wolf359 opened this issue on Sep 24, 2002 ยท 109 posts
pdxjims posted Tue, 24 September 2002 at 10:21 AM
Beta testing can be a complicated art. Beta testers are often under appreciated and over blamed for problems in new software. The terms of the beta testing aren't known, so we don't know if specific testers were assigned specific areas for testing. Also last minute changes for problems found by the beta testers may not have ever been tested a second time. Of course, the larger the test base, and more supervised the testing is done has a lot to do with how well the final version is. There was a LOT to test in P5. Usually with a system this complex, a set of testing requirements is done in advance, with standardized forms for problem reporting, follow-up, retesting by both the same tester and a second tester for confirmation. Remote beta testing without constant contact has to be managed even more tightly. A good beta test for a system like this should take a minimum of 2 to 3 months to allow for fixes and retesting. I think CL was feeling a LOT of pressure from Poserland to come out with the product A.S.A.P. Not to mention the financial needs of a relativly small company. While I think the testing needed to be more thorough, I can't fault the testers with the problems in the product. CL did drop the ball on this one, but I think it the needs of their company required an early release. I'm VERY happy with P5, in spite of the bugs. The big things I wanted are all there, I'm handling the learning curve, and I'm pretty sure most of the bugs I've found will be fixed in the next few weeks (since most of them are simple coding errors probably made after the testers had found the main set of bugs). Things like Don's left shoulder and library lookup problems really aren't that bad, or tough to fix. I would be happier if there were a standard bug reporting system in place, with a status of the fix and estimated correction date at the CL website for that specific problem (as well as pulling in some of us to test individual fixes). CL's biggest problem, as well as their greatest asset is the Poserland community. We bang their software to death, tearing it apart to see how far we can stretch it. They should use us more, and in a more organized fashion. However, I do have a concern about the CL/Daz relationship. I know that Daz is making a Poser-like product for release someday, but right now they have the most detailed Poser figures, with the most 3rd party products available for Poser. Lets face it, without Michael and Victoria, Poser wouldn't be nearly as big as it is. CL and Daz should have worked harder together to make those 2 characters P5 fully compatable. I think its a real pity that Daz didn't have a Michael and Vicky V3 that was fully P5 compatable ready when P5 came out. Both companies would have profited greatly from that. There isn't even a free skull cap for the Millenium family available. So, don't blame the beta testers. They did the best job they could with the supervision and experience they had. Rant over. I may have been in programming too long. Maybe I'll go apply for ajob at Starbucks...