Talk:Thomas Tallis

From ChoralWiki
Jump to navigation Jump to search

Automating Thomas Tallis page

A proposed automation can be found at User:Bcjohnston523/Automating Thomas Tallis. Interested to know what you think, positive or negative. Barry Johnston, Bcjohnston523 (talk) 04:30, 6 January 2016 (UTC)

Hi Barry, you did a very good job there! It's just a pity that DPL isn't able to align all columns equally spaced. —Carlos (talk) 04:16, 7 January 2016 (UTC)
Thanks, Carlos! I have since added sortmethod=sortkey, so that an entry like A new commandment is alphabetized under N. I have included all entries in my test page, which brings up some questions:
  • Should works entries generated by dpl appear in italics?
  • In order to automate "Contrafacta" I had to create a new category, Which I called "Category:Contrafactum", the singular (mostly following the Wikipedia page). Should this page have been in the plural?
  • How should we deal with references, such as "*Procul recedant somnia - see Te lucis ante terminum"? Such do not appear in the alphabetized dpl list.
  • There are three "works" currently in the Sacred Music in Latin list, that do not have pages, they are not links. I am not sure what these are?
  • Larger works are not included yet in the automation, so they would have to be managed manually.
According to the dpl manual, columns can be equally spaced by defining a table class in Common.css (I don't know if that exists in this wiki).
Should I continue to improve this (for example, to include larger works in automation), or rather pursue improvements to the SortWorks template? (see the forum discussion.) In any case, I need to wait a while to see what other users think. Barry Johnston, Bcjohnston523 (talk) 16:32, 7 January 2016 (UTC)

Some de-automation

The great disadvantage I see with automation is that it can give a misleading picture of a composer's total oeuvre, and doesn't identify the red-links that might inspire contributors to fill gaps. After consulting Early English Church Music 12 & 13 (ed. Ellenwood 1971) I've reformatted much of #Sacred music in English; adding Christ rising to EECM's 10 "Anthems" seems unproblematic, but the reattributed This is my commandment (Thomas Tallis) would still look better with an annotation imo. Richard Mix (talk) 03:23, 29 November 2020 (UTC)

I find great disadvantages to de-automation:
1. All composer pages have a 'List of choral works' section. Everytime a composer page is created, it admits the first hosted work(s). So the list is and remains generally uncomplete, but allows users to easily know if the work they are looking for is HOSTED or NOT. Such lists don't pretend to be neither a list of vocal and instrumental works, nor a complete list including unhosted works.
2. It's easy to quickly find a work in a unique alphabetical title sorting, not among a works list 'sorted' by genre and/or language and/or voicing, like the Orlando di Lasso one.

Sure, one may also browse a works list to look for a liturgical usage, a specific voicing, or a specific language, so ALL lists are useful, don't let's remove anyone.

3. The SortWorks function implemented by Max and Barry ensures that, if an editor creates a new work page, it appears instantly on the corresponding composer page AND, if a work page title is modified, the change also appears instantly on the corresponding composer page. So, removing 'SortWorks' should imply you'd now check every morning of your life if a work has been introduced or a work title modified during the night on your manual lists. Not for me, thanks.
4. 3,000 of our front composer pages have just SortWorks in the list of choral works section, other 300 have another sorting. Many 'other sortings' are already just one click away, outlined within a yellow rectangle like on the present Tallis' page. The more user-friendly would be to find the same things in the same place, throughout all the 3,300 composer pages, isn't it? Claude (talk) 07:41, 29 November 2020 (UTC)
Thanks for the reformat, Richard. I have a few questions and comments.
a. You seem to be dividing the page into "Single (or individual) works" and "Larger works"? If so, "Larger works" can be removed from the SortWorks commands in the "Single works" section by adding &&!Larger works – I have done it.
b. Please create pages for the other services – The Dorian Service and The Choral Service. Does "Magnificat and Nunc dimittis (Dorian)" belong in the Dorian Service?
c. Where did the extra lines in Parker's Psalter come from? I see only nine tunes in that publication.
I appreciate your remarks about the red ink being a stimulus for users to fill gaps – but perhaps that function belongs to Larger-work pages (or publication pages). Especially for composers having as many single works as Tallis. I think the Thomas Tallis page should be completely automated, or completely not: trying to manage a partially automated page is difficult for both of us, and also for other users who don't understand. An automated page's structure might look like:
Single works
Sacred music in Latin
Sacred music in English
Secular music
Larger works
(list of all larger works, without showing their contents)
After you create pages for the other services, I will create a new automated page (separately, not changing Thomas Tallis) so you can see what I mean. A manual list of all Tallis' work would be useful, no doubt, either as the main page or a subpage. — Barry Johnston (talk) 22:39, 29 November 2020 (UTC)
Without our philosophical differences (why wouldn't the goal be a complete list?) Claude and I would just have to find something else to talk about :-)
I got started on this by wanting to understand The whole Psalter translated (1567): to see a Nine Compositions header on the composer page with more than 9 entries is, well, startling. How would automation show that the same 'Tunes' are used for different texts? (Obviously one could merge pages, I suppose) The 'Dorian' service has 9/10 items: 3 morning canticles, 4 or 5 items to make the rest of a morning service including communion, and the 2 canticles for Evensong; I wasn't quite sure how to number them with #'s.
I also think a list of larger works that doesn't list contents is less useful than it could be; that's why we ended up with 3 composer pages each for Brahms and Mendelssohn. Automation does have considerable value as a check against lists-in-progress like the text pages (which then still need hand updating), and in managing under-construction complete works like Lasso and Palestrina. Tallis & Byrd (I certainly won't have got to Laetania (William Byrd) without Gradualia I (William Byrd)) are so close to finished that a CheckMissing function seems better suited. Richard Mix (talk) 01:21, 30 November 2020 (UTC) I'll just add that I don't have the resources at hand right now to make a list of Tallis' Latin works.
About "Nine Compositions", the same situation exists in the psalters of Esty, Ravenscroft, etc., where five or six (or more) psalm texts are assigned the same tune. My opinion is that we ought to title these by their original titles (Psalm 1, Psalm 68, etc.), but the 19th-century hymnology movement insists on titling them by first line, so then pick the first one and find some way to explain (I used tables on the publication page, not the composer page) that the same tune was used for other psalms. If you try to show all of the texts on the composer page, the result is a useless mess. — Barry Johnston (talk) 04:35, 30 November 2020 (UTC)
Hymnology hasn't definitively settled tune naming, so we still have variously HERZLICH TUT, O HAUPT & PASSION CHORALE (a convention I wish we'd adopted is small caps for tune names). For Gibbons, the second hymn tune is almost always named SONG 3. I've done that sorting on the publication page The Hymnes and Songs of the Church (George Wither), and, for me at least, that's been a useful mess, though not finished by any means. Richard Mix (talk) 21:29, 30 November 2020 (UTC)
You put your sorting on the Publication page, not Composer page, which was part of my point. For part of the day I worked on Thomas Tallis/Works – what do you think? I'm sure there are mistakes; please feel free to correct, or put notes in the Comments column (at the end of each line in the code). Perhaps I started with the wrong list; if so, please point me to the right one. — Barry Johnston (talk) 03:04, 1 December 2020 (UTC)