appear on IMSLP, or say Petrucci ML , shortly after they appear on your website.

thanks
peter

[jfarrington]

We have no problem with people putting music we've put into our repository into IMSLP. In fact, we've discussed the idea of mirroring our material, but the folks who run the institution's repository don't have time right now to make the necessary modifications. But we're more than happy for people to take scores from URResearch and upload them to IMSLP. Spread the wealth!

Jim

Does this mean that a parallel submission from your side to both your repository and to IMSLP will never be the case?
Then, what is your opinion about IMSLP users trying to mirror every submission to Sibley's online library? It is quite easy noticable that a lot of works appear on IMSLP, or say Petrucci ML , shortly after they appear on your website.

thanks
peter

[imslp]

Like Jim mentioned, IMSLP has been talking with Sibley about mirroring the entire Sibley archive. I had proposed that some changes be made on Sibley's side such that mirroring will be entirely automatic, but Sibley's tech people are in the middle of setting up a new framework for the entire system, and so cannot implement random features yet, understandably.

I have, however, thought up an alternate system where Sibley submissions to IMSLP are streamlined and automated to the greatest extent possible. I've already mentioned this to Leonard as a possible project to work on, and we should get something set up relatively fast (as in, the next few weeks hopefully). This way, we will have a manageable backlog even when we switch to a completely automated system.

By the way, Leonard will soon be the main programmer for IMSLP. I'll still be overseeing stuff and double checking of course, but I'm sure Leonard would be able to contribute much more to IMSLP's infrastructure than I can, especially considering the amount (or lack) of free time I have for programming specifically (as opposed to dealing with other aspects of IMSLP).

[Peter]

I suppose the rss feed is quite usable in this aspect.

[imslp]

Well, the biggest problem is the difference between the structure of Sibley and that of IMSLP. Sibley is entry-based, whereas IMSLP is work-based. The computer cannot be asked to put each entry into the correct work page. This is why I proposed a technical change on Sibley's part (I basically was asking about the possibility of Sibley submitters submitting to IMSLP first, which will bounce the submission back to Sibley, as human submitters can classify entries into the correct work page much easier than computers).

However, an alternative is to streamline the Sibley-score submission process on IMSLP so much that it makes duplicate submission (both to Sibley and IMSLP) much less trouble. The idea is that users will only need to identify the work page the entry should go on, and the system will do the rest (download the file, get the info, etc). Technically it is a little more complicated, but doable.

Ultimately RSS can be useful, but at this point we need to get the infrastructure in place first.

[pml]

Given that the IMSLP work pages can move around a bit, a persistent URL ought to use either the IMSLP or PMLP numbering systems,

e.g. for IMSLP numbers, http://imslp.org/wiki/Special:ReverseLo ... arch=28540

This provides a persistent URL to a specific edition, whereas the following provides a persistent URL to a particular work page using the PMLP number:

http://imslp.org/wiki/Special:ArticleIDRedirect/61174

The advantage of the latter is that the browser is redirected straight to the target page, whereas the former URL presents a list of pages linking to that file.

Regards, PML

[imslp]

A note that PMLP means "Petrucci Music Library Page", i.e. the ID number of the page itself. It would survive moves and maybe even delete/recreates, but items on the page can still move around (i.e. to other pages). IMSLP # is the real persistant index for a particular entry, and it should be fairly easy to make a persistant link that is better than just the search link. But a few other things first

[Leonard Vertighel]

Following up my earlier post: SMA is now gone from the Yale library website, but they didn't add IMSLP, which I find slightly odd. Do we have people from Yale here, or anyone who feels in a position to suggest it to them?

[Peter]

A significant omission is also in the Duke Uni page, which is quite extensive, and does list SMA.

[Vivaldi]

Wonderful news!