User talk:CodeCat

Archives: 2009-2010 · 2011 · 2012
Start a new discussion


Thread titleRepliesLast modified
PIE root shapes114:27, 24 October 2016
Can you check a P-Gmc. reconstruction I made?120:16, 22 October 2016
The i is not barred011:23, 22 October 2016
Inheritance from Koine217:37, 20 October 2016
pronunciations at bezie013:34, 20 October 2016
Gloywos314:42, 18 October 2016
The i doesn't become barred016:26, 17 October 2016
Template:wikipedia and foreign languages004:22, 17 October 2016
Seriously, stop614:25, 14 October 2016
This is getting ridiculous.618:26, 12 October 2016
ine-pro115:50, 12 October 2016
Catalan bot015:03, 9 October 2016
bron214:54, 8 October 2016
sister-wife008:02, 8 October 2016
Gildas120:18, 5 October 2016
ē?219:48, 5 October 2016
"not a root"519:05, 5 October 2016
Code feasibility question 412:49, 4 October 2016
Accelerated adjectives015:58, 3 October 2016
Revert to ابری321:01, 2 October 2016
First page
First page
Previous page
Previous page
Last page
Last page

PIE root shapes

I was reading through cat:Proto-Indo-European irregular-shape roots and I noticed *wreyḱ-, *wreyt-, *mlewH-, and *mreǵʰ- are considered irregular. According to w:Proto-Indo-European root#Sonority hierarchy, these should be considered legal, and LIV certainly has many such roots. Should we change over to a classification system that distinguishes R and M?

JohnC502:23, 24 October 2016

I'd prefer a system that distinguishes them only in these particular cases, ml, mr, wl, wr. We currently already do this with s: *sed- is just considered CeC, but *speḱ- is sCeC.

CodeCat14:27, 24 October 2016

Can you check a P-Gmc. reconstruction I made?

For the second element of 𐍆𐌰𐌹𐌷𐌿𐌲𐌴𐌹𐍂𐍉. I think that would be the correct Proto-Germanic form. Want to be sure tho, as I nommed it for FWOTD.

Kleio (t · c)19:52, 22 October 2016

I'm not sure about the German and Dutch cognates. You'd expect gier- in Dutch and geier- in German.

CodeCat20:16, 22 October 2016

The i is not barred

Proto-Celtic ū gives an unbarred i. It would only be close central unrounded if there was an a before it, giving Proto-Celtic axt.

UtherPendrogn (talk)11:23, 22 October 2016

Inheritance from Koine

Are you the person to ask? {{inh}} does not accept "grc-koi" as an argument - neither does {{m}} although {{etyl}} does. Is it possible to show inheritance from Koine - but Ancient Greek should be the language linked. Thanks

Saltmarshσυζήτηση-talk15:01, 20 October 2016

It works for me.

CodeCat15:35, 20 October 2016

aaaaa! I was using Greek characters - "grc-κοι" not "grc-koi" - sorry I bothered you!

Saltmarshσυζήτηση-talk17:37, 20 October 2016

pronunciations at bezie

Sorry to bother you with this, but I couldn't get the .ogg file at bezie to work. Would you mind putting it under the right etymology? Tia.

Lingo Bingo Dingo (talk)13:34, 20 October 2016

Are you saying Matasović is wrong in having Gloywos and not Gloiwos? UtherPendrogn (talk) 14:01, 18 October 2016 (UTC)

UtherPendrogn (talk)14:01, 18 October 2016

We just had a discussion at the talk page of WT:ACEL, where we decided to denote diphthongs as oi, ai, ou etc. This is nothing but a spelling change, but you'll need to convert Matasović's notation to ours from now on whenever you create entries.

CodeCat14:02, 18 October 2016
Not much of a discussion if not everyone is involved.
UtherPendrogn (talk)14:27, 18 October 2016

Your inability to keep track of this project is not our fault.

JohnC514:42, 18 October 2016

The i doesn't become barred

Proto-Celtic ū gives an unbarred i. It would only be close central unrounded if there was an a before it, giving Proto-Celtic axt.

