Toki Pona Forums

So lately I've been trying to find all the lexicon (classic, removed, proposed, and new words), usages/definitions, and writing systems of toki pona, and I realized. It's been a long time since jan Pije made his lessons, and the language has grown since then. Words have acquired new usages. New words have been coined. New alphabets have been made, with several real writing systems being adapted as well. So what would kulupu pi toki pona think about all of the scattered bits of information out there (around the forums, wiki, wikibooks, wikipedia, jan Mato's sites, jan Kipo's site, and whatever other sources I don't know about) being gathered up and used to update jan Pije's lessons? Then there'd be a single, comprehensive resource instead of having everything scattered all around the internet.

I will be the loudest to applaud the first to put the bell on the cat.

mi kalama mute kin tawa jan ni: jan li open pali.

I plan to work on it between Christmas and New Year.

I plan to set up a Wiki on wikia.com, and to copy the currrent status of BJ Knight's book into it. This will be a lot of work because I will have to merge my revised German edition with the original edition. Besides, I have to convert the format to Wiki pages.

I will write a program which converts the Wiki pages to one big XML file, which in turn will be converted to LaTeX, HTML, or whatever. (For me, this is the fun part because I'm a programming amateur.)

The idea is to convert the work on the book into a collaborative project. We can use the discussion pages on Wikia to solve open issues in the book. Regularly, a PDF and a webpage are updated.

Possibly we can publish a mature version eventually (which can be considered stable for at least a couple of years).

Why not making new lessons ?

Kuti wrote:Why not making new lessons ?

Generally, it is better to take existing material as a starting point. Unless of course if it is totally rubbish, which is not the case here.

But Knight's book must be heavily revised in my opinion.

Wikia is a riddle wrapped in a mystery wrapped in an inappropriate localization to toki pona. I consider myself among the top 50 of toki pona reader/writers and I can't make sense of how to do anything on wikia.

There is a wiki books area for toki pona-- it has avoided deletion so far. Wiki books allows for long format printing, e.g. suitable for sending as a huge doc to a printer or an ebook. Wikia is optimized for a big set of heterogeneous hyper-linked docs that would take a million years for a reader to organize and print out.

http://en.wikibooks.org/wiki/Toki_Pona

The 1st author was planning to just copy the jan Pije lessons. I was going to reboot it and do a conlang documentation project in the style of Thomas Payne's "Describing Morphosyntax" outline, which is suitable for a field linguist to describe a language. Unfortunately, I stalled out on that project- laziness and too many questions that just lack an authoritative answer, such as how best to describe toki pona's valence increasing or decreasing operations (i.e. when can a semantically in/transitive verb add/drop an "e"-phrase) A potential improver of the jan Pije lessons will have the same challenge. jan Kipo addresses the challenge by sometimes just making decisions based on intuition and community experience. I am still a bit reluctant to do that because I don't want to take over the design of the language (let the community work out the solutions and express it in the corpus) and if I am going to do that much designing, I will design a brand new language.

There is also a russian wiki books that is just the translated PJ lessons in wiki books format-- so the idea of a wikibook version of the jan Pije lessons are a I don't have the link to to.

Yes, Wikibook may be even better-suited. However, I insist on being able to read the wiki pages automatically and convert them to XML. Therefore, the wiki pages must stick to a couple of rules. It may be impossible to enforce this on Wikibooks. The reason is that I don't want the book being locked up in Wikibooks. However good the layout of Wikibook's PDFs may be -- it is an automated conversion with all its flaws. Besides, it should be possible to have the book nicely embedded into other websites.

BTW, our hometowns are sister cities!

I think they are all media wiki applications or branches thereof. I suspect the technical limitations with respect to export to XML would be similar to all (wikia, wikipedia, wikibooks).

The key to successful conlang projects is to keep the project size to something that can be done in a few evenings. No point in accidentally starting an oxford english dictionary project, which lasted longer than the lifes of all the project initiators.

If the collected documents can be turned into one long printable document, that is a realistic goal. Making it exportable to an open format, Latex, etc. is just going to reduce the odds of completion. [Unless one is doing this specifically to learn about creating books in a variety of formats! I've done lots of projects with toki pona where I was really just trying to learn a bit about a new technology, like jquery or the like]

janMato wrote:I think they are all media wiki applications or branches thereof. I suspect the technical limitations with respect to export to XML would be similar to all (wikia, wikipedia, wikibooks).

It's not about technical limitations but policy. On Wikia, I would be admin (not the only one of course) and could enforce the additional rules. On Wikibooks, everyone could convert the pages into a format that cannot be processed by my tools anymore.

janMato wrote:The key to successful conlang projects is to keep the project size to something that can be done in a few evenings. No point in accidentally starting an oxford english dictionary project, which lasted longer than the lifes of all the project initiators.

My plan is to merely create a revised edition of Knight's book. I don't think that we need to add further chapters, although this would be to be discussed.

janMato wrote:If the collected documents can be turned into one long printable document, that is a realistic goal. Making it exportable to an open format, Latex, etc. is just going to reduce the odds of completion. [...]

The tools for processing the raw form of the book into other formats is the simplest part of the plan. I have a lot of experience in this field.

Ok, like I said, I applaud who ever puts the bell on the cat. I'm too lazy to follow through with even a fraction of the tp projects I'd like to do or participate in.

What deficits do you have in mind regarding the current lesson set?

Some things that come to mind
- The "li pi" construction is just bad. I still think it is a jan Pije innovation.
- Community proposals have been created for time, calendars, number systems, spelling, writing/scripts, and so on. I suspect jan Pije would say, we don't need no stinking calendars, number systems, etc-- but the rapid rate of their creation by the community implies otherwise.
- There are 20-30 odd apocryphal base words that the jan Pije lessons don't address, although there is the one "old words" page on his site
- set phrases (or what ever they are called today) are not really addressed, and this would be a lot of work
- the lessons have only the minimal glossary. A full word list would be a fairly large document that would look like my "improved word list" or jan Kipos dictionary.
- As a pedagogical document, it lacks progressively longer documents, ie. a reader section. On my website I did make a first stab at a readability score for toki pona text and ranked the scores for all the documents in my accumulated corpus. The metric has flaws (esp with respect to scoring poetry) but is faster than manually reading/counting words.