Page 1 of 2 12 LastLast
Results 1 to 10 of 12

Thread: DeviceKit-disks Renames Itself To UDisks

  1. #1
    Join Date
    Jan 2007
    Posts
    13,415

    Default DeviceKit-disks Renames Itself To UDisks

    Phoronix: DeviceKit-disks Renames Itself To UDisks

    As Alan Coopersmith pointed out after mentioning the X.Org plans to move away from HAL, the DeviceKit-disks project has renamed itself. DeviceKit-disks is now to be known as udisks...

    http://www.phoronix.com/vr.php?view=Nzc2NA

  2. #2
    Join Date
    Aug 2007
    Posts
    437

    Default

    I'm totally supportive about this. One of the biggest strength of OSS / Linux world is people is willing to, and have the guts to make fundamental changes to things that needs to be changed. If sure creates a short term pain, but in the long run this is a good thing. Far better than the Microsoft's backward compatible to every single bug approach. Look at Win32API, it's designed about 20 years ago and still have to be supported in latest Windows 7. That's a shit load of burden Microsoft is laid on itself.

  3. #3
    Join Date
    Jul 2008
    Location
    Greece
    Posts
    3,762

    Default

    That's why people pay them.

  4. #4
    Join Date
    Mar 2009
    Location
    Hellas
    Posts
    1,000

    Default

    Quote Originally Posted by RealNC View Post
    That's why people pay them.
    Yup their benefit and their curse altogether...

  5. #5
    Join Date
    Sep 2006
    Posts
    714

    Default

    If it's just for the sake of a name change it's the most goddamn stupid thing I've ever heard in my life. That is if it works the same and applications can interact with it the same way it's completely irrelevant that it is not using separate daemons anymore and thus the namespace change is just mind numbly stupid.

    There is justification for a change if the change is a improvement in basic design, but otherwise change is bad.

  6. #6
    Join Date
    Nov 2007
    Posts
    1,024

    Default

    Quote Originally Posted by drag View Post
    If it's just for the sake of a name change it's the most goddamn stupid thing I've ever heard in my life.
    Yup. Next release cycle the devs will decide that udisks makes it sound too much like a kernel integration library and rename it idisks. Then the next development cycle they'll realize that sounds too much like an Apple product and rename it to ldisks. Then the cycle after that they'll decide to rename it to something catchy like PhazerGunDiskTechnologyPro.

    But hey, at least it'll be _less confusing_!

    Open Source at its finest.

  7. #7
    Join Date
    Oct 2008
    Posts
    2,908

    Default

    While I agree that breaking a bunch of code just for a name change is really stupid, let's be fair here. The release notes still have this highlighted near the top:

    NOTE NOTE NOTE: This is an unstable release of DeviceKit-disks, all API is subject to change.
    And it's brand new with barely anyone really using it yet, so now is definitely the time to make changes like this rather than waiting around.

    Most people wouldn't complain if a new project still in alpha status changed some of it's API, it's just the fact that this project is so central to the Linux system and the fact that what it's replacing has already entered a feature freeze that people take notice.
    Last edited by smitty3268; 12-02-2009 at 10:51 PM.

  8. #8
    Join Date
    Aug 2007
    Posts
    437

    Default

    Well, if it wasn't for the sake of perfectionist OSS would not have gone this far.

    Why rip the whole goddamn graphics stack while the previous one works fine and we would just patch it and patch it till it fulfills the need?

    Name changing, if done on early stage can greatly reduce future confusion. Remember why electrons have negative charge, and it comes out from negative pole of a battery and goes in to the positive pole, while the electric 'flow' is the other way around? People tends to be damn lazy and in the end live with the shit they left over forever. If you are happy with this, Windows suits you better, they just keeps patching, building one layer on top of another but never go back to improve the basement. .NET is based on Win32, even a lot of its libraries reflects that, is that what you want to see, to use? Then Windows is better for you again.

  9. #9
    Join Date
    May 2009
    Posts
    10

    Default Portability?

    reliant upon libudev and libgudev
    I thought DeviceKit (as it were) had an interest in producing portable code. That all the OS specific code was abstracted and thus allowing other OS's (such a FreeBSD) to easily leverage the common infrastructure.

    It appears to me that, being so closely tied to udev that this is no longer possible?

  10. #10
    Join Date
    Aug 2009
    Posts
    2,264

    Default

    Quote Originally Posted by FunkyRider View Post
    Why rip the whole goddamn graphics stack while the previous one works fine and we would just patch it and patch it till it fulfills the need?
    Because X has suffered from good FLOSS graphics drivers, fast FLOSS graphics drivers, more capable FLOSS graphics drivers and wait for 3 years to get your GPU do 3D FLOSS graphics drivers.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •