Closed Apr 5, 2014 Last updated 8 months ago

    A while ago we built the elicitation app and another app for lettin…

    A while ago we built the elicitation app and another app for letting users build their own vocabularies. This milestone is to clean up those projects and repackage them as a user-driven language learning app where the users get to build their own corpora for language learning using anything they would like for images audio and video and text. Due date: when mary-ann comes to montreal @gretchenmcc please edit the due date when you know when it is. "provide screencasts and tutorials explaining the usage of our tools, how they may apply to language-learners’ situations, and why they are particularly useful for learners of under- resourced languages, providing key information for languages that may have few other text and audio records to refer to." Screenshots of the first iteration are here: https://play.google.com/store/apps/details?id=ca.ilanguage.anomiaforme Screenshots of this iteration Tablet uses fragments side by side: Phone uses list: And a separate detail screen (here with speech recognition showing)

     
    88% complete
    Closed Oct 29, 2017 Last updated 10 months ago

    EMELD recommends that data be stored in a format that is easily acc…

    EMELD recommends that data be stored in a format that is easily accessible, i.e. which can be easily searched and read by humans as well as crawled by search engines. Make data visualization formats to appear in response to searches on user public corpus pages like https://www.lingsync.org/lingllama/lingllama-communitycorpus with views for both humans (interlinear gloss) and machines (json).

     
    100% complete
    Closed Oct 23, 2017 Last updated 10 months ago

    The search algorithms needed to create second language learning mat…

    The search algorithms needed to create second language learning materials would maximize the variety of grammatical constructions rather than simply returning a random result or the most frequent constructions. This maximization of variety would also be useful for searching publicly accessible web-pages and blogs for linguistic data. Doing a regular internet search for certain grammatical terms often returns uniform results without variety. Different ways of using the term are lost due to their low frequency. ... when searching for intransitive verbs in Mi’gmaq the results will return forms with first person, second person and third person subjects. It will vary whether the subject is animate or inanimate; plural, dual or singular. It will vary speakers (male and female) and sessions so that a language learner is exposed to a variety of pronunciations. The algorithms will exploit cluster analysis, forming equivalence classes of expressions that will be organized into neighbourhoods. Distance between the neighbourhoods will be measure in terms of how many morpheme-by-morpheme translations are shared within a single equivalence class as compared to those across equivalence classes. Two equivalence classes will be close if they share many morpheme-by-morpheme translations. The algorithm will yield an array of entries where each entry is a representative member of an equivalence class and the array of entries maximizes the distance between the represented equivalence classes. Such algorithms are particularly useful with languages that are highly agglutinative or polysynthetic, such as Mi'gmaq, in which inflectional patterns of words and morphemes depend on other grammatical factors such as animacy, number, person and evidentiality. Furthermore, most of these highly agglutinative languages lack textbooks and therefore are in desperate need of material for second language learners. This need is all the more urgent since many of these languages are in threat of extinction. The lesson creation application will take the result of its search algorithms and create lessons by playing audio files for the second language speaker, while also displaying text and providing appropriate translations. .. Tasks: Create lesson creation application (LCA) (~200 hrs); Evaluate LCA with language teachers and second language learners. Milestone: Functional lesson creation application with code and documentation published on GitHub; publish the evaluation result on the website Success criteria: Qualitative analysis based on teachers and students’ experiences in terms of the tool’s usability and helpfulness to their education. Risks: The application may not yield sample sentences are that are directly useful for language learning. To avoid this eventuality, the project will extensively beta-test in consultation with learners, teachers, and native speakers in order to produce the best product possible.

     
    100% complete
    Closed Mar 25, 2017 Last updated about 1 year ago

    Update corpus pages so they too are using wiki markup Re-do corpus…

    Update corpus pages so they too are using wiki markup Re-do corpus pages in angular so we don't need to use have to teach interns jade Add members to corpora Add lexicon browser to corpora

     
    100% complete
    Closed Mar 25, 2017 Last updated over 1 year ago

    *Allow search of a single field or a subset of available fields. *U…

    *Allow search of a single field or a subset of available fields. *User-friendly regex search; i.e. a menu that allows users to set regex values like “at beginning of word/phrase, preceded/followed by x, etc.” *Plug into new searchdev API.

     
    100% complete
    Closed Sep 2, 2015 Last updated over 2 years ago
     
    100% complete
    Closed Feb 2, 2016 Last updated over 2 years ago
     
    100% complete
    Closed Sep 2, 2015 Last updated almost 3 years ago

    use deprecated map reduces faq for instructors leave corpus button save button stays in menu

     
    100% complete
    Closed Sep 2, 2015 Last updated almost 3 years ago

    praat working on mac server again

     
    100% complete
    Closed Jul 31, 2015 Last updated about 3 years ago
     
    100% complete
    Closed Mar 2, 2015 Last updated over 3 years ago

    The server was giving 500 errors even now and then when larger corp…

    The server was giving 500 errors even now and then when larger corpora were being accessed by users. We are using a tiny cheep server, we more RAM, and we also need to retire the spreadsheet app because it doesnt use pagination and triggers a number of unnecesary database indexing.

