IMSLP talk:Score submission guide/Layout of Work Pages

Hi Snailey. Just to avoid any misunderstandings, I've reverted your last edit. The reason is that we shouldn't use "Complete Score of 1st movement" as an example, because "Complete Score" is intended to mean the whole work, not just a section. For a movement by itself we'd use "I. Allegro" (or whatever the tempo happened to be), "III. Scherzo", and so on.

I liked your point about "Horns 1 and 2" meaning something different from "Horns 1, 2", but this is something we need to think more carefully about, as it could have unintended consequences. Can I get back to you on this? -- P.davydov 17:09, 23 February 2009 (EST)

Well, I understand. But we have already covered complete score above. Perhaps we ought just remove it.-- Snailey Yell at me Email me 12:21, 26 February 2009 (EST)


Pedantry, and forcing square pegs into round holes

Rant mode.

Normally, I wouldn't have anything important to say on this discussion, since I tend to think the layout of a page should be intelligible at first glance, before considering such matters as whether it conforms exactly to the suggested site-wide style manual. Obviously, having a page which reads as an intelligible whole is not such a concern as ensuring consistency of style.

I was motivated to turn up here, however, by a rather wrong-headed group of edits, which I propose to dissect in detail. The page in question is for Benvenuto Cellini, the opera by Hector Berlioz. Not an overture, an opera.

I've recently put a good deal of effort into the Berlioz pages, not only by adding a variety of new scores, but by attempting to bring the pages up-to-date with modern scholarship. That this is (lamentably) necessary is due to there still being people around, for example, who think Les Troyens was composed as two operas, not one.

A discussion was held over at the forums on the wrong-headedness of the use of the composer's full baptismal name Louis-Hector, prior to the composer name actually being migrated. Opus numbers have long been deprecated as a standard of nomenclature since the publication of Holoman's catalogue. So the division of works pages follows the catalogue; the basic principle is, one catalogue number = one works page. This system seems to work very well for the Holoman catalogue, and other intelligently designed catalogues such as Köchel's and Deutsch's, compared with the illogical Searle catalogue of the works of Liszt.

Most of Berlioz's compositions are well served by having a single works page to catalogue number, though as usual there are exceptions. Some works exist in multiple revisions made at different times. Some works are collections incorporating other distinct works. In any case, there is a published reference work to use as a starting point for intelligently listing Berlioz's works within the IMSLP page structure.

The page for Cellini did this: the principal work is an opera, however there are no scores of it uploaded to IMSLP presently, hence a prominent note was made to this effect; in actual intention, this is an open request for someone to spot the very obvious gap in the repertoire, and find a way of filling it, since the full and vocal scores are not easy to get hold of.

However, there is also a subsidiary work, the concert overture, which very justly is one of the major, frequently performed works. So essentially there are two works under discussion here (in the case of Les nuits d'été, there are two principal versions and as many as six, in practice, twelve subsidiary works). Hence, instead of the prosaic Complete Work I offered "Complete opera, H 76" as a level-two heading. The page title doesn't say Benvenuto Cellini, opera; therefore this is as good a place to call the work an opera as any, rather than having to scroll down all the way to where it says so in the general information section.

Likewise, the major subsidiary work - the concert overture - is given a level-two heading, which was as good a place as any to note the different catalogue number H 76B, and the fact that it was the overture, not the opera, that inherited the opus number Œuvre 23. Again: specific, useful information that one would rather not have to scroll all the way down the page in order to read.

Under each of those two major headings - based on the Holoman catalogue - the normal system of level-3/4/5 headings should work without any confusion or major cognitive dissonance whatsoever.

