Forum History

What do you think about the changes? Something to add?
-- Risto

That's a good idea, Jens. The alphabetical list can be made easily from the AllPages view. Let's have a separate page for each letter: A for entries beginning with A, etc.
I added the alphabets to the SideBar already. That's the best place for it. We can have the categorical listing elsewhere or simply remove it. Let's start a new practice: Every time when you add a new word, add link to it to the corresponding alphabet page. If you add zone, link it from the Z page.
-- Risto

I was thinking about the front page. If we want to make Mulivo user-friendly, one idea would be to have a title, and then a list words that people can click on, in alphabetical order. I know the globe is cool, and so I don't want to get rid of it, so maybe we could put it somewhere else. But for navigation sake, I think the best would be to have the words alphabetically, and then rearrange the sidebar to make it easy to go to various places directly, like the list of languages, word lists by category, etc. I don't think it would be very difficult to make an alphabetized list; I could probably do it with MS Word. What do you think?
-- Jens

Oops, you're right. Maybe I'm getting to the age where I need some glasses. . .
-- Jens

I counted five bars after the language name just like in all other rows. So it should be correct.
-- Risto

Risto, when you changed the Model page, you changed the entries for verbs for Arabic, etc., back to only found vertical bars. Did you do that on purpose, or is it a mistake? Unless there are more bars, those entries will be short, I think.
-- Jens

The template for verbs/adjectives was wrong for Arabic and three other languages. I added the extra marks, but maybe it has to be done in a different way.
-- Jens

Yep. Most languages have ISO-codes. Esperanto's code is epo and Sumerian's sux. Some creoles and pidgins have codes too, but I don't know does Patois a.k.a. Antillean Creole have a unique code. cpf is used for "other French based creoles and pidgins", but I guess that patois or patwa will do in this case.
-- Risto

Risto, I have a question about the addition of new languages, which I think is a great idea, BTW. This is kind of an elementary question, but are there ISO codes for all languages? What about Sumerian or Esperanto or some fairly obscure language? What should we do if somebody wants to add a language that doesn't have an ISO code? My suggestion in that case is to alphabetize it under the name of the language.
-- Jens

I am also unable to finish things... And I think a mailing-list is a good mean for us to share our ideas.
-- Florent

Well said! Mulivo is a good example of the effect of cooperation. In the last couple of years I have tried to collect global word list with several methods and tools but didn't achieve much, because I constantly changed my methods and software tools. In Mulivo we reached quite soon mutual understanding of the covered languages and the way of presentation. The result is that at the moment Mulivo includes more data than all of my individual attempts combined.
Maybe we should continue this discussion in email. We could start using the Mulivo mailing list for that purpose if you want to have archive of messages for reference.
Are you in too, Florent?
-- Risto

Risto, I definitely agree to cooperating and then testing. I have exactly the same syndrome, inability to finish things. Working with other people definitely helps to complete things. We should talk about how to do it, i.e. procedures, and once we've decided on a process, start working. Rather than always arguing theoretically against the proponents of the "big three," we should start our own and show that we can do something.
-- Jens

I will take a closer look at it later and write you email then. I think that Mulivo will be used by several people. Florent has his longtime language project Luni which has evolved to Tanimu recently. I got a longtime language project too, with varying names. The only published sketch so far was Lone in 2002 but I have continued developing it all the time on the side.
There is something in common in all the three projects: analytical grammar and global vocabulary. We could join our efforts and put together a single language, test it by using it in practice. Then we can start producing learning and reading material for the public. We really should do it!
In spite of constructing my own grammar for the last 5½ years I'm still open to all options and ready for compromises, because it seems that I'm not able to finish anything on my own. I tinker too much with various ideas on the conceptual level and try them too little in practice.
-- Risto

I just wanted to let you guys know that I started a wiki page that has my own language work on it. The page is here. It's really a scratchboard more than anything else, but it does make liberal use of Mulivo so this is perhaps an example of the use of what we're doing. Feel free to give me any comments. But it's pretty disorganized.
-- Jens

The words are at the moment partly unsorted in our dictionary. I'm going to start organizing them in another way by adapting roughly the classification from Roget's Thesaurus. A classification tree is needed because it helps to keep track of covered concepts.
-- Risto

I agree. I ordered Dictionnaire Français - kiKongo ya leta a few days ago to complete the Bantu section. A few more contributors would be nice to have too, but it's hard to find them, because most people aren't auxiliary language crackpots like us. ;-)
-- Risto

I am reorganizing my website, and I just added a link to Mulivo. The best thing we should do, IMO, is to continue our work so the wiki will become more and more attractive.
-- Florent

Sorry, Jens, I don't have access to such data. It is not a feature of this wiki. I would like to see some statistics too.
Mulivo has been up and running only for a little more than one month now. So I would estimate that there aren't a lot of visitors yet. At the moment the only links to Mulivo on the web are at LangX and my personal website. More is needed if we want to get more attention. Florent has a website too, a link there would be nice, so we would get the attention of the francophones. :-)
Feel free to advertise Mulivo anywhere!
The advertising will go other way too. I intend to set-up a page to Mulivo that provides links to all the languages that have benefited from Mulivo and publicly acknowledge the influence. So far, there are no such languages, but that is about to change this year.
-- Risto