     
    100% complete
    Closed Mar 20, 2015 Last updated over 3 years ago
     
    100% complete
    Closed Mar 2, 2015 Last updated over 3 years ago

    Removing more map reduces we arent using to reduce index size, upda…

    Removing more map reduces we arent using to reduce index size, updating all online protoype apps to the most recent code for the authentication refactor (so that android users can login, aparently some people are using it)

     
    100% complete
    Closed Mar 2, 2015 Last updated over 3 years ago
     
    100% complete
    Closed Mar 2, 2015 Last updated over 3 years ago
     
    100% complete
    Closed Mar 2, 2015 Last updated over 3 years ago

    Update collections view to backbone 1.1.2

     
    100% complete
    Closed Mar 2, 2015 Last updated over 3 years ago

    This is a major release rewrite of templates to use the corpus fields

    This is a major release rewrite of templates to use the corpus fields app is localized in english and french rewrite of audio upload and audio recording ability to edit more corpus info and more

     
    100% complete
    Closed Jul 18, 2014 Last updated over 3 years ago

    https://chrome.google.com/webstore/detail/jfmpeiicncingobdejgmmcamknndpbbi @jrwdunham @alvations the chrome extension for Cherokee, Inuktitut, Kartuli and Cree is done. It is only as good as the Wiktionary is good for that language. later we can add a function to let the user create the new wiktionary page for new words, and create new measures of quality for the existing words. And those working dictionaries can work towards improving the wiktionary together. Cherokee 194,426 entries http://chr.wiktionary.org/wiki/Special:AllPages (most are fake!) Inuktitut 251 entries http://iu.wiktionary.org/wiki/Special:Statistics Kartuli 7,363 entries http://ka.wiktionary.org/wiki/სპეციალური:სტატისტიკა Plains Cree (incubation) 0 entries http://incubator.wikimedia.org/wiki/Wt/cr

     
    100% complete
    Closed Sep 24, 2012 Last updated over 3 years ago

    Lots more instructions, new activity feeds and simple user interfac…

    Lots more instructions, new activity feeds and simple user interface bug fixes.

     
    100% complete
    Closed Oct 16, 2014 Last updated over 3 years ago
     
    100% complete
    Closed Oct 8, 2014 Last updated over 3 years ago

    Release v1.92 to all users Release lexicon map reduces & updated custom map reduces

     
    100% complete
    Closed Dec 26, 2014 Last updated over 3 years ago

    HTML5 audio/video recording (plus import of any sort of file to an document)

     
    100% complete
    Closed Nov 17, 2014 Last updated over 3 years ago

    Prototype uses the bower component for glosser, and more export options (xml, json)

     
    100% complete
    Closed Nov 17, 2014 Last updated almost 4 years ago
     
    100% complete
    Closed Jun 4, 2014 Last updated almost 4 years ago

    This was Xianli's summer project, to put an image into the UI so that users could transcribe old texts.

     
    100% complete
    Closed Oct 8, 2014 Last updated almost 4 years ago
     
    100% complete
    Closed Oct 8, 2014 Last updated almost 4 years ago
     
    100% complete
    Closed Oct 8, 2014 Last updated almost 4 years ago
     
    100% complete
    Closed Oct 4, 2014 Last updated almost 4 years ago

    Make widget that can embed on blogs (e.g. migmaq.org which uses wor…

    Make widget that can embed on blogs (e.g. migmaq.org which uses wordpress) and websites which displays a public, searchable, live-updating datalist of datums as selected by a user from their LingSync corpus.

     
    100% complete
    Closed Oct 4, 2014 Last updated almost 4 years ago

    Desiderata: -prioritizes user-created glosses --> overrides everything

    Desiderata: -prioritizes user-created glosses --> overrides everything -prioritizes longer morphemes: xxxx > xx-xx, morphemes at least 2 symbols long -prioritizes more frequent parses -context-sensitive -user decides how aggressive glosser is -generative lexicon -- navigate, clean, gloss -Hunspell - open source spellcheck -sensitive to who entered data (user, all users, web, date of gloss, outliers) -don't segment words beginning with capital because they're probably proper names -learns user preferences/history - timestamp /gloss /correct -bootstrapping - hook Tasks: Computer-assisted human annotation using automated morpheme segmentation tool, morpheme-gloss alignment checker, and morpheme-gloss consistency checker (~100 hrs); evaluate time cost and error reduction rate in annotation with vs. without the segmentation and checking tools; create auto-complete/correct tool (~150 hrs, continue to M11). Milestones: Publish the time cost and error reduction evaluation report on the website (M 10); Functional and accurate auto-complete/correct tool with code and documentation published on GitHub (M 11). Success criteria: Evaluation results in 20% time cost and 50% error rate reduction with the segmentation and checking tools compared to without the tools. To be tested by linguists and language consultants . Risks: Tools may over- or under-report glossing errors, resulting in an increased human workload due to compensating for tool error. This likelihood will be reduced by frequent communication between programming and annotating parties, so that potential misreporting gets fixed as soon as it occurs and human annotators avoid time-wasting.

     
    100% complete
    Closed Feb 17, 2014 Last updated almost 4 years ago
     
    100% complete
    Closed Sep 19, 2013 Last updated almost 4 years ago

    The first release of the spreadsheet app which was designed for quick spreadsheet like data entry.