This however did not suit the style pedants, and so the page had to be made to CONFORM:

  • Item. It is forbidden to refer to a work using a genre-based description such as "opera". The only permissible word is "work". Useful information must instead be buried at the bottom of the page in the General Information section.
  • Item. It is forbidden for the file description parameter of a file to actually describe anything useful about the file, such as:
    • provenance of the score: the fact that the score belongs to the "Old Berlioz Edition", has no relevance;
    • title of the work on the score itself: the fact that the editor followed the example of the composer by putting a freaking huge title right across the entire width of the first page of the score, "Grande Ouverture de Benvenuto Cellini", has no relevance;
    • listing the break-down of movements within a score which would otherwise unhelpfully be described as "Complete Work"; has no relevance.
  • Item. If a level-four heading, e.g. "Overture" is thus exactly the same as each of the "File Descriptions", it is forbidden to remove or alter either the heading or the description to negate the appearance thereby created, of almost complete redundancy.
  • Item. The use of the phrase, "Typeset by ... using Sibelius" is obviously potentially confusing. Thus all such references must be disambiguated by adding the word "Software", even in contexts where the potential for ambiguity is virtually non-existent.
  • Item. It is not good enough that the orchestral part is entitled "Violoncello" on every page of the score; for standardisation this must be changed to "Cellos".
  • Item. The terms "Orchestral Parts" (i.e. for orchestral music) and "Instrumental Parts" (i.e. for chamber, ensemble or other music forms) are needlessly complicated and must be replaced by "Parts".

Okay, so all of this is completely over the top, but perhaps instead of worrying over whether every single word matches an illusory hope of complete standardisation across every genre, every style of work, every type of work page, the pedants pushing this crusade might consider that the words that are actually written, might have been put there for a reason.

Rant over. Philip Legge @ © talk 20:35, 25 February 2009 (EST)

Hi Philip. I'm sorry you're upset. What you say is true to some extent, except that we're faced with hundreds of different-sized pegs and awkwardly-sized holes. As Wikipedians can testify, this may be workable on a small-scale, but eventually it can become bewildering to new users who need to find information quickly. This was my experience, and after six months as a frustrated (and many years' experience as a music bibliographer), I made some suggestions that I hoped would be helpul. These were publicised on the forum and refined and rewritten over a period of for several weeks, and as a long-time and valued contributor to IMSLP your views would have been very welcome.
I hope that my edits last night didn't give the appearance of the proverbial bull in a china shop, but they were all done with good intent, and "for a reason", to borrow your phrase. For example, there are reasons why the use of "Cellos" was agreed instead of "Violincello" (was there just the one instrument before?), or "Parts" instead of "Orchestral Parts" (should there be separate headings for solo and instrumental parts?). The structures and levels of headings have also been devised with careful thought as to how they will work throughout the site, and what information will be conveyed in the contents box at the top. (See User talk:P.davydov/standards‎ for the discussion history).
If I inadvertantly removed anything whose significance I didn't appreciate, then please point it out so that I will know better in future. (I thought that I'd retained the note about the "Old Berlioz Edition" in the "Misc. Comments" section, but apologies if that wasn't the case). While there may be a case for having the overture to Cellini as a separate work page from the opera, I'm not going to argue that here.
Please rest assured that the last thing I (the "pedant-in-chief"?) or anyone else wants to do is trample insensitively on other people's hard work. If this is how it's coming across then I will stop right now, without any hard feelings, and use the time for other projects. But while IMSLP is already a great thing, there is the potential for it to become an encyclopedic and truly indispensible resource for all serious music lovers. This can't happen unless we lay down solid foundations now, which is the object of this exercise. You're more than welcome to contact me publicly or privately via the Forum or my homepage so that we can try to resolve any problematic issues — P.davydov 04:19, 26 February 2009 (EST)
Fair enough. All I saw, having edited the page barely a month ago, was that a huge amount of the useful, significant information had simply been eradicated by the act of attempting to have the page conform to a bland and unvarying standard.
For example, from the point of view of a conductor, it is important to have a score highlighted as "Old Berlioz Edition" versus "New Berlioz Edition" or some other edition entirely, rather than the information being buried under "Miscellaneous Notes"; what is "miscellaneous" about the fact that a particular score belongs to a 19th century tradition – and one, in terms of the Breitkopf & Härtel scores, which was actually foreign to Berlioz's orchestral ideas – or to the 20th century? You did, by the way, retain the "Old Berlioz Edition" marking in the "Misc. notes", whereas I tend to view that the edition of the score is somewhat more important and worthy of a bold description. That's why headings are made bold, so that important information stands out.
It is well known that the Old Berlioz Edition scores are a quite distant second best, and yet because the new editions from Bärenreiter are more difficult and expensive to obtain because of copyright and hire issues, and not so widely and cheaply available, they are still frequently used.
From the point of view of the orchestral librarian, is it significant to differentiate between Violoncello, Violoncellos, Violoncello e Contrabasso, Violoncelli e Contrabassi? Absolutely. Explicit divisi notated in an orchestral part should be indicated by the plural; inessential divisi (or ones that can be accomodated by multiple stops) by the singular. Titles and descriptions are sometimes made "non-standard" to reinforce a point.
However, I suspect that most users don't read much of the small print below the bold lettering that describes headings and filenames: the fact that a score might have been printed at Leipzig: Breitkopf & Härtel, (n.d.) plate no., blah blah blah – is not important to most visitors.
On the other hand, if you want to get hold of an opera, does it matter if the heading says "Complete work"? No; but on the other hand, if you only want the overture, then differentiating the idea of "Complete work" = opera from that of "Complete work" = overture does become an issue. So why go to the trouble of eradicating information such as "Complete Opera", because you need a one-size-fits-all term? Why is "Parts" thought to be a better term for generic reasons, than the more contextual designations "Orchestral Parts" or "Instrumental Parts"?
Yes, the many Wiki pages across the site are inconsistent and could be standardised. But why do so at the cost of removing useful information that has already been presented there. You could better spend effort to tidy the pages that really do need some proper organisation, rather than stripping the pages that need only a modicum of improvement. My €0.02... Philip Legge @ © talk 05:52, 26 February 2009 (EST)

