Template talk:R:M&A

Translators in t:cite-book

@Smuconlaw So, this book is actually a translation by Auden of Carl Meissner's Lateinische Phraseologie. Is there a way to add Auden as the translator and not just as a second author? —JohnC5 19:24, 13 June 2016 (UTC)

He translated and, with permission, augmented it; hence, he is both translator and a sort of co-author, but the latter very secondarily. Is there some kind of adequately brief parenthetical that can convey this relationship? Ah, made first2 and last2 to work; I tried but was not able. Isomorphyc (talk) 20:34, 13 June 2016 (UTC)

My suggestion would be to use |author2= like this: "|author2=Henry William Auden, transl.". Alternatively, "|last2=Auden" and "|first2=Henry William, transl." could also be used. — SMUconlaw (talk) 21:51, 13 June 2016 (UTC)

@Smuconlaw: That's a fair idea. Also, I notices you just use semicolon-delimited authors in the template, when I thought the preferred method as “Author 1, Author 2, & Author 3”. If you're interested, I created a template {{format list}} to simplify this logic. I could also get it to change the delimiter too, I guess. What do you think? —JohnC5 01:32, 14 June 2016 (UTC)

The use of semicolons was already in the template; I didn't alter this. I think there should be a wider discussion (for example, at the Beer Parlour or the Grease Pit) if this is to be changed. However, I note that the {{cite}} and {{quote}} templates use commas to separate the various elements of a citation such as the author's name, the book title, and the place of publication. It might be a bit confusing if commas are also used as a list separator. — SMUconlaw (talk) 01:58, 14 June 2016 (UTC)

Oh, I didn't mean that you altered it—it's just odd to me that we use them. I'm thinking of the APA format (which is the one off of which ours is based?):

Liddell, H. G., Scott, R., Jones, S. H. S., & McKenzie, R. (1940). A Greek-English Lexicon: A New Edition Revised and Augmented Throughout by Sir Henry Stuart Jones, with the Assistance of Roderick McKenzie. 2 Vols. Clarendon Press.

or Harvard format:

Liddell, H.G., Scott, R., Jones, S.H.S. and McKenzie, R., 1940. A Greek-English Lexicon: A New Edition Revised and Augmented Throughout by Sir Henry Stuart Jones, with the Assistance of Roderick McKenzie. 2 Vols. Clarendon Press.

I've never seen a format in which the authors are separated by semicolons. Does such a system exist? —JohnC5 02:21, 14 June 2016 (UTC)

No idea. I'm happy to go with whatever the consensus is. — SMUconlaw (talk) 02:24, 14 June 2016 (UTC)

The semicolons looked rather unorthodox to me also. I wouldn't have suggested it, but I would support changing it. Nice Oxford comma option. Isomorphyc (talk) 02:29, 14 June 2016 (UTC)

@JohnC5 On this subject, should the template name be changed to R:Meissner or R:Latin_Phrase_Book? Do you have another suggestion? I picked the original inaccurate name essentially for a reason parallel to the bibliographic problems we are now having. Isomorphyc (talk) 14:50, 14 June 2016 (UTC)

@Isomorphyc: I was wondering the same thing. Is there any standard abbreviation in the literature? I might go for {{R:M&A}} or {{R:la:M&A}}. —JohnC5 14:54, 14 June 2016 (UTC)

@JohnC5: I cannot find one; it is not a particularly academic resource. I'm leaning towards R:M&A unless there's a preference for something else. Isomorphyc (talk) 15:37, 14 June 2016 (UTC)

@Isomorphyc: Go for it. Though update all links to it or leave a redirect, please! —JohnC5 15:39, 14 June 2016 (UTC)

@JohnC5: Done. I took what I hope is not too much of a liberty to update your mention of it in the Beer Parlour. Isomorphyc (talk) 15:52, 14 June 2016 (UTC)

Formatting errors

I wanted to standardize the formatting in the References section in gurges, since one reference (namely the M&A ref) had a formatting difference when compared to the others. Its URL was linked to a bracketed number one[1] rather than the lemma of the entry.

