Does the category walker really works well?

[Notenschreiber]

Thank you Feldmahler and Kalliwoda, with this searchbox it works well! I should have hit on it by myself.
But nevertheless it may be good to have kicked off this discussion, because the problem remains, that my three double reed players
could not find all the suitable pieces at IMSLP. This would be the case, if all compositions which can be performed (seriously) by only two oboes
and bassoon are tagged either to belong to "For 2 Oboes, bassoon" or to "For 2 Oboes, bassoon (arr.)" . For example: All works, original or arranged, for
2 Oboes and B.c. would then be listed in one of these two categories. The same policy would facilate the search for all possible instrumentations.

[Davydov]

I take your point Notenschrieber, but there's no way to anticipate every possible combinations of instruments that could make up a continuo part. Your friends' best option would be to look at all works involving 3 players on this page:

http://imslp.org/wiki/Special:GenreWalker/For_3_players

... and works for 2 players with continuo on this page:

http://imslp.org/wiki/Special:GenreWalk ... h_continuo

... in order to find promising combinations, of which there seem to be a few.

[steltz]

@Notenschreiber: Please note that a "Concerto" for 2 oboes and bassoon cannot be tagged as a Trio because the word "Trio" is not in the title. The tagging system works by the Uniform Title system, which means the nouns in the title are the tags. Tagging by the number of people involved in a chamber work is also not reliable, since a work with continuo can be played by variable numbers of people.

@Feldmahler: I meant collapsible in the sense that the large categories are:

Work types
Instrumentation
Featured Instruments
Languages

These all fit on one page, so there isn't an issue of thinking something isn't there, or having to go to page 9 to get to a category. But when you click on one of these categories, the number of sub-categories are so large that you get the multiple pages.

If you click "For 1 player", you get on page 1:
Work types
Keyboard
Chamber-Instrumental
Orchestral
Vocal
Featured Instruments
Languages
Periods
Composers
Unknown

Page 2 starts again with Work types and all of the other headings also appear again.

If I remember correctly this was the easiest way to make sure that the most populated pages appeared first in the list, but it can be misleading if what you are looking for only has a couple of works uploaded.

What if the headings only appeared once, and when you click on them, a drop down box appears, or a new page or tab opens with all of the sub-categories for that header.

In Notenschreiber's example, he would click Chamber-Instrumental, and "2 oboes, bassoon" would appear next to "2 oboes, bassoon (arr)" because that category wouldn't be split between many pages -- there wouldn't be some on page 1, some on page 2, some on page 3, etc.

If this is very tricky to implement, since Notenschreiber seems to be happy with the suggestion given, maybe it's better to leave my suggestion. It was just the only way I could think of to get similar instrumentations next to each other when one has many more uploaded works than the other.

[Notenschreiber]

@Steltz: Why is it not possible to tag - for example - the Zelenka sonatas with "For 2 Oboes, bassoon", not as a unique tagging, but as one among others.
It is not comprehensible, that only the Heinichen Trio is to be found in this category.

[imslp]

Steltz: Are you asking for the CW results to be sortable by alphabetical order?

Ideally we could return all the results for a particular section, but that would probably take forever. For example, if currently it takes 1 second to load a CW page, and there are 10 CW pages in total for that CW category, returning all the results for a particular section would take 10 seconds (there is no benefit to ignoring the other sections, as they come free). Of course, this is assuming that there is only 10 CW pages. It is actually a bit of a miracle that the CW system runs as fast as it does right now (it took a lot of tweaking).

In light of this, the closest possibility to what you are suggesting is to sort by alphabetical order, which is the same speed as the current CW. However, you will get a whole lot of categories with only one page in it, and you will probably need to click many many times to get to the page you want.

Maybe the search box has mitigated the problems well enough? (Serious question.)

[kalliwoda]

@imslp:
I had not used this search box before, until you mentioned your search results.
Now I realize how an extremely powerful tool especially for instrumentation searches it is. What is missing is some guidance/instruction to prod people to use it!

