Wiktionary:Grease pit/2016/November

A lot of entries in this category use {{bor}} which appears to be inappropriate since many of those words weren't borrowed directly from Old French into Modern English, but inherited from Middle English. Should we correct all those instances of inappropriate uses of the template, and ultimately delete the category? ばかFumiko￥^talk 06:08, 4 November 2016 (UTC)

When we have chains of derivation, sometimes going back to unattested languages, we usually categorize the entry into each language in the chain. I look at the situation and say that the problem is that for many of these the Middle English term is missing, possibly because it is unattested or not inferred in a scholarly source. For others the Middle English term may have a spelling identical to one of the Modern English spellings. I don't see why we would want to delete the category reflecting one of the links in the chain. DCDuring TALK 10:55, 4 November 2016 (UTC)

I've been considering this problem more, and I think it may actually be more appropriate to call them borrowings even if they were borrowed into an earlier stage of the language. —CodeCat 13:05, 4 November 2016 (UTC)

I've got it!!! We should redirect the categorization to Category:English terms derived from Old French, which doesn't have the implication of being limited to the etymon in the immediately preceding language. DCDuring TALK 14:05, 4 November 2016 (UTC)

Isn't that the currently accepted practice, i.e. to use {{der}} for indirect borrowings? If you mean literal redirection, then that's not a good idea. --WikiTiki89 14:09, 4 November 2016 (UTC)

Middle English is English, though. Per CodeCat. Renard Migrant (talk) 12:13, 5 November 2016 (UTC)

To be fair, terms borrowed into PIE would also be considered borrowings into English. Because PIE eventually became English. —CodeCat 12:19, 5 November 2016 (UTC)

"Middle English is English, though. Per CodeCat." No, it's not, because we're talking about strict technicality here regarding the use of {{bor}}. If a term was borrowed into Middle English, it does not qualify for using {{bor}}. ばかFumiko￥^talk 03:54, 6 November 2016 (UTC)

Only by you, apparently. Following your reasoning to its conclusion, a single borrowing potentially counts as a borrowing into hundreds of languages, and there's no such thing as a borrowing between related languages. Chuck Entz (talk) 04:12, 6 November 2016 (UTC)

I need to think some more about that second part. Chuck Entz (talk) 05:38, 6 November 2016 (UTC)

Excuse me, but am I the only who misunderstood this instruction?

"This template is intended specifically for loanwords that were borrowed during the time the borrowing language was spoken (periods as defined by scholarly consensus and/or Wiktionary's own definitions). It should not be used for terms that were borrowed at an earlier stage, which should use {{derived}}.

For example, if English has a word that was borrowed from Anglo-Norman French in the 12th century, then the Modern English entry would not use {{borrowing}}, because the word was not borrowed in modern English times: English as defined on Wiktionary begins at the year 1500, so the 12th century does not fall under the modern English period, it falls under Middle English. The Middle English entry for that word would use the {{borrowing}} template." ばかFumiko￥^talk 05:47, 6 November 2016 (UTC)

I understand that. My thought is that we should change that, so that terms borrowed into English, Middle English, Old English, Proto-Germanic and Proto-Indo-European are all considered to be borrowings in English. My reasoning is that it makes a lot of sense to consider for example change a loan, because it's obviously not a native word inherited from time immemorial. —CodeCat 13:18, 6 November 2016 (UTC)

Any idea what's going on here? These showed up in CAT:E about a week ago. As far as I could tell at the time, there were no recent edits to either of the above, Module:hierarchy nor to Module:category tree/poscatboiler/data, and only a single innocuous edit to any Module:category tree/poscatboiler/data submodule. Of course, with the use of recursion on the contents of all of those submodules at once, it might not take much. Chuck Entz (talk) 05:15, 6 November 2016 (UTC)

Do the lists look as they did before?--Dixtosa (talk) 07:38, 6 November 2016 (UTC)

I don't see such an error message now. Could this have been a transitory effect of today's multiple changes to Module:hierarchy? @Chuck Entz: If those edits not visible to you, I wonder why. DCDuring TALK 14:15, 6 November 2016 (UTC)

Since those edits were after and in response to my post above, there's no reason to wonder why the state of things described in that post doesn't reflect them. I have no idea how the output of Module:category tree/poscatboiler/hierarchy was supposed to have looked before, but it no longer has a stack overflow error, which is a big improvement. Chuck Entz (talk) 15:05, 6 November 2016 (UTC)

My mistake. I was looking at UTC times in history and EST on this page. DCDuring TALK 18:07, 6 November 2016 (UTC)

At ] This diff by User:ReidAA in December 2014 is one that introduced "heading" as second argument of {{lb}}. It seems to occur with senses that have subsenses. There seem to be some 60 entries with this.

