Author Archives: builder

jQuery Mobile Team

jQuery Mobile Team Meeting – Jun 12 2013

  • Attending: John Bender, Ghislain Seguin, Gabriel Schulhof, Alex Schmitz, Jason Scott, Ralph Whitbeck, Anne-Gaelle Colom, Jasper de Groot

  • stick to 3-4 months release cycle, every 2-3 weeks maintenance release

    • keep community informed via Twitter, etc.
    • Jasper will update the roadmap wiki page and add dates to Github milestones

  • weekly meetings

    • shorter meetings
    • agenda for things that need to be discussed, updates only in the document/minutes

  • testing (high priority!)

  • roadmap

    • new slider widget 1.5
    • deeplinking 1.5
    • filter widget 1.5
    • fetchlink - this will be possible as result of other changes; add demo instead of feature
    • continue widget review 1.5+

  • test devices

    • managed by Alex
    • Alex brought test devices from test lab at FG for team members
    • there is budget to buy devices for testing if needed

  • Ghislain demonstrated code coverage

  • John updated Gabriel and Alex about navigation

  • John, Ghislain and Jasper talked with the UI team about working together and eventually merge both projects

    • we are going to use one -dev IRC channel (#jqueryui-dev)
    • UI/Mobile team meetings on regular base
      • first meeting: wed June 26, 12:00 EST
    • To-do’s:
      • all UI component widgets should work on both frameworks
      • UI is going to use Mobile’s solution for SVG icons
      • Pointer events for both UI and Mobile
      • 1 CSS framework

  • Note: No weekly IRC meeting on June 13, 2013 because of the jQuery Conference

jQuery Mobile Team

jQuery Mobile Team Meeting – Jun 06 2013

  • Attending: Todd Parker, John Bender, Jasper de Groot, Anne-Gaelle Colom, Gabriel Schulhof, Alex Schmitz, Ghislain Seguin

##Todd

  • We’re focused on 1.4 development
  • Decided to go with SVG icons based on Grunticon after testing because of it’s compatibility with ~100 devices/browsers: View test results
  • Decide on final icon sizes (one vs. 2 size options), black/white, if we dont’ default to the disc bg, setting icons at the swatch level
  • Ideally, we'd let you set per theme swatch: icon color (black/white), icon opacity (0-100%), disc (y/n) and disc color (any hex)
  • New set from Glyphish with ~50 icons will be part of 1.4
  • Team will be meeting in Portland next week

##John Bender

##Jasper de Groot

  • have been working on “next” unit tests; will finish updating last tests tonight
  • been busy with preparations for JQM tutorial and talk at Dutch Mobile Conf (today and tomorrow)
  • will be on vacation right after Portland conf until June 26

##Anne-Gaelle Colom

  • No update... recovering from back problems

##Gabriel Schulhof

  • Fixed tests in ”next” - left some for Jasper (those that deal with theme inheritance)
  • Found some instability in the event sequence based tests - really hard to reproduce.
  • Some bug massaging (#5978,#6014,#6057)
  • Planning on running custom select jQuery selectors through jsperf.

##Ghislain Seguin

  • No update

##Alexander Schmitz

  • New Flip switch widget

  • triage
  • working on getting tests passing again in ext-panels branch
  • working on demos of ui widgets in jqm
  • worked with jorn a bit on cleaning up jenkins
  • fixed random failure in toolbar test
jQuery UI Team

jQuery UI Team Meeting – Jun 05 2013

jQuery Core Team

jQuery Core Team Meeting – Jun 03 2013

Attending: timmywil (married), gibson042, DaveMethvin, m_gol, orkel, gnarf, scott_gonzalez (style discussion)

link jQuery 1.10.2

link jQuery 2.0.3

link 1.10.2/2.0.3

link Can we wait until after Portland for a new release?

  • let's try … gotta monitor the duplicate tix

link Intel wants to help with some optimization issues

  • Rich Winterton profiled with VTune, saw bad JITting behavior
  • But...are any on hot paths?
  • Maybe we can meet with him in person next week?
  • Collab with Intel App Framework

link jQuery 2.1/1.11 changes

  • Discuss at in-person Portland meeting
  • Some thoughts
    • Make Deferreds and Callbacks optional modules
    • Deferred used in .ready() only, never documented
    • Use AMD to do our module builds? snover has a branch
    • needs updates to grunt build
    • optional dependencies (Deferred in .ready) are tricky
    • Does defaultDisplay really solve a problem?
    • Only reliable way is to append, THEN hide. That doesn't cause extra reflows does it?
    • Our "visible" check causes reflows, can we fix that?
    • Need to use display:none vs .offsetHeight/Width
    • Would fix issues like
      visible bug
    • Either changes the meaning of visible, or requires climbing the tree to determine visibility
    • Avoid offsetHeight/Width completely
    • Always a whole number
    • Screws up zoomed results and subpixel rendering
    • Attach data directly to elements for faster teardown