Risto, I was wondering, but do you have data on accesses to this site? If you do, it would be great if you could post the figures here or somewhere else from time to time, just so we can see how accesses are going.
-- Jens

Hyvää uutta vuotta! 新年好！
-- Risto

今年もよろしくお願いします.
-- Jens

Bonne année!
-- Florent

It works for me too.
-- Jens

That's OK now for me.
-- Florent

On a 1024 pixels wide screen, it would be OK with a picture of about 670 px.
-- Florent

I have the same problem as Florent, with a 15" monitor. I'm not sure exactly, but I think if the picture were about 1 cm smaller it would fit into the screen (with the browser taking up the full screen).
-- Jens

Can you tell me how big picture would fit into your screen, Florent? I can edit the picture tonight after I get back home. The picture is there to arouse web surfers' interest on the first sight and provide preliminary insight on what Mulivo stands for.
-- Risto

The picture looks good, but is too big (at least for my screen), so it is displayed below the sidebar.
-- Florent

What do you think about the new picture at the front page? Each of the bubbles reads "language" in various languages. I couldn't fit all the Mulivo languages there.
-- Risto

The new Telugu words are well legible in my Firefox browser. Probably you don't have correct fonts to display Telugu script. I have the same problem with the Ethiopic script for Amharic language at work. Firefox displays only question marks for it. But I got an Ethiopic installed at my home computer and everything works well there.
If you have problems writing IPA symbols, you may as well use the rough IPA, (e.g. S for ʃ (English sh) and tS for tʃ (English ch), and the rest of us will replace them with the correct IPA characters.
Come to think of it, I think we should write somewhere a crash course to IPA, because certainly some people who visits this site are not familiar with it.
-- Risto

It's great to see that there are now some Telugu words in the dictionary. Unfortunately, though, my browser can't read the script. I wonder if there's something that has to be done to make it legible, or is it just a browser issue? Atually, there are other languages I can't read on this browser (Firefox on Win2K), such as Korean, Hindi, and Tamil, so it may just be a problem on my part. Also, although it may be challenging at first, it would be wonderful to have the IPA readings for the words. But in any case, it's great to see that we're getting more contributions.
-- Jens

Welcome to both Arvind and Rao. It's great that we now have contributors who have knowledge of South Asian languages and Europe and also Chinese and Japanese. It would be great to find people with knowledge of Southeast Asian languages as well as Africa.
-- Jens

Good idea, Florent. I was lead to the distinction of village/town/city by the fact that these words are differentiated in some languages though some others (including Finnish) have only village and town/city. But I agree with you totally! Villages belong to countryside and towns and cities are both urban. So we have to combine town and city pages together.
-- Risto

I find the distinction village/town/city fuzzy, also very variable, depending of the country. I imagined to define village as a group of buildings with agriculture on its territory, city/town as the same without agriculture.
-- Florent

That's fine by me. Also dash (-) could be used for the same purpose.
-- Risto

Here is another issue to consider. There are certain words, especially grammatical words, that do not exist in some languages. A good example is definite and indefinite articles. I think we can just place N/A in the spaces where there are no words. What do you think?
-- Jens

Looks good. Hopefully people will get interested and we will get new contributors too. It's good to spread the word about this project, so that language designers are aware of it. It's OK that my email address is included there as well.
-- Risto

By the way, I created a Letter that I've sent to some yahoogroups to try to get people interested. If anybody wants to make changes, please feel free to, and to send it to webmasters of interesting pages or anybody else who might be interested. Risto, I didn't put your e-mail address in because I didn't think it would be nice to spead your address around the Internet without asking you.
-- Jens

They are case sensitive so forum and Forum refer to different pages.
-- Risto

Just a simple question, but does anybody know if the file names are case sensitive or not? I'm asking because for example, the name of this file is "Forum." I'm wondering, if I ever create a definition for the word "forum," will it conflict?
-- Jens

You are right about (1). I can imagine it could scare away some non-technically oriented people if they have to create new pages from scratch. This wiki is quite simple but it doesn't have WYSIWYG editing. But the alphabetical listing which indicates incomplete pages is a necessity. It shouldn't be too difficult to maintain either because the "recent changes" page helps to keep track on changes.
-- Risto

My way of thinking is that we should create new pages as quickly as possible, because then new people who join can add things without having to worry about copying the template and being the "first" to make a new definition. But I do see your point about completing entries. I think there are two ways to approach it:
(1) we only have expertise in certain languages, so we should try to start new entries with the languages that we know, and try to get other people to add in later.
(2) The project is most useful when we have complete entries, so we should concentrate on trying to finish the entries we have started.
My personal preference is for (1), because I'm hoping to find new people to "fill in the blanks." But if it's better to take the second approach, I'm willing to try that too.
In any case, though, I do agree with the idea of making a vocabulary list with the format using x to indicate entries that are incomplete.
-- Jens

