1 |
Greg KH posted on Thu, 21 Feb 2013 11:55:34 -0800 as excerpted: |
2 |
|
3 |
> On Thu, Feb 21, 2013 at 07:33:48PM +0100, Ulrich Mueller wrote: |
4 |
>> >>>>> On Thu, 21 Feb 2013, Greg KH wrote: |
5 |
>> |
6 |
>> >> Ulrich Mueller (ulm) wrote this on the 16th: |
7 |
>> >> |
8 |
>> >> > Look into the WHENCE file and be horrified. Taking just the first |
9 |
>> >> > ten items (of a total 114): |
10 |
>> >> > |
11 |
>> >> > Unknown license (3 times) |
12 |
>> |
13 |
>> > Which ones specifically? |
14 |
>> |
15 |
>> Driver: snd-korg1212 -- Korg 1212 IO audio device Driver: kaweth -- USB |
16 |
>> KLSI KL5USB101-based Ethernet device Driver: dvb-ttusb-budget -- |
17 |
>> Technotrend/Hauppauge Nova-USB devices |
18 |
> |
19 |
> As these originally came from the kernel source tree, they are "by |
20 |
> default" ok. |
21 |
|
22 |
Well, not entirely. It's exactly this sort of issue that's the reason |
23 |
many people were uncomfortable with firmware in the kernel at all, and |
24 |
why all the ongoing work to separate it out. |
25 |
|
26 |
>> >> > GPL, but without source (3 times) |
27 |
>> |
28 |
>> > Really? Which? |
29 |
>> |
30 |
>> Driver: ambassador -- Madge Ambassador (Collage PCI 155 Server) ATM |
31 |
>> NIC. |
32 |
>> Driver: snd-maestro3 -- ESS Allegro Maestro3 audio device Driver: |
33 |
>> qla1280 - Qlogic QLA 1240/1x80/1x160 SCSI support |
34 |
> |
35 |
> Some of these came from the kernel source tree originally, others don't, |
36 |
> but they all imply that the GPL really isn't for the firmware itself. |
37 |
> Odd. |
38 |
|
39 |
Again... precisely why this stuff's being gradually kicked out of the |
40 |
kernel in the first place. |
41 |
|
42 |
>> >> > "All rights reserved" |
43 |
>> |
44 |
>> > That's not an issue, unless it is alone, is there something else in |
45 |
>> > the license as well? |
46 |
>> |
47 |
>> Driver: snd-ymfpci -- Yamaha YMF724/740/744/754 audio devices |
48 |
>> |
49 |
>> According to WHENCE, it is: |
50 |
>> "Copyright (c) 1997-1999 Yamaha Corporation. All Rights Reserved." |
51 |
>> Nothing else. |
52 |
> |
53 |
> That's a copyright notice, not a license, so I don't know what to |
54 |
> suggest :) |
55 |
|
56 |
See above... |
57 |
|
58 |
>> >> > BSD, without source |
59 |
>> |
60 |
>> > There's no problem with that. |
61 |
>> > |
62 |
>> Driver: advansys - AdvanSys SCSI |
63 |
>> |
64 |
>> Right, and it's the only one out of the first ten that we're allowed to |
65 |
>> redistribute. |
66 |
|
67 |
=:^) |
68 |
|
69 |
>> >> > Right for redistribution not granted |
70 |
>> |
71 |
>> > Huh? Which? |
72 |
>> |
73 |
>> Driver: smctr -- SMC ISA/MCA Token Ring adapter |
74 |
> |
75 |
> Token ring drivers were dropped from the kernel already, so this isn't |
76 |
> an issue. |
77 |
|
78 |
It's only an issue to the extent that we're still shipping it. |
79 |
|
80 |
>> >> > "Permission is hereby granted for the distribution [...] as |
81 |
>> >> > part of |
82 |
>> >> > a Linux or other Open Source operating system kernel" |
83 |
>> |
84 |
>> > What is wrong with that? We happen to be distributing a Linux |
85 |
>> > operating system. |
86 |
>> |
87 |
>> Driver: keyspan -- USB Keyspan USA-xxx serial device |
88 |
>> |
89 |
>> We distribute it in a separate package. And it doesn't say "part of an |
90 |
>> OS" but explicitly "part of a kernel". |
91 |
> |
92 |
> Ah, that's because at the time, that's the way it was originally |
93 |
> distributed. Given that the company isn't around anymore, I don't think |
94 |
> this is going to be an issue :) |
95 |
|
96 |
But what about whoever bought up the rights? In practice, that's |
97 |
precisely when many of these things BECOME an issue, when a new owner |
98 |
decides they can monetize... |
99 |
|
100 |
|
101 |
In general, this is an ongoing problem for the entire community. It's |
102 |
gradually being straightened out, but it's a years-long, likely decades |
103 |
long, project. |
104 |
|
105 |
In practice, as long as nobody's suing, and because the overall trend is |
106 |
to clean things up, that's why most distros kind of wink and nudge and |
107 |
don't talk about it much. |
108 |
|
109 |
But it's also one of the BIG reasons for the "firmwareless kernel" |
110 |
projects. While they're not entirely practical for most people on their |
111 |
own, they DO serve the purpose of demonstrating that it's possible under |
112 |
limited circumstances and measuring how far we have to go... |
113 |
|
114 |
-- |
115 |
Duncan - List replies preferred. No HTML msgs. |
116 |
"Every nonfree program has a lord, a master -- |
117 |
and if you use the program, he is your master." Richard Stallman |