     
    100% complete
    Closed Jan 18, 2013 Last updated about 4 years ago

    We are adding a module for creating free Android/iPhone/Chrome seco…

    We are adding a module for creating free Android/iPhone/Chrome second language learning apps based on fieldinguist's databases. This module will be able to use FieldDB to create new lessons and units, and display the exercises in fun new dashboards specially designed for second language learning. This module useful for researchers working with communities as a way to give their research back to the community. Our first step in this process was creating an Android Prototype for the the Migmaq team. The Module shall: make it easy to turn an existing corpus into an App encourage users to get their language consultants to decide if their data can be licensed under Creative Commons early in the process so that they dont need to re-record all their data if they want to turn it into an app for the community. add additional fields to corpus, and datum to make them usable as a gamified app which can detect user's progress and encourage learners to use the app. add audio processing such as slow down the audio stimuli (very handy for language learners) let the language users create their own vocab and stimuli (and share with peers or classmates if they choose)

     
    100% complete
    Closed May 8, 2014 Last updated about 4 years ago

    This is a release for small fixes found during the Spring 2014 semester and before the ETI workshop to reduce registration bugs.

     
    100% complete
    Closed Aug 25, 2013 Last updated about 4 years ago

    Angular has better 2-way databinding. The app has outgrown backbone…

    Angular has better 2-way databinding. The app has outgrown backbone... we will take a look at switching to Angular if we have other projects which work well in Angular. The conclusion was: the spreadsheet app was written in angular the activity feeds were written in angular the activity heatmap was written using jquery the search widgets were written using jquery

     
    100% complete
    Closed Jul 18, 2014 Last updated about 4 years ago
     
    100% complete
    Closed Jan 20, 2014 Last updated about 4 years ago

    Often one of the products of a field work grant is a dictionary whi…

    Often one of the products of a field work grant is a dictionary which can be given back to the community. This milestone is to set up Wiktionary integration, to make it easier to publish a community driven and community editable living dictionary with the data collected in a corpus.

     
    100% complete
    Closed Jun 14, 2014 Last updated about 4 years ago

    With the batumi team we are making a kartuli speech recognizer. htt…

    With the batumi team we are making a kartuli speech recognizer. https://play.google.com/store/apps/details?id=com.github.opensourcefieldlinguistics.fielddb.speech.kartuli In this iteration we will use the learnx app to build a speech recognition training module for the kartuli app which will let the user train the recognizer to their voice and vocabulary.

     
    100% complete
    Closed Jun 4, 2014 Last updated about 4 years ago

    automatically correct errors, draw attention to inconsistencies, and

    automatically correct errors, draw attention to inconsistencies, and fill gaps in the data 6 steps which in practice work iteratively: (i) automated segmentation of unsegmented words into morphemes, (ii) alignment of morphemes and glosses, (iii) human annotation of a small set of data, (iv) extraction of morphemes to a lexicon server, (v) detecting inconsistent annotation and alerting the annotator(s), and (vi) assisting human annotation by automated completion/correction.