I found a Thai dictionary from Internet: [http://www.thai2english.com]. It can be adjusted to use IPA for transliteration, which is quite useful for us. Also the original script can be copy-pasted from there. So the situation is improving.
There is now 380 pages in Mulivo, mostly vocabulary tables. Maybe we should slow down creating new pages and instead concentrate on completing the incomplete word lists. In addition we could create alphabetical lists of words where we also mark the complete pages. At the moment there is none, but it will be useful in the future, when there are hundreds of words and some lists are empty and some are full. The list could look like this (the x marks incomplete lists):
sand x
sea x
sun
-- Risto

I got a small dictionary of Thai with something like IPA, so I can input some of those words. I'll try to write up a short blurb that we can send to people who might be interested.
-- Jens

Maybe you're right. Even the English version hasn't attracted many people yet. ;-) But I'm starting to figure out why there aren't many egalitarian auxiliary languages out there: it requires a lot of devotion and hard work to make one.
I would be happy to see Burmese and Thai even with only romanizations. At the moment the languages which we have neglected most are Bengali, Marathi, Panjabi, Burmese, Thai, Javanese, Mandingo and kiKongo. I don't expect to find specialists to handle those languages any time soon, so we should figure out a way to deal with it ourselves. If you can find good and extensive on-line dictionaries for any of those languages (preferably with romanization) that would be highly appreciated.
-- Risto

In my book, there is a little vocabulary for Fula and kiKongo, with the original orthography. I have only pronunciation for Tai and Burnese. Burnese is especially painful to write, even with unicode.
About the French interface, I am OK. I can create a french translation of homepage and other main pages. Tell me what you wish exactly. However, I am not sure it will attract many people...
-- Florent

Jens, hopefully it will happen. Mulivo will be on-line for years so sooner or later some people will find it and lend us a hand with languages that we don't know so well at the moment.
Florent, there's most of my links. I have't remembered to add all there. Then I got some dictionaries at home and looks like I should order dictionaries for Fula, Thai and Burmese because their fields are constantly empty here. Also a good source is needed for kiKongo (ya leta) because my Angolan source in Portuguese isn't that good. Sometimes I wonder, would it help to attract more contributors if Mulivo interface was also in French. Because there are English and French blocks of linguistic influence.
-- Risto

I think it's also better not to put question marks in the entries. We are not experts in many of the languages, so I think we should just fill in the blanks with the information that's available. The nice thing about a cooperative project is that later, somebody who is more knowledgable can go back and clean up things that are mistaken.
-- Jens

Do you wrote about these links? Also, Wikipedia is interesting. Some words have entries in many languages, using the original script.
-- Florent

It's OK. I was just thinking that it's best to type in words that are sufficiently reliable. The quotation marks made me wonder is the data reliable. But I suppose that a book like that is OK. I have used many word lists and dictionaries from Internet. They seem reliable, but what do I know about Bambara, Persian, Tamil, etc? So in many cases I can't guarantee that the words that I type in are entirely correct.
I could post you the links to some Internet dictionaries that I have been using. I got already hands full with book dictionaries, so it would be good if you could help me to collect data from Internet. Are you interested?
-- Risto

I used question marks as a reminder for me that I have not found the info. Also, I have a book with a small vocabulary in many languages, but the pronunciation given is far from being phonetic. So, these pronunciations need to be double-checked. But, OK, it was bad ideas. I am sorry to have annoyed you...
-- Florent

Florent, what's this idea to put question marks to empty cells? Do you want that initially every table is filled with question marks? Empty spaces indicate already that we don't have the information available.
And what's the purpose of quotation marks? Do you want to emphasize uncertain words and IPA transliterations?
In any case, I think it's better discuss it here first before starting to make significant changes to the presentation.
-- Risto

OK, the "expert" decides to use /ʀ/. (After listening some sound files, I think my own R is /ʁ/...)
-- Florent

You are the expert, Florent! :-)
-- Risto

From Wikipedia:

And it's true! I am not sensitive to the differences between rhotic sounds. For me, japanese R is a kind of L, all other Rs are roughly the same. So, Risto, I can use again /ʀ/, if you wish so.
-- Florent

I noticed that Florent has begun to use /r/ instead of /ʀ/. Perhaps it would be more appropriate to keep on using /ʀ/ because that is the real sound in French. Similarly we cannot start using /l/ or /r/ for Japanese /ɾ/ because they would represent one important feature of Japanese pronounciation incorrectly. Of course, if /r/ is a dialectal variant of /ʀ/ in France, then this symbol would be justified.
-- Risto

In theory, French distinguish between ɔ/o, ɛ/e, a/ɑ, ans also œ/ø. But in practise, the pronunciation varies from people to people, and distinctions are blurred. To deal with variety of french pronunciations, linguists often use metaphonemes /A E O Ø/. So it won't be a problem to use a simplified system.
-- Florent

How is this as an idea? I made a link from the Japanese entry in the language code table that gives the transliteration guide. I gave it the name jpnipa. Perhaps making one for each language would be a good step. And I could add an explanation that this is not intended to be IPA for linguistic studies.
-- Jens