Can anyone figure out what the intent was? Did something in {{lb}} or its supporting modules change? DCDuring TALK 19:37, 9 November 2016 (UTC)

I think it means something like "these subsenses have the following characteristic in common". I personally feel this is misuse of subsensing. — Ungoliant ^(falai) 19:40, 9 November 2016 (UTC)

Oh, I think I remember now. Thanks. I believe this is intended to mimic some of the groupings of definitions in dictionaries such as MWOnline. While some have explicit senses under which the subsenses are grouped, others do not, probably because although they are clearly related, they defy any wording that encompasses all the subsenses. It was thought that we needed this as a way to have the senseless-subsenses structure.

It seems wrong as implemented. A gloss without the "heading" label would be sufficient. If no suitable wording can be found, then perhaps a non-gloss definition can be. If not then we can just eliminate the subsense grouping.

The 60 entries can be manually reviewed then to make sure there is a sensible gloss or non-gloss definition. DCDuring TALK 19:54, 9 November 2016 (UTC)

At probo, the references for the Italian section are left dangling at the end of the Spanish section. I can't remember how to get it placed at the end of the Italian, I tried using some Wikipedia-style references tags, but it hasn't worked. --Derrib9 (talk) 14:33, 10 November 2016 (UTC)

 Done. The slash comes after the word "references" and before the greater-than sign, unless you want to open and close the "references" tag separately. Thus either

<references/>

or

<references> ... </references>

but since nothing ever needs to be enclosed within the "references" tag, the first option is simpler. —Aɴɢʀ (talk) 14:53, 10 November 2016 (UTC)

Yep, that's the one. Cheers, Angry Bird --Derrib9 (talk) 11:10, 13 November 2016 (UTC)

I have been struggling with creating a headword-line template for Chichewa nouns. I'd really appreciate it if someone more technically adept could make it for me, once I give them the rules for plural generation. (I'd also like someone to help me create an IPA module, but that's not as pressing.) —Μετάknowledge^{discuss/deeds} 04:06, 11 November 2016 (UTC)

OK, what are the rules? DTLHS (talk) 04:15, 11 November 2016 (UTC)

Thank you so much. I've not started typing up the rules yet; I'll put them on your talk-page to avoid cluttering the GP more, if that's okay. The rules for the IPA module are here, and I can write up testcases as soon as they are needed. —Μετάknowledge^{discuss/deeds} 04:25, 11 November 2016 (UTC)

Would this be possible? So that we could see what people have searched for before. Also, adding a conspicuous box to allow the user to submit the word for creation after searching for a nonexistent title - "Hey, it looks like you entered a word in the Latin / Cyrillic / Devanagari / Hangul / Han script, which has not yet been created. Your query has been recorded OR Please click here to submit it to the request list (allowing the user to add comments too)."

Besides, how about a reverse transcription / transliteration search? It can be done via reverse transliteration modules, some of which have been written, such as Module:mnc-Latn-translit and Module:pi-Latn-translit. Alternatively, it could be done via regularly updated transcription/transliteration-word lists. For example, I have a list of currently generated Pinyins (and the pages they are found on) from multisyllabic Chinese entries of en.wikt (45,000+), and it could be parsed when Pinyin is selected as the type of transcription/transliteration used as input. Wyang (talk) 11:07, 11 November 2016 (UTC)

I can foresee one problem: filtering out searches that are misspelt, or have typos. Otherwise I'm in favour. DonnanZ (talk) 11:16, 11 November 2016 (UTC)

I like the reverse search idea. We could avoid the need for out-of-sync romanization entries and more easily search for entries without romanization entries. I wonder if Wikimedia Labs + JavaScript could be used to do this. The request button would also be pretty handy (although I wonder if it will lead to spam from unregistered users...). —suzukaze (t・c) 11:27, 11 November 2016 (UTC)