     
    100% complete
    Closed Jun 4, 2014 Last updated about 4 years ago
     
    100% complete
    Closed Jun 4, 2014 Last updated about 4 years ago
     
    100% complete
    Closed Jun 4, 2014 Last updated about 4 years ago
     
    100% complete
    Closed Jun 4, 2014 Last updated about 4 years ago
     
    100% complete
    Closed May 19, 2014 Last updated about 4 years ago
     
    100% complete
    Closed Dec 2, 2013 Last updated over 4 years ago
     
    100% complete
    Closed May 6, 2014 Last updated over 4 years ago

    Exploring the Interfaces 3: Prosody and Constituent Structure McLin…

    Exploring the Interfaces 3: Prosody and Constituent Structure McLing is pleased to announce that the call for the third Exploring the Interfaces workshop has just been posted! Here are the details: Exploring the Interfaces (ETI) 3 will take place at McGill University from May 8–10, 2014. This workshop will be the last of three workshops organized by the McGill Syntactic Interfaces Research Group (McSIRG) as part of a multi-year grant to study linguistic interfaces. Following ETI 1 (Word structure) and ETI 2 (Implicatures, alternatives and the semantics/pragmatics interface), the topic of ETI 3 will be Prosody and Constituent Structure. In particular, ETI 3 will deal with issues surrounding prosodic and phonological evidence for syntactic constituent structure, with a focus on verb-initial languages. Goals of the Workshop: To bring together researchers working on issues at the syntax-phonology interface (e.g. syntactic constituency, prosodic effects on word order) from the perspectives of syntax, prosody, and phonology/phonetics To bring together researchers working on a variety of different languages, with an emphasis on languages with default verb-initial word order To encourage communication and discussion about methodologies that can be used for the empirical study of prosody and the syntax-phonology interface Invited Speakers: Judith Aissen (UC Santa Cruz) Sasha Calhoun (Victoria University of Wellington) Lauren Eby Clemens (Harvard) Emily Elfner (McGill) Jim McCloskey (UC Santa Cruz) Norvin Richards (MIT) Joey Sabbagh (UT Arlington) Kristine Yu (University of Massachusetts Amherst) Methods Tutorials: In addition to the regular session, we will have two tutorials on local technological tools for fieldwork, with special reference to fieldwork on prosody: Tutorial 1: Automatic Acoustic Alignment in Underdocumented Languages Tutorial 2: LingSync: An Online Tool for Field Work This conference is supported through an FQRSC team grant on Linguistic Interfaces and funding from a SSHRC Grant on Prosody and Constituent Structure. Organizing Team: Emily Elfner, Jessica Coon, Lisa Travis, Michael Wagner Student Organizers: Michael Hamilton, Henrison Hsieh, Yuliya Manyakina Call for Papers: Abstract submission deadline: February 28, 2014 Notification of acceptance: March 10, 2014 Conference: May 8-10, 2014 In addition to eight invited speakers, we are accepting abstracts for a limited number of additional talks (30 minutes + 10 minutes discussion) and posters. We particularly welcome papers which address the following questions: What can prosodic and phonological evidence tell us about syntactic constituent structure? To what extent do syntactic, phonological and prosodic evidence agree with one another regarding constituent structure? What is the role of prosody in determining word order? Can prosodic and phonological evidence be used to help distinguish between competing syntactic accounts of how word order is derived? We welcome abstracts dealing with these topics in any language, but would particularly welcome abstracts on verb-initial languages in keeping with the theme of the conference. Abstracts should be anonymous and no longer than 500 words (including examples, but not counting title or references), and should be submitted in PDF format on the following easychair site: https://www.easychair.org/conferences/?conf=eti3 Please indicate on the form whether you would prefer an oral presentation, a poster presentation, or whether either would be acceptable. By default, we will first consider you for an oral presentation. Additionally, we hope to have some funding available to supplement travel costs for student presenters. Contact eti3.mcgill@gmail.com with questions.

     
    100% complete
    Closed Apr 13, 2014 Last updated over 4 years ago
     
    100% complete
    Closed Apr 6, 2014 Last updated over 4 years ago

    Testing app in the field (Batumi, Georgia)

     
    100% complete
    Closed Mar 12, 2014 Last updated over 4 years ago
     
    100% complete
    Closed Feb 8, 2014 Last updated over 4 years ago
     
    100% complete
    Closed Jan 17, 2014 Last updated over 4 years ago

    In this milestone (part of other releases in v1.90) we updated all …

    In this milestone (part of other releases in v1.90) we updated all the email contents to reflect the current app terminology and also added more verification and checking for when users registered usernames which wouldnt work well with web urls (each user has a share url, so its best if their user name works well in a url)

     
    100% complete
    Closed Feb 8, 2014 Last updated over 4 years ago
     
    100% complete
    Closed Feb 8, 2014 Last updated over 4 years ago

    The CURL call for papers discuss the use of linked data for low res…

    The CURL call for papers discuss the use of linked data for low resource languages, and how we can help groups of people collaborate on low resource languages by exporting linked data, and also converting existing resources to the process of producing linked data

     
    100% complete
    Closed Jan 28, 2014 Last updated over 4 years ago
     
    100% complete
    Closed Jan 17, 2014 Last updated over 4 years ago

    This release makes it possible for v1.36 users to use the spreadshe…

    This release makes it possible for v1.36 users to use the spreadsheet app, among other small bugs and improvements. This was a big release, that had to be tested over a period of 1 week (jan 10-jan16) Results: all users are on the same server, all users can use the spreadsheet app, all offline users can back up their data in preparation for an only only release...

     
    100% complete
    Closed Aug 25, 2013 Last updated over 4 years ago

    This was really busy milestone. Josh added more features to the cor…

    This was really busy milestone. Josh added more features to the corpus pages, Tobin added more features to the authentication service to make it possible for the spreadsheet app to create new corpora and register new users and louisa added auto generated latex trees.

     
    100% complete
    Closed Aug 25, 2013 Last updated over 4 years ago

    In this milestone we investigated Solr and Elastic search, and ulti…

    In this milestone we investigated Solr and Elastic search, and ultimately choose elastics search which was able to support the flexible data schema of the corpora. Josh set up an elastic search instance and also some client side scripts which use it from the corpora's pages. The search indexer uses the public user to authenticate, so only data in public corpora (basically just lingllama) are discoverable.

     
    100% complete
    Closed Jan 26, 2013 Last updated over 4 years ago

    In this milestone we made it possible for the online app and the an…

    In this milestone we made it possible for the online app and the android app to let users login to the public database and show sample data. As well as refactored the connection between backbone and couchdb to cut pouchdb out.

     
    100% complete
    Closed Jan 17, 2013 Last updated over 4 years ago

    In this release we updated the mcgill users to use the testing server because the mcgill server still wasnt able to support CORS.

     
    100% complete
    Closed Sep 19, 2013 Last updated over 4 years ago

    A whole bunch of stuff.