It seems that /ɺ/ and /ɻ/ are inspired by the fact that these sounds are written with "r" in the official romanizations. /ʐ/ represents retroflex z. /ɻ/ represents retroflex English /ɹ/. So both are good. In that case I have mistaken because I have used /ʃ/ for /ʂ/ in Mandarin. The manner of articulation is different but the sound is almost the same.
On the other hand, Mandarin is the only Chinese dialect which differentiates /ɕ/ and /ʂ/. The others have /ʃ/ or merely /s/ instead. /ɕ/ appears only before /i/ and /y/ and /ʂ/ appears before other vowels. So in practice these sounds could be written with one letter (I suggest /ʃ/) because there is no possibility of confusion. If we write /ʃin/ you know that it means /ɕin/, Pinyin "xin". If we write /ʃan/, you know that it means /ʂan/, Pinyin "shan".
I agree with both of you that we should use simplified version of IPA. Then we have to create a page where we explain the romanization conventions used in Mulivo.
What comes to /ɔ/ and /o/, /ɛ/ and /e/, /a/ and /ɑ/, they should have separate representations only in such languages where they differentiate meanings. For example Mandarin has /jɛ/ but no contrast to /je/, so in my opinion we can use /je/ here.
Apart from Volapük, I don't know any other international auxiliary language which has differentiated between /e/ and /ɛ/. And honestly, I have never been able to hear the difference between these two in English. The same goes for /a/, /ɑ/ and /ʌ/. To my ears they are all the same, at least in normal speech.
-- Risto

About Japanese "r", I used /ɺ/, but I don't know if it's correct. Japanese "r" is between (English) "r" and "l". Personaly, I hear and pronounce it "l", because French "r" is too different.
-- Florent

About chinese "r", my dictionary uses /ʐ/. Also, don't forget that IPA notation are usualy phonemic and rarely phonetic. English dictionaries don't shows aspirated "p", nasalization of vowels (like in "bank"), etc. We should define a level of detail to use. It would make sense to merge /i/ and /ɪ/, /y/ and /ʏ/. Some languages distinguish between /ɔ/ and /o/, /ɛ/ and /e/, /a/ and /ɑ/, but they could be merged too.
-- Florent

Risto, I agree with your suggestion at the end with regard to English. I think that we should generally use a "minimalist" approach to IPA. Our purpose is to allow people to understand how the word is pronounced generally, so it's important for example to know that the "ch" in French is different from the "ch" in English. But for example, I don't think in Chinese that there is a need for us to distinguish between "q" and "ch" (pinyin) because in any IAL, those would almost certainly be the same sound. Maybe eventually we should make our own style manuals. Actually, I haven't been able to even find any guide for transliterating Japanese into IPA. But in any case, I think our purpose is to make sure that language designers don't make silly mistakes, such as between "j" and "y" in European languages, and is not to make small distinctions. Anyway, I'm happy to use the ɻ for R in Chinese.
-- Jens

We should fix some rules how we use IPA to write some languages. This occured to me, when I saw that Jens had used ʑ to represent the Mandarin "r". I have used at least in one occasion ʓ. When I tried to search for which one is correct, I noticed that Wikipedia suggests ɻ. Also I have seen ʐ used though I can't remember where. The sound, as it was tought to me by a native Chinese, is the same as in English but the tongue is curled backwards. I really don't know which is the correct IPA symbol, but it would be good to use only one symbol in Mulivo.
Then there's another kind of problem with English. Normally the IPA spellings do not show aspiration in words like [pɪt] but they differentiate painfully accurately different vowels. As far as I know, there is no contrast between [pɪt] and [pit]. So why should we differentiate it here? Is it helping language makers anyhow? After all, optimal IAL would have small phoneme inventory with no such distinctions.
-- Risto

Actually, I registered this site with Langmaker about a week ago, so if they do update it, they should eventually list us. But if it isn't being updated, then who knows.
-- Jens

I changed the style of lists. In the default style, the dots are made with GIF images, and there were some display problems (at least for me...). Tell me if they are problem for you with the new style.
-- Florent

Unfortunately, Langmaker hasn't been updated since 08/30/2005... I hope it's a temporary stop.
-- Florent

Yeah, that's why I wanted to use something else than Lucida, the font looks horrible in italics. But I guess we have to live with it unless there is some better font available. I have used Gentium sometimes, but it's not very beautiful either on screen.
Good news! Mulivo is now found with search engines like Altavista and Google. Let's just hope that a lot of people will find here. Perhaps we should also announce Mulivo at LangMaker. What do you think?
-- Risto

Done! I saw a little problem: the class prefixes in italic look bad on tables.
-- Florent

The sidebar looks OK to me now. I changed my mind about Arial Unicode MS. It includes also the Chinese and Japanese characters but they are less legible than my other fonts. So it is fine by me if you go back to the old order. Sorry to bother you.
I agree also that this Forum here is better place for discussions than the new discussion option.
-- Risto

Arial is first now. But the sidebar is rather big here, I don't understand why...
-- Florent

At the moment Lucida is displayed but I would prefer Arial. Please put Arial first in CSS. The text on the SideBar is very small. Otherwise it's OK.
-- Risto

I tinkered a bit the style. I specified the fonts: "SIL Doulos IPA", "Arial Unicode MS", or "Lucida Sans Unicode". How is it for you? Is the sidebar too big?
-- Florent