Provided that the searching script is specified during the search and there is a bijection between the set of transliterations and a set of words in the original script it can easily be done with JS only. --Giorgi Eufshi (talk) 11:58, 11 November 2016 (UTC)

That's great. What about using reverse transliteration modules? This is the easier option for scripts with good bidirectional transliteration faithfulness, such as Devanagari. Can the searched transcription / transliteration be passed to the module for reverse transliteration, and the output be used for searching? And what about the ways in which the reverse search function could be added to the existing search - perhaps an "Advanced Search" next to the search icon (would it be possible?). Wyang (talk) 12:13, 11 November 2016 (UTC)

Modules can be used (using API:ExpandTemplates) but that will require one more API request per keypress.

The easiest way to put the functionality here is to use a separate search field.

One clarification: I assumed autocomplete was meant to be possible without it is even easier. --Dixtosa (talk) 11:02, 13 November 2016 (UTC)

Also, check out this mw:Extension:UniversalLanguageSelector. I wonder why we do not have it installed while other wiktionaries have. --Dixtosa (talk) 11:58, 13 November 2016 (UTC)

Didn't know about the UniversalLanguageSelector tool before - it seems to have been installed on en.wikt (?), accessable via Languages on the left and in the search bar as a small keyboard icon. I tested using Tibetan and Devanagari and it is working well. It is a good way to allow people familiar with the script to search for a term, but unfortunately this doesn't support many other scripts (such as Japanese kana, Korean Hangul, Vietnamese Quoc Ngu), and the function may need to be made more conspicuous. Another way to give non-Latin script support is to develop a separate search field, so that people not fully familiar with the script and the native keyboard can search by romanisations. I have no clue how to code for the search functionality and incorporate it into the layout, but I can help developing the reverse transliteration modules (such as for Sanskrit, Japanese (partially available in Module:ja#romaji_to_kata)) and transcription data modules (for Chinese pinyin). Wyang (talk) 21:01, 15 November 2016 (UTC)

I do not think the first idea can be done with out facilities only. You can request the feature at meta:2016 Community Wishlist Survey. --Giorgi Eufshi (talk) 11:58, 11 November 2016 (UTC)

In the Cyrillic edit tools, the characters Ć ć appear, but these are actually Latin letters. They should be replaced with their Cyrillic equivalents. —CodeCat 15:21, 11 November 2016 (UTC)

So, you want them to be replaced with С́ с́ in MediaWiki:Edittools? Just checking that that's the right encoding. —Μετάknowledge^{discuss/deeds} 00:26, 16 November 2016 (UTC)

Yes. —CodeCat 00:49, 16 November 2016 (UTC)

The small letter must be с́, not с. See Sje--Anatoli T. ^{(обсудить}/^вклад) 00:58, 16 November 2016 (UTC)

@Metaknowledge Done, don't change it. --Anatoli T. ^{(обсудить}/^вклад) 01:00, 16 November 2016 (UTC)

Sorry, I removed the accent from Metaknowledge's post so I could see if the underlying letter was the right one, and forgot to undo it. I put it back now. —CodeCat 01:02, 16 November 2016 (UTC)

It's OK. I recommend downloading SC Unipad (free). You can decompile or convert characters (such as diacritics, etc) and see a lot of info about them. You can also enter a big variety of characters from different scripts. --Anatoli T. ^{(обсудить}/^вклад) 01:08, 16 November 2016 (UTC)

I use Linux. —CodeCat 01:21, 16 November 2016 (UTC)

You can use Windows emulators on Linux.--Anatoli T. ^{(обсудить}/^вклад) 01:28, 16 November 2016 (UTC)

I'm sure there are tools available for Linux as well. I wrote my own python script for this, it was pretty easy. --WikiTiki89 15:44, 16 November 2016 (UTC)

@Crom daba The Cyrillic equivalent of this term is с́утра, which has the letter с with a combining diacritic. The diacritic is removed however, so that the link points to сутра, which is the wrong entry. —CodeCat 15:25, 11 November 2016 (UTC)

Is there a way to make entry_name patterns capture vowel + diacritic combinations but leave с́ alone? Module:languages/data2 seems to replace both 'Ȁ' and dbreve, isn't this reduntant? Crom daba (talk) 15:40, 11 November 2016 (UTC)

