Wolof/Final project

From LING073
Revision as of 00:18, 11 May 2023 by Rlim2 (talk | contribs) (→‎Evaluation)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Keyboard suite

For the final project, three downloadable keyboard suites were created: one for macOS, one for an apple handheld device, and one for an android handheld device. It is to my knowledge that both the keyboard for apple and android devices also work on tablets.

These keyboard suites were created in collaboration with an L1 Wolof community consultant, and specific changes were made to suit her preferences. While not indicative of the entire scope of Wolof users, the keyboard suites were made with an actual user in mind, and lend to the state of translanguaging between French, English, and Wolof that many speakers employ.

The iPhone keyboard suite was adjusted from the English QWERTY layout. Since Wolof contains 5 different types of diacritics, and Wolof speakers very often code switch between French and Wolof (especially considering the state of French loanwords), all vowel keys have been adapted to include swipes with each diacritic: â, ã, à, á, ä. Although not all vowels have all variations of each diacritic, because of the aforementioned code-switching, the larger range of swipes allows for French to be easily typed as well. The Çç swipe was added to the Cc key for the same reason. Given that the loanwords in Senegal and Gambia differ based on colonial history (French and English respectively), this keyboard allows for English, French, and Wolof to be employed simultaneously.

The Senegalese Franc, Euro, and Gambian Dalasi symbol were added in the tertiary numeral page to reflect the language's user population

The android keyboard followed the same layout.

The development of the macOS keyboard relied on our previous project which involved deadkeys for the â, à, á, ä. Because ã is actually the only letter in Wolof to have an umalat, a deadkey was not created for it. This key was specifically requested to be kept by our language consultant, who would previously have had to switch to a Spanish keyboard for the letter, or copy paste the symbol from the internet. The addition of the velar nasal "ŋ" was consistent across all keyboard suites, and determined to be it's own independent key instead of a secondary or tertiary form of "n"


lowercase of the proposed iphone keyboard suite
uppercase of the proposed iphone keyboard suite
numerical page of the proposed iphone keyboard suite

Code

link to the github repositories: https://github.com/srheal/Wol_Keyboards

Evaluation

Keyboard effectiveness was evaluated by WPM and presence of pain in the hands and wrist (for macOS suite) and thumbs/fingers (for the mobile suites)

This method of evaluation is rudimentary and has many limitations, the biggest being that our language consultant is only one person, and therefore data from a single source cannot be called reliable. Secondly, WPM relies more on a user's familiarity with a keyboard layout than the actual ergonomic placement of keys, and since we do not have an extended period of time for our tester(s) to become used to the typing method, our most reliable form of evaluation will be the user's own feedback and critiques.

WPM of our language consultant on the macOS keyboard was 63WPM, which is quite impressive for someone using a completely unfamiliar keyboard entry that does not match the physical keyboard. The WPM on an Android and iPhone were quite similar, with 60 WPM and 57 WPM respectively, with a reported lack of joint and wrist pain after about 5 consecutive minutes of copying text from the Wolof Wiki (which includes French loan words) It may be important to note that our language consultant does work in an office setting daily, which could account for the typing speed.

The end products received altogether positive comments, with the acknowledgement of a lack of time to get used to the format. The greatest critique was the steps needed to access such a keyboard as opposed to built in systems for higher resourced languages such as French. Due to our limitations as mere students and not tried and true software developers, our ability to optimise our language consultant's experience ends here, but with hope that the project, and projects for many other under-resourced languages with such heavy populations, will be picked up by more experienced hands in the future.,