Hi again Philip. When asked, most people seem to agree that having a consistent across-the-board set of standards is a good thing, but then they also ask for exceptions in particular instances where they think that bending the rules can be justified. This is perfectly natural, and something I'm certainly not immune to myself. It does, however, present a difficulty in finding a happy medium between ruthlessly imposing an iron-clad set of rules on the one hand, and having everyone doing their own thing on the other. I'm finding that the navigation of this narrow channel can prove rather hazardous, but I think it's worth giving a go :-)

To addressing some of the specific points you raised:

  • "From the point of view of the orchestral librarian, is it significant to differentiate between Violoncello, Violoncellos, Violoncello e Contrabasso, Violoncelli e Contrabassi?" — Absolutely, and in this case "Violoncello" in the singular was being used to refer to the whole cello section. That's why it was changed to the plural. Similarly we speak of "cello concertos", not "violoncello concertos", hence the preference to use "cellos" throughout IMSLP.
  • Headings — we are effectively limited by MediaWiki to three levels of headings (3, 4 or 5 equals signs) on work pages, and have to cut our cloth acordingly. The new heirarchy has been designed so that the headings will appear in the same sequence in the contents boxes at the top of each pages; it provides the same type of information for all categories of works, and allows for files to be added without having to move headings that were already on the page. In the case of the heading for "Old Berlioz Edition", I'm tempted to believe that specialists who already know about the idiosyncracies of this edition will be carefully reading the publication details anyway, while it might mean very little to anyone else. But I'd agree that the bilbiographical descriptions could be improved throughout, and this is somethng we should look at further down the line.
  • Complete Work vs. Complete Opera — on the face of it this seems quite reasonable; except that around half the scores in the "Opera" category weren't designated as "Operas" by their composers. There were more properly "Singspiels", "Lyrical Scenes", "Masques", or any number of varied designations. The questions then arise whether we should term them all "Operas" regardless, or use "Complete Singspiel", "Complete Lyrical Scenes", etc. A more practical alternative is to use "Complete Work", which can also be used for any other type of composition or literary work. I appreciate that this may seem a bland catch-all approach, but in future it will allow IMSLP to provide advanced search facilities where the user can readily identify complete works where they are available.
  • Parts — this heading was arrived at after some debate; if we used "Orchestral Parts" (or "Orchestra Parts" as it appears on some pages), then the point was made that this would raise difficulties in the case of concertos, where the solo instrument part would strictly speaking have to go under a different heading. The term then evolved to "Instrumental Parts", until the question of solo voices and choral parts in vocal works was raised. Hence the generic "Parts" (which can also be used consistently without requiring specialist knowledge of the instrumental forces required for particular work). The alternative of varying the heading depending on the type of work involved could also restrict the possibility of future searches, and preclude the option of having "tabs" on work pages to differentiate between full scores, vocal scores, parts and arrangements (which I understand is under consideration).