I found here how to change the CSS. You have only to upload a file named "wiki.css" containing your custom style. I will try to grab the current CSS to modify it.
-- Florent

My browser does the same. I paid the fee for premium service and one of the extra features is customisable CSS, but I haven't found any information how it could be done. In fact, it seems to be on the upcoming features list. Have you noticed the new "Discuss" button on each page?
-- Risto

I would like a better display of IPA characters, if possible. On my browser, there is a mix of Verdana and IPA fonts. Is there a mean to modify CSS? If yes, I would be able to change the font(s) used.
-- Florent

Remember to use the five column template for the verbs. It has places for verb and noun. Also it should be used for adjectives, since adjectives are basically verbs in some languages. In that case there is place for adjective/verb and noun. That would also clarify situations like these in English:

It is easier to identify which is root word and which is derivation.
-- Risto

Thanks for the improvements, Florent. It was getting tedious to scroll down this page every time. IPA keyboard is a fine tool and it deserves well to be added to the Instructions.
-- Risto

What do you think about the IPA keyboard? Maybe we could add the link on the Instructions page?
-- Florent

I think we should write our messages here in the reverse order. New messages would be displayed on the top of the page, and then be of easier access.
-- Florent

I have been advertising Mulivo in some news groups. So far one person has been interested. Hopefully she will join us.
If you know any favourable place where you could advertise Mulivo, feel free to do so!
-- Risto

I linked "lead" to Pb in order to avoid this kind of confusion. The chemical symbols are unique so it is more systematic to use them than iron, lead, gold, etc.
For the verb to lead you should use the link lead. But in that case it should be listed only in Actions not in Names.
I noticed similar problem with right which was intended in one page as the opposite of left, but the other meaning "right to do something" was used in the referred page. Then I created new link righthandside for the opposite of left but hid it under different presentation, right, with the vertical bar.
English words are sometimes ambiguous and it is important to disambiguate them one way or the other.
-- Risto

That looks better, thanks. Actually, it's a bit of a problem, because apparently the "lead" is a link to "pb," so now we have "to lead" classified under "Pb." How difficult would it be to change the links so that pb goes to "lead," "au" goes to "gold," etc.? Although actually it's not that big a deal, because it doesn't happen that often. But the same problem will come up with "iron," which also means a tool for ironing clothes in English. And for "mercury," which is also the name of a planet.
-- Jens

I changed the presentation for Pb. Tell me what you think.
-- Florent

Something that just came up. There are words in English with the same spelling but totally different meanings. For example, lead (the verb) and lead (the metal). Maybe the best way to handle this is to use 1 and 2, and then put a note at the top noting there is more than one definition? I'll do it for lead, and let me know what you think. (the other possibility would be to have more than one entry, with some way of marking the difference).
-- Jens

Oh yeah. Good that you reminded me. I'm too used to write plain HTML.
-- Risto

Dictionaries are far from being perfect. However, they do a good job until we find a native speaker to correct it. Also, Risto, I saw you use html to do italics, like this: <i>word</i>, but wiki syntax is simpler: ''word'' (2 apostrophes before and after the word).
-- Florent

Also, just for interest, there is a word "ifuku" in Japanese, which is a loanword from Chinese, and written with the same kanji was "yifu." In Japanese it's usually used for the industry rather than everyday clothing. But clothing stores are usually called "yofuku" stores, which means "western clothing." And when people say "get dressed," they say "ki-gae," with ki being the same as in kimono and gae meaning "change." And like Florent said, "kimono" is usually reserved for the traditional dress.
-- Jens

First, I agree we have to be careful about what we use and don't use. In English, toga usually means the Greek clothes, and robe would be a better term to start from in English. As you said, one problem we'll run into is that none of us are experts at certain languages, so it's hard to decide what terms are general, etc. There are also problems that will come up just because we are using English to start with, rather than concepts. In any case, I think the idea of an alphabetical listing would be great if it's easy to do.
-- Jens

In that case my multilingual dictionary is wrong about that term in French. So I used another dictionary to verify that "toga" really means a long robe in Polish. That was the exact definition. But sure, the problem is that we do not master all the languages that are included in Mulivo and we have to resort to dictionaries which reliability we are not able to judge. Like I don't really know, is "toga" ever used in Polish because there are other synonyms.
It would be good, if the words could be sorted in order of frequency of use. That would be helpful. Unfortunately I can't do it with my dictionary languages (except in cases where the dictionary gives only one alternative which probably is the most common term).
By the way, doesn't Japanese have the term "ifuku"? I remember seeing it somewhere and connecting it immediately with Mandarin "yifu". Perhaps it has yet another meaning, though the Mandarin word means simply all kinds of clothes.
Anyway, it's fine by me if you add or remove words from the lists or rearrange them. Wiki is for collaboration. Just be careful with functional links that refer to existing pages. It's better not to remove such links completely. We could create a trash bin for that kind of links.
By the way, I have been thinking of creating an alphabetical listing of all the links. What do you say? Would it be helpful?
-- Risto

