thexxos/ithkapp

redirect

(The)XXOs

I am a second year student at the University of Adelaide in South Australia, with an interest in videogames, coding, geohashing, and conlangs.

...and I have a tendency to not finish projects that I start lol

ithkapp's Issues

Case not showing on adjuncts when switching in from verbal formative

See title

Possible vision for the future direction of Ithkapp

I see a lot of potential with Ithkapp.
I see Ithkapp as a IDE for writing ithkuil. If fact, making that the slogan may help communicate what ithkapp is to new people: "Ithkapp, the Ithkuil IDE!"

WRITING IN ITHKUIL:
I'm currently seeing learning how to write in ithkuil as going through the following learning levels:

1. Being able to write by having to go through the description for each option for each slot, in order to determine what you want
2. Being able to write by going through the description for each option for each slot, looking for the option you want (and already have in mind).
3. Being able to write by going through just the names for each option for each slot, looking for the option you want (and already have in mind). (examples: "objective" or "communicative")
4. Being able to write by going through just the abbreviations for each option for each slot, looking for the option you want (and already have in mind). (Example: "OBJ")
5. Understanding the conversion of the concepts of each option into the Latin ithkuil spelling of the word.

Ithkapp does a great job of providing a interface that helps people write in ithkuil no matter which of the above levels they are at. (And actually makes learning the last level (I.E. understanding the Latin ithkuil spelling) unnecessary for writing. (At least until one is so fast at writing ithkuil that they would be faster with a text editor then using ithkapp (I.E. currently no one :))))

READING ITHKUIL:
However, the ithkapp interface could also be useful for reading ithkuil at all the various above listed levels as well.

Ithkapp already has a neat "import from JSON" option

I did a loopback test of taking a ithkuil sentence from ithkuil.net, translating it into abbreviations/names (via the parsing software at https://github.com/ngoriyasjil/IthkuilGloss/ ), then selected those options in ithkapp to try to recreate the original sentence (up to shortcutting differences), then finally saved the files as json.
These JSON files are example sentences that people can read, and whenever needed the reader can easily look up the definition of things like "OBJ"/objective.

If a optional field for a "secret" was added to the JSON format, those files could be turned into a puzzle/learning game, where the user would load a sentence (written by someone else) into ithkapp via JSON, and by looking at the roots/stems, and options chosen (and reading the associated explanation for each), the user tries to figure out what the ithkuil sentence means. Once the user thinks they know what it says, they can reveal the secret which the author had put in the file that would tell the reader what the english translation actually is.

Using the above "looback test" procedure of the parsing translator into abbreviations/names of options, selecting those options in ithkapp, then trying to match the original ithkuil sentence (up to shortcutting differences), seems to work for almost all words. There are a couple words that I haven't been able to get to some things that don't seem to perfectly match up though. They could be a problem with me (the most likely cause), the parsing translator, with ithkapp, and at least one of them is caused by a typo in ithkuil.net

I'll try to add the JSON files here, and you are welcome to add them to the ithkapp repo as examples if you would like, (but you may want to fix the couple that don't match perfectly. And note that I have spoiled the "secret" by putting the english translation right in the filenames :)

Also, if instead of JSON, if ithkapp also had the ability to load Latin ithkuil sentences, that would be awesome. However loading Latin ithkuil sentences would require writing a ithkuil parser... so assuming that's out of reach, maybe ithkapp being able to load the output of the parsing translator listed above (put link here again) would be a possibility?

Being able to load ithkuil sentences via a reasonably direct method like that would also be useful for writing. A user could find a sentence similar to what they want to say, load that sentence into ithkapp, and then start modifying it.

NOTES ON IthkuilGloss:
If you are interested in seeing IthkuilGloss, the readme says partly how to install it, but is incomplete. Here are some steps I did to get it working that were not in the readme:
after doing the instructions in the readme (including changing it from "discordbot" configuration to "CLI" configuration) and installing java and maven, I did:

• mvn compile
• mvn package

Then created a 1 line script called "ithkuil-gloss-word.sh" to run it:
echo "?full $1" | java -jar ./target/ithkuilgloss-1.1.0-1.0.0-jar-with-dependencies.jar

• Then did: ./ithkuil-gloss-word.sh "eutruloewû"

BUILDING COMMUNITY:
Ithkapp is useful
I don't think enough people know about Ithkapp and think you need to promote it.
I recommend making:

• a full announcement on reddit
• a announcement on discord
• a link on ithkuil.place
• and you should probably add a link to the ithkuil Wikipedia page, as well as the ithkuil wikiversity book too!

Feature Request - Show roots/stems after you type them in

I propose that after a user types in a root, that Ithkapp look up the root in https://docs.google.com/spreadsheets/d/1JdaG1PaSQJRE2LpILvdzthbzz1k_a0VT86XSXouwGy8/edit?usp=sharing and display the root underneath.

This would help a person to be sure they typed in the correct root

w/y phonotactics

Formatives and modular adjuncts with Aspect+FAC/CCN and dual referential adjuncts all allow the option of using -y- in place of -w-.
While I couldn't find anything about this in the documents, I should probably add a check for what the preceding (and maybe following) vowel is and decide on -w- or -y- based on what it is (to avoid -uw- and -iy-).

Proposal: Hover over text

Currently, when the mouse over the chunks (chunks like "OBJ") of the currnet word that is written at the bottom of the screen, it displays the "things that are assumed to be there and not required to say". This is good info.
I'm proposing though that when you hover over a part like "OBJ", that it show you the definition of "OBJ" in the hover text (same definition as in the entry "big version" of the selection form).

I believe that this would allow people to grasp what a entered word means, quite rapidly.

Also, while this could be useful on it's own, this will be required for my next idea which I'll be putting in a separate report.