Besides the above issues, many pages contain a great deal of invaluable information from contributors that appears in the "Comments" section of the IMSLP work page, or in the parts of the IMDBP data which are unfortunately not visible outside the "General Information" section normally displayed. There must be ways in which this type of data can be better utilised, and other important details (such as the key signature of a work) might be added to the basic information shown.

I'm currently in the process of working alphabetically through the opera category, implementing the standards as recently agreed, and my editing of the Benvenuto Cellini page was done as part of this process, and please rest assured that it wasn't singled out as being in need of special attention. I apologise if I was insensitive to all your past hard work, which was not my intention, and if there is anything I'm unsure about in the other Berlioz pages then I promise to seek your advice first. But I hope you'll agree that overall it is a worthwhile task, and something that has to be done before things get so badly disorganized that it impedes the future development of the whole project — P.davydov 11:34, 26 February 2009 (EST)

  • Being limited to 3 level headings was the reason I originally questioned why there needs to be a top level heading "Music Scores" on every page - this could be assumed and removed (and books etc. could have top level headings that make sense for them.)
  • I do think it is very important to have a consistent, conforming, and well thought-out page layout - the reader will get used to seeing it, and be able to see what there is and find what they're looking for much more easily. I've been making suggestions for the vocal scores - i.e. the person setting this up may not have any special insight into what is most important to know about something they're not used to using (i.e. the language(s) that a vocal score uses).
  • I agree that a "Complete Work" heading could be useful on all pages, even if only excerpts have been uploaded so far - just to make clear that the score you can't find isn't buried somewhere, and to let people know that it needs to be uploaded. This is different from leaving out the "Vocal Score" category in an instrumental work.
  • It also makes sense to use the generic "Complete Work", to avoid arguments about what some pieces are supposed to represent.
  • An extract, such as an overture, with it's own catalogue number could be given a top level heading and listed before the Arrangements and Transcriptions.
Operalala 13:11, 26 February 2009 (EST)

Hi Operalala. Many thanks for your suggestions. I'm told that there's nothing we can do about the "Music Scores" heading for the moment, so it's something we'll have to work around for now. I agree with the rest of your suggestions, except I feel strongly that the contents box at the top should only show what we actually have, and that we shouldn't put "Complete Work" if it hasn't been added yet. In a case like Benvenuto Cellini, it might be better to put the overture on a separate page, as it has a different publication history and catalogue number than the rest of the opera (although thematic catalogues can differ considerably in the way they deal with versions of works).

I've started to include sub-headings for the languages for vocal scores, putting them at "level 5" (the same for languages of books). But there are some cases where it isn't clear which language is being used, and the file downloads are so slow at the moment it's difficult to check. Is this something you'd be willing to help with? — P.davydov 14:51, 26 February 2009 (EST)