@Notenschreiber:
Now that you mention the Zelenka Sonatas, I strongly disagree with your idea. Performers should be aware that they could use works for slightly different instrumentation, especially in the baroque and early classical period. Anyway, if the alternate instrumentation possibilities are part of the title, they will already be tagged accordingly.

Even if there are no alternatives mentioned in the title, performers will know that you can play flute and oboe pieces easily on a violin, but not all of them vice versa on an oboe. But this will always be some matter of personal preference, and not all baroque Trios for example sound convincing if the continuo is played on a bassoon - and three of the 6 Zelenka "Triosonatas" are actually quartet sonatas with a bassoon part that is different from the continuo line, which may imply that the others would also benefit from a doubling of continuo instruments.

[Notenschreiber]

@Kalliwoda. I know the Zelenka Sonatas very well and I know of course, that the instrumentation "2 Oboes and bassoon" is not suitable for all of them.
But there comes up a basic difficulty of our tagging system in connection with the category walker: Not pieces are tagged, but categories. If you narrow
a search for example "for 3 players", then comes a list with duos, quartets, etc. This is confusing at the first moment, but the reason is clear: these
categories contains also pieces for three players.
To come back to Zelenka, if one searches compositions for 2 ob, bs and the category "6 sonatas (jan dismas zelenka)" is tagged accordingly, he is
directed to this category and can decide which pieces are suitable. In any case the single result "Heinichen" as answer of the described search is not
adequate.
If you use the search box with the entry "2 Oboes", you find the category "For 2 Oboes, continuo", which contains the 6 sonatas of Zelenka.
Of course this is a way to find more possibilities for the instrumentation 2 Ob., bs. , but it would be nicer to find them by an entry which coincide exactly
with the wanted instrumentation.

[Notenschreiber]

I had a closer look to the category "For 2 Oboes, continuo" and I realized, that the tagging goes in the direction, which I would prefer:
The pieces in this category are mostly not pieces especially for oboes, but could well performed by them.
I had also a look to the sheet music of the Heinichen trio, where indeed the score looks like a trio for 2 Ob. and bs. This - and the title -
may be the reason for the exceptional tagging of this piece compared with the other compositions for 2 Ob. and continuo. But I am not
sure if this is justified.
I admit, that there are perhaps to many possibilities of performing the continuo (with cello or basson or contrabass, with or without
cembalo, with organo, etc.) in order to take account of them while tagging the categories.

[feldmahler]

@kalliwoda: I agree that it should be advertised more. In terms of usage it should be rather easy to use, but with one wrinkle that people probably don't know about. Normally it searches for each word (separated by spaces) in each category name, but if you put in a slash (/) it will search each section delineated by the slash. For example, "2 oboes/bassoon" searches for "2 oboes" and "bassoon", whereas "2 oboes bassoon" searches for "2", "oboes" and "bassoon". Otherwise it is really a simple text search.

@notenschreiber: Considering the complexity of the tagging system and the (sometimes) subjective nature of the result, I would leave the final decision to the librarians. It may be reasonable to disagree, but barring some systemic problem (or if the tagging is blatantly wrong), I would let the librarians have the final say on the issue, since they are the ones who do the tagging work.

[Notenschreiber]

I would like to add, that the Heinichen Sonata at IMSLP (from SLUB) coincides with the "Sonate für 2 Oboen und B.c in c-moll" (Schott).

[steltz]

@imslp: the fact that there are so many categories with only 1 or 2 pieces in them was the reason that I thought a drop down box (with alphabetical categories) would help, but in any case, it looks like the problems experienced in this particular instance have more to do with tagging, so I think perhaps this idea should be shelved at least for the moment, since there are problems with implementing it and it might not solve the problem as well as something else.

@notenschreiber: you might want to play pieces for 2 treble instruments and continuo with 3 people, but other people might want to play them with 4. The basic problem (or flexibility, depending how you look at it) wth continuo parts is that they can mean anything you want them to mean. A good lute player can play them. We can't tag for 3 players, 4 players, lute players, guitarists, harpsichordists, in other words all possibilities, because it creates an infinite number of tags and a lot of clutter on the page, which nobody wants to look through because it's cluttered.