UtherPendrogn (talk)16:26, 17 October 2016

Template:wikipedia and foreign languages

Was the new addition of transliteration/romanization (for example, at ประเทศเกาหลีใต้ ) to the link intentional?

suzukaze (tc)04:22, 17 October 2016

Seriously, stop

@Angr, CodeCat, Anglom, Victar, JohnC5

I sincerely doubt Anglom got every ordinal number wrong. Stop trying to sabotage those pages since you don't like me, it's petty and not frankly very constructive. In the meantime I've managed to reverse your edits.

UtherPendrogn (talk)13:37, 14 October 2016

I restored them again. If you revert once more I will report you for creating bad reconstructions.

CodeCat14:15, 14 October 2016

Anglom created those reconstructions. I'm reverting them back to the correct forms.

UtherPendrogn (talk)14:17, 14 October 2016

Why would these forms be correct with iθ when all others have ɨθ? We should be consistent.

CodeCat14:18, 14 October 2016

Dude, fucking stop. Stop editing things you're not even sure about, for fuck's sake. Anglom added those forms, take it up with him.

And don't threaten me. You really are scum, but this is an all new low. I'll keep on reverting them to the CORRECT forms until you stop.

UtherPendrogn (talk)14:21, 14 October 2016


CodeCat14:23, 14 October 2016

This is getting ridiculous.

Are you going to stop RFDing random shit? This is getting ridiculous. Fael is attested in Old Irish.

UtherPendrogn (talk)16:34, 11 October 2016

RfV is for when there are doubts over whether the term exists. And I have doubts.

CodeCat16:41, 11 October 2016

And why is your opinion superior to others? UtherPendrogn (talk) 17:50, 12 October 2016 (UTC)

UtherPendrogn (talk)17:50, 12 October 2016

And why is your opinion superior to others? UtherPendrogn (talk) 17:50, 12 October 2016 (UTC)

UtherPendrogn (talk)17:50, 12 October 2016

And why is your opinion superior to others? UtherPendrogn (talk) 17:50, 12 October 2016 (UTC)

UtherPendrogn (talk)17:50, 12 October 2016

And you didn't just rfv, you also deleted my correct meaning of the name, then edited again to stop me from adding it back.

UtherPendrogn (talk)18:04, 12 October 2016

This is one of the best edits I've seen. I hadn't included some information (since I didn't know of it), which JohnC5 and Semperblotto filled in. You just seem to change inconsequential things. Why do you go around nitpicking on my edits? A lot of it is unwarranted too. If I put "cel-noun" you change it to "head|cel-noun", while if I put "head|cel-noun" you change it to "cel-noun".

UtherPendrogn (talk)18:26, 12 October 2016

In any PIE root (lets say Reconstruction:Proto-Indo-European/ten-) there are sometimes terms derived with the correspondent explanation (innovative of, stative, -grade) but there are others that are just given as an extension (with hyphen), like -onts, -os/-es, -tis, -tlo, -tus, -s-li, -eh2, -tós. Shouldn't this extensions already exist and then be linked (for example, the Reconstruction:Proto-Indo-European/-tós of *tn̥-tós if it's the case) or be explained like that "suffix denoting extent of time)"? I mean, is there an official repertoire of those extensions/suppletions applied before the suffix?

Sobreira ►〓 (parlez)10:40, 12 October 2016

Sometimes, the suffix of a particular PIE descendant's word is not really traceable to a known PIE suffix. While we do know a fair bit about PIE morphology, we (as in, the worldwide linguistic community) are still learning more as we go and figure out things.

We could give the known suffixes names, but it's not always easy to think of appropriate ones. Both *-tis and *-tus are action nouns, but I'm not aware of a distinction in use or meaning between them, so all we can do is call both "action noun". Likewise, *-nós and *-tós are pretty much interchangeable in function (in Germanic and Slavic, obviously so), and I wouldn't know what to call them either. They tended to become past participles in the later languages, but they were not participles in PIE as they were suffixed only to roots, not to verb stems. They could also have meanings that would not be appropriate for a simple past participle.

