Linux Kernel 3.1 RC 2 Released 209
sfcrazy writes "Linus Torvalds has announced the release of Linux kernel 3.1 rc2. He said '300+ commits for -rc2 is good, but please make me even happier for -rc3 by ONLY sending me real fixes. Think of it as "fairly late in the -rc series," because I really want to compensate for the merge window being fairly chaotic.'"
Re:Damn, this feels like Firefox. (Score:5, Interesting)
Yeah, the *2.6* SERIES is quite different even from itself. Meanwhile the 2.0, 2.2, and 2.4 series had at least mostly had stabilized API/ABIs during the time of their existance, occasionally getting features backported (Thinking about USB,Wifi, and a few filesystem module primarily there). 2.6 however was having constant and incompatible changes ever 5-10 minor numbers. Devfs droppage, incompatible udev changes (Ever tried updating a system only to have it temporarily hosed because you had the wrong udev version running and all your device entries are now wrong??), constant gfx abi breakage (see nvidia/ati drivers constantly being 2-5 minor nums behind, and then having to drop older support for maintainability).
While a jump to 2.8 for the aforementioned features stabilizing would make sense, with a 2.9 dev branch started to restandardize 'stable' versus 'experimental' changes the jump to 3.0 was entirely unwarranted and just more of the me-tooness that linux seems to be have been heading towards for a good 5 years now. Honestly the only thing holding me to linux at this point is a lack of desire to have to repartition my disks using bsd slices, and a lack of alternative open source OSe that are actually robust enough to boot on all my hardware. (I have reference spec dual processor 440FX systems, the same chipset emulated by qemu, and despite being developed on it, ReactOS, Haiku, Solaris, and a few others never make it out of their first stage bootloaders, on IDE, SCSI, or SATA. Disappointing to say the least.)
Combined with the current Gnome BS (Which anyone who has tried running it on an system dated '04 or earlier will attest to.), there's not a lot of motivation to use linux over alternatives such as Windows, or a Mac/Hackintosh OSX box. The latter two might be slow, but nowadays with a 'desktop' GUI, so is the former. And it seems like the bureaucratic messes running these 'foundations' are so focused on 'features' and 'moving forward', that they've forgotten that one of great strengths of UNIX has always been it's long term compatibility.
For another example of this fubar'ing, Go look at GNU coreutils, and as an example, try running the old Loki linux game demos on it. Gee, don't work too well? They decided to deprecate and remove the - feature of head and tail, leading to breakage of numerous scripts dating back how many decades? Additionally, while I may be wrong, the line number feature they replace it with hadn't even EXISTED back in those days, and so for the sake of (whatever rationale was used) they broke it, knowing full well it would cause lots of peoples software to break in unexpected and possibly silent manners.
Would you trust this sort of mentality with YOUR long term software needs?
(And no, contrary to the belief set forth I am not a shill for MS or Apple. In fact I have a rather low opinion of both. I just happen to also hold many of the unilateral development decisions pushed by 'benevolent dictators' (not just Linus! Go look at glibc for another example!) in utter contempt due to their throwing the baby out with the bathwater, especially given the ever increasing bloat in many of the applications, libraries and kernel (C'mon, seriously, removing backwards compatibility while adding *10* extra features that add a meg of code 9/10s of people will never use while removing the one feature they will?!?!).
I'll just end this rant by asking: 'How many of you have been bit by one of the aforementioned issues, and what is your take on the modern 'MBA' mentality that seems to be creeping it's way into the open source ecosystem?'
Re:Damn, this feels like Slashdot (Score:3, Interesting)
What's with all the slashdot users recently, going fucking stupid about version numbering? Who cares what the versions are called: 3.10, 3.11.30 3.A03930. As long as the software works and the users (developers and end users alike) are able to interact with the software, what's the big issue?
Its evidence an underlying problem whereby projects are focusing their attention more on PR gimmicks and the 'gee-whiz' factor of version numbers than actually producing good software.
In the case of the Linux kernel I don't think that applies, after all the 2.6 kernel lasted many years and it is highly probable that 3.x will now do the same. With Firefox (and some others) however, the versioning itself is absurd and the new features being added in each version reflect the aforementioned attitude: "Hey lets rewrite the UI again instead of fixing longstanding problems". Not to mention that when these things are brought up to the powers-that-be at Mozilla they're summarily ignored showing a growing distance between the people running FF and the everyday devs and addon writers who have made it successful.