The librarians are aware of the possibilities, but the single tag "bc" covers all of them, and the decision was to tag just for the continuo and leave people to decide how many people and what instruments would constitute the continuo part. Your best bet is to add "2oboes, continuo" into the categories you want to search.

As to using "trio" where it isn't in the title, we have the category "for 3 players" so it is redundant to use "trio" for anything that uses 3 players.

Unfortunately, if you want to add continuo into the mix, there are 3 categories you have to search: 2oboes bassoon, 2oboes bassoon (arr), and 2oboes continuo.

Having said all of that, there may very well be errors in some of the tags -- I can't believe we've tagged nearly 40,000 pages without at least a couple. Also bear in mind that we rely on what people enter on the work pages, and they don't always give complete or accurate information. Also, sometimes later information changes the tag. I had to re-tag something a few days ago that I initially tagged, because the original uploader put it as piano accompaniment, but someone subsequently found an earlier manuscript, and it was continuo. It wasn't a mistake at the time, but became one later.

If you question a tag for something please bring it up under in the Categorization forum and we will have a look at it.

[Notenschreiber]

After all it would be the best to eliminate the tag "For 2 Oboes and bassoon" from the Heinichen Sonata in order to avoid the expression, that
this sonata is the only one at IMSLP (besides arrangements) which is possible for this instrumentation. As I pointed out the Heinichen sonata is not
an exeption and the category "For 2 Oboes and bassoon" should be empty.
If somebody is looking for original compositions which can performed by these three instruments, he must have a look at "2 Oboes" at the search box
and find out, what else is suitable.

[steltz]

First of all, I just want to make sure I'm looking at the right piece - it is the Trio Sonata in C minor, SeiH 259. I downloaded the manuscript score for this, and also quite a while ago, the parts. The manuscript (autograph?) score clearly says 2 oboes and "basson". The parts say "basso", which usually happens in continuo parts. For this reason, this work was tagged for both possibilities. (This doesn't normally happen unless there are different authoritative taggable possibilities in different places.)

At the moment, I am reluctant to remove the 2ob bn tag because the title specifies it, but if it isn't an autograph manuscript, and some authoritative source gives something different, then we should consider it. I need to start with who wrote this manuscript.

Does anyone know if the Heinichen collection of scores at SLUB Dresden is an autograph manuscript collection?

[Notenschreiber]

Yes, I mean the Trio Sonata in C minor, SeiH 259. (Of course, this is the only one tagged 2 Ob bs).The manuscripts at IMSLP are part of the famous
"Schrank II" collection of the Dresdner Konzertmeister Pisendel (http://www.schrank-zwei.de/). One of them is written by the copyist Grundig with his
typical very clear handwriting, the others from other copyists of the Dresdner court. (http://opac.rism.info)
There are several publications of the Triosonata
(http://haynes-catalog.net/works/show/13 ... -627647478).
I already mentioned the one of the Oboenbibliothek of Schott.(http://www.schott-musik.de/shop/1/show,80665.html).
I want to mention at last, that it is in any case clear, that in a baroque trio with two discant voices and one bass voice the latter plays the role
of the continuo. The bass voice may be more or less virtuosic, but there are no specific trios for two oboes and bassoon (only) in the baroque area.
This comes later in the classic period, for example the trios of Salieri.

[Notenschreiber]

I wonder if I should continue the dicussion about the tagging of the trio sonata of Heinichen. On the one hand this a question of
minor importance, an the other hand if the status quo is maintained this throws a bad light on the expertise of the tagging team.
Everybody who is familiar with baroque music see at once that the fact that one copyist liked to call this composition
"trio for 2 Oboes and bassoon" is completly unsufficient to assigne this trio the outstanding property to be the only composition
specifically for this instrumention. If one doesn´t stop reading the manuscript immediatly after the title on the top of the first
page it is easy to notice, that the bassoon part is a typical continuo part carrying the harmonies. The notation in three lines
named Oboe 1, Oboe 2, bassoon is the same as for example in the autograph of first trio sonatas of Zelenka. I strongly advice
to change the tagging.