link Style guide updates

  • Grouping parens need space? (i.e., ALL parens need space)
    • If so do all { need trailing newlines?
    • e.g. var x = { test: true } would be invalid
    • Do all ? and : need surrounding space?
    • We tend to do b ? v1 : v2;
    • But we do { test: true } no leading space on colon
  • Dave to make a pull request against the style guide for comment

link Open tickets triage

jQuery Mobile Team

jQuery Mobile Team Meeting – May 30 2013

  • Attending: Todd Parker, John Bender, Jasper de Groot, Anne-Gaelle Colom, Gabriel Schulhof, Alex Schmitz

link Todd

  • Work on 1.4 continues
  • Thorough testing of SVG (Grunticon) vs. icon fonts is almost done.
    • Initial testing results here
    • Grunticon works on every browser we tested, PNG fallbacks are working great. One bug identified in FF 3-3.6 with the fallback has been logged and already fixed.
    • Icon fonts fail in some key devices (WP7, Opera Mini, old Android) and yield false positives to feature tests.
    • Verdict: I think we should use this over icon fonts for max. compatibility.
  • Portland conference: who’s going? Jasper, Gabriel, Alex, Anne-Gaelle

link John Bender

  • Content widget: latest updates are in issue #5427
  • addresses the pagebeforechange double event
  • change == load + transition
  • you get the string or the object in pagebeforechange depending on what was passed in
  • change is just a wrapper around load and transition
  • all the page* events that come from nav are deprecated as of 1.4, use content* now (contentbeforechange)

link Jasper de Groot

  • Done with refactor in branch "next"
  • Found solution for theme inheritance in popup, collapsible content, etc.
  • Now working on updating tests, docs, etc.

link Anne-Gaelle Colom

  • ported the old meeting notes to meetings.jquery.org (55)

link Gabriel Schulhof

link Alexander Schmitz

jQuery UI Team

jQuery UI Team Meeting – May 29 2013

jQuery Core Team

jQuery Core Team Meeting – May 27 2013

Attending: m_gol, jaubourg, dmethvin, gnarf, orkel

Official Agenda:

link jQuery 1.10.1/2.0.2

link jQuery 2.1/1.11 changes

  • Discuss at in-person Portland meeting?

link Open tickets triage

Ecma/TC39

Ecma/TC39 Meeting – May 23 2013

link May 23 Meeting Notes

(by Erik Arvidsson)

John Neumann (JN), Allen Wirfs-Brock (AWB), Eric Ferraiuolo (EF), Erik Arvidsson (EA), Luke Hoban (LH), Doug Crockford (DC), Yehuda Katz (YK), Sam Tobin-Hochstadt (STH), Alex Russell (AR), Dave Herman (DH) (calling in), Bernd Mathiske (BM), Andreas Rossberg (ARB), Mark Miller (MM), Tom Van Cutsem (TVC), Istvan Sebestyen (IS)

link Promises vs Monads

MM: ...Continuing from yesterday

AR: https://github.com/slightlyoff/Futures/blob/master/Promise.idl

STH: Don't like resolve but not willing to die on this hill.

AR: DOM has a bunch of ad hoc APIs to do promise like things.

YK: Mozilla is also actively working on APIs using promises.

AR: A lot of methods today return void so we can change these to return a promise. This is forward compatible.

AR: then does recursive unwrapping...

link ES6, ES7, ES8... Mark's Strawman Roadmap

LH: The important part is not the features but the process.

AWB: Can things be decoupled?

LH: These kind of structural questions are the important part

MM: Suggests "concurrency" to be the main theme.

AWB: Thought about the event loop. All we need is a processing queue... put things in the front and the back.

DH: Only need to add to the back.

AWB: OK.

STH: The callback is called at some later point.

AR: Don't think we need to specify the order.

STH: If we are going to specify promises etc we need to be able to specify things in detail. We can be loose in ES6 and then come back in ES7 and provide a more tight spec.

DH: We could specify the pending events as a set or something. Not sure if there is a consensus that we want a fast small ES7. Not opposed to a modularized approach.

AR: Are there any browsers that are not shipping stable ES6 features today.

YK: Yes. V8.

AWB: Where we have problem today is that there is a lot of interdependency.

MM: These ("concurrency") are coupled together to the event loop

AWB: We can do it as a separate non 262 spec

DH: Opposed to a separate spec. Introduces versioning confusion.