Structurally, the "Music Scores" heading arguably makes sense, as it is at the same level as "Recordings" and "General Information". Unfortunately, the wiki syntax doesn't provide a way to mark an "appendix" section on the page, so I'm not sure how you would properly structure the page omitting the "Music Scores" heading. --Leonard Vertighel 15:36, 26 February 2009 (EST)
Is this something you'd be willing to help with?
Sure - I've been adding languages to pages I come across. Where is it not clear what languages the scores are in?
Some of the Handel operas, for example (I've only looked at the ones beginning with 'A' and 'B' so far. P.davydov 05:06, 27 February 2009 (EST)
I'd like to see clear linking (both ways) between excerpts with a life of their own and their parent works. I.e. perhaps a heading in link form where the excerpt might be expected to be found in the parent work.
If the excerpt or 'derived work' (such as a suite) is on a separate page, then we can use cross-references at the top of each page, e.g. as on Tchaikovsky's "The Nutcracker (ballet)" and "The Nutcracker (suite)" P.davydov 05:06, 27 February 2009 (EST)
Operalala 20:06, 26 February 2009 (EST)
I was going to stick my two kopecks in but I see P. davydov has already mentioned giving the overture its own page. This seems reasonable to me, since it does have its own catalog number and a different publication history. Also, there's somewhat of a precedent with Tchaikovsky - where there are separate pages for the Voyevoda Overture and opera. Actually, there is a reason explicit publication data is important: Such info prevents certain publishers from making ridiculous claims about items being under copyright, which happens far too much. Carolus 22:32, 26 February 2009 (EST)
I can see no particular reason for the concert overture to be placed on a separate page, even with some different publication history, when there will actually be few versions of the complete opera that can be legally submitted here: Choudens' 1880s full and vocal scores, or possibly facsimiles of the numerous extant manuscript sources. Besides, it goes against the "rule" that Holoman put both the opera and the overture under the same catalogue number, 76.
There is another issue with the semantics of the heading of "Complete Work": who decides whether the work is complete? The composer? The editor? The publisher? Or the IMSLP wikignome du jour? There are also several categories of works which may be described in a performance sense as "complete" (e.g. there is a "complete" version of Puccini’s Turandot) but not in a musical or artistic sense (e.g. the "complete" score by Choudens is not actually complete, with major parts of Acts IV and V having been cut).
I will acknowledge the issue with headings, that although they may be bland, functional, and uninformative, it may make sense for them to be consistent if there are particular search mechanisms in place, or different future methods to view IMSLP works pages à la tabbed browsing. But my understanding was that if Composer X called his work an operetta, a singspiel, a cantata, an opera buffa, or even called it "an act of artistic suicide", such hair-splitting wouldn't really matter, so long as for the purpose of categorisation, the genre was specified as one of a more limited set, so that the item was added to the appropriate category. The whole advantage of a Wiki is that you can use generic categories to unify under multiple classifications a heterogenous collection of works; which is currently being ignored.
If there is to be virtually no flexibility in terms of level 3/4/5 headings, then I fail to see the reason for not allowing greater flexibility in the choice of what is displayed as the File Description, which is the main item to catch the eye when looking at any individual score or batch of scores (unless, say, there is a thumbnail picture associated with the score, like this one). Instead, I note that the style guide says this:
The sole purpose of the File Description field is to show whether the file concerned represents the whole work or just an extract from it.
I would have to state my complete and profound disagreement that there is a sole purpose, when often the File Description is frequently the first thing to hit the eye when looking at an IMSLP score. It stands to reason that if there is something important that is worth stating about the file that it be mentioned there, rather than the same information be buried in the Miscellaneous Notes along with all of the other miscellaneous stuff! This is not an issue of "bending the rules", this is an issue of (not) presenting information as well as making it more difficult to navigate the website, besides dumbing down the expectations of visitors to the website. Regards, Philip Legge @ © talk 23:37, 26 February 2009 (EST)
I think that style-guide line about the "sole purpose" of the File Description field is quite obsolete and should be changed. As for a separate page for the overture, you're right about there not being very many items that could be potentially uploaded for the opera, so the real question here is what will be least confusing for the prospective user. It's very difficult - perhaps actually impossible - for all works to fall into a single organizational pattern. That's why we were going back and forth for several days in attempting to come up with a rubric that would work in the majority of cases and at the same time retain some flexibility. The ultimate purpose - always - is to make things as clear and concise for the user as we can. There will no doubt be cases where - despite our best efforts - the clearest and most concise layout or hierarchy for an item will demand that we violate the "rules." Carolus 00:05, 27 February 2009 (EST)
I think some people are assuming that there is one definition of "what will be least confusing for the prospective user". I could be totally post-modern and say something like "there are as many definitions of what is confusing as there are prospective users", but I think there are probably a limited number of different techniques for using the site. For example, if someone came here looking for Mozart's Coronation concerto, I very much doubt s/he would simply click on the "Random page" several thousand times until the right page turned up. :-)
Nevertheless, I hope I'd find some agreement with the propositions that there are aspects of IMSLP's categorisation that are very good, while some are abysmal. For example, have a look at this template, and try to quickly locate the correct Mozart concerto: (a) for three pianos; (b) the "Coronation" concerto; (c) the one in C minor.
Obviously the current version of the template is a vast improvement. On the other hand when you visit the Category page, you're back to the same problem as before. This is why, when you look at each of the pages for the piano concerti, that the first thing about the heading of the Old Mozart Edition full score that jumps out at you, is not "Complete Score". Instead: (a) Concerto in F for Three Pianos and Orchestra, K. 242 »Lodron-Konzert«; (b) Concerto in D for Piano and Orchestra, K. 537 »Krönungskonzert«; (c) Concerto in C minor for Piano and Orchestra, K. 491.
What I'm worried about is that in the interests of "standardisation", we might end up at the point where you have to download the score in order to actually discover that it's (a) the one with three pianos, (b) the one Mozart played at the 1791 coronation, (c) the one in C minor; because we're so afraid of actually putting useful information in front of the users.
You might think this is a ridiculous example, but have a look at the edit date on the template, as well as on the concerto pages themselves: those changes were only implemented a month ago. Before 2009, if you went to any of the Mozart concerto pages, the only place telling you that simple, basic information, was to be found in the General Information box. (Or not! I didn't touch any of the pages over at IMDBP, although I should have, because they are a scandal also. Life is short...)
If anything, trying to find the right Mozart symphony if you know the key or the Köchel number, but not the symphony number (between 1 and 55, somewhat arbitrarily assigned by Breitkopf) is worse: no key signatures, virtually no Köchel numbers. If this is the case for Mozart, what hope for any number of less influential composers? In my (not so) humble opinion: pathetic.
Despite this last statement: regards, Philip Legge @ © talk 03:07, 27 February 2009 (EST)