:)

Case accessor affixes and 4th-series vowels are unavailable for affix-as-root words

Identified by @ryanlo713 on Discord and GitHub

root considered invalid

Thanks for writing this

some roots (like roots starting with q) seem to be considered invalid by the system.

My guess is that this is caused by changes in the new version of ithkuil that was released earlier this year.

Write a better README

This is something that I'm fairly certain that I should do before making this repository public, because at the moment it's not very in-depth.

Settings + External Juncture

Settings and External Juncture no longer work.

Disabled Modals

Clicking on an unselected option in a modal where the category that the modal is related to is disabled will change the dropdown but not select it in the modal until the category is reenabled.

show glosses for case accesor and inverse case accesor/stacking affixes

title. Also while i'm here, do you use the case accessor afixes or normal case slot for derivational meanings?

New Ithkuil v1.0

As the final version of Ithkuil IV has been released, I will need to update my system to reflect that.
Here are the updates:

• -äi- and -öi- to -üo- and -üö- respectively
• Standalone C_A form -řļ- to -ň-
• Combination referential Slot 3 forms -x-, -xx-, -lx-, -rx- to -x-, -xt-, -xp-, -xx- respectively
• Slot 8 Aspect consonant forms -w/y-, -hw-, -hlw-, -hly-, -hnw-, and -hny- to -w/y-, -hw-, -hrw-, -hmw-, -hnw-, -hňw- respectively
• Complete rework of Illocution, Expectation, and Validation (as Expectation has been removed)

Empty dropdowns

When a dropdown is undisabled or unhidden without its contents being modified by the user first, then the dropdown will lose its contents (but behave functionally as having the default option).

PROPOSAL: Allow exporting to URLs

Currently there is a download button for a json file.

I'm proposing a button that would create a link. Something like:

<a href=http://ithkapp.github.io/{something}> {ithkuil sentence you wrote} </a>

So you could send it to someone, they would see the ithuil sentence, but if they click on it, the come to the ithkap site.

Now, I'm going to propose one way that the idea might work, but i'm proposing the concept, not the implementation :)

What if we minified, then compressed the json file, then uuencoded it, or some other way to turn it into a URL safe string, and

<a href=http://ithkapp.github.io/?sentence={uuencoded,compressed,minified json file of the sentence}> {ithkuil sentence you wrote} </a>

So now you could use ithkapp to write a ithkuil sentence, then send the link to someone, and if they couldn't read it, they could click on each word they didn't know, and quickly browse the meaning of each option for the word (reference to issue #21).

I doubt it would be considered polite to send all your messages out to everyone that way, but I'm thinking it should be workable for anyone with a passing interest in ithkuil as a recipient. (which gives more practice, which builds more interest in ithkuil, etc)

:)

Note: Possible implementation snag: I do not know what the URL string size limit is on modern webservers.

External Juncture Calculations

Section 1.5 of the official Morpho-Phonology document:

1.5 External Juncture

When a word ending in a consonant-form (i.e., either a single consonant or a multiple consonant conjunct) is followed in the same breath-group by another word beginning with a consonant-form, it is usually necessary to append a vowel either to the end of the first word or the beginning of the second word, so as to avoid confusion as to which word the word-final and/or word-initial consonants belong to. This is accomplished by ensuring that appropriate word-initial and/or word-final vocalic Slots (e.g., Slot II, Slot IX) are filled.

See new Section 5.8 regarding how to indicate juncture between sentences.

This is not currently supported.

Better Ca and case stacking options

Make it more obvious as to how case-stacking and C_A-stacking via affixes works by having the affix option boxes change their contents when a case-stacking affix or the C_A-stacking option is added/selected

Minor feature - "done checkboxes"

A possible minor feature would be to have a checkbox on each answer. (I.E. one for plexity, one for similarity, one for connectedness, etc.) The checkbox would merely be a way to let your future self know that you have gone through all the options for this answer and think you have gotten it right.

I.E. anything without a checked checkbox would still be on your "todo" list while creating the word.

Combination Referentials

In Ithkuil IV, it is possible for any referential slot to have more than one actual referential.
That is not currently possible in Ithkapp.

(Potentially repurpose affix OptionBox code?)

Edit: this has been implemented, check first comment to see what I still need to do

Slot VII shortcuts not showing when shortcutting is enabled but no shortcutting is applied

If there's a situation where the shortcutting is set to "Yes, if possible" but shortcutting is inapplicable, if there are any Slot VII shortcuts to be applied they won't but they still will be removed from the slot, resulting in a word that just doesn't have anything indicating that the shortcut exists.

T3 Affix interpreted as Referential Shortcut when with T4 Affix(es) together in the same Slot...?

Slot IX glottal stop alternate locations

Special Note from Section 3.9.1 of the Morpho-Phonology document:

SPECIAL NOTE: V_C values for cases 37 through 68 are the same as for Cases 1 through 36 (minus vowel-tier No. 8) with the addition of a glottal-stop; note that a glottal-stop cannot appear in word-final position (use the rules of Sec. 1.7. when adding a glottal-stop to V_C). To shorten V_C for cases 37 through 52, the glottal-stop may instead be added to any other vocalic form in the formative after Slot II (i.e., to V_R in Slot IV, V_X in Slots V or VII, or V_N in Slot VIII) as long as the formative does not have a Slot IV/VI a+C_A shortcut per Sec. 3.2, and as long as the Sec. 3.8.1.2 rule on moving the C_N Mood/Case-Scope affix to Slot VI has not been applied.

This is not currently supported.

Sentence menu is very broken

When attempting to change words in the sentence menu, it will often replace the word you want to change to with either an empty word or the word that was just selected - I may need to rewrite the word switching code.