On Thu, 21 Feb 2008, Thomas De Schampheleire wrote: > On Thu, Feb 21, 2008 at 10:18 AM, Robert P. J. Day > <rpjday@xxxxxxxxxxxxxx> wrote: > > > > On Thu, 21 Feb 2008, Thomas De Schampheleire wrote: > > > > > On Wed, Feb 20, 2008 at 9:07 PM, Brandon Philips <brandon@xxxxxxxx> wrote: > > > > On 13:17 Wed 20 Feb 2008, Robert P. J. Day wrote: > > > > > i just picked up a logitech quickcam communicate STX webcam for my > > > > > fedora 8 system, and was pointed in the direction of: > > > > > > > > > > http://mxhaard.free.fr/download.html > > > > > > > > > > from where i grabbed the gspcav1-20071224.tar.gz tarball, > > > > > uncompressed, build module, installed and the webcam simply worked. > > > > > so the question is -- is that software an *alternative* to what's in > > > > > the kernel right now in terms of spca-related drivers? > > > > > > > > The gspca author has never submitted the driver to the mainline Kernel. > > > > That is why it isn't in regular 2.6 releases. > > > > > > Indeed. There has been a discussion on this on the gspca mailing list > > > a while ago: > > > http://lists-archives.org/spca50x-devs/01167-adding-gspca-to-the-mainline-kernel.html > > > > > > The authors have several objections against mainlining the driver... I > > > don't think this will change soon. > > > > for curiosity's sake, i'm going to peruse that discussion to see what > > the objection is but, based on a quick glimpse, i have a bad feeling > > ... > > http://sourceforge.net/mailarchive/message.php?msg_id=20060720124839.64364.qmail%40web34707.mail.mud.yahoo.com > > > > "This module will probably _never_ be in mainstream kernel, since > > development of kernel and module works at very different speeds. > > spca5xx contains alot of (relativly) untested code which would never > > be permitted in the kernel, and when finally permitted the code in > > question would probably be obsolete." > > > > i'm sorry ... you don't want to mainline the code because you don't > > want to *test* large portions of it? are all of the other arguments > > that feeble? yeesh. perhaps someone can put the children to bed and > > let the grownups run things over there. > > Be careful when making such remarks. I think that parts of the > 'untested' code the quote is referring to, add support for new > webcams. Since the number of gspca developers is very low (mainly > Michel Xhaard, plus few others), one cannot expect that they have all > this hardware. Testing is then not possible, and one has to rely on > some users giving test reports. > Although I agree that mainlining the code would be a good thing, and > can produce more test and bug reports, I think making remarks like > this will not help convincing the developers. while your points are well taken, that's still not a convincing argument for refusing to mainline what is certainly going to be popular code. if development is really proceeding that quickly, then the proper approach would be to at least mainline the basic infrastructure and the drivers that appear to be well-tested and reliable, and make the more recent content available perhaps as a kernel patch. that way, everyone's happy. but i have no sympathy for anyone whose argument seems to be, "we're just too busy for that." and i doubt that my opinion is going to sway anyone one way or the other. it seems that the developers have already made up their minds on this one. rday -- ======================================================================== Robert P. J. Day Linux Consulting, Training and Annoying Kernel Pedantry: Have classroom, will lecture. http://crashcourse.ca Waterloo, Ontario, CANADA ======================================================================== -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