Hi Philip and Carolus. Thanks for your last comments. I agree with Philip that there are fundamental issues that need to be addressed with regard to the existing page templates, IMDBP, etc. However, I'm optimistic that we can come up with solutions if we concentrate our efforts. After all, what choice do we have if the site is to expand much further? It's inevitable that any consensus reached will run counter to some of our own personal preferences, and there are some things in the recently-agreed standards that I'm not particularly happy about, but I have to accept that the arguments for them are stronger than for my alternatives.

The key, I think, is to look forward to a time when IMSLP has all possible scores, parts and arrangements, and then to envisage how this "ideal" site would look. This is what Carolus and I attempted to do recently, and I can understand how the some of our recommendations might seem odd in cases where there are currently only one or two scores for a particular work. But it will also make it easier to see what is missing, and (hopefully) encourage people to fill in the gaps.

I mentioned above that one of the ideas being discussed is the idea of having several 'tabs' or sub-sections on work pages, which would enable people to look just at the full scores, vocal scores, arrangements, or specific versions of works. I can see great advantages in doing this, but with 14,301 works and 26,767 files at the time of writing, the process of restructuring the pages would inevitably have to be automated. The software bot then carrying out this process will have to rely on the pages following a reasonably consistent layout, looking for certain "===" level headings for instance, or for key words and phrases to decide where the sections begin and end. But look at any random sample of pages at the moment, and you'll find that this just isn't possible.

This is just one way in which IMSLP's growth will be stunted unless we agree on some basic universal fundamentals. I strongly agree with Philip that there must be room for the kind of additional information he mentioned with the Mozart examples (perhaps under the "Misc. Comments" heading, assuming that IMSLP users won't so dumbed-down that they won't read anything unless it's in a large bold heading). And we also need a debate on which languages should be used for work titles, as Philip and others have suggested, and on expanding IMDBP, etc. Now that the Forum seems to be back up to speed, perhaps it would provide the best way to start these discussions... P.davydov 05:06, 27 February 2009 (EST)

