Archive for March 15th, 2006

Foresight Linux, the champion of bugs

My friend Thom is jumping up and down for Foresight Linux usually, and so I decided to have a second look yesterday, downloading a VMWare image of their latest 0.9.4 ‘stable’ release that included Gnome 2.14 in it.

I found about 30 bugs in about one hour of using the system. Some of them are usability bugs, some are just personal irks, but about 15 of them are real hardcore bugs.

In fact, I am a proud finder of a huge security hole that I discovered on Gnome’s fast-user-switch 2.14.0 applet. This is the first security bug I ever found, IIRC. Anyways, I emailed James who is the maintainer of the applet, he replied soon enough and he will possibly look into this (I’ll be watching).

But what really bugs me is that Foresight Linux’s bugs I found are easily detectable with a simple SMOKETESTING. Don’t these people care about their project? Don’t they see these reproducible errors, crashes, freezes, whatever? I have to state the following in capitals, I am sorry:

IF A PIECE OF SOFTWARE DOESN’T PASS SMOKETESTING, DON’T BLOODY RELEASE IT.

No matter if your software is marketed as “bleeding edge” or not, there is a difference between having some bugs here and there and not pass smoketesting. This is the difference between a Beta/RC version and an eternal Alpha. Who wants to use Alpha software? Surely not someone who respects him/herself.

Apple now loves SPEC?

I have an Apple developer account and I received today an email from Apple asking devs to become “ADC Select” members and get a Macbook discount. Well, here’s the thing. It really surprised me that they had this fine print on their email: “*Based on estimated results of industry-standard SPECint and SPECfp rate tests. SPEC® is a registered trademark of Standard Performance Evaluation Corporation (SPEC).”

You see, when Apple was in the PPC architecture, they were avoiding publishing SPEC benchmarks just like the ‘devil avoids the holy water’ (this is a Greek saying btw). Apple would just not release SPEC results, even if ALL other CPU and machine manufacturers would. But now that they are on the Intel platform, they don’t mind doing so and mentioning SPEC.

This exposes Apple as huge liers to the public when they were saying back in the day via ads that the G3 is faster than the PII or that the G4 was faster than the P4. In reality, especially the G3, it was even slower than a PII-Celeron at the same clock. Internally at Motorola, the G3 was never meant for a high performance CPU in fact, but Apple had no choice but to use that CPU for their Mac line or they would have been out of business.

Of course, Intel is not much better in that regard when they were saying that the PentiumII would make your internet connection… faster. Or AMD’s claims over some of Intel’s models. Everyone lies. Pick your poison. It’s just that I don’t think that anyone has lied as much as Apple has the past few years regarding CPU speeds.

Burning down the house

Remember my adventures with my Powerbook’s connector? Well, it seems that the Magsafe is not really that safe either…