In Cyrillic, а̏ is made out of two characters. Each character is replaced individually. The same applies for acute accents too. So the code is really just removing the diacritic as a separate character. —CodeCat 15:50, 11 November 2016 (UTC)

Could someone please create "Category:en:Philanthropy" properly? I have no idea how to do this. — SMUconlaw (talk) 18:20, 14 November 2016 (UTC)

Philanthropy would have to be added at Module:category tree/topic cat/data to one or more of the existing topics at Category:List of topics. Where would it be appropriate? To judge from the categories that Wikipedia's w:Category:Philanthropy belongs to, I'd say CAT:Ethics, CAT:Economics, and CAT:Political science as a first guess. What do you think? —Aɴɢʀ (talk) 19:20, 14 November 2016 (UTC)

I think "Economics" and "Ethics" are suitable, but am doubtful about "Political science". Is there a way to view the entire category tree? — SMUconlaw (talk) 19:47, 14 November 2016 (UTC)

It's not actually a tree, which makes it rather complicated to view. —CodeCat 20:08, 14 November 2016 (UTC)

That sounds like a good project for someone with the right kind of technical talent: rendering the category tree with collapsible levels of detail. Have any other wikis done it? WP, de.WP, de.wikt, ....? DCDuring TALK 23:32, 14 November 2016 (UTC)

We already have a module to make a tree of languages (and those actually are a tree). The difficulty is with categories is that they're not a tree and can even loop back on themselves. A category could be a subcategory of itself, directly or indirectly. —CodeCat 00:02, 15 November 2016 (UTC)

The item in question does seem to be in a topical hierarchy. Even if the hierarchy is "impure", there is almost certainly a main hierarchy with accessory elements (loops).

The important thing is to demystify category structure and allow more contributions to and use of the categories in a consistent manner. If an active contributor is having trouble or even just finding it not worthwhile to research, clearly there is a failure in the system. DCDuring TALK 02:03, 15 November 2016 (UTC)

Oh, sorry. I missed the point of the current system. It is supposed to force users to come begging to their betters for help if they don't chose to master our baroque template/module systems, lest they start treating this thing like some kind of wiki. DCDuring TALK 02:11, 15 November 2016 (UTC)

De beste stuurlui staan aan wal. —CodeCat 02:22, 15 November 2016 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Google Translate says: "Best mates are ashore". Ok I guess.

Anyway, about a decade ago, we had {{nav}}, a template in which we would fill a lot of parameters for each category. Category:es:Animals had this specific code:

{{nav|lang=es|langname=Spanish|current=Animals|parent=Nature|parent2=Taxonomy|parent3=Zoology}}

I think our current approach is better than {{nav}}, at least. If there are any other ideas, I'd like to hear them. (though I know some issues have been discussed multiple times, like the jargon/subject distinction) If we used {{nav}} or just plain wikitext, apparently it would be a pain to keep our 565 "Category:<lang>:Animals" categories consistent. I'm assuming that consistency is desirable. --Daniel Carrero (talk) 02:29, 15 November 2016 (UTC)

It's a Dutch proverb which we do, in fact, have an entry for. Here's a discussion of the best cultural translations into English. —Μετάknowledge^{discuss/deeds} 03:42, 15 November 2016 (UTC)

Being a critical user is hard work, too, especially going against a mindset as insular and non-user-oriented as the one prevailing here. DCDuring TALK 13:18, 15 November 2016 (UTC)

DCDuring, actually I like the idea that every addition to the hierarchy is discussed, because one bad naming can proliferate causing many articles to to be changed (for example Category:Georgia (State) where state can be ambiguous). So I would even impose a protection to the modules allowing only administrators to edit them on a per-request basis. If we were to look through the hierarchy we would find many controversial categories/namings. Category:Wines and Category:Wine is one good example. or I am just making a bid deal out of it. Just like you are. Een storm in een glas water ya know... --Giorgi Eufshi (talk) 05:44, 15 November 2016 (UTC)

How many of them have been discussed? Or is it just those that came after the tablets were brought down from the mountain. Whose additions have been discussed?

