New auxlang: Atlas

[Rodiniye]

Covering 75% of the world's speakers would be an achievement for a language that only has over 500 roots, considering that the 16 languages in Atlas have pretty much an equal participation in the vocabulary (I would not say that the language is represented if only two or three words were coming from that language).

As I said, it is difficult to establish a system in order to create the vocabulary, but as I said I think the one used by Atlas is fair. It could have been done in other ways but I don't think it would have been a better solution, just a different one.

In an of itself, there is nothing wrong with your manner of choosing vocabulary. As aokoye says, it's the claim of "representativeness" or "fairness" that causes the problem.

Why should one person be less equal than others, simply because they're one of a population of thousands instead of millions? Equality must look at everyone as an individual, not as a member of a group of a given size. Hell, as a Spaniard yourself, you surely understand that many Basque speakers would resent the idea that Romance vocabulary somehow represents them...?

I am very aware and concerned about minorities, especially language related. Not only Basque, but I am from Barcelona so I speak Catalan, a language that was actively pursued not so many years ago and banned. I included Catalan and Basque roots in Rodinian (I love Basque actually, it is an extremely interesting language for those of you who love grammar and language structures). Atlas is another story. As I said, I could have selected 6000 roots, one for each language (or 12.000, 2 each, etc), but it would not be fairer. M it would tick a box, but it would be useless and nobody would be able to identify a single root in the language. As a minority-language-speaker, I fully understand that an auxlang might only want to include some languages, because it makes things easier for learners, generally speaking. Another story is for instance Esperanto, where only European languages are really taken into account. That is another story and possibly one of the reasons why it "failed" (this can be discussed, but it certainly did not succeed as an auxlang).

Per cert, veig que parles Català, com és això?

[Rodiniye]

I am a bit useless with computers sometimes, I have downloaded an app for windows but still I cannot open the file. Any suggestions about what program should I use or how can I open it? explanation for dummies please!

Download the file which is "AtlasPopupDictionary.zip"

Unzip it. This produces one main file "AtlasPopUp.jar", plus a directory called "files". Inside that directory are the two text file dictionaries.

Assuming your computer is Java-enabled, just double-click on "AtlasPopUp.jar".

If your computer is not Java-enabled, you can download and install the Java Runtime Environment from here:

Download Java Runtime Environment

Have others been able to install and use the Atlas Popup Dictionary successfully?

Ez aiket i andar datorqihazu. Wenzu novi i al-duninvhissu, ta wi engezet al-kelimkitabu de di.
It worked in another computer. New post in the blog, and I have added your dictionary.

Thanks Tommen!

[zenmonkey]

Another story is for instance Esperanto, where only European languages are really taken into account. That is another story and possibly one of the reasons why it "failed" (this can be discussed, but it certainly did not succeed as an auxlang).

Which was the scope of Esperanto - it was never billed as a language that was inspired by all the languages of all the world. As to it having failed? What expectations do you have for your efforts - do you think that Atlas will somehow acquire 1000 native speakers? Do you intend to try and supplant Esperanto's position?

Today Esperanto is spoken in over 100 countries, Zamenhof was well aware that it may take much time, maybe even many centuries, to get a language to grow to the goals of 'universality'.

Heck, It even has it's own asteroid.

I'm not a fan (nor do I have plans to learn it) - but I wouldn't say it has failed.

[Xmmm]

I can't take Esperanto seriously, ever since I read Graham Greene's The Confidential Agent where it is parodied as Entrenationo.

However, it was embraced by anarchists and used in the Spanish Civil War. It has a poet nominated for a Nobel prize, a film starring William Shatner, a Duolingo course, and a magazine for the blind.

Esperanto is the bitcoin of conlangs.

[Rodiniye]

Which was the scope of Esperanto - it was never billed as a language that was inspired by all the languages of all the world. As to it having failed? What expectations do you have for your efforts - do you think that Atlas will somehow acquire 1000 native speakers? Do you intend to try and supplant Esperanto's position?

Today Esperanto is spoken in over 100 countries, Zamenhof was well aware that it may take much time, maybe even many centuries, to get a language to grow to the goals of 'universality'.

Heck, It even has it's own asteroid.

I'm not a fan (nor do I have plans to learn it) - but I wouldn't say it has failed.

I can't take Esperanto seriously, ever since I read Graham Greene's The Confidential Agent where it is parodied as Entrenationo.

However, it was embraced by anarchists and used in the Spanish Civil War. It has a poet nominated for a Nobel prize, a film starring William Shatner, a Duolingo course, and a magazine for the blind.

Esperanto is the bitcoin of conlangs.

Guys I wrote "failed" and specified that maybe it did not failed. I have a huge admiration for Esperanto, but it had many things that were far from perfect and it has not become the international language, that's all. That being said, it is by fair the most important conlang/auxlang and it is an inspiration for others, everyone is well aware of that. I live in a middle class neighborhood in BCN, and there is an Esperanto club meeting in a nearby train station, or at it least it was years ago. That is why I used the " with the word "failed", because it has been the most successfull attempt in history.

