ChoralWiki talk:Texts of works not yet added

From ChoralWiki
Jump to: navigation, search

If we persist in putting on this page texts of works not yet added, then we shall very soon be way over the 32 kilobyte recommended upper limit for page sizes. The page is already at 31 kilobytes. Perhaps we should agree on some other sort of solution? -- Chucktalk Giffen 04:41, 24 February 2009 (UTC)

  • Posted by: Vaarky 09:17, 24 February 2009 (UTC)
 Help 

I'm not sure I see the point of having text for pieces for which there are no scores. Easy enough to transcribe off the score, no easy way to store them (you either get a large page, create a score page which only has lyrics and then disappoints users, or have to create a parallel hierarchy for lyrics. Being a lyric server is not the purpose of CPDL and I think we start drifting afield if we start storing lyrics for works that are under copyright and can't be uploaded (Vaughan Williams?) or even scores that simply happen not to be on CPDL yet. I don't feel strongly about this either way, but don't really see the point.

I take your points; I certainly agree that CPDL should not become a "lyric server".
Chuck: I think this 32kb page "limit" is rather antiquated - these days, many Wikipedia articles are much larger than that. Even some really old browsers will still be able to edit such pages, especially if the user only wants to edit one section at a time. Anyway, that's all rather academic since I think I've moved all the texts here that I originally intended to move here, ie. I don't think there are any "text/translation only" score pages left.
Vaarky: I agree. We shouldn't be encouraging users to submit texts where there are no editions. I raised these concerns when a user (I think John Henry Fowler) started adding texts for works where we had no editions listed. He seemed to get quite defensive so, not wishing to offend him, I backed down. I hoped what I've done here might be seen as a compromise.
The following two pages: (1) & (2) should be treated in the same way, in my view, though their creation by User:Depravo raises another point - should we start indexing scores available at IMSLP? In my opinion, we should not, since IMSLP is much larger than CPDL and the indexing would take a long time. A better idea, in my opinion, would be to add links to IMSLP composer categories from our composer pages where there are at least a couple of vocal/choral works by that composer available at IMSLP. What are your opinions? --Bobnotts talk 15:28, 26 February 2009 (UTC)
Rob, I agree that we should not be in the business of providing works pages or "editions" that are actually just links to IMSLP (or other) scanned scores. ChoralWiki is not an indexing service. If anything, they might be listed either as "See also:" or "External links" items ... on relevant composer pages (or on works pages with one or more other editions). The mere existence of the present page had me worried that it might be seen as a repository for users to add texts to just about any choral/vocal works suiting their collective fancy - I was worried way beyond the 32kb limit. By the way, although many Wikipedia pages are indeed very long (it is, after all, an encyclopaedia - we are not), one frequently sees suggestions for such long pages that either the page be broken up into smaller pages or that larger sections have their own pages which are then simply included (or possibly summarized) in the main article. -- Chucktalk Giffen 16:14, 26 February 2009 (UTC)

Reply by: Vaarky 16:31, 26 February 2009 (UTC)

 Help 
Indexing IMSLP: As a chorister, I love the idea of CPDL indexing choral works on IMSLP so people don't have to learn multiple interfaces. It would get out of date but that way people who don't check IMSLP have more of a shot at knowing what music is available to sing.
However, that muddies the water in terms of Copyright if CPDL creates the traditional score info page, since that makes representations about copyright. We have not had a direct upload to CPDL, and therefore have had no interaction with the editor uploading the score making direct representations to us. It's also more work to index stuff that can be found elsewhere--hard to keep up with, and I'm not sure how easy their interface makes it to discern what new scores were needed and need indexing since the last batch that was indexed.
Given the copyright question and the preferences expressed above that we not index, I think it makes sense to put a link to IMSLP. I think it's better to put it right within the section of the composer's page rather than under external links--"Additional scores by this composer are available on the (full name of site)--this is similar to the way we add the Wikipedia link in the bio section. Ordinarily it's good to put misc external links in that section, esp. to warn them more visibly that they're going offsite at that point, but Wikipedia and IMSLP are known reputable sites and can be treated separately. OTOH, from an interface perspective, doing this might encourage users to similarly put links (even to their own site) that simply say "add'l works by this composer can be found at (my site)".
So maybe putting a note just below the scores saying "See External Links section for information about more scores" might be the best compromise.
Ambivalent.