There are some simple rules for our category names that we seem to follow, mostly implicitly: eg, "plural names for categories of coordinate terms", "not too long", "not too specialized", "visible categories preferably fitting in a hierarchy", "language-specific categories grammar, orthography, etc. in own subsystem". There are plenty of areas in which a little room for purpose-built categories were, are, and would be appropriate (probably would have been, too). DCDuring TALK 13:18, 15 November 2016 (UTC)

I don't think our topical category tree was properly discussed or designed, it's a mess. I may choose to be optimistic and believe that they are a work in progress and could go in a better direction sometime, though. I do love the fact that we have a templated/Luacized system in which if you change the description and categorization of "Category:??:Animals", it reflects automatically in all language versions.

I take DCDuring's point that we are not having a system in which new categories require to be discussed before they are created, because people with Lua skills are able to create new categories at will. To be fair, editing the category tree is not too hard. DCDuring, don't you know how to use the system yet? --Daniel Carrero (talk) 21:17, 15 November 2016 (UTC)

I'm trying to keep a beginner's mind with respect to some parts of our system. My ignorance qualifies me better to guess how some newer, less technical users might view that system. DCDuring TALK 02:30, 18 November 2016 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Sooo ... are we having a discussion on whether "Category:en:Philanthropy" should be created? It seems to already be in use. — SMUconlaw (talk) 12:24, 17 November 2016 (UTC)

It's in use, but has an invalid label. I'm willing to create a topic Philanthropy for the tree that isn't a tree, but only if doing so doesn't result in Drama. —Aɴɢʀ (talk) 13:18, 17 November 2016 (UTC)

Swahili monosyllabic verbs such as la (“eat”), ja (“come”), fa (“die”) retain the infinitive prefix -ku- in some tenses to avoid stressing the tense prefix. The affirmative conjugation template sw-conj-aff does something odd with the progressive (-na-) tense: the singular forms have infinitive -ku- but the plural forms lack them. I don't believe this is right, from what I know about Swahili, but I don't know enough to be sure whether this tense should have -ku- across the board or not. And I certainly don't know how to change a template. --Hiztegilari (talk) 13:56, 15 November 2016 (UTC)

The Swahili conjugation templates are really, really crappy. It's been a long-term goal of mine to completely retrofit them; they need to be corrected and Luacised, and I'm not fully capable of that myself. Having just gotten quite a lot of help from DTLHS, I'm hesitant to bother him again. The truth is, we need a standard format that cover enough (but not too much) that we can modify for other Bantu languages. @CodeCat, maybe? —Μετάknowledge^{discuss/deeds} 18:44, 15 November 2016 (UTC)

The Zulu table we currently have isn't much better. The big issue is that there's simply too many forms to list. There's already a lot of potential forms if you include the subject concord, but when you start adding object concords as well, the number of forms explodes. With 15 possible subject and object concords, you end up with 225+ forms in one table, which is way too much and risks losing the useful information among the noise. —CodeCat 18:53, 15 November 2016 (UTC)

I like the idea of autocollapsed tables within the table. But another way of doing it is to have explicitly labelled exemplar paradigms (e.g. showing object concords only for the simple present tense and putting a note at the bottom that says that they can be inserted with all the other paradigms in the table. —Μετάknowledge^{discuss/deeds} 19:01, 15 November 2016 (UTC)

That's ok in theory, but at least in Zulu the concords can interact with the surrounding sounds, in particular vowel coalescence between the concord and the following verb stem. So it's not as simple as "insert concord here". —CodeCat 19:08, 15 November 2016 (UTC)

Yeah, that's common in Bantu languages. I've considered the idea of putting those rules at the bottom, too, but it's messy. I think that argues for showing more combinations, by means of autocollapsed tables-within-tables. Do you want to try that idea, or do you have a better idea? —Μετάknowledge^{discuss/deeds} 20:34, 15 November 2016 (UTC)

Please create the family:

m = {
	canonicalName = "Northeast Bantu",
	family = "bnt",
}

And add the following codes to it:

bou
dig
doe
ebu
guz
gwr
ikz
jit
ked
ki
kiv
kiz
ksb
kuj
kya
lg
luy
myx
nim
now
nym
nyn
nyo
pbr
rag
reg
rim
ruf
rw
suk
suw
sw
swb
sxb
ttj
vun
wni
zak
zdj
ziw

