1 |
On Thu, Oct 11, 2018 at 9:25 PM Sarah White <kuzetsa@××××××××××.ovh> wrote: |
2 |
> |
3 |
> assuming "commit authors are copyright holders" shouldn't be |
4 |
> trusted blindly (the git log doesn't "solve everything") |
5 |
> |
6 |
|
7 |
Sure, but neither can any other conceivable log, and anything that |
8 |
could be put in such a log could also be put in a commit annotation, |
9 |
which IMO will track it a heck of a lot better than a random text |
10 |
file. |
11 |
|
12 |
> |
13 |
> Does this mean "gentoo authors" will appear in court when |
14 |
> there's infringement? This is not a rhetorical question. |
15 |
> |
16 |
|
17 |
Simply being named in a copyright notice creates no particular |
18 |
obligation to appear in a court. |
19 |
|
20 |
If there is some kind of copyright lawsuit, then presumably those most |
21 |
associated with the code in question could be called as witnesses, |
22 |
assuming whoever is doing the suing can be bothered to track them |
23 |
down. |
24 |
|
25 |
Ultimately though if Gentoo gets sued by somebody, the burden of proof |
26 |
is actually on them to prove that THEY owned the copyright. Gentoo's |
27 |
burden would probably be to show that reasonable care was exercised |
28 |
over accepting code, and that infractions were dealt with in a |
29 |
reasonable manner when brought to our attention. |
30 |
|
31 |
Really though, that is about the best we can do anyway. If we want to |
32 |
suppose that somebody can make us do more than what is "reasonable" |
33 |
then we might as well give up on FOSS. |
34 |
|
35 |
> |
36 |
> This is a very useful notice, because if someone has a snapshot |
37 |
> tarball, or other non-git copy of particular source files, the |
38 |
> lack of a proper notice is a legal problem (see below) |
39 |
|
40 |
Just stick it in git. If somebody wants to know the history, they can |
41 |
look it up in git. If somebody wants to go redistributing snippets of |
42 |
the file, then complying with the law is their problem, not ours. |
43 |
|
44 |
> |
45 |
> - This ignores the purpose: copyright notices are to make |
46 |
> sure when someone gets "free code", they don't assume it's |
47 |
> free for any/all purposes with zero restrictions: |
48 |
> |
49 |
|
50 |
Our notices already make this clear, regardless of whose name is listed. |
51 |
|
52 |
> the infringer could say the copyright status (copyleft |
53 |
> uses copyright law for enforcement) wasn't apparent due |
54 |
> to the lack of a clearly formatted copyright notice... |
55 |
|
56 |
The required format of a copyright noticed under US law is explictly stated: |
57 |
|
58 |
(b)Form of Notice.—If a notice appears on the copies, it shall consist |
59 |
of the following three elements: |
60 |
(1)the symbol © (the letter C in a circle), or the word “Copyright”, |
61 |
or the abbreviation “Copr.”; and |
62 |
(2)the year of first publication of the work; in the case of |
63 |
compilations, or derivative works incorporating previously published |
64 |
material, the year date of first publication of the compilation or |
65 |
derivative work is sufficient. The year date may be omitted where a |
66 |
pictorial, graphic, or sculptural work, with accompanying text matter, |
67 |
if any, is reproduced in or on greeting cards, postcards, stationery, |
68 |
jewelry, dolls, toys, or any useful articles; and |
69 |
(3)the name of the owner of copyright in the work, or an abbreviation |
70 |
by which the name can be recognized, or a generally known alternative |
71 |
designation of the owner. |
72 |
|
73 |
> ... so something generic like "gentoo authors" can be |
74 |
> difficult to enforce - I've not seen case law on this. |
75 |
|
76 |
Our copyright would be completely enforceable even if we had no notice at all. |
77 |
|
78 |
However, if Gentoo sued somebody for infringement, then the defendant |
79 |
could attempt to claim that the infringement was innocent (ie they did |
80 |
not know the work was copyrighted). IMO that would be an uphill |
81 |
battle. If the court decides our notice complies, then they are |
82 |
required by law to not give any credence to such an argument. If they |
83 |
decide that it doesn't completely comply, then they would still weigh |
84 |
the argument, and how plausible is it that a court will buy that you |
85 |
didn't know it was copyrighted when Copyright 2018 Gentoo Authors is |
86 |
on the first line of the file? |
87 |
|
88 |
And none of this will ever matter at all unless Gentoo files a |
89 |
lawsuit. Right now we can't seem to file our taxes. How likely do we |
90 |
think it is that Gentoo will be filing a lawsuit as the plaintiff? |
91 |
The notice doesn't matter at all if somebody sues us. |
92 |
|
93 |
> |
94 |
> I believe licensing is the reason for GCO, not copyright |
95 |
> attribution. Language about committer, acked or signed-off, |
96 |
> and copyright holder VS licenses and GCO, all within the |
97 |
> same |
98 |
|
99 |
Sure, that is completely true. The GCO/DCO/etc has nothing to do with |
100 |
attribution. |
101 |
|
102 |