[zenmonkey]

Guys I wrote "failed" and specified that maybe it did not failed. I have a huge admiration for Esperanto, but it had many things that were far from perfect and it has not become the international language, that's all. That being said, it is by fair the most important conlang/auxlang and it is an inspiration for others, everyone is well aware of that. I live in a middle class neighborhood in BCN, and there is an Esperanto club meeting in a nearby train station, or at it least it was years ago. That is why I used the " with the word "failed", because it has been the most successfull attempt in history.

What expectations do you have for your efforts?
Do you hope to have 100 speakers? A million?

What is success for you?

[Rodiniye]

Guys I wrote "failed" and specified that maybe it did not failed. I have a huge admiration for Esperanto, but it had many things that were far from perfect and it has not become the international language, that's all. That being said, it is by fair the most important conlang/auxlang and it is an inspiration for others, everyone is well aware of that. I live in a middle class neighborhood in BCN, and there is an Esperanto club meeting in a nearby train station, or at it least it was years ago. That is why I used the " with the word "failed", because it has been the most successfull attempt in history.

What expectations do you have for your efforts?
Do you hope to have 100 speakers? A million?

What is success for you?

Success is to be satisfied with what you have done, and I am proud of the final result*. The rest... time will tell, or not.


*this is not as easy as people might think. Creating a whole conlang/auxlang/artlang and feeling really proud of it is not easy at all. I have done small projects in the past (many years ago) and once put together all of them did not look right. I sometimes find pieces of paper belonging to them and I laugh. So reaching the point where you have found what you wanted, and what had in mind is what you see in the text, is a really cool feeling. Rodinian had that, but people opened my eyes and the result is even better. So having that feeling about something that has been terribly difficult to make is already a success.

[tommus]

Although I am not learning a whole lot of Atlas at the moment, I am having a great time playing with it on the computer. Because it is so regular (allegedly no exceptions), a computer program can handle it quite easily.

Numbers:
If you look in the Atlas Grammar on page 7-8, you will see the structure of the numbers, and how they are modular. Example:

4053: vorkel-piatdas-san

So to generate a bilingual numbers dictionary, you just have to concentrate on the parts between the hyphens. Because of that, there are only about 140 words (the parts between the hyphens) needed to construct all the numbers from 0.00001 to 9 billion. So if you see "vorkel-piatdas-san", you or a popup dictionary just have to look at or double click each part between the hyphens to get four thousand-fifty-three.

Small words, such as prepositions, conjunctions, etc.:
I just had the computer build a small words dictionary from the small words that Rodiniye has in the Atlas Grammar. That produced 116 words.

Plurals, male, female, male plurals, female plurals:
According to the Grammar, only animal nouns are supposed to have male and female variants, by adding the interfix -at or -er. However, I took some poetic licence and generated male and female versions of all nouns (yes: animals, vegetables, concrete and abstract). A language like Atlas that lives on prefixes, suffixes, interfixes, etc. and prides itself on having no exceptions (and you can build the words you need), why restrict male and female to animals. For example, parts of flowers can be male and female. So can lots of other concrete and abstract things. You could have "female intuition" or a "male toilet", etc. So, in any case, I let the computer generate all those, plus the plurals of the neuter, male and female versions. That made a dictionary of another 1500 words. Now you may say that many (or most of those) are nonsensical, and you would be right. But if the purpose was to have an Atlas >>> Other Language popup dictionary, the nonsensical ones will never appear because they normally wouldn't be in the text you were translating. But they are waiting in case someone thinks some such combination is a perfectly good word.

To follow up a bit on the above paragraph, perhaps Rodiniye would consider some small modifications to his vocabulary. For example, the word "woma" is translated as "man, woman", but in a computer translation of the male version (womera), and the female version (womata), the following could appear:

woma >>> man, woman
womera >>> man, woman [male]
womata >>> man, woman [female]

I think it would be better to have the neuter version something like this:

woma >>> human, person
womera >>> human, person [male]
womata >>> human, person [female]

The same would go for similar pairs such as (daia = uncle, aunt),( baia = sister, brother), such as:

baia >>> sibling
baiera >>> sibling [male]
baiata >>> sibling [female]

There are probably those who would say that this amount of detail is perhaps too much for the language forum. But I think it might be useful to have some detail here to illustrate the power of a language with a small number of root words, lots of prefixes and suffixes, absolutely regular rules, etc in this age of computers and the Internet.

So with those additional words, the internal dictionary for the pop-up dictionary gets expanded to about 2,200 words in English (2100 in Spanish: I need help with the Spanish translation of the hundred or so "little words"). So here is a new version of the Atlas popup dictionary. It displays all 2,200 words, even the nonsensical ones.

