Skip to main content

Notice: This Wiki is now read only and edits are no longer possible. Please see: for the plan.

Jump to: navigation, search



Attending: Oliver, Paul, Eugene, Chris, Joanna

Any issues with last weeks summary?

  • No


Oliver asks if we are we in good state?

  • Paul notes we are 3/4 thru development cycle
  • Joanna notes that there are 76 open defects.
    • Asked people to clear as many as could before test pass
    • Joanna noted that the Intel China team is making good progress
      • Good progress on some critical defects
      • Asaf is helping out with some reviews


  • Thusfar there is a tutorial proposal
  • There is a Profiler outreach talk proposal (short)
  • There is a Test talk proposal
  • Intel still plans to submit another talk if they can get permission from management to do so

Profiler Outreach

Eugene did not get feedback from the three interested parties last week

  • He has sent out email to them asking for followup...
  • He hopes for followup in next few days.
  • Oliver will follow up as well and express interest from the TPTP PMC for feedback

Harm had made comment about some test failures that have popped up for profiler.

  • We need to make sure that bugs are filed as they affect usability.
  • Any test failure should have a bug filed for it
  • Paul notes that reporting tool provides for adding defect number which turns into link in but report
    • This might be a good way for us to verify that all failures have bugs filed


There is some debate in Eclipse over purpose of the simultaneous release train.

  • Some want to make sure that it is international
    • It is an open question why all projects are internationalized?
    • Some projects have errors internationalized but all of their documentation is English.
  • In some cases "big" vendors want internationalized support which simply adds barrier of entry cost to smaller vendors
  • Oliver wonders why internationalization is all the range and why should we not pick some other scenario for a release
    • For example "memory leaks".

To join the Galileo train there are some steps to perform

  • Must make an intent to join known
  • Must provide plan details in standardized xml document format
  • Must agree to watch cross product inbox, attend certain meetings,
  • Have written rampdown policy and follow full IP approval processes and timeline

There are some requirements about use of APIs and around message bundles.

  • Paul notes that this is not a big deal for us.
    • He notes that message bundles are pretty easy and there are assorted tools.
    • This is only a problem if you have a complete project that has used strings thruout, conversion is a pain.
    • However if one is largely using message bundles already, there are tools to find outliers and resolve

Components must be able to be enabled and disabled. We discussed whether this meant each individual plugin needed to be individually enabled and disabled. We believed that it should be sufficient to fully enable/disable TPTP as a whole.

  • Eugene will write up a paragraph on enable/disabling of TPTP as a whole for "deployment" section of list

The only open question nthat we had: Do we have any translationability for Galileo? Typically, you do not do this for a point release and we are simply releasing 4.5.3 and not a 4.6

  • We did not close on an answer to this question

Back to the top