> GLEP (#76) adds a lot of confusion because copyright |
103 |
> is barely mentioned, and not in a clearly defined way. |
104 |
|
105 |
What is unclear about it? |
106 |
|
107 |
It used to be much more prescriptive. However, that was considered to |
108 |
be too inflexible, and other prominent projects (like Linux) don't |
109 |
seem to require this. The importance of a very specific notice also |
110 |
did not really seem to be sufficient to create controversies when |
111 |
contributors felt they could not comply with a very specific |
112 |
requirement. |
113 |
|
114 |
> Copyright should be treated as a separate issue from GCO. |
115 |
|
116 |
Well, both deal with copyright, but I agree that the notice is |
117 |
separate from the GCO, and they are in fact mentioned separately. |
118 |
|
119 |
> |
120 |
> Do "gentoo authors" file a lawsuit when there's infringement? |
121 |
|
122 |
You do not need to be listed in a copyright notice to file a lawsuit. |
123 |
You merely need to have authorship. |
124 |
|
125 |
> How does jurisdiction work when the only thing which can |
126 |
> be known for certain is: "someone claimed the commit they |
127 |
> wrote was FOSS/Libre & they signed-off with a GCO line" |
128 |
|
129 |
Ultimately anybody wanting to file a lawsuit has to prove the |
130 |
underlying facts. If you want to sue somebody for copyright |
131 |
infringement you have to demonstrate to the court that you wrote |
132 |
whatever you're suing over. |
133 |
|
134 |
If we were in the business of selling software and were more likely to |
135 |
be suing people regularly, then I'd certainly agree that a lot more |
136 |
rigor could be used to capture proof of ownership. This would |
137 |
presumably pay for itself as a cost of doing business. However, that |
138 |
isn't the kind of organization we are in. Mostly we just want to show |
139 |
reasonable care, and to be responsible in general. That helps keep us |
140 |
from getting sued. I'm skeptical that Gentoo would ever sue anybody. |
141 |
|
142 |
> Does this mean GCO sign-off lines obligate the contributors |
143 |
> to respond whenever FOSS/Libre legal issues come up? |
144 |
|
145 |
No. Your obligation to respond to a court is established in your |
146 |
local laws. In many places you can be called as a witness against |
147 |
your will if you have knowledge of a case, whether you sign anything |
148 |
or not. I don't think that is particularly likely to happen here, and |
149 |
it is especially unlikely outside of your local jurisdiction, and most |
150 |
courts do have rules to avoid placing unreasonable burdens on |
151 |
witnesses. |
152 |
|
153 |
IMO signing the DCO/GCO/etc probably reduces the likelihood of being |
154 |
called to testify simply because your testimony is already a matter of |
155 |
public record (well, maybe aside form some notarized statement |
156 |
affirming that you signed it). |
157 |
|
158 |
> If there was a FLA policy in place, and gentoo formally held |
159 |
> itself out to protect (as a fiduciary) any FOSS/Libre interests |
160 |
> of the contributors; gentoo needs to hold the copyright, and |
161 |
> more importantly: invest in policy and planning to legally |
162 |
> protect FOSS/Libre interests when any infringement occurs. |
163 |
|
164 |
Actually, the FLA as written by the FSFe explicitly does NOT assign copyright. |
165 |
|
166 |
> the protection should be proper: a real entity. changing |
167 |
> the language from foundation to authors and treating it |
168 |
> like it's still a copyright assignment is pointless |
169 |
> unless the simplified attribution still assigns the |
170 |
> copyright to the gentoo foundation hold copyright. |
171 |
|
172 |
The intent is not to require assignment of copyright. We were |
173 |
actually considering rolling out the FLA in parallel (voluntarily), |
174 |
but this was felt to be making the GLEP even more complex. |
175 |
|
176 |
> GLEP 76 shouldn't try to be an umbrella for multiple things. |
177 |
|
178 |
I guess we should consider that when we approve it. Oh wait, it is |
179 |
already approved. I guess when you author the next one you can take |
180 |
that into account. :) |
181 |
|
182 |
It isn't perfect. IMO it belongs together. However, others might |
183 |
disagree. It only took about half a decade to finish. I'm sure |
184 |
somebody willing to put enough time into it will surpass it. Then |
185 |
they can go on to real challenges like filing the Foundation's taxes. |
186 |
|
187 |
In any case, I'm not convinced that copyright notice is really worth |
188 |
THAT much fighting over. I wasn't a huge fan of "Gentoo Authors" |
189 |
either, but in the end I realized that notices are fairly overrated. |
190 |
I think that the GLEP accomplishes what it ought to on this front. |
191 |
Any requirement you add to it is just going to create another group of |
192 |
devs who feel they cannot comply with it. Heck, the current policy |
193 |
basically allows almost any notice that complies with US law and there |
194 |
is still some concern. |
195 |
|
196 |
-- |
197 |
Rich |