Linked by Thom Holwerda on Mon 17th Aug 2009 09:34 UTC, submitted by moochris
BeOS & Derivatives I had been following the mailing list for the Haiku project the past week with growing interest. The topic of discussion? Why, the alpha release, of course! What needs to be done, who needs to prepare what, and most importantly, what schedule are they going to settle on? Well, after numerous insightful back-and-forths, the community has settled on a schedule.
Thread beginning with comment 379029
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[4]: Good and bad
by WendelFree on Mon 17th Aug 2009 15:10 UTC in reply to "RE[3]: Good and bad"
WendelFree
Member since:
2009-08-17

"
Why do you assume I assumed? I have actually studied and hacked on the Haiku code (mostly interested because I had intimate knowledge of the NewOS code base.)


Because you're assuming that the behavior you're seeing holds true for everyone else, which is very much not the case.
"

AGAIN: I don't assume anything. I base my assertions on knowledge about the code, not merely the three examples of problems I gave.

I know it hurts to hear that the project you work on has problems, but I do this with the best intension.

In fact, I'm preparing a patch for a bug in the PCNet32 driver as we speak ;)

Reply Parent Score: 1

RE[5]: Good and bad
by anevilyak on Mon 17th Aug 2009 16:28 in reply to "RE[4]: Good and bad"
anevilyak Member since:
2005-09-14


AGAIN: I don't assume anything. I base my assertions on knowledge about the code, not merely the three examples of problems I gave.

I know it hurts to hear that the project you work on has problems, but I do this with the best intension.

In fact, I'm preparing a patch for a bug in the PCNet32 driver as we speak ;)


I have no problem with you saying it has problems, what I have issues with you quite unambiguously making it sound as if the code as a whole sucks while only giving vague assertions with no substance to back that up, especially when your entire backing evidence is pretty much "It doesn't work on my machine". "I'm intimately familiar with NewOS" is utterly meaningless. Given that SMP for instance works fine on my A64X2, as well as various people's Core 2 Duo/Core 2 Quad (and in some cases multi-Core 2 Quad), it decidedly sounds like an issue with your specific hardware, not an endemic problem of the code as a whole. If you insist otherwise, then provide details.

Reply Parent Score: 3

RE[6]: Good and bad
by WendelFree on Mon 17th Aug 2009 16:47 in reply to "RE[5]: Good and bad"
WendelFree Member since:
2009-08-17

I have no problem with you saying it has problems [...]


Uh, ok. It sure sounds like it thought.

, what I have issues with you quite unambiguously making it sound as if the code as a whole sucks while only giving vague assertions with no substance to back that up, especially when your entire backing evidence is pretty much "It doesn't work on my machine". "I'm intimately familiar with NewOS" is utterly meaningless.



Not really. It means I know what I'm talking about (I do). To be clear: I have tried Haiku on two physical machines, along with VMWare. In VMWare with a two-cpu setup on my core2duo, the PCNetII driver locks up, and I know why (this is the patch I was talking about.) Generally, that driver is extremly badly written and serves as another example.

The code as a whole doesn't necessarily suck, but there are serious issues with the subsystems I have looked at regard security and stability. I can provide details why your TCP/IP stack is full of holes in both those camps (and I intend to submit this as PR's.)

Edited 2009-08-17 16:56 UTC

Reply Parent Score: 1