PDF engineering meeting notes, 2012-04-12

[Bill Walker]

meeting notes, PDF.js 2012-04-12

TASKS:

* Deal with Add-on Blocker [bwalker]
TODO: schedule meeting with Dave Townsend, Justin Wood

* Build new UX that matches FF [artur]
TODO: implement current mockup with existing viewer codebase; replacing jQuery UI in mockup with sometime else
TODO: be sure to do keyboard shortcuts and tab order

* Make Viewer Unit Testable [artur]
TODO: find contributor [jviereck]

* Devise strategy for i18n of bundled Add-on in FF [bwalker]
TODO: schedule meeting with dietrich and axel
Dietrich and Myk looked at l10n for Add-ons v. FF
TODO: See TestPilot -> Feedback Mode
STATUS: NOT STARTED
see also https://wiki.mozilla.org/Firefox/BundlingAdd-ons

* Devise UX and mechanism for enabling / disabling / prioritizing multiple PDF viewers [bwalker]
TODO: schedule meeting with madhava, asa
TODO: we've already got UI for picking handlers for media types, can we reuse that?
standalone viewers v. plugins v. add-ons
Alert the user the first time they encounter a PDF
- would you like to try PDF.js?
- if YES, disable Acrobat and use US
NICE: defer to other readers when we get into trouble
make it easy to go to something else when we fail
STATUS: NOT STARTED

* Devise Aurora uplift criteria for PDF rendering issues [bdahl]
TODO: base this on the corpus report?
TODO: do we really need to implement JBIG2? most scanned documents use JBIG2 now

* MIT license header missing? [yury, https://github.com/mozilla/pdf.js/issues/1528]

* Devise bug tracking guidelines for integration issues v. rendering issues [ask scoobidiver]
TODO: meet with them to ask about shadowing in github
TODO: get other folks

* Implement text selection and searching that is well-integrated with Firefox [artur, bwalker, shorlander]
TODO: work with Asa and Madhava and Stephen to be specific
Q: should we use built-in search?

* Implement a reasonable printing experience for PDF's [bwalker, jviereck]
TODO: talk with Asa abour requirements

* Implement a reasonable experience for form filling [?]
TODO: figure out what requirements are? [bwalker]
Q: do people want to fill form and print? fill form and save?

* Implement partial downloading / render (see https://github.com/mozilla/pdf.js/issues/1108)

* resume bot testing on linux [artur]
- install new bot on linux

* get accessibility review of new UI [artur]
blocked by implmenting new UI

* decide on task tracking tool [??]

* Implement shorlander's new UI, get code review [artur]

* Add mochitest cases for the new UI, get code review [artur]

* learn about Talos, try running with PDf.js [bdahl]
possible followup: can we add our own test cases to Talos?

* investigate using bot to capture nightly performance log [artur]