     
    100% complete
    Closed Sep 16, 2013 Last updated over 4 years ago

    This release contained numerous user experience fixes for the proto…

    This release contained numerous user experience fixes for the prototype search remembers the term transparent datalists if app is transparent (so user can have a dark app) corpus terms using wiki markup (yay!) comments using wiki markup (yay!) a transliteration bot for inuktitut updates in the enteredByUser and modifiedByUser fields for both apps

     
    100% complete
    Closed Jan 14, 2014 Last updated over 4 years ago

    This release was a hotfix for v1.82 where we removed the route so t…

    This release was a hotfix for v1.82 where we removed the route so that if you were at the register/login screen there would be a clean url. Unfortunatly that same route was used to list the copora so we put the route back, but called it corpora_list instead of spreadsheet_main (in preparation for the spreadsheet to become the main app)

     
    100% complete
    Closed Sep 6, 2013 Last updated over 4 years ago

    In this release we added a 5 minute timeout on recording (one of th…

    In this release we added a 5 minute timeout on recording (one of the testers said she clicked on record and walked away with out realizing so it recorded 10 minutes of audio and uploaded it.) and we also added onbeforeunload to help users from loosing data and forgetting to push save.

     
    100% complete
    Closed Nov 1, 2012 Last updated over 4 years ago

    In this milestone we created a read only dashboard, to be used potentially for the public urls. We also did some server maintenance.

     
    100% complete
    Closed Jan 19, 2014 Last updated over 4 years ago

    This release contained numerous small user experience fixes which c…

    This release contained numerous small user experience fixes which came from jonathan and testing the interaction of the spreadsheet and prototype apps to ensure that both were producing consistant data which could be read by the other.

     
    100% complete
    Closed Jan 17, 2014 Last updated over 4 years ago

    In this milestone we moved all the stable users to the testing server. This was the first step towards updating the stable users so they could use the spreadsheet app.

     
    100% complete
    Closed Sep 16, 2013 Last updated over 4 years ago

    In this milestone we found some bugs in the new auth service which …

    In this milestone we found some bugs in the new auth service which was updated to much of the corpus and user creation client side (previously the app was offline, so the corpus creation had to be done client side, and then only executed server side when the user synced their preferences). We also added Swagger to the auth api and started a refactor of the auth api to make it much more modular and robust. The refactor was never finished, but did result in a branch of the auth service which can install itself (handy for setting up new dev servers).

     
    100% complete
    Closed Mar 4, 2013 Last updated over 4 years ago

    In this milestone we added CORS support to the auth and corpus webs…

    In this milestone we added CORS support to the auth and corpus webservices to ensure that the new Chrome App manifest 2 (which no longer had whitelisting for CORS) wouldnt break the app. It was exciting...

     
    100% complete
    Closed Jul 2, 2013 Last updated over 4 years ago

    In this release we added gravatars to the users who could be added …

    In this release we added gravatars to the users who could be added to a corpus, this way users can see faster if they are potentially adding a user they dont know to their corpus.

     
    100% complete
    Closed Jan 20, 2014 Last updated over 4 years ago

    In this milestone we added a button on the share pages where the vi…

    In this milestone we added a button on the share pages where the visitor could open the corpus using the spreadsheet app, because we had added gravatars with links to the share pages we wanted to make sure the user could get back to the app.

     
    100% complete
    Closed Sep 6, 2013 Last updated over 4 years ago

    In this milestone we created a small web spider using Node which he…

    In this milestone we created a small web spider using Node which helped us import bible translations. Henderson 2013 discussed some of the benefits of working with bible text many of the bible translators really focus on writing a text which will 'speak' to the community members, which means the text is not stilted or formal but often closer to natural language there are hundreds of translations, making it often the only large corpus source for many dialects The same spider of course can be used on other resources to help supplement field corpora with additional contexts.

     
    100% complete
    Closed Sep 16, 2013 Last updated over 4 years ago

    In this milestone we created a sample cleaning bot to show how user…

    In this milestone we created a sample cleaning bot to show how users could semi-automatically clean their data, so they could worry less about data validation and data quality while letting multiple team members enter data.

     
    100% complete
    Closed Sep 6, 2013 Last updated over 4 years ago

    In this milestone we added a new heat map to corpora pages to help …

    In this milestone we added a new heat map to corpora pages to help users gain interest in their corpora, and/or demonstrate activity to members outside their corpus team (ie community or other interested parties who don't have access to the data itself).

     
    100% complete
    Closed Dec 26, 2012 Last updated over 4 years ago

    This was the last stable release for the offline chrome users (and …

    This was the last stable release for the offline chrome users (and was used for the period of Dec 2012 to July 2013). In this release we added the new illustrated guide to the dashboard and other fixes which made this the final stable offline release. This release was superceded in Jan 2014 by Spreadsheet Release 1.90

     
    100% complete
    Closed Jan 28, 2013 Last updated over 4 years ago

    We always wanted the tags to look a lot more like popular tag patte…

    We always wanted the tags to look a lot more like popular tag patterns like those on Pinterest and YouTube, but the way we implemented them the first time was blocking us from doing this. In this milestone we turned tags into a normal datumField, and added an autocomplete based on tags in the corpus. While not awesome, it's one step closer to a better user experience (not to mention searchability, when the tags were not a datum field, they were not being indexed.).

     
    100% complete
    Closed Jul 31, 2012 Last updated over 4 years ago

    In this milestone we did a large refactor of how models were saving in the backbone app (the Protototype).

