[MEI-L] Instrument settings and tunings

[Richard Lewis]

Dear Axel,

Apologies for leaving this thread cold for so long.

We have some example encodings in our Github repository.

<https://github.com/TransformingMusicology/mei-tmus/blob/master/examples/italian.xml>
<https://github.com/TransformingMusicology/mei-tmus/blob/master/examples/baroque.xml>

The italian.xml example uses the <instrName> element. We haven't
specified what the @family attribute might contain and I'm not sure
now that I really had a specific use-case in mind when I included
it. I'm not aware that @code carries any specific semantics so it
doesn't, AFAICT, conflict with our (admittedly vague) @family.

The baroque.xml example actually conforms to an earlier version of the
schema for <instrDesc>! Although, it does have a more complete example
of <instrConfig>.

One thing to point out regarding our examples is that we're not
encoding information from the sources in our <instrConfig>. This is
actually information about musical practice which comes from scholarly
authorities. This is partly why, as we mentioned in our MEC paper,
we're interested in providing Linked Data versions of this kind of
<instrConfig> information (and tweaking our schema to allow that), so
that such non-source derived assertions can be backed up.

Best,
Richard

On Wed, 21 Oct 2015 13:12:10 +0100,
Axel Teich Geertinger wrote:
> 
> Hi Richard
> 
> Oh yes, now I remember! I am sorry! I was actually there to see your
> presentation, but the question had not occurred to me at that point
> :-)
> 
> Could you post a few examples here on the list for illustration and
> further discussion? I see in your customisation that you can also
> have an <instrName> element inside <instrDesc>. Do you have an
> example showing your use of <instrName> along with other content in
> <instrDesc>?
> 
> What values do you give the @family attribute in <instrName>? Is
> there a danger of doubling or conflicting information given in @code
> on <instrVoice>?
> 
> Best wishes,
> Axel 
> 
> > -----Oprindelig meddelelse-----
> > Fra: mei-l [mailto:mei-l-bounces+atge=kb.dk at lists.uni-paderborn.de] På vegne af Richard Lewis
> > Sendt: 21. oktober 2015 12:29
> > Til: Music Encoding Initiative
> > Emne: Re: [MEI-L] Instrument settings and tunings
> > 
> > Dear Axel,
> > 
> > We did some work on the problem of encoding instrument
> > configurations earlier this year. We actually came up with a
> > customisation which is published on github:
> > <https://github.com/TransformingMusicology/mei-tmus>, specifically,
> > <https://github.com/TransformingMusicology/mei-tmus/blob/master/schemata/instruments.odd>.
> > 
> > We took the approach of extending <instrVoice> to include
> > <instrDesc> and <instrConfig> elements. The <instrDesc> allows a
> > more structured description of an instrument, while <instrConfig> is
> > intended for instrument-specific set up. In our specific case we
> > then provided a further customisation for lute tablature which
> > extended <instrConfig> to allow elements for describing the tunings
> > of the courses.
> > 
> > Our intention was that this generic provision could be extended for
> > exactly the kinds of things you describe: organ registrations,
> > electronic instruments.
> > 
> > We didn't make any provision for your second requirement, however,
> > that of changes of instrument configuration during the course of a
> > piece.
> > 
> > We presented this at MEC 2015 but since then we haven't really
> > revised it. I think it's still quite experimental and really could
> > do with some more testing and criticism.
> > 
> > Best,
> > Richard
> > 
> > On Wed, 21 Oct 2015 09:01:53 +0100,
> > Axel Teich Geertinger wrote:
> > > 
> > > Dear list
> > > 
> > > I have a couple of questions related to the recent discussion about 
> > > the model for describing instrumentations in <perfMedium>. Sometimes I 
> > > need to add more detailed information to an instrument's name, such as 
> > > transposition or tuning. I simply write this information into 
> > > <instrVoice>, like
> > > 
> > > <instrVoice>Clarinet (A)</instrVoice>
> > > <instrVoice>Timpani (C, G)</instrVoice>
> > > 
> > > Of course I can add more complex information too: the settings for a 
> > > prepared piano or electronic instruments; scordatura; or even organ 
> > > registrations (if they remain the same throughout the piece).
> > > My first question is: Is it acceptable to have a rather detailed text 
> > > within <instrVoice> (or <perfRes>, if we decide to rename it), or 
> > > should we have some other place for a detailed description of 
> > > settings/tunings/transpositions?
> > > 
> > > My other question is: how would this kind of information be encoded in 
> > > the score? I know about the transposition attributes on <staffDef>, of 
> > > course; the timpani case is trivial too. But I wouldn't know how to 
> > > encode things like organ registrations, scordatura, electronic 
> > > instruments settings within <music>. Plain text? Or are there some 
> > > special element for instrument settings?
> > > I could also put it this way: Are these settings simply to be regarded 
> > > as playing techniques like pizzicato or col legno (which I don't know 
> > > how to encode either, however...)?
> > > 
> > > Best,
> > > Axel
-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Richard Lewis
j: ironchicken at jabber.earth.li
@: lewisrichard
http://www.richardlewis.me.uk/
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-