Reply by: Chucktalk Giffen 20:35, 26 February 2009 (UTC)

 Help 

To me, the most bothersome part about Rob's two examples is that each provides two links, the second to the IMSLP works page (which would be what we should use), and the second a direct link to the PDF (bypassing the IMSLP copyright warning). I agree with Vaarky's suggestions about how to link on the composer pages.

BTW, it is possible that the recent problems encountered with the automatic back-up scripts were caused by the large files copied from IMSLP that have been recently added (Verdi's Requiem has been also uploaded - size is more than 20 MB). They may cause a memory overflow when they are compressed. I'm testing a workaround to this problem. So, from an IT viewpoint, I would be happy if large scanned files copied from IMSLP are deleted from CPDL. I like IMSLP (and the proof is that I've donated to them an amount that is comparable to my support to CPDL), but I think CPDL should not duplicate IMSLP and vice versa. --Choralia 21:04, 26 February 2009 (UTC)

Reply by: Chucktalk Giffen 23:04, 26 February 2009 (UTC)

 Help 

Amen to that, Max. Thanks for your information.

It seems we have two questions here : 1. Texts-without-score : They aren't useful without music for our choral project. When scores will be added, then we'll add the text on the work page. Were these texts edited by CPDL contributors? So, if we don't want to delete them, let's just archive them. BTW, it isn't a shame to be also a "lyrics server" if we display texts of the public domain, with the corresponding scores: some are gems of world poetry, some are available only on CPDL, in the whole world. 2. Links with IMSLP : there are two limits : a. copyrights, only public domain has to be linked. b. PDF from music edition software against PDF from scans. This double check will take time but is mandatory. What kind of link? We have three kind of links : pdf (5598), extpdf (393) and net (4946) links. Let's try to get more extpdf/extmid links and minus 'net' links, where sometimes it's difficult to reach the score (I'm speaking of Abel diMarco). A link to IMSLP in general would be to no avail, only useful as extpdf/extmid links. - Claude 08:24, 27 February 2009 (UTC)
I agree with most of what you're saying there, Claude, but I'd like to point out one serious objection of mine to your suggestion that we should aim for more extpdf/extmid score links (ie. more direct external links to files, or as I like to call them, "deep links").
[Here beginneth the essay!] Firstly, these deep links are difficult to maintain. For example, a website change its structure or disappear completely from that domain name, to appear somewhere else (as we have seen with Adrian Cuello and Rod Mather). In this situation, we'd have to change every single score link (as we did with the Mather editions) manually. To avoid this, someone suggested the excellent Template:Website to standardise things. Now if we want to link to an external site in the future, we add it to the template then use only the template for those score links. Should the website move, we only have one place to change the address - in the template.
Secondly, deep linking to a resource which one does not own or have permission to deep link to is considered bad "netiquette" (net etiquette) because it uses another site's bandwidth without actually linking to one of their web pages and giving them visitors. Even when/if we have permission to deep link in this way, I still believe that it's a bad idea for the above reason and also in most cases, users of CPDL will find out more information about a work if they visit the external site rather than being sent directly to the file. In the case of the Mutopia Project, for example, there is lots of useful information on their equivalent of our "score page", eg links to PDFs of works in alternative paper sizes. See this page for example, which was originally linked to like this.
In an ideal world, of course, all the resources on CPDL would be hosted by CPDL but that's just not going to happen. I accept that there are advantages to have direct deep links, that it's less hassle for users trying to navigate an external site, but I believe the arguments I've made above outweigh that small advantage.
Where the contributor has added deep links when they add the edition, I then find it difficult to change those links - it's clearly more convenient for them if the links from CPDL are direct. However, I believe we should discourage it as a general practice. [Here endeth the essay!]
Some of the rest of you may be thinking that I sound like a broken record (see JHF's talk) but I think this issue is quite important. Sorry for babbling on! --Bobnotts talk 10:44, 28 February 2009 (UTC)