CodeCat15:50, 12 October 2016

Catalan bot

Hey CodeCat. As I haven't mentioned it for a few months, now is a good time.... Fancy running your Catalan verb bot again? Or maybe donating the code to someone else?

Derrib9 (talk)15:03, 9 October 2016

Proto-Celtic sn becomes n, not nn. Only nsn or nd would become nn.

UtherPendrogn (talk)14:38, 8 October 2016

How do you know?

CodeCat14:51, 8 October 2016

All the other Proto-Brythonic lemmas.

UtherPendrogn (talk)14:54, 8 October 2016


I was trying to translate the word sister-wife into PB. Would this be it?

Proto-Celtic: swesūrwrakī Proto-Brythonic: hwɨhirwrėg Welsh: chwaerwraig

UtherPendrogn (talk)08:02, 8 October 2016

Gildas stems from Keiljos and Deiwos, or Keiljodeiwos. How in the world did it become Gildas and not Gɨlduɨw? Is Gildas latinised?

UtherPendrogn (talk)20:15, 5 October 2016

It's possible. I don't know.

CodeCat20:18, 5 October 2016

What happens to Proto-Celtic ē in Proto-Brythonic?

UtherPendrogn (talk)19:04, 5 October 2016

It becomes uɨ.

CodeCat19:43, 5 October 2016


UtherPendrogn (talk)19:48, 5 October 2016

"not a root"

Re *ḱeres-. Do you mean that stem didn't exist (says who?) Or do you mean it's morphologically composite? If the latter, where should this information go?

4pq1injbok (talk)01:11, 5 October 2016

True roots only have one vowel in them. So with its two e's, this isn't a root. The descendants also seem somewhat dubious, there's no s at all in the stem of the Lithuanian or Sanskrit terms, and the second e is missing in the Russian. All in all, there's just too much wrong to warrant keeping it.

CodeCat01:38, 5 October 2016

I agree on all points that this form is low-quality as an IE reconstruction. What surprises me is the practice that an individual's editorial judgment is good enough to reject a reconstruction that appears in a reputable source.

Also, if the policy were that being securely traced to a root (with just one phonemic vowel) was necessary for a reconstruction to deserve an entry, I'd have to call that too dogmatic. That would make us get rid of the entry for *kʷetwóres!

4pq1injbok (talk)03:53, 5 October 2016

(For example, I could spend months getting rid of all the dubious stuff in Karulis' etymological dictionary that we currently quote in Latvian entries. So far I've mostly held back...)

4pq1injbok (talk)04:04, 5 October 2016

I have my doubts about the Latvian stuff as well.

CodeCat13:27, 5 October 2016

It's ultimately a good policy, since if something is changed incorrectly, it will eventually get changed back to something correct. One day those Latvian entries will be correct. That's the aim of wiktionary, the way I see it. Keep beating things with a hammer until they are good.

UtherPendrogn (talk)19:05, 5 October 2016

Code feasibility question

Hello, I have a code feasibility question. Let's say you have words (like those in the category "French verbs") each with a certain set of properties (in the form of categories, like 1st group, 2nd group, transitive, reflexive,...). Would there be an easy way to auto-generate a table of words×properties with a check when the combination is applicable?

Word Property1 Property2 Property3
word1 x x
word2 x
word3 x x

This would basically require to iterate though the content of a category ("French verbs") and for each entry, iterate through the categories it belongs to ("1st group", "transitive"...). Thanks a lot for your help!

Julien Daux (talk)00:58, 4 October 2016

It's certainly possible, but not purely with anything the wiki software has. You'd need an external program written in a language such as Python (with the pywikibot module) that does the work for you.

CodeCat01:07, 4 October 2016

To be clear, you could either use one of the dump files ([1]- I'm not sure which one specifically has category membership information) or the API. There's not really any way to do it live- you would have to run a bot periodically.

DTLHS (talk)01:14, 4 October 2016

I see... If I understand correctly, the python bot would run from my machine.. I'm quite surprised to see that Lua or other module running server-side don't provide primitives for iterating through the content of a category or through the categories a page belongs to? Will look into that pywikibot, then. Thank you!