     
    100% complete
    Closed Jan 28, 2013 Last updated over 4 years ago

    In this release we made the Prototype app run online. Chrome 24 mad…

    In this release we made the Prototype app run online. Chrome 24 made some signficant change which meant we would no longer be able to use PouchDB and IndexedDB to store data offline. This is the release where the beta users and stable users split for a long time (1 year Jan 2013-2014).

     
    100% complete
    Closed Jun 4, 2013 Last updated over 4 years ago

    In this milestone we cleaned up the code and the tests. We were hoping to do more screencasts for future developers but we cancelled them because we ran out of time.

     
    100% complete
    Closed Sep 16, 2013 Last updated over 4 years ago

    In this milestone we made the search widgets use fuzzy search by default (request from Claire).

     
    100% complete
    Closed Oct 11, 2012 Last updated over 4 years ago

    In this milestone we made FieldDB run offline on an Android using TouchDB.

     
    100% complete
    Closed Aug 25, 2013 Last updated over 4 years ago

    In this milestone we set up an audio web service which accepts audi…

    In this milestone we set up an audio web service which accepts audio or video from the Android Elicitation app and automatically generates Praat TextGrids using the Prosody Lab Aligner. Prosody lab aligner: https://github.com/kylebgorman/Prosodylab-Aligner Audio Web Service: https://github.com/OpenSourceFieldlinguistics/AudioWebService Android Elicitation Session App: https://github.com/OpenSourceFieldlinguistics/AndroidFieldDBElicitationRecorder

     
    100% complete
    Closed Jul 2, 2013 Last updated over 4 years ago

    A new user interface which lets you enter data as if you were in Ex…

    A new user interface which lets you enter data as if you were in Excel. Released for the Computational Field Linguistics Workshop at McGill (May 27, 2013): http://migmaq.org/workshops/2013-computational-fieldworkshop/ http://litebeta.lingsync.org This user interface was designed to be quick for data entry (unlike the prototype which had numerous user experience issues, and was confusing.)

     
    100% complete
    Closed Nov 1, 2013 Last updated over 4 years ago

    In this milestone we added csv export (by request from some prospec…

    In this milestone we added csv export (by request from some prospective users, we think). We also set up the infrastructure we needed to build client side export widgets that can be embedded in any app, any webpage (ie if a research group wants to make their data available on their website).

     
    100% complete
    Closed Dec 9, 2013 Last updated over 4 years ago

    Louisa's first step in her project is to annotate some data to get …

    Louisa's first step in her project is to annotate some data to get used to the language. To help her, we could all annotate 5 datum each, so not only will she have experience annotating, but also end up with a pretty useful baseline for the glosser and morpheme segmenter :) Proposal: We can turn this into a user study! Control linguist non-expert using prototype app: hiskanog (a linguist, but never done Inuktitut) Control linguist non-expert using spreadsheet app: senhorzinho (a linguist, but never done Inuktitut) Target RA non-expert using spreadsheet app: louisa-beilig (a budding linguist, but never done Inuktitut) Control linguist expert using spreadsheet app: richard (a linguist specializing in Inuktitut) Control non-linguist non-expert using spreadsheet app: jdhorner (a non linguist, never used the app, never worked on Inuktitut) Not useable data: linguist non-expert using spreadsheet app: cesine (a field linguist, has done some Inuktitut years ago Any one else want to participate? To do this, we should make sure we can get some sort of measure of dificulty, alexis used time... so we could modify the spreadsheet and prototype apps to add a measure of time spent for activities... (ie, 1 hour spent segmenting and glossing..) Thanks to all who participated :) We were able to find out some very good details from the flow of annotating Inuktitut. We made some new bots already to handle most of the data prep, what is left are layout issues. Some take homes: In general, the current templates we have for entering data in either the prototype or the spreadsheet app are simply not comfortable for corpus data agglutinative languages (long sentences in Inuktitut). We need something that is IGT Both: a Large part of morpheme identification is done outside the app. maybe it doesn't learn fast enough? I had guutiullu (guuti-ul-lu glossed as "God-?-and") in 4 datums but the glosser did not predict the segmentation nor the gloss. over prediction happened in a few places too. In this example the utility shows that allomorph-wise, ullunut breaks down into ullu - nut, but that the possible morphemes are actually ubluq or ublu as well as nut or nnut. Perhaps they could be linked as well? In S. Allen's paper, on PDF page 6, (paper page 72, Table 2), there's a wonderfully useful table. Maybe add a feature to the app that allows users to specify, in one action, those types of markers so that the auto parser can more easily handle them from the start? Spreadsheet: The font is too large for long utterances/morphemes/gloss lines. It may also help to make it monospace, so characters would more easily align. Specifically for annotating, (perhaps a new layout choice, from the two that currently exist) the input boxes should stretch across the entire screen. Input boxes should be able to count hyphens and spaces in order to make each row align with its partner rows. (see screenshot in #1204 for a better visual representation of this) Prototype: datum field boxes are way too small for the font size, there isn't enough margin on top and bottom of the texts. sometimes i thought i put the cursor in the text and press delete because i wanted to remove the dash-, then the screen went back to the previous page, without saving the glosses.