Also create the following family:

m = {
	canonicalName = "Southern Bantu",
	family = "bnt",
}

And add these languages:

loz
nso
rng
st
tn
ts
tsc
two
ve
vmw

And also the family bnt-ngu. —CodeCat 22:23, 16 November 2016 (UTC)

Southern Bantu is a sure thing, but Northeast Bantu not so much. Considering that it is not completely agreed upon and in any case, no accepted protolanguage has been reconstructed, I'm really not sure what the value of adding it is. —Μετάknowledge^{discuss/deeds} 22:49, 16 November 2016 (UTC)

What about a smaller subset like Great Lakes or Northeast Coast? —CodeCat 23:20, 16 November 2016 (UTC)

Both of those are definitely valid. PNEC has been reconstructed, too; I'm not aware of systematic Proto-Great Lakes reconstructions, but I'll have to check the literature. Incidentally, Proto-Sabaki has been reconstructed as well (no surprise there, considering how heavily Swahili has been studied).

Hi @Chuck Entz, CodeCat, Metaknowledge, Wyang, JohnC5, Daniel Carrero I know it is probably a bit rude for me to ask, and I apologise for the broad {{ping}}; but would it be possible for me to be whitelisted for template editing using the new permissions level? I get requests for bot tasks which are really more naturally done with protected modules, and I have been annoying the administrators too much with third hand module edit requests. I am also thinking of looking for a way to include phonetic transcriptions in Module:links in a language agnostic way. I don't plan to do anything permanent without the agreement of all interested parties at a minimum, and possibly with a vote, if any interested party would like one in addition. But I would like access to the module to be able to test draft functionality. My planned edits to Module:links will be a more concrete variant of Chuck Entz's most recent proposal for a parallel transcription system; but I would like to defer discussion of that topic for the time being. Instead, I will start a new thread when I have a draft module (or set of modules), which I will be able to summarise in more concrete language. Thanks, Isomorphyc (talk) 16:10, 19 November 2016 (UTC)

Edit: Also pinging @Wikitiki89. Isomorphyc (talk) 16:14, 19 November 2016 (UTC)

 Done. (my flood flag is active so my message won't appear in the recent changes) Have fun editing templates and modules! --Daniel Carrero (talk) 16:16, 19 November 2016 (UTC)

Thank you for being so fast! Isomorphyc (talk)

• I hope this serves to encourage people other than me to nominate template editors... —Μετάknowledge^{discuss/deeds} 18:07, 19 November 2016 (UTC)

Wiktionary:Votes/sy-2016-11/User:Gamren for admin is not listed at Wiktionary:Votes/Active and I dunno how to do it. I swear creating votes used to be easier, didn't it? Please fix if you can. Equinox ◑ 18:56, 19 November 2016 (UTC)

It's not going to show up magically. You have to add it. —Μετάknowledge^{discuss/deeds} 18:58, 19 November 2016 (UTC)

When was it easier? Before, you had to add it in Wiktionary:Votes; now you have to add it in Wiktionary:Votes/Active. Incidentally, back then you would have to scroll the votes page to know what are the current votes, which were not organized by end date. --Daniel Carrero (talk) 19:04, 19 November 2016 (UTC)

I tried adding it but got the red broken template text. I thought the colon in the page title might be messing it up. I dunno, nothing ever works. Equinox ◑ 03:34, 20 November 2016 (UTC)

Very little on the technical side here is forgiving. Users seem to be divided into categories like: developers, power users (ie, readers of templates and modules), documentation readers, other contributors, all others. Only the first two are effectively served. DCDuring TALK 18:01, 27 November 2016 (UTC)

Going to look at {{votes/layout}} for documentation is no help. Such subordinate templates should have a documentation page that links to a page with either substantive documentation or hand-holding templates that enable contributing users to do things like start a vote. In this case WT:Votes would fit the bill.