For the record, I was not suggesting that IMSLP users wouldn't take the time to read publisher details or comments, though I suppose this assumes a certain type of method in using the site that is unconstrained by pressure of time. What I was suggesting was:
File Descriptions, being given prominent headings in bold, should be descriptive. For all sorts of reasons this might include extra information rather than the totally bald statement of "Complete Work".
Misc. Notes, being relegated almost at the bottom after an editor, publisher's info and copyright field, and sometimes with an Urtext box or some other distraction immediately below, these comments should be properly miscellaneous, as in a mix of comments that cannot be put anywhere more fitting because they are not sufficiently important.
For example, a full score with several important instrumental parts missing (I can think of one well-known and frequently performed work by Schubert) should not have that information buried under "Misc.", or to be unmentioned at all as an "assumed" fact, because the traditional version leaves the parts out and an "urtext" version might put them back in:
Misc. Comments: no trumpet and timpani parts
That is what I have argued all along about intelligibility. Important information (a subjective viewpoint, I am aware) should not be relegated to a miscellany, or removed altogether. Best regards, Philip Legge @ © talk 17:46, 27 February 2009 (EST)
Hi, Philip. You make some good points which merit further discussion. So for the moment I'm concentrating on the headings/heirarchies of the pages, and only alter the descriptions and misc. notes if they're plainly wrong or misleading. With 300 pages done there's only another 14,000 or so remaining <gulp>, and I need a good night's sleep before trying to sort out the various versions of Don Carlos... P.davydov 18:17, 27 February 2009 (EST)

FYI, I just noticed wiktionary uses tabs within each article. --Peter talk 13:34, 27 February 2009 (EST)

Hmm, interesting... Would you know if this is the type of thing being considered for IMSLP? Thanks — P.davydov 16:33, 27 February 2009 (EST)
No, just an idea. Whatever we would place under that second tab, it would become unvisible for a lot of unexperienced users who would just never look at what clicking a tab in this first line would do.
  • I took a look at Handel operas A-B, and didn't see any vocal scores...
Sorry, no there are only full scores for these works. I left a note on your talk page to tell you that I'd done the rest (now into the 'G's).
  • I'd like to see excerpts and derived works linked from the point on the parent work's page where a reader would expect to find them as a part of the whole. I think readers in this wiki would be more likely to glance at the Page Title, then scroll down to their desired score, and will be less likely to notice parenthetical links above the title. Even in other wikis I know I don't notice these links until I realize I'm not on the right page, and in this wiki, you could have people scour the whole page for a missing work, and not notice a link at the top.
I thought that this was already the case, but if not then could you give me an example of what you mean? Thanks — P.davydov 11:49, 28 February 2009 (EST)
Using an example mentioned above, The_Nutcracker_(ballet),_Op.71_(Tchaikovsky,_Pyotr_Ilyich), the link to the "suite" is at the top of the page, rather than within the the article, under arrangements, where a reader would be more likely to look for the suite arrangement. The link may be at the top of the article, but if you don't know beforehand that the suite is a separate article, a misc. link at the top of the page will be very easy to overlook. Operalala 19:41, 28 February 2009 (EST)
Ah, now I see what you mean. OK, I'll see to it... P.davydov 04:27, 1 March 2009 (EST)

Order of publications within one subsection

Not trying to mingle in the above discussion, I think the issue of the order of the different publications that reside in the right same subsection, has not been discussed yet (the guideline only mentions "chronological order"). This might not be most important at this time, but as above look at a time where IMSLP will contain all known public domain editions of a single work - think about the possible length of the pages of Beethoven's piano sonatas where Carolus already started to reorder.