Julien Daux (talk)01:37, 4 October 2016

I think it may be because it adds the possibility of nondeterminism to modules. What if a module checks if the current page is in a category, and if it's not in the category, it adds the category, but if it's already in the category it doesn't add the category?

CodeCat12:49, 4 October 2016

Accelerated adjectives

Hello CodeCat,

Will you update the creation rules for adjectives to use this?__

Gamren (talk)15:58, 3 October 2016

Revert to ابری

Greetings, I see that you reverted my changes to this term. I provided reasons for the changes, but I don't see that you addressed my points. Can you please explain your position?

The word used in all early sources, including Ottoman Turkish employ the term ابری abrī not ابرو. See the main WP article for Paper Marbling which cites the earliest Turkish text, dated after 1615, entitled "Tertîb-i Risâle-i Ebrî (ترطیبِ رسالۀ ابری)

This is why I added the label for "Early Ottoman" and "Late Ottoman", as there is a distinct change. The word ابرو ebrû doesn't appear until the 19th century Redhouse Dictionary, 2nd ed. That is "late Ottoman".

Furthermore, ebru is NOT an English word! It is used by some contemporary English-speakers does not automatically make it English. , but it is Turkish, and it means "marbled paper".

Jemiljan (talk)14:00, 30 September 2016

Did you read the edit summary I added? You messed up all the transliterations of Persian words so that they no longer conform to our standard format.

CodeCat14:03, 30 September 2016

Thank you for your reply as I wasn't aware of this singular standard. Nevertheless, it seems that Wiktionary has arbitrarily decided to use a mixture of different standards based on the personal preferences of an ad-hoc group of editors. This is extremely problematic.

While I did see your summary, I also note that there is some debate on this "system", such as it is, on the associated discussion page! The contributors observe the incompatible use of a mixture systems. That mixture with different standards has shown up in this particular article. As you can see in the first post, one person asks if they really must change all of the macrons to circumflexes! Then the last person asks that the standards here be reviewed that it be put to a vote, which never happened.

Secondly, you unilaterally reverted corrections of obvious mistakes and changes that are in fact in line with the standard. Others were incorrect translations that had nothing to do with the Wiktionary diacritics, such as translating as well as transliterating ابری as "ebru". It's incorrect on both counts. Ebru is not an English word. The correct translation is "marbled paper".

Please see the talk page in which I review the standards, the inconsistent application, and puzzling translation of Persian into Turkish rather than English.

Jemiljan (talk)16:16, 30 September 2016

I agree it's problematic to use inconsistent transliteration systems. I don't know much about Persian but WT:Persian transliteration indicates the usage of i and â rather than ī and ā, so you should probably stick with that, and change macrons to circumflexes. I don't agree that we should leave macrons alone; if the standard is to be the usage of circumflexes, we should change the entries that way. Awhile ago I went through and corrected all the Arabic entries to use a single transliteration system.

As for ebru being an English word or not, there is a good deal of Google evidence indicating that it is (e.g. search for "ebru technique" or look at the Wikipedia articles on "ebru" and "paper marbling"). Wiktionary is descriptive rather than prescriptive, meaning it documents what people actually say rather than dictating what they should say. Since people do say "ebru" in English, it's fine to use this in English translations.

As for your comments about transliteration vs. transcription, Wiktionary somewhat misuses the term "transliteration" to refer to romanization in general, and tends to prefer a more phonetic rather than strictly written approach, which is why various Arabic letters are mapped to the same Persian letter. The idea is to help readers who aren't very familiar with the foreign script; those who are more interested in etymology are more likely to be able to handle the foreign script. Note however that there is currently a long-running discussion about transliteration vs. transcription in certain languages (e.g. Tibetan and Burmese, but also Thai, Persian, etc.) where the two differ significantly, with some proposing to use one in some circumstances and the other in other circumstances and others proposing to use the two in tandem, or to stick with the current more phonetic approach.

Benwing2 (talk)20:55, 2 October 2016
First page
First page
Previous page
Previous page
Last page
Last page
Return to the user page of "CodeCat".