All that developers need to do is think about more classes of users than their own. OTOH, if the intent is to limit vote creation to votes created by worthies without it looking as if that were the intent or to provide a playpen, the current system is fine. DCDuring TALK 18:24, 27 November 2016 (UTC)

How can I find all requests for ety, or all requests for pron? In general, if I want to find a category with requests in it, where do I start? Equinox ◑ 03:34, 20 November 2016 (UTC)

@Equinox: Category:Entry maintenance subcategories by language, perhaps? —Aɴɢʀ (talk) 15:02, 20 November 2016 (UTC)

Hello. I added the English Wiktionary as a search engine on my Mozilla Firefox secondary (search) bar. But it has not it's own icon, making it basically the same as any other search engine that has no icon. Couldn't you people fix this, so that the Wiktionary is added with it's favicon? Thanks!--200.223.199.146 10:36, 25 November 2016 (UTC)

If you search for "internet slang", the page Appendix:English internet slang doesn't show in the search results. I think it would be very good to make it show as the first result in that search. Ark25 (talk) 14:11, 25 November 2016 (UTC)

Please add the following to bnt-pro:

	sort_key = {
		from = {"", "", "", "", "", "", "ɪ" , "ʊ" , ACUTE, GRAVE},
		to   = {"a"   , "e"   , "i2"  , "o"   , "u2"  , "n"   , "i1", "u1"}},

—CodeCat 17:35, 25 November 2016 (UTC)

@CodeCat: Done. —JohnC5 17:48, 25 November 2016 (UTC)

I have been receiving this error for a few days now. Can't use Firefox any more on any computer:

Sorry! We could not process your edit due to a loss of session data. Please try again. If it still doesn't work, try logging out and logging back in..

I have tried to log out and log back in. --Anatoli T. ^{(обсудить}/^вклад) 03:33, 29 November 2016 (UTC)

I've always gotten that when I leave an edit window open for a long time and then click "save changes". It can be prevented by clicking "show preview" first, then "save changes". I don't usually bother with that, because I've found that simply clicking "save changes" a second time will work like it should have worked the first time. Of course, if you get this every time, even after a minute or two, it may not be the same thing. Have you done anything with the settings for how long to keep cookies or browser history or the like? Chuck Entz (talk) 04:02, 29 November 2016 (UTC)

Trying clearing your cache and cookies. --WikiTiki89 04:12, 29 November 2016 (UTC)

I get it all the time now, no matter how small and quick an edit is. Cleared cache and cookies. Wiktionary has been forcing me to reset my short and "unsafe" password. I wonder if it has to do with that. --Anatoli T. ^{(обсудить}/^вклад) 04:25, 29 November 2016 (UTC)

I don't think it's that. I've been ignoring that too and have not had this problem (but I use Google Chrome). --WikiTiki89 16:22, 29 November 2016 (UTC)

Is your FF latest? Have you tried using Chrome?--Giorgi Eufshi (talk) 05:39, 29 November 2016 (UTC)

My Firefox is up-to-date. I have my favourite plugins on Firefox, which are unavailable or bad on Chrome. I don't like the way browsing history works on Chrome either. I can find sites I often go to much faster on Firefox. --Anatoli T. ^{(обсудить}/^вклад) 05:46, 29 November 2016 (UTC)

Try checking that your system time is correct (if not then fix it, and then clear your cache and cookies again). --WikiTiki89 16:22, 29 November 2016 (UTC)

I have this problem as well. —CodeCat 18:58, 29 November 2016 (UTC)

Are you guys using any privacy/password add-ons? This seems like it has to be a cache/cookie issue, and sometimes add-ons handle such things. - TheDaveRoss 19:42, 29 November 2016 (UTC)

I don't. I have no system time problem either. Both my work and home computers are affected - not able to save anything in Wiktionary using Firefox. --Anatoli T. ^{(обсудить}/^вклад) 05:41, 30 November 2016 (UTC)

Could it be your IP provider or interaction between FF and the IPP? DCDuring TALK 11:59, 30 November 2016 (UTC)

I don't think this could be the problem. I use different providers at home and at work. CodeCat seems to have the same problem at the same time. --Anatoli T. ^{(обсудить}/^вклад) 12:05, 30 November 2016 (UTC)

I created a Phabricator bug after checking in with the IRC crew. Feel free to add to the bug if I missed anything. - TheDaveRoss 14:03, 30 November 2016 (UTC)

And then found out there was another ticket already in progress, follow this ticket instead. - TheDaveRoss 14:09, 30 November 2016 (UTC)