Atlas Popup Dictionary Version 3.0

[tommus]

Having praised Atlas for its consistency and regularity, and the huge contribution that is to computer programming related to Atlas, I now want to point out some challenges. The challenges may be very hard to address. However, the payoff could be huge. For example, in this age of the Internet where language learners and language users rely heavily on machine translation (such as Google Translate and Bing Translate), I think it is "essential", let me repeat "essential", for Google, Bing etc to pick up Atlas as a language they support. Google has done so with Esperanto and I'll bet that has been a huge advantage. So what I am suggesting is that the more regular Atlas can be, the better it will be (by far). One set of rules. No exceptions. And very, very easy to parse, to recognize nouns, verbs, adverbs, etc. When you can do that, a computer programmer loves it. And Atlas needs Google to love it.

But:

The good news is that all animals words end in -a, vegetables in -o, concrete nouns in -u, abstract nouns in -e. So far so good.

More good news: verbs in the past tense end in -t, in the present in -s, and in the future in -z.

The bad news starts with non-complete verbs ending in -a, complete verbs in -e, and habitual verbs in -i (so far -i is OK).

So you see the problem. Animals and non-complete verbs both end in -a. Abstract nouns and complete verbs both end in -e. So a computer (or a person) now cannot always know whether a word is a noun or a verb. There must now be additional "processing", more memorization of vocabulary and more learning of rules, more considering the words in context, etc. In other words, much more complexity.

And that is just nouns and verbs.

At the moment, the small words (prepositions, conjunctions, etc. are all over the clock in terms of what they end in. Wouldn't it be nice if all prepositions ended in the same letter. Likewise for conjunctions, etc. Perhaps these small words could be handled without too much difficulty because they are small and few in number. But they add complexity whereas they could have added simplicity.

Atlas could really excel in simplicity if all different categories of words could have a unique ending.

[Rodiniye]

Having praised Atlas for its consistency and regularity, and the huge contribution that is to computer programming related to Atlas, I now want to point out some challenges. The challenges may be very hard to address. However, the payoff could be huge. For example, in this age of the Internet where language learners and language users rely heavily on machine translation (such as Google Translate and Bing Translate), I think it is "essential", let me repeat "essential", for Google, Bing etc to pick up Atlas as a language they support. Google has done so with Esperanto and I'll bet that has been a huge advantage. So what I am suggesting is that the more regular Atlas can be, the better it will be (by far). One set of rules. No exceptions. And very, very easy to parse, to recognize nouns, verbs, adverbs, etc. When you can do that, a computer programmer loves it. And Atlas needs Google to love it.

But:

The good news is that all animals words end in -a, vegetables in -o, concrete nouns in -u, abstract nouns in -e. So far so good.

More good news: verbs in the past tense end in -t, in the present in -s, and in the future in -z.

The bad news starts with non-complete verbs ending in -a, complete verbs in -e, and habitual verbs in -i (so far -i is OK).

So you see the problem. Animals and non-complete verbs both end in -a. Abstract nouns and complete verbs both end in -e. So a computer (or a person) now cannot always know whether a word is a noun or a verb. There must now be additional "processing", more memorization of vocabulary and more learning of rules, more considering the words in context, etc. In other words, much more complexity.

And that is just nouns and verbs.

At the moment, the small words (prepositions, conjunctions, etc. are all over the clock in terms of what they end in. Wouldn't it be nice if all prepositions ended in the same letter. Likewise for conjunctions, etc. Perhaps these small words could be handled without too much difficulty because they are small and few in number. But they add complexity whereas they could have added simplicity.

Atlas could really excel in simplicity if all different categories of words could have a unique ending.

Tommen maybe it was not clear in the grammar book, but ALL verbs end in -t, -s, -z.

Aspect is an interfix. So you have:

Wi denk-E-t (I think-FINISHED-Past) => I thought.
Wi vis-A-t (I see-NON-FINISHED-Past) => I was seeing.
Wi stud-I-t (I study-HABITUAL-Past) => I studied (I used to study).

So you have, for nouns, adjectives, verbs and adverbs:
- nouns end in -a,-o,-u,-e. (plus -n for plural).
- adjectives end in -i.
- adverbs end in -em.
- verbs end in -t, -s, -z.

You even have determinant correlatives (see grammar) ending in -ar, or connectors ending in -ex (for example, a connector saying "first of all" would be "ekex" (EK-one, EX-connector).

Yes, prepositions and conjunctions have free endings, same as personal pronouns, but computers would be able to handle them well, because they are fixed (in Rodinian they were not) and have a very clear and specific meaning, which is one of the reasons as well prepositions do not have more than one meaning each.

Talking about prepositions, because all verb complements (apart from direct objects) take a preposition in Atlas, computers could analyze the sentence quickly and easily. For instance, in the English sentence (I went there last week), it is not clear (if you do not know English) that "last week" is a time complement. In Atlas: Wi veset i ce u letzi wike, it is clear that "there" is a complement of place ("i" preposition before) and "letzi wike" is a time complement. So it is easier to identify this, which makes Atlas easier to learn too.

Thanks!!