AWB: Roll up

DH: Think of all the extra overhead.

STH: Big difference with 402 since it was run by different people.

LH: Lack of confidence in new features has been an issue for implementers. Good exceptions were Object.observe and Proxies where the wiki contained a mostly complete spec.

AWB: We need to have wiki proposals be deltas to the spec.

TVC: We could have "stable" wiki pages. These would have complete spec deltas.

DH: Very concerned about over modularizing.

AWB: We need to find a way to work faster and be less monolithic.

DH: Agree. ES6 process has blocked implementation work.

LH: We are not committed to our designs.

STH: We are not resolving issues until we start to spec. We are not getting feedback until engines starts to implement.

EA: The problem is that we didn't start to spec things until very late. We had agreements on features long before there was any spec drafts for them.

YK: More from our champions before we get to concensus.

ARB: Lots of the proposals were very vague.

AWB: The more complete spec you bring to tc39 the better chance you have to reach consensus.

ARB: Lack of early spec leads to lack of early implementations...

AWB: ...which leads to lack of feedback.

LH: Not more work, just doing the work earlier before things pile up too much.

DH: Need to look at the dependency graph. Hold of the work of later feature.

ARB: We need to higher bar before we accept proposals.

MM: What we agreed to 2 years ago was that the features are the one we want to spend work on speccing.

LH: Less features to bite of.

DH: A lot of us have a hard time not getting too engage in too many features.

YK: if we focused more effort on managing the overall complexity instead of getting stuck on a lot of technical discussions (and nit picking).

DH: Object.observe and Proxy moved fast but are fairly isolated features

TVC: Didn't involve syntax.

AWB: With ES6 we had a long backlog.

DH: A language will have smaller and smaller complexity budgets as it grows.

AR: ES future needs events

DH: Since this is Mark's wishlist people will throw in their pet features.

MM: This is the direction I am going to work.

LH: There is a page on the wiki outlining the goals.

LH: Looking for 2 things: Something that would allow earlier implementations. Have not brought proposals (over the last 2 years) because we have been blocked by ES6.

LH: When is the appropriate time to bring new proposals to TC39?

AWB: We are free to do what we want. We can issue 6.1, 6.2 etc or technical reports which would serve as a recommendation.

DH: We cannot exclusively work on ES6.

YK: Time at f2f is the most important. Champions can go off and do what they want.

DH: Suggests adding non ES6 items to the agenda. We will prioritize the non ES6 stuff we can get to given our limited time.

YK: We should reinstate the rule that agenda items needs links to wiki pages.

YK: Spec language is good but examples at the top are a must.

ARB: Add step after proposal. For example "stable" or "spec" which a proposal gets promoted to once there is a spec draft, good enough to start implementing.

DH: Strawman: Anything goes.

YK: Proposals used to mean approved.

DH: 3 sections: strawman, proposal, spec/candidate. Keep strawman. Work on improving as a proposal, and when mature enough promoted to next level.

jQuery Mobile Team

jQuery Mobile Team Meeting – May 23 2013

  • Attending: Todd Parker, John Bender, Jasper de Groot, Anne-Gaelle Colom, Gabriel Schulhof, Alex Schmitz

link Todd

  • No meeting last week
  • Team is focused on 1.4 work
  • New icon set is coming along nicely
  • Spoke to Dragan who may be able to help us with docs and FAQs, he’s been doing great work on Stack Overflow support

link John Bender

link Jasper de Groot

  • Almost finished with “next”, just few more small things to look into examples of new markup: http://view.jquerymobile.com/next/demos/test/theming/
  • One issue: inheritance of popup theme or collapsible content theme
  • Going to update unit tests, demo center, API docs, 1.4 changelog wiki one widget at the time
  • Also want to create test pages (kitchen sink)

link Anne-Gaelle Colom

link Gabriel Schulhof

  • Starting work on widget cleanup with Alex

link Alexander Schmitz

  • Fixed ability to view directories on jquerymobile.com
  • gnarf added me to infrastructure so any needs we have there i can help with now
  • merged toolbar widget into master
    • page.sections is gone
    • toolbars now work outside pages
    • true persistent toolbars
  • listviews now work outside pages
  • First pass at panels outside pages.
    • in branch ext-panels
    • so far everything seems to be working just need some css tweaks
  • looked into pointer event polyfills
    • will replace vmouse events except for vclick
    • ui starting https://github.com/jquery/jquery-pointer-events
    • looked into other fast click solutions : ours seems to be the most robust currently but we need to make it work with pointer events
    • Going to look into making this update Kinblas agreed to help with anything i need regarding vClick and vMouse
  • Triage & bug fixing