I didn't know what a "dhoti" is... The french for "toga" is "toge", but is only used for antique roman clothes, so it's a rather uncommon word. "Kimono", in Japanese, is a generic word for traditional japanese clothes, without pants. Kimomo = ki-mono = wear-object (着物). It refers generaly to the ceremonial clothing. For occidental clothes, the word "youfuku" (洋服), or only "fuku" (服) is used.
-- Florent

The definition of "toga" in my dictionary is "a loose robe worn by clergymen, lawyers, teachers etc". If this is not common usage in English then we could include the word that more commonly represents this concept: robe or gown. Though I notice that "toga" is used for this concept in the Romance countries. So togas didn't die off with the Roman empire.
As for "dhoti", I suppose similar clothes are used also elsewhere than in India and they might be called with different names. That there is no word for it in English doesn't exclude the possibility that e.g. there's different word for it in Tamil, Thai or Indonesian. We should not let our prejudices show when selecting words. For example the word "coat" is completely useless in some languages of the hot climate zones. If we exclude dhoti, we should exclude coat as well.
As far as I know, kimono is a general term in Japanese which refers to all kinds of clothing, not only the certain kind of ethnic dress of geishas. And pizza is just one type of pie. It is called "pisa" in Mandarin which is suitable for IAL with simple phonology.
-- Risto

I took the words "dhoti" and "toga" out of the names list. I don't think we need words there that clearly belong culturally to one place. For example, it doesn't matter is some country has a different word for "kimono," we should still retain the Japanese word to refer to a Japanese ethnic dress. And for food as well, I think "pizza" should be "pizza" since it's the Italian word for the Italian food.
-- Jens

Yes, I think maybe not marking it is the best choice. After all, we're not doing this for a study in phonology, so I think the important thing is that we get the general pronunciation of a word right, not the exact phonology. So I won't bother with the devoicing.
-- Jens

We should keep it simple here. Too much detail in the pronunciation can become quite a burden. We have already made some simplifications here. For example we do not mark stress and tone, though they are important features in many languages. Also we haven't marked grammatical gender. We have done so and we should continue to do so because grammatical gender, tone and irregular stress patterns are undesirable features for IALs. Also devoicing would be a useless complication. Perhaps we can make the choice not to mark it?
-- Risto

About the Japanese devoicing, you can look at Wikipedia article. Phonetics can become hideously complex...
-- Florent

Bantu languages use prefixes to classify nouns. They are not part of the stem but they have effect on the meaning of the word in total. For example in Swahili mlima = hill, walima = hills, kilima = mountain, vilima = mountains.
First I was going to separate the stem from the prefix with a dot or a dash, but that is not the actual usage. Italics is a neutral way.
-- Risto

Florent, I understand the problem you have with French there, although actually it's a bit different, because in the Japanese case, it's just a softer sound. Shita sounds a bit like shta, but there is actually a sound there. With the French, the problem is that the sound changes depending on the following word, right? So for example, in "trois pommes" it is just pronounced /tRwa/, whereas in "trois amis" it is written /tRwaz/. So probably the notation should be different for those two cases. I think that using the () is a good idea for the problem with French. So for the devoicing issue, what about putting the vowel in italics? There are words in the list in some African language that have italics (like the Zulu word for "mountain". I guess that Risto put those in. What do those italics mean? It's possible that that is signifying devoicing? But I'm happy to use smaller letters as well.
-- Jens

I am the same problem with French "mute e" and liaisons. There are variations depending of the rythm of speech, the dialect, the age of speakers, etc. I propose we note them between parentheses: /S(i)ta/, /des(u)ka/. In French, I will do the same, for example: trois (3) = /tRwa(z)/.
-- Florent

Maybe Florent knows the official way. I can only suggest using smaller type (desuka) with the <small> tag of HTML.
Are these sounds similar to the Manadarin "si" and "shi" where the "i" is not pronounced at all. They are almost like syllabic consonants (with the tone, of course).
-- Risto

I have an elementary question on IPA notation. How should I notate a vowel that is devoiced? In Japanese words like "shita" and "desuka" for example, the i and u get reduced to something that has almost no sound.
-- Jens

For your information. I added three languages to the template. Mandingo was listed above but somehow I did not notice it earlier. I also added two Bantu languages: kinyaRwanda (the largest in number of native speakers) and kiKongo (the lingua franca in coastal Central Africa).
There won't be more additions anymore from my part because the coverage is now very good.
-- Risto

Yeah, that list of words amazed me too. I knew previously that the Slavic and Indian words were connected (Indoeuropean) but it was surprising to see similar cognates in Chinese languages as well, and in the Horn of Africa too. All these words, miel-mel-mar-mjod-mat-madu-(damma)-mitsu-mi form a continuum. Finnish "mesi" (genetive "meden") is related too, as well as Turkish "bal" where the initial M has become B.
There is the so called nostratic theory of the common origin of all languages, it hasn't convinced me earlier, but all these words for honey are undoubtedly related. It is very exciting to find ancient, even prehistorical international words. I hope we will find more of these. If you notice any other, please let me know too.
-- Risto

One interesting thing about this project is that it could be applied not only to IALs but also could be used by researchers studying etymology. It's sort of interesting to see the similarity for the word "honey" between Asian languages and Slavic languages, for example. In this case it may well be just coincidental, but in any case this could be a resource for people trying to work back to protolanguages.
-- Jens

I'm happy to keep the discussion here. It actually makes more sense, because if a question comes up while I'm editing, I can quickly post it. I started the mailing list, but I don't think there's any reason to use it at present.
-- Jens

If you know any good place where to publicize, feel free to do so. Now that our method for arranging the data has developed (it looks quite scientific to me) we are ready to have more contributors. I am happy to enter the data for several languages but I know that native speakers would do better job than me with a dictionary. Also with more people we get more work done faster.

