svn commit: r327005 - in head: sbin/ipfw sys/sys usr.sbin/watch

[Pedro Giffuni]

On 19/12/2017 21:55, Rodney W. Grimes wrote:
>>
>>> On Dec 19, 2017, at 19:15, Rodney W. Grimes <freebsd at pdx.rh.CN85.dnsmgr.net> wrote:
>>>
>>>> Author: pfg
>>>> Date: Tue Dec 19 22:40:16 2017
>>>> New Revision: 327005
>>>> URL: https://svnweb.freebsd.org/changeset/base/327005
>>>>
>>>> Log:
>>>>   SPDX: These are fundamentally BSD-2-Clause.
>>>>
>>>>   They just omit the introductory line and numbering.
>>> I again must assert that it would be better to not apply an SPDX than to
>>> apply one that is not an exact match for the license.
>>>
>>>
>> Not being a lawyer, I would normally agree, however:
> This isnt about any legal issue.

Only legal issues matter. If there is even a hint of a legal basis to 
revert the change, I won't object.

>> 1) SPDX IDs are only advisory: we always keep the exact license text, which is what has legal value.
> And we should do our best to provide the most accurate advisory we can,
> and we know that this is not a direct copy of the BSD 2 clause, so making
> advice that it is, IMHO, would be poor advice.
>
>> 2) The license is detected by license scanners as BSD and it has two clauses so the description fits.
> And a human reading it sees it reads like a 2 clause but does not match a 2 clause exactly so
> how do I trust any of this SPDX stuff as being done with some ration of sanity.
>
>> FWIW, according to SPDX lawyers, the numbering is not relevant and it would appear to me that the phrase:
>>   "Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:? doesn?t add any information to the two clauses, which read:
> Again, not a legal issue, an issue of this is not an exact BSD-2-Clause so I
> do not believe that we should tag it as such.  Basically if we are doing this
> "close enough" thing it means the SPDX tags are actually pretty useless for
> anyone trying to do a legal evaluation cause they are just going have to
> completly redo what was done in adding the SPDX tags, and if that is the
> case we should seriously consider just what value do these have in the
> tree?

I see it the other way around:

If the text matches exactly the common license text, then the tag is 
redundant (which is not necessarily wrong but doesn't imply much value).

Yes, there is some amount of judgement being done on my part: it would 
certainly be wrong from to tag some file as BSD-2-Clause when it is an 
MIT license under the reasoning that they are "basically" the same thing 
but I think I am being sufficiently reasonable by setting the best 
possible match. As I said the SPDX guys do consider non-substantial 
differences are OK.

What I will do .. just to be safe, is to check with the SPDX guys to see 
what they think.

Thanks,

Pedro.

>> Redistributions of source code must ?
>> Redistributions in binary form must ...
>>
>> Pedro.
>>
>>
>>
>>>> Modified:
>>>>   head/sbin/ipfw/altq.c
>>>>   head/sbin/ipfw/dummynet.c
>>>>   head/sbin/ipfw/ipfw2.c
>>>>   head/sbin/ipfw/ipfw2.h
>>>>   head/sbin/ipfw/ipv6.c
>>>>   head/sbin/ipfw/main.c
>>>>   head/sbin/ipfw/nat.c
>>>>   head/sys/sys/msg.h
>>>>   head/sys/sys/snoop.h
>>>>   head/usr.sbin/watch/watch.c
>>> ...
>>>
>>> -- 
>>> Rod Grimes                                                 rgrimes at freebsd.org