Difference between revisions of "Processing.js/ConferenceCalls/Feb-11-2010"

From CDOT Wiki
Jump to: navigation, search
(Proposed Agenda)
(Minutes)
 
(4 intermediate revisions by 2 users not shown)
Line 3: Line 3:
 
===Attendance===
 
===Attendance===
 
*[[User:asalga|Andor Salga]]
 
*[[User:asalga|Andor Salga]]
 +
*[[User:mlam19|Matthew Lam]]
 
*[[User:Dhhodgin|Daniel Hodgin]]
 
*[[User:Dhhodgin|Daniel Hodgin]]
 +
 +
*[[User:ASydiK|Mickael Medel]]
 +
 +
*[[User:David.humphrey|David Humphrey]]
 +
 +
*[[User:Anna.sobiepanek|Anna  Sobiepanek]]
 +
 +
*[http://twitter.com/Corban Corban Brook]
 +
* guest: Anthony
  
 
===Proposed Agenda===
 
===Proposed Agenda===
Line 9: Line 19:
 
* New Project Members
 
* New Project Members
  
* 0.5 Update
+
* [https://processing-js.lighthouseapp.com/projects/41284-processingjs/milestones/current 0.5 Update]
  
 
* Merging from Anna's branch
 
* Merging from Anna's branch
* Discuss Dave's test framework if it's going to be mandatory for 0.6+
+
* Discuss Dave's [https://processing-js.lighthouseapp.com/projects/41284/writing-automated-tests test framework] if it's going to be mandatory for 0.6+
  
 
* PGraphic
 
* PGraphic
Line 25: Line 35:
  
 
===Minutes===
 
===Minutes===
 +
* new project members not present
 +
* deadline to get stuff into 0.5 is tomorrow (Fri. 12) @ noon. Anything not in at that time will be moved to 0.6. Since there is no peer-reviews needed everyone should be tackling some bugs. Pay special attention to ensure that no duplicate tickets are filed in Lighthouse (use the search feature before adding anything).
 +
* once 0.5 is released everyone has to merge it with their master branch. I will send out an email to remind everyone. I will aslo write out a blog to pass on my Git knowledge. To make things easier check out the https://processing-js.lighthouseapp.com/projects/41284/project-management-guidelines
 +
* Dave has written up a "how to" for the unit tests: https://processing-js.lighthouseapp.com/projects/41284/writing-automated-tests Try to do it yourself, contact Dave for a one-on-one if you have trouble. Also, get a list together of all of your visual tests that you wrote and gave it to Dave. These test will need to be converted to unit tests. From now on unit-tests are mandatory and visual tests are just gravy. However, if the function you are testing doesn't return anything visual tests are mandatory.
 +
*PGraphic is not needed for 3D. Its basically a back buffer that is used to store pixels off the screen. createGraphic() already makes a new context PGraphic will just need to toggle the contents from the back buffer to the visible canvas. PGraphic will be looked at again once PImage is completed
 +
*smooth() & nosmooth() are basically setting a <canvas style> called imageRendering. Firefox 3.6 is suppose to recognize this style but so far Matthew is getting an error: unknown property imageRendering. Lighthouse ticket for this is going to be blocked. Dave will file a firefox bug and  Anna will track the progress.
 +
*Anna is going to clean up the zenit Processing.js wiki page. Project Tests and Processing.js 0.1 Code status sections will be removed. 3D Implementation Roadmap will be moved to Lighthouse.
 +
* P3D is Processing's software renderer. Since WebGL hardware/software rendering is configured by the user in about:config, we won't have access to this setting. We should create the 3D context if the user uses P3D or OPENGL in their size() call and just let the browser decide how to do rendering.
 +
* We should look into profiling PJS sketches, especially ones which use a 3D context. A profiler may give us some clues as to what WebGL calls could be made more efficient.

Latest revision as of 17:15, 18 February 2010

Meeting Feb 11, 2010

Regrets

Attendance

Proposed Agenda

  • New Project Members
  • Merging from Anna's branch
  • Discuss Dave's test framework if it's going to be mandatory for 0.6+
  • PGraphic
  • smooth/ nosmooth
  • Zenit wiki cleanup
  • P3D V.S. OPENGL
  • Profiling

Minutes

  • new project members not present
  • deadline to get stuff into 0.5 is tomorrow (Fri. 12) @ noon. Anything not in at that time will be moved to 0.6. Since there is no peer-reviews needed everyone should be tackling some bugs. Pay special attention to ensure that no duplicate tickets are filed in Lighthouse (use the search feature before adding anything).
  • once 0.5 is released everyone has to merge it with their master branch. I will send out an email to remind everyone. I will aslo write out a blog to pass on my Git knowledge. To make things easier check out the https://processing-js.lighthouseapp.com/projects/41284/project-management-guidelines
  • Dave has written up a "how to" for the unit tests: https://processing-js.lighthouseapp.com/projects/41284/writing-automated-tests Try to do it yourself, contact Dave for a one-on-one if you have trouble. Also, get a list together of all of your visual tests that you wrote and gave it to Dave. These test will need to be converted to unit tests. From now on unit-tests are mandatory and visual tests are just gravy. However, if the function you are testing doesn't return anything visual tests are mandatory.
  • PGraphic is not needed for 3D. Its basically a back buffer that is used to store pixels off the screen. createGraphic() already makes a new context PGraphic will just need to toggle the contents from the back buffer to the visible canvas. PGraphic will be looked at again once PImage is completed
  • smooth() & nosmooth() are basically setting a <canvas style> called imageRendering. Firefox 3.6 is suppose to recognize this style but so far Matthew is getting an error: unknown property imageRendering. Lighthouse ticket for this is going to be blocked. Dave will file a firefox bug and Anna will track the progress.
  • Anna is going to clean up the zenit Processing.js wiki page. Project Tests and Processing.js 0.1 Code status sections will be removed. 3D Implementation Roadmap will be moved to Lighthouse.
  • P3D is Processing's software renderer. Since WebGL hardware/software rendering is configured by the user in about:config, we won't have access to this setting. We should create the 3D context if the user uses P3D or OPENGL in their size() call and just let the browser decide how to do rendering.
  • We should look into profiling PJS sketches, especially ones which use a 3D context. A profiler may give us some clues as to what WebGL calls could be made more efficient.