@CodeCat, Atitarev If you have cleared cookies and are still having issues, there has been a request that you capture the header data for edits which cause this error. Also, if you could create a step-by-step process which documents what you do which results in the error message. - TheDaveRoss 14:25, 30 November 2016 (UTC)

When editing Wiktionary:Sandbox the following error message is displayed:- "Module error: The function "list=?'"`UNIQ--item-0--QINU`"'?" does not exist.". This does not affect editing of the sandbox. SemperBlotto (talk) 07:31, 29 November 2016 (UTC)

@SemperBlotto: The edit text of Wiktionary:Sandbox is unfortunately located in Template:Sandbox, although they are not related (the latter is a sandbox for templates). Therefore, the module error is in Template:Sandbox. @Daniel Carrero: Do you have any way to remove the edit text? --kc_kennylau (talk) 16:22, 9 December 2016 (UTC)

Fixed. WT:Sandbox was invoking Module:sandbox, which is a module for test purposes. So I removed the module from that page. (but feel free to readd the module if someone was testing the module) --Daniel Carrero (talk) 16:40, 9 December 2016 (UTC)

@Daniel Carrero: I thought it was invoking Template:Sandbox... --kc_kennylau (talk) 18:39, 9 December 2016 (UTC)

@Kc kennylau: I'll explain better what I said above. WT:Sandbox is transcluding Template:sandbox to show the introduction text, which seems perfectly fine. When SemperBlotto opened this discussion to notify us of the module problem, WT:Sandbox was invoking Module:sandbox too, which is just a separate, unrelated test module whose contents are unpredictable. Anyone can edit the module to do tests. For some reason, the module seemed to be causing the module error, so when I removed it from WT:Sandbox, the module error stopped appearing.--Daniel Carrero (talk) 19:29, 9 December 2016 (UTC)

@Daniel Carrero: Can you link to the revision of the sandbox that was invoking Module:sandbox? --WikiTiki89 19:40, 9 December 2016 (UTC)

Damn. My mistake. I removed the module from Template:sandbox, not from WT:Sandbox. --Daniel Carrero (talk) 19:43, 9 December 2016 (UTC)

That doesn't explain why the error was showing up at Wiktionary:Sandbox. And anyway, why are we concerned about the sandbox having errors? The whole point of it is to test things out. --WikiTiki89 19:49, 9 December 2016 (UTC)

@Daniel Carrero: Can you verify the origin of the edittext of Wiktionary:Sandbox? Who set it to Template:Sandbox? --kc_kennylau (talk) 01:21, 10 December 2016 (UTC)

It looks like Template:Sandbox used to be the testing template and Template:sandbox used to be the editnotice but they ended up getting merged: Template talk:sandbox. —suzukaze (t・c) 01:27, 10 December 2016 (UTC)

Is there any setting that will make the edit box smaller? It's too tall, with the other toolbars and stuff that loads at the top. I want to be able to see the "watch this page" tickbox without having to scroll. Equinox ◑ 08:21, 29 November 2016 (UTC)

@Equinox in the preferences -> editing section the "rows" and "columns" affect the size of the edit box. - TheDaveRoss 13:40, 29 November 2016 (UTC)

I had to log in again, as is customary after a month or so, but now I can't show translations etc. I'm being told in preferences I need to enable cookies, but how were they knocked out? DonnanZ (talk) 11:03, 1 December 2016 (UTC)

I think the cookies are working, but the hide / show facility isn't working anywhere, and NEC isn't working either for me. DonnanZ (talk) 12:26, 1 December 2016 (UTC)

Hide/show and NEC not working sounds like a Javascript problem. Maybe clearing your browser's cache will help. — Ungoliant ^(falai) 12:56, 1 December 2016 (UTC)

I think I've tried everything, used restart and shutdown instead of sleep, checked JavaScript (enabled), logged in and out several times. I can't change preferences either, with the message "You must have JavaScript and cookies enabled...". Other odd behaviour: the images for the WOTD snowdrop are blanked out, but come up when clicked on. I can enter translations but can't review them or click on them. DonnanZ (talk) 13:27, 1 December 2016 (UTC)

I seem to have resolved the issue by logging out, then I was able to show a translations box full of translations. Logging back in again it now works, and the message at preferences has disappeared. How weird! DonnanZ (talk) 14:47, 1 December 2016 (UTC)