So here we go, a proposition

  1. composer's manuscript
  2. copyist's manuscripts in chronological order ?
  3. first printed edition(s)
  4. further editions in chronological order

issues coming up in this proposal:

  • composer's revisions that are published at a considerable interval after the first edition. are they placed on top under the first printed edition, or down the list in chronological order.
  • where do we place reprints in this chronological order
  • in the case of arrangements and transcriptions within a certain instrumentation, do we order chronologically or on the arranger's name?
  • typesets
    • new typesets: date of creation/publication?
    • typesets that are based on a single historical edition (and claim to be an identical copy) ?

a little bit relating to this discussion; is it worth to have or to actively delete the following files on the server:

  • a facsimile reprint where e.g. front matter or plate numbers have been removed (eg Dover reprints)
  • a simultaneous issue under different publishers, with differing front pages and plate numbers
  • multiple scan versions of the very same edition
a) where the high quality file has a very large size and the lower quality a significantly lower size (e.g. 25 MB versus 4 MB)
b) differing scan versions with only minor differences in file size, the latter not being indicative for the quality (e.g. 2 MB versus 5 MB)

--Peter talk 06:35, 27 February 2009 (EST)

Hi Peter. In most cases a simple chronological arrangement of sources and editions is likely to reflect the sequence you proposed (i.e. composer's manuscript; copyist's manuscripts; first printed edition(s); further editions), with typesets last coming last. However, there may be difficulties where the date of a manuscript or date of an edition isn't known; and particularly in the case of reprint editions where the original source is unidentified. As for choosing files for deletion, I'm steering well clear of that hornets nest!  :-) P.davydov 08:33, 27 February 2009 (EST)
This is not the only issue, when imagining a future IMSLP that has virtually any version of any piece of the basic repertoire (for works outside the frequently played repertory, I imagine there will tend to be fewer versions).
Ensemble music normally involves performers working from different parts and scores, and those materials need to be compatible. For a variety of reasons there are a sizeable proportion of compositions that were never left in a suitable final state, or went through numerous revisions, or are traditionally performed in a different manner to the intention of the composer, even if we are lucky enough to know what those intentions might have been.
There are numerous works where for one reason or another the vocal scores are badly matched to the orchestral parts, having anywhere from a bar missing here to an entire section there; or different sets of orchestral parts and full scores with contradictory rehearsal figures/letters. These are the sorts of things one tries to "avoid like the plague"!
It is not clear to me either how the page structure will resolve this problem of multiple versions of a similar but not quite identical pieces of music in close proximity. Obviously one would not have, for instance, the 1919 version of Sibelius' Fifth Symphony rubbing shoulders with the orchestral parts of the 1915 version: you would treat the 1915 version completely separately. For cases that are not so clear cut, where would the dividing line fall? Would the average IMSLP wikignomes be expected to know? Would the end-users?
Mmm, questions... Philip Legge @ © talk 07:02, 27 February 2009 (EST)
Good questions indeed, which have troubled many a music librarian over the years! I'm not sure it's necessarily IMSLP's role to come up with definitive answers to these centuries-old issues, but it underlines the importance of providing accurate information on the sources of scores (and there are far too many where the field for "Publisher information" is completely blank). Our new work page guidelines have introduced standard headings to differentiate between versions of works (see Tchaikovsky's Romeo and Juliet, for example), which is hopefully a step in the right direction. But if the person uploading a score doesn't know that it might exist in more than one version, we'll have to rely on the unfailing helpfulness and wisdom of fellow IMSLP contributors :-) P.davydov 08:33, 27 February 2009 (EST)

other fields

I added some information about other fields, but just now realized that this might better go into a general style guide and not on this page.--Peter talk 17:09, 27 February 2009 (EST)


I almost start liking this neurotic fixation on the manual of style. Cadenzas are now put under arrangements and transcriptions. To me it has nothing to do with either - it should have a level 1 heading.--Peter talk 10:06, 22 July 2009 (UTC)

when to, and not to, use the parts bar/include something in the pages with parts category

i think one guideline - since i think there has been some confusion and/or disagreement - is that the ===parts=== header should be used in at least many cases, though not necessarily all, where the piece is 'ready to play' - we don't yet have a category that exactly means that, score and all parts available for an orchestral work e.g. - for an ensemble (again, i suppose it's possible to print out 20 copies of a score for a small-orchestra work 1 for each player, but parts are more convenient with more notes per page and less page-turning, etc. not the best reason, but a reason; i'm sure there are yet better.) Eric 17:47, 26 July 2011 (UTC)