Hi Eden, thank you for your interest in the Apertium project and the Swahili-English pair!
We have added the 'Apertium African' project to the project ideas page to show special interest of Apertium in African languages, but most of the tips and suggestions given on the 'Adopt a language pair' project page [1] apply to you as well. The text of the proposal is important, of course, but what in my opinion is even more important, and what prospective mentors want to see is your ability to develop the language pair in question. In practice, this means passing the coding challenge as well as possible. The 'Apertium African' project's coding challenge doesn't mention that particular text, but for 'Adopting a language pair' applications the standard coding challenge since several years now has been to extend apertium-x-y (with lemmas, word paradigms, translation rules) so that it translates the given story [2] as much as possible. I suggest that you translate that English story into Swahili and use it as a test case for your translator, and start working on it soon enough so that prospective mentors can see your progress. Another good idea is to get an account on the wiki and draft your application there (and add links to your Github repos with apertium-swa-eng and apertium-swa). Ask mentors on IRC if you need an account on the wiki. There is an apertium-swa-eng repository [3] already set up, but it is missing many required pieces for it to compile successfully. The simplest option for right now is to bootstrap apertium-swa and apertium-swa-eng using [4][5], initialize Git repositories in these directories, and host them under your account on Github. pass the '--analyser=hfst' option to apertium-init.py when initializing apertium-swa. As for the English morphological transducer/tagger, you should clone apertium-eng and use that. As for writing the application, [6] is a must read. If you have any further questions, talk to us on IRC! Best, Ilnar Selimcan [1] http://wiki.apertium.org/wiki/Ideas_for_Google_Summer_of_Code/Adopt_a_language_pair [2] https://sourceforge.net/p/apertium/svn/HEAD/tree/branches/xupaixkar/rasskaz/ [3] https://github.com/apertium/apertium-swa-eng [4] http://wiki.apertium.org/wiki/Apertium-init [5] http://wiki.apertium.org/wiki/How_to_bootstrap_a_new_pair#With_no_existing_monolingual_packages [6] http://wiki.apertium.org/wiki/Top_tips_for_GSOC_applications On 2/28/19 12:53 PM, Eden Grace wrote: > Hi, > I will simply copy the email I sent to Professor Forcada: > > My name is Eden Grace and I'm a first year Computer science student at > the university of Alberta in Canada. > I'm interested in creating a usable version of the English-Swahili pair > andI would like to request your guidance on what a good proposal would > look like. I believe I already meet all the requirements( GNU/Linux > advanced user, bash, XML editing, python), except maybe for git(but I > can easily pick it up), and since I'm originally from Africa, my > knowledge of Swahili is advanced. My goal is to hopefully be a long time > contributor by creating and integrating as many African languages as > possible. > For the time being, I am mining online dictionaries of Swahili and > English to create a bilingual dictionary and submit a pull-request in > week or two. > > Thank you > Sincerely, > Eden. > > > _______________________________________________ > Apertium-stuff mailing list > Apertium-stuff@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/apertium-stuff > -- GPG: 0xF3ED6A19 _______________________________________________ Apertium-stuff mailing list Apertium-stuff@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/apertium-stuff
