On 17/01/2007, at 6:54 PM, Naoyuki Ishimura wrote:
Yes, your view on TCM test cases are the same with me. All of test cases except using TestTool one need to be done by the tester who understand the target language for checking the semantic of the localized strings. That is the original purpose of introducing TCM test cases.
I recall Rafaella talking on the l10n list about a testtool script to produce screenshots from each action. Using that, could we provide both options?
1. run TCM cases manually2. run TCM cases automatically, which provides a screenshot of each menu/dialogue to be checked manually for correct language (or perhaps even automatically compared with each target menu/dialogue?)
I think there is still utility for automation in these tasks.I would also like to mention at this point that I created a community questionnaire for the 2.1 release, out of the release-sanity scenario. It made it possible for people intimidated by TCM, or who were simply unable to handle that much English, to perform those actions and provide feedback on them. The questionnaire is currently in Vietnamese, but I could be done in any language.
I put the questionnaire up on the wiki, and also posted it to our project mailing lists.
I have found that TCM and testtool still embody a significant barrier to increased participation, particularly by people who do not have a lot of time or English. We need to lower this barrier, or we will still be trying to do many people's work with a very few people.
from Clytie (vi-VN, Vietnamese free-software translation team / nhóm Việt hóa phần mềm tự do)
http://groups-beta.google.com/group/vi-VN
PGP.sig
Description: This is a digitally signed message part