     
    100% complete
    Closed Dec 3, 2013 Last updated over 4 years ago
     
    100% complete
    Closed Dec 2, 2013 Last updated over 4 years ago
     
    100% complete
    Closed Oct 2, 2013 Last updated almost 5 years ago
     
    100% complete
    Closed Oct 1, 2013 Last updated almost 5 years ago

    @senhorzinho has been working on an android production experiment a…

    @senhorzinho has been working on an android production experiment app with @kondrann here: https://github.com/OpenSourceFieldlinguistics/AndroidSpyorNot

     
    100% complete
    Closed Oct 1, 2013 Last updated almost 5 years ago

    Goal *Create a sample Android App that does any perception experiments

    Goal *Create a sample Android App that does any perception experiments Research questions *usable for phonetics studies (is the population segment targeted by the Android app is a better representative as opposed to the undergrad student population normally targeted by these types of studies) *usable for discourse studies and pragmatics (creating less invasive environment as opposed to the artificial lab surroundings) Materials *Spyornot data https://github.com/AccentsResearchLab/GamifyPsycholinguisticsExperiments/tree/master/analysis *Json (experimental design) *Materials and images Procedure *Re-write the experimental code in Java *Come up R hypothesis and expected outcomes *Test these hypothesis *Make adjustments to the app based on the test outcome Further Work *Create screen casts *Create tutorials for other linguistics to use the code References *https://github.com/OpenSourceFieldlinguistics/AndroidFieldDBElicitationRecorder *https://github.com/AccentsResearchLab/GamifyPsycholinguisticsExperiments *https://github.com/iLanguage/OPrime

     
    100% complete
    Closed Jun 8, 2013 Last updated almost 5 years ago

    Using such information as access and error logs, database calls and…

    Using such information as access and error logs, database calls and file storage, create and implement dashboards and information queries that will help us better understand server requirements and users' needs.

     
    100% complete
    Closed Sep 2, 2013 Last updated almost 5 years ago
     
    100% complete
    Closed Jun 5, 2012 Last updated about 5 years ago

    We will write the tests before the code, so that what the code is s…

    We will write the tests before the code, so that what the code is supposed to do will be clear. Tests essentially tell you the entry and exit conditions of each function, and what types of fringe cases the code is supposed to treat. Tools: Jasmine Phantom Eclipse Approach: behavior driven development (BDD)

     
    100% complete
    Closed Jul 30, 2013 Last updated about 5 years ago
     
    100% complete
    Closed Jul 11, 2013 Last updated about 5 years ago
     
    100% complete
    Closed Jul 10, 2013 Last updated about 5 years ago

    Primary feature: upload video function

     
    100% complete
    Closed Jul 2, 2013 Last updated about 5 years ago

    Release an alpha version of an elicitation session Android app for the Master Apprentice program in Listuguj.

     
    100% complete
    Closed Jun 4, 2013 Last updated about 5 years ago

    Details to come... Maybe complete the user study with the fieldinguistics students Do the prosody lab aligner integration Demo of data cleaning with Carol's functions

     
    100% complete
    Closed Jul 17, 2012 Last updated over 5 years ago

    In this iteration we want to have the whole cycle of data from an o…

    In this iteration we want to have the whole cycle of data from an offline chrome extension, and up to the server and back again. This depends on having a clear design, and clear testing code and must be done by the end of June, or we wont have enough time to test the software before launching it in August. This needs to meet the basic requirements: simple collaborative make your own categories tags/keywords (e.g. antipassive) search unicode offline anywhere ipa smart import .csv The How and Why: simple (cute design) friendly (the software needs to be localizable, it will start in English, French and Spanish since we will be launching it in Guatemala) collaborative (need teams and users) make your own lines (have a design page so they can set their lines and add new ones, and change the order, and decide which lines are rare) tags (have tags and be able to use the tags for search, mostly for "constructions" like "passives" or other things that may or may not really exist but have been used in the literature.) search (progressive search, intersect and union search) unicode (all the pages are unicode, technically all the input data will be unicode automatically. needs testing..) offline (use local storage in the chrome extension, sync when the user clicks sync, in later iterations, try to sync in other ways) anywhere (need teams and users, and a central server to server the various browsers, chrome is supported, but others like Opera and Safari will work) ipa (we really need ipa support, but the ipa keyboard is tricky, this needs looking into) smart (we will set up the logic for the software to guess morphological parsing, and glosses, this should be a plugin so that the "smart" can be from anyone's existing morphological parser) import .csv (it will be super userful to import csv data sets while testing and developing so that we dont have to enter data manually to test the system) non proprietary structured open data

     
    100% complete
    Closed May 5, 2013 Last updated over 5 years ago
     
    100% complete
    Closed Mar 18, 2013 Last updated over 5 years ago
     
    100% complete
    Closed Mar 15, 2013 Last updated over 5 years ago
     
    100% complete
    Closed Feb 12, 2013 Last updated over 5 years ago

    Let's have a bug fixing fiesta, a hackathon style weekend fueled by…

    Let's have a bug fixing fiesta, a hackathon style weekend fueled by pizza and fun! A few small teams learning and teaching each other data manipulation, html, javascript whatever they decide to work on. Master of ceremonies will be @gretchenmcc leading Usability and @cesine leading HowTos, helpers will be @hisakonog and @Kedersha, with guest appearances from some of our friends in computational linguistics and montreal techie types :) Add issues to this milestone if you think it could be tackled by a someone with a bit of guidance :)

     
    100% complete
    Closed Feb 3, 2013 Last updated over 5 years ago

    This milestone contains all issues that must be completed before beta testing the app for use in a Field Methods course.