How about that mailing list, Jens? Or are you happy with this discussion forum? Actually this is quite good place because everything is archived immediately and I can see new entries from Changes. If there will be more discussion, we should start new discussion threads. The threads could be named along these lines: ForumNewTopicForDiscussion. So all discussions would start with Forum to separate them from the real data on this site.
-- Risto

Just a quick question, but should we do anything to publicize this site? On one hand, it's not really complete yet, so might be difficult as a resource, but on the other hand, it would be nice to have more people involved, especially people who know some of the South Asian, Middle East and African languages.
-- Jens

I think it's OK. We can just separate the words with comma both in original script column and in IPA column. Hindi and Urdu have both borrowed the Arabic word but they haven't given up their own word for island. Since both are found in dictionaries, then they should be included in the list. General rule would be to permit synonyms which are commonly used, in maximum 2-3 words per language. Does it sound reasonable?
I suggest we use more columns for action words. The first two columns are for nouns, the other two for verbs. See doubt and fight for examples. It would be useful for those languages where the noun differs from the verb (unlike usually in English). We could make a template for this too.
-- Risto

"Island" needs cleaning up; there are two sets of words for Urdu and Hindi.
-- Jens

Looks good to me. I created one (hand). I have one ore suggestion, which should be easy to implement. Why don't we put notes at the bottom of the page if necessary. I was thinking of things such as mentioning that certain words are "prepositions" in some languages but "postpositions" in others. Or that articles do not exist in some languages, for example. And for example, how should we deal with things like sentence ending particles? English doesn't have them, so there wouldn't be a category if the dictionary is based on English. But some IAL designers might want to include them.
-- Jens

I made a Model for new pages. The table includes the most of the above listed languages. Wikipedia has list of languages by number of native speakers. I took the largest Indian languages from there and Bhojpuri didn't end up to the list.
The list got pretty long already, but there's so many languages in the world. If you think that there is urgent need to extend the list, feel free to make changes to the model and inform us too. Jens is right that it is easy to change the list in the future. So let's start adding words and make changes when it's needed.
-- Risto

Sure, Quechua could be added. It seems that vocabulary lists can be found for that. Another idea also, maybe as a later project, would be to add classical languages like Sumerian, Sanskrit, Latin, etc., which might be of interest to people trying to create an IAL based on some idea of a "human protolanguage."
-- Jens

I had listed Quechua (runasimi) as an isolate language. I have read somewhere that it has 20 million speakers, though oppressed and divided by dialects. By the way, Quechua is similar to Chinese and Korean in that it differentiates aspirated/unaspirated but not voiced/unvoiced. The other Native American languages are unfortunately very small, as far as I know.
(Addition: Guarani is spoken by 7 million people according to Wikipedia.)
-- Risto

I look forward to seeing your revisions, and probably can agree to anything that is pretty much along these lines. Another thing we should think about is, the list seems to lack any languages as all from Native Americans. They are generally small languages, but maybe Kiche (Maya) should be put in symbolically at least. One problem, of course, is that if we include too many small languages, we may not be able to find anyone to fill in the vocabulary, and so it will be meaningless. We should remember that we can always add languages later, so it's probably best to start out with too few rather than too many.
-- Jens

Yes, the words should be rearranged. The words are from the Essential World English list, which has the English approach to it. I already transferred all the tool verbs to the names/nouns section. In English you can use "bath" and "brush" as verb and noun, but in many other languages it is not so straight forward. Categorizing them (and "snow") as nouns is better.

Your list of the languages is quite good, though how can Amharic have more speakers than Oromo? Both are Ehtiopian languages and it is uncontested that Oromo is the largest. Also I must criticize that there is only three Sinitic languages but four Romance languages, though the number of Sinitic speakers is twice as much. Also Slavic and Dravidic got two languages each but Bantu, which is about the same size, only one language.
Anyway, I agree with most of the list and all of the listed languages should be included in Mulivo. Just some additions are needed.
-- Risto

Another thing we need to think about is how to organize the words. I do like the idea of doing it conceptually, but there are problems. For example, the word "snow" is included in actions, which may be correct in English, but at least in Japanese you have to say "the snow falls," and I think also in Mandarin the term is "falls snow." (xia xue?). Maybe this would simply require some rearrangement. What do you think?
-- Jens

I don't know if this will help, but here is a list adapted from the Linguasphere 1999-2000 survey, so it's a bit outdated. I think it comes to about 40 languages, so there is room for a few more (like Finnish, maybe). :) The list is a bit long, sorry.

