<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
On Fri, Jun 18, 2010 at 01:24:12AM -0400, Michael Stone wrote:<br>
> 1. Will sugar-0.90 be better if it gets more and more serious testing<br>
> between now and its release?<br>
<br>
Better than it otherwise would be, yes. In terms of quantity of latent<br>
defects on release. We'd also have a better idea of the defects it is<br>
released with that won't be fixed before release.<br>
<br></blockquote><div>Yes.<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
> 2. Might we get more meaningful testing done if we all agreed to<br>
> agree, on a roughly weekly basis, on *what to test* -- that is, on a<br>
> "sugar HEAD"?<br>
<br>
Yes, the testing would be more meaningful if it was anchored to some<br>
version concept, even if it was a hash. What this begins to sound like<br>
is rapid releasing during development. 0.89.n this week, 0.89.n+1 next<br>
week.<br>
<br></blockquote><div>We meet every Saturday morning (NZ time 11am) and would like to know say
Thursday what to download on Friday for testing, or a similar pattern
each week. Include any specifics of what you want tested, more detail the better. Testing pattern is good - release on Thursday, test on Saturday with results same day, analysis of tests and ask questions Sunday, and carry out improvements and fixes before release on Thursday, and so on. It is worth noting that if released Friday American time we cannot test for another whole week. <br>
<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
> 3. Might we get more comprehensive test results if "sugar HEAD" were<br>
> directly testable against the real software in use on all platforms of<br>
> interest? [2]<br>
<br>
Yes, test results would be more comprehensive.<br></blockquote><div><br>We can test XO-1.0, XO-1.5, and SoaS easy enough, we have a bit of
hardware range (Acer, Asus, HP and 3 Mac laptops but the Mac testing is
just too hard). Easy installation is required because often we are doing the installation during the testing time (in a bar once a week while all the XOs are together). We are having success with reflashing XOs as a way of installing the new version. We have limited internet during testing so doing a 100MB yum upgrade on each laptop is not possible. <br>
<br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<br>
> 4. Would you like to participate in a roughly-weekly social event<br>
> designed to accomplish these goals? (That is, for example, in a weekly<br>
> Sugar Labs "Friends in Testing"-style exploratory testing<br>
> opportunity?)<br>
<br>
Yes, subject to timezone considerations. ;-}<br></blockquote><div> </div><div>We go on IRC at 11am on Saturday mornings (NZ time) and are online for a few hours. After that it is sporadic, when I remember to start IRC client basically. I would be quite happy to do another IRC on Sunday (NZ time) to follow up anything done.<br>
Our testing is pretty social, it is in a cafe/bar in Auckland and in Wellington, with food and coffee, lots of chatting (about other FOSS education stuff too). <br><br><br>Tabitha</div></div><br>