     
    100% complete
    Closed Jan 8, 2013 Last updated over 5 years ago

    Ultimately it should be possible to download your data to a new com…

    Ultimately it should be possible to download your data to a new comptuer/browser using chrome or firefox and even use the data offline without being in a chrome store app. There are some technological roadblocks, this mile stone represents them. We decided to go with a couchapp, currently only online, but eventually with the possibility of using pouch on the same url to have the data offline (ie for iPad) Here is an online app example: https://ifielddevs.iriscouch.com/public-firstcorpus/_design/pages/index.html

     
    100% complete
    Closed Dec 26, 2012 Last updated over 5 years ago

    The Migmaq team wants us to set up the web services on their depart…

    The Migmaq team wants us to set up the web services on their department servers so it can plug in to the ProsodyLab's Aligner. This milestone represents all the issues that need to be done to get their server up and running.

     
    100% complete
    Closed Nov 6, 2012 Last updated almost 6 years ago
     
    100% complete
    Closed Oct 3, 2012 Last updated almost 6 years ago

    We are adding a module for creating free Android/iPhone/Chrome seco…

    We are adding a module for creating free Android/iPhone/Chrome second language learning apps based on fieldinguist's databases. This module will be able to use iField to create new lessons and units, and display the exercises in fun new dashboards specially designed for second language learning. This module useful for researchers working with communities as a way to give their research back to the community. Our first step in this process is creating an Android Prototype for the the McGill Migmaq team. The prototype shall: show a listen and repeat style exercise be cute and user friendly run on Android be installable from the Android Market be created from the team's existing xml files Here is the resulting prototype: video demo: http://youtu.be/MOG22jpAPPM installable app: https://play.google.com/store/apps/details?id=com.github.opensourcefieldlinguistics.android.lessons&hl=en

     
    100% complete
    Closed Sep 26, 2012 Last updated almost 6 years ago
     
    100% complete
    Closed Jul 14, 2012 Last updated almost 6 years ago

    In this iteration we will focus on getting all the handlebars and V…

    In this iteration we will focus on getting all the handlebars and Views roughly implemented, and getting the control flow in place for Page load Enter new datum Import data Search

     
    100% complete
    Closed Aug 5, 2012 Last updated almost 6 years ago

    We want the software to have been tested for at least 1 month before bringing it to the CAML workshop in Guatemala.

     
    100% complete
    Closed Jul 18, 2012 Last updated almost 6 years ago

    We would like to recruit some people to use our app a few weeks bef…

    We would like to recruit some people to use our app a few weeks before we launch on August 1st in Guatemala. We are targeting July 1st for open Beta Testing. If you would like to join the Beta Testers, you can fill in this form and we will give you access to the app. https://docs.google.com/spreadsheet/viewform?formkey=dFlXWENNd1FXRmxPOTl1NVVqemc2ZlE6MQ

     
    100% complete
    Closed Aug 16, 2012 Last updated about 6 years ago

    Team features

     
    100% complete
    Closed Aug 16, 2012 Last updated about 6 years ago

    In this milestone we will add in the team and collaboration features:

    In this milestone we will add in the team and collaboration features: team activity feeds letting the user add corpus contributors and collaborators (its already working, just needs a user interface) urls to share your corpus with the public

     
    100% complete
    Closed Aug 1, 2012 Last updated about 6 years ago

    In this milestone we put all the issues that must be done, before w…

    In this milestone we put all the issues that must be done, before we can tackle the Everything Working milestone !. Authentication of existing users 2. Registration and Authentication of new users 3. Replay of data updates when pouch syncs with iris couch (through the auth server) 4. Sending credentials down to the offline app to decrypt confidential data

     
    100% complete
    Closed Jul 21, 2012 Last updated about 6 years ago

    We want to put in some logic to have the app train a glosser overnight so that users can reduce the amount of predictable info they have to do in the glossing.

     
    100% complete
    Closed May 22, 2012 Last updated about 6 years ago

    We need a initial architecture, which will be filled out by the testing code. We can think of this as an "empty black boxes" stage.

     
    100% complete
    Closed May 13, 2012 Last updated over 6 years ago

    We need a completed design, with buttons and graphics before we sta…

    We need a completed design, with buttons and graphics before we start dividing up the coding. Having a cohesive design before coding will help us code only what needs to be coded and help the communication about what each button/function is supposed to do (ie, the input and output and work flow will be well defined, so the implementation should be more clear). Tools: Gimp Eclipse Chrome Store Dependancies: a name branding color schemes a mission statement

     
    100% complete
    Closed May 9, 2012 Last updated over 6 years ago

    mecathcart made a clickable prototype in HTML using images to show …

    mecathcart made a clickable prototype in HTML using images to show functionality like export. This stage was to get something concrete to talk about, to expand and reduce and define the limits of what we want to build.

     
    100% complete