Module talk:lo-headword

Alt forms marked "dated"

Hey nice to see the work on Lao lately.

But I've just noticed the new {{lo-alt}} template is marking spellings with the ຫມ and ຫນ full spellings, the ones without the ligatures ໝ and ໜ, as "dated".

Unless there's some source for this it just doesn't seem to be true. I've been to Laos a few times now, most of the provinces, and it still seems to me pretty much arbitrary whether a word is written in full or with the ligature. This goes for signs on buildings, including government buildings, forms, dictionaries, just about anything I can think of.

It's exactly the same for ຫລ vs ຫຼ as far as I can tell.

There are other spelling variations which do seem dated or obsolete, such as:

• ຽ at the end of a syllable instead of ຍ.
• using the subscript ລ with any letter other than ຫ
• using implicit vowels a la Thai
• using a single consonant as both the final consonant for one syllable and the initial consonant of the next syllable

I don't really speak Lao though, just enough to be a tourist and read signs and menus. So I could be wrong and would be interested to see if there is some official statement from the Lao government etc on this topic. — hippietrail (talk) 15:28, 22 June 2016 (UTC)

It seems that the government does not care about her language; she does not have any agent to nationalise her spellings. I just say that the dated convention can be found in historical documents, and ໜ ໝ ຫຼ only found in modern use. The actual obsolete terms are ones that resemble Pali-Sanskrit or even Khmer forms (which Thai still do). Why we can see full letters around may be that they are easier to type. BTW, I have never seen any official documents from Lao either.

What I have done lately is not finished. It would be not too late to adjust the module. --Octahedron80 (talk) 02:04, 23 June 2016 (UTC)

There have been official rules about the language before. The spelling used to be much more like Thai with inherent vowels, more silent letters, and even some letters which are not just obsolete but are not even in Unicode. After the communists took power they did a language/spelling reform which actually turned the script into an alphabet from an abugida, since every vowel is spelled out in full. Another thing they did was to make ຣ obsolete in favour of ລ, even though these days nobody seems to care about that rule, including the government. It was all supposed to make spelling more phonetic - spell it as you say it. SEAlang's Lao dictionary actually draws on two sources, one from before the reform and one from after.

On Windows it is trivially easy to type ໝ, ໜ, ຫຼ, and ຣ. On Mac on the other hand there is still no keyboard layout even in OS X El Capitan! So every letter is equally difficult to type (-: Mac can't even render the Lao word for "water" correctly ...

Now one thing we could do is count how many of each spelling variant are used in a big electronic Lao corpus like the Lao Wiktionary. One problem is that Lao doesn't use spaces between words and I'm not aware of a free Lao word segmenter. Or if there is one how accurate it is. In fact even segmenting into syllables turns out to be hard due to so many old spellings still out there.

I might see if it's possible to make some kind of Googlefight for Lao spellings.

Have you been doing your own programming work by the way? I should have some code for Lao analysis still lying around if you're interested. All my code in for nodeJS these days.

— hippietrail (talk) 04:59, 23 June 2016 (UTC)

Hmm well Google treats ຫນ and ໜ identically even in words that are quoted, which would make a Googlefight difficult to implement. — hippietrail (talk) 05:14, 23 June 2016 (UTC)

The alt method is moved to Module:lo-alt. --Octahedron80 (talk) 05:21, 25 October 2017 (UTC)