On Sun, 3 Apr 2005, Phillip Kerman wrote:
Here's the bottom line and unfortunate truth: slimserver is not stable.
5.4 is/was pretty good. Yeah, it had bugs, but no more than, say, Windows Media Player.
6.0, on the other hand, is unusably bug-ridden -- but this really shouldn't be a surprise, as the Slim people ripped out the guts of the thing and reworked it. Based on knowing how unsmoothly that usually works in any project, past experience with SlimServer releases, and looking at the bug reports before 6.0 was deemed "release quality," it had to have been clear to most people on this list that SlimServer 6.0 would effectively be a beta product well after its official release.
It'd be nice if Slim stopped releasing things before they were ready; but until then, the sensible thing to do if you want stability is to let other people be the beta testers -- don't buy new products from Slim right away, don't upgrade your server to a new release for a while, that sort of thing.
But, I hear Michael's point: we're all so patient... I'll even be more patient... in fact, I'm sold on the product--so I sort of have to be patient. But what can we do here? Are there QA tools that can be applied to open source software that aren't being used? Is this a natural outcome of open source (I don't think so)?
Slim could just stop releasing stuff before it's done. It was pretty clear that 6.0 wasn't really a release version, but it got called one anyway. If we were in a situation where final betas looked rock-solid, but the actual releases revealed bugs from the more diverse setups of a wider-ranging group of users, well, that's a tricky situation. But the situation we're actually in, where the betas are known to be flaky and buggy but are bumped to production anyway, is pretty straightforward to solve: Just don't do that.
(And yes, I'm aware that from Slim's perspective, this isn't that simple. They have to get their new product out there to drive new revenue/remain competitive in a fast-paced market, can't do it without the new software, and therefore need to accelerate releases. Still, I think in the long-term, releasing stuff before it's finished is a lousy strategy. I evangelize the hell out of Squeezeboxes, but there's no way I'm going to tell anyone to buy a Squeezebox 2 for the near future.)
-- Mike Kozlowski http://www.klio.org/mlk/
_______________________________________________ Discuss mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/discuss
