Limits to what can be done without source

[Andy Green]

Les Mikesell wrote:
> On Wed, 2006-05-17 at 15:10, Andy Green wrote:
> 
>> Except that users of binary blobs are often in for long waits to get a 
>> new version with their bug fixed.
> 
> As opposed to?  When will we see an SATA driver that passes errors
> correctly up to the md layer so a software raid would be
> safe to use?  Firewire drives were broken for a long time by an
> update in FC3 and I'm not sure it's completely back yet.  Long
> waits aren't unique to binaries.

But they are unique to binaries for simple fixes.  The SATA driver case 
may well be sorted in later kernels already... as a wise man said ''If 
something was ever done right and doesn't work now, that's the problem. 
  There are other problems in doing it right the first time.''

>> Hum OSX runs on a tiny landscape of platforms, all documented and 
>> understood by the OS vendor.
...
> That other vendor covers a lot of equipment too. They aren't bug
> free, but I don't think you're going to win a current comparison
> at the device driver level on either features or reliability. I've
> intentionally limited the hardware I've tried to run under
> Linux and still have had problems with mainstream things like
> adaptec scsi controllers and Dell's mylex raid cards.

Well I don't want to get into defending Linux against Brand X, I am sat 
here using Linux in effective ways and it is highly reliable.  If a 
given thing is broken it's nothing that can't be worked around in the 
same way workarounds are needed for broken areas in everything else I 
use.  With a FOSS solution if I need to get dirty with a workaround then 
I can find source everywhere I am interested and sometimes that makes 
all the difference, particularly when it is running on custom hardware. 
  With a binary solution I am just another customer in line with a 
trouble ticket.

>> What I have had -- am having right now typing this -- a good experience 
>> with is the Xorg nv driver.  For me it works very well and has always 
>> done so for the functionality it claims to support.  It seems a good bet 
>> to me that if nVidia donated the sources it currently keeps private into 
>> that project then things will go on in the same smooth way but with 
>> complete support. 
> 
> I'm inclined to believe nVidia's claim that they can't disclose
> that information because of contracts with their suppliers.  So
> that's not one of the choices.

Neither of us know what nVidia's choices are, but my point was that the 
FOSS equivalent in Fedora has a good track record and works smoothly, 
whereas working with the deliberately opaque binary causes much support 
grief.  Were nVidia to donate sources as suggested I don't see the nv 
driver being destabilized but solving the whole issue.

>>  Not least because Fedora would ship with it all in 
>> the box.
> 
> That part is all politics since nVidia permits redistribution
> of their binaries and it is fedora's choice not to do so.

Yes as it stands Fedora bans such nonfree confections.  I guess one can 
imagine nobody got involved with FOSS to increase the reliance of the 
world on nVidia's binary blobs except maybe nVidia people themselves.

We don't seem to be disagreeing about much.

-Andy
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 4492 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://listman.redhat.com/archives/fedora-list/attachments/20060517/faffd5ad/attachment-0001.bin>