Language Table

Sinitic
Putonhua 1000m
Wu 85m
Cantonese 75m

Germanic
English 1000m
German 125m

Indic
Hindi 900m
Bengali1 250m
Panjabi 85m

Marathi 80m
Bhojpuri 60m

Romanic
Spanish 450m
Portuguese 200m
French 125m
Italian 70m

Slavic
Russian 320m
Ukrainian 60m

Semitic
Arabic 250m

Hesperonesic
Malay-Indonesian 160m
Jawa 80m

East Asia
Japanese 130m
Korean 75m

Dravidic
Telugu 75m
Tamil 65m

South Asia
Vietnamese 75m

Transasian
Turkish 70m

Bantuic
Swahili 60m

Daic
Thai 45m

Bauchic
Hausa 40m

Iranic
Persian 40m

Irrawaddic
Burmese 33m

Amharinya
Amharic 25m

Benuic
Yoruba 25m

Atlantic
Fula 16m

Uralic
Hungarian 15m

Kushitic
Oromo 14m

Mandic
Manding 13m

Hellenic
Greek 12m

Tamazic
Berber 12m

-- Jens

Yes, I was thinking along similar lines. I think we should limit it to a reasonable number, but then agree that there should be at least one language from each family. This depends on definitions, of course, but since this isn't a scientific project, we have some leeway. I was personally thinking of maybe the top 30 in terms of population, plus the largest language from each family that doesn't reach the top 30. That would probably make a reasonable number.
-- Jens

That's a sensible idea. Of course we will allow only a limited number of languages to the list. Which languages exactly? There is a list of the 50 most widely spoken languages. But it is inaccurate and out of date.

I have studied this matter for a long time and I have also studied the United Nation's population growth projections to find out, which will be the largest languages in the future. Also it is important to see the big picture, to look at the matter on the level of language families. For example, though the largest Bantu languages of Africa have only up to 15 million speakers, the total number of Bantus is 300 million and the Bantu languages are essentially quite near to each other, namely, they have a lot of common vocabulary. So we have to allow some Bantu languages, though not a single Bantu language (perhaps except Swahili) reaches the top 50.

I know pretty well which languages to include to Mulivo. I can also provide reasonable justifications for each of them. Let me work on it a couple of days and I will post it to this site.
-- Risto

It's OK with me to do it that way, actually it's a very logical way. The problem is, if we allow thousands of languages, then those in the bottom group may have to scroll down through pages and pages to get to the right language. Why don't we organize it by language groups, but then limit it to languages with a certain number of speakers? Say the top 50 languages or something like that?
-- Jens

I prefer grouping according to language families and cultural groups (as done in the new version of language codes), because it would establish a link to this scientific theory, which was an important inspiration for starting this project in the first place.
It would be good if we could highlight the major languages somehow, but I don't know can it be done with this wiki easily.
Capital letters for language codes is OK by me, though I believe that ISO standard uses lower case letters. They are all the same to me and both work well with tables/graphs.
-- Risto

I put in another suggestion, for a listing by number of speakers. To be honest, I see advantages to the different systems, so I'm not sure which is best.
(1) I like the system of listing by speakers, because it automatically puts the major languages at the top, and makes it easy to see which words are most common. However, it doesn't allow us to see which words are shared by many languages, unless we look carefully.
(2) I like listing the similar words on one line, as per the original proposal, because it helps us see which words are shared by many languages. However, it's possible that a word could be shared by 20 very small languages, but the number of speakers of Chinese could be enough to overcome that.
(3) The system of using a table and putting similar words together seems nice also, except that it makes it impossible, if we have many languages, to easily find what the word is in a certain language. Using the system in (1), if you want to find out words in French, for example, you can always look at the number 6 position. Increasingly small languages would be at the bottom of the list, so it's always possible to add smaller languages.

Finally, I do like the idea of using capital letters for the country codes, simply because it makes it clear that the country codes are not actually letters to be pronounced. So maybe we should go with that. In any case, we can change them at a later date again.
-- Jens

It works just fine. Well done!
I am at work now and cannot check my private email at the moment. What do you think of the different alternatives to arrange the languages, IPA and native scripts? There are samples in apple.
-- Risto

I just installed IPA modifications to my OS X system, and just tried it. I changed the entry in "arm". Could you check to see if the font is OK? I still have to learn the IPA for Japanese...
-- Jens

I typed html, but it's fixed now.
-- Florent

For some reason the Unicode characters that you have typed appear as HTML codes in the Edit mode, but my characters are plain Unicode. Did you type HTML or does your browser do it automatically. In that case I suggest that you switch to Mozilla Firefox or Opera, which handle Unicode natively.
Because it will be a problem in editing the pages if all characters specific to IPA, Russian, Japanese, Chinese, etc. appear as numbers in the edit mode.
-- Risto

I downloaded some better IPA fonts, so it works well now.
-- Florent

I can read nasalized ɛ in there. The nasal sign is not exactly over the letter, but that is the font's fault. The input is correct.
-- Risto

I have problems to insert IPA, like this: ɛ̃
-- Florent

 Create your own educational wiki!HelpPublic, Premium Early Adopter wiki Privacy Policy RSS