When I viewed the reference in both HTML and ePub formats, I also noticed that the content of the entry in the reference and that in Wiktionary differed.

M&A:
gurgitibus hauriri—to be drowned in the eddies.

gurges:

• to be drowned in the eddies: gurgitibus hauriri

So, it would be great if somebody could make any of the following changes to the way the template functions:

1. Change Template:R:M&A's output to match the order, punctuation. & formatting of the M&A's entries.
2. instead of putting the URL in square brackets by itself, link it to either the entire Latin phrase of the entry or the Latin word within that phrase that is the inflected form of the WTWW entry's headword.

Sorry all I can do is point this out! Thank you in advance. —Geekdiva (talk) 22:02, 14 August 2017 (UTC)

Memory issues

Just flagging up that this template is massively memory-inefficient - it added 17MB to e when I re-enabled it (since it had been commented out), which currently makes it the most memory-hungry page we have. It's not critically high, but Module:R:M&A definitely needs rewriting. Pinging @Benwing2, @Surjection, @Chuck Entz, @Erutuon @This, that and the other. Theknightwho (talk) 07:59, 9 January 2024 (UTC)

@Theknightwho Please comment this out again. It uses some huge tables that need rethinking. Benwing2 (talk) 08:52, 9 January 2024 (UTC)

@Benwing2 While it definitely needs rewriting, I don't really see what benefit there is to commenting it out, since there's still about 8MB to spare. Theknightwho (talk) 08:58, 9 January 2024 (UTC)

@Theknightwho Because the way you are going, there will soon be no memory to spare. 8MB is far too little and having crappy modules like this enabled discourages people from fixing them. Benwing2 (talk) 09:04, 9 January 2024 (UTC)

@Benwing2 Alright - I've commented it out. Theknightwho (talk) 09:17, 9 January 2024 (UTC)

@Theknightwho: All of the Perseus modules are studies in massive overkill: the data modules are planetary in scale, but sharded to disperse everything evenly so memory usage can't be optimized. And all of this is dragged in with every page load to assemble and format content that doesn't ever change. It was nice to collate all that data to start with on what content is available to link to for each page, but it's just silly to keep generating everything from scratch over and over again. It would be better to have a bot replace these dynamic monstrosities with their output hard-coded as parameters into a simple reference template. Chuck Entz (talk) 08:59, 9 January 2024 (UTC)

@Chuck Entz I agree. See Module:User:Theknightwho/en-pron/data from line 91 as an example of how you can organise complex data in a single table, instead of subdividing it into a billion small ones (which is what causes memory use to become huge due to the way Scribunto works). You just need to organise it sensibly. Theknightwho (talk) 09:05, 9 January 2024 (UTC)

@Theknightwho: Why bother with a data module at all? Just picking a random example: Latin dies has this in the Further reading section. You could replace it with a single {{quote-book}} followed by 35 instances of {{ux}}. Once you load in the parameters, the display will be indistinguishable from what we have now with the existing template, except for the colapsibility. Chuck Entz (talk) 09:25, 9 January 2024 (UTC)

@Chuck Entz Yeah, I agree - there's no obvious advantage to having it all stuffed in a data module here. I suppose it can automatically draw down everything that contains a given word, but that's definitely bottable. Theknightwho (talk) 09:33, 9 January 2024 (UTC)

I don't get this template. If the phrases are worth including, they should be incorporated into the entry as collocations! I guess that is "hard work" though. This, that and the other (talk) 09:25, 9 January 2024 (UTC)

A bot could do it easily. Chuck Entz (talk) 09:27, 9 January 2024 (UTC)

Module:R:M&A/ix to phrase would take less memory if it were converted to an array and were looked up by number. The indices would have to be incremented so that "0" looks up index 1 in the table. — Eru·tuon 21:04, 9 January 2024 (UTC)