On Tue, Jul 17, 2012 at 8:22 PM, Freddie Chopin <[email protected]> wrote:
> W dniu 2012-07-17 16:14, Andreas Fritiofson pisze:
>> Speak for yourself, I would certainly not consider avoiding giving
>> negative review because of a rule like this.
>
> "Rule" is a wrong word here - I'm not a native speaker in English, so
> maybe I have problems expressing what exactly I meant by this patch -
> it's really just a "last resort" solution when the person who gave
> negative review cannot be reached and does not re-review the updated
> change. It should not be seen as a "hard rule" which gets executed with
> sub-second precision after the timeout passed...

Well, I too meant rule in a looser sense. One you have the possibility
to use. Or take another route. Or wait a bit longer. You wrote "may
use" so I think it was clear.

> The only thing MPSSE patch changed in rest of the code is use of
> libusb-1.0 for stlink and jlink (probably) - not a big problem imho.

And that is only if it's configured in with --enable-ftdi.

>> Seeing that the community lacks the energy to review (architecturally
>> and not style-wise) significant patches (and contributing in pre-patch
>> discussions), I'm kind of put off by the work required.
>
> I don't care much about the style, as long as it's not a complete mess (;

Style and details are important, but for the first version of a
(large, non-trivial) change, a contributor wants feedback primarily on
the big ideas of the change, not on details. Style can be easily
adjusted whenever. A complete rewrite is not fun to have to make after
several rounds of style-fixing work.

> First of all I'm not a PC programmer. I also don't think it's the right
> thing to do to change someones patches - in "major" things, "minor"
> stuff would probably be OK (like order or patches, some typos, marking
> parts of code as tested when they were marked as experimental, etc.).

Well, I agree (both as reviewer and author) that it's a bit taboo to
mess with other's patches. On the other hand patch sets in gerrit can
be viewed as suggestions. Anyone can make a suggestion for a slightly
different version of a patch. It's not necessarily the latest patch
set that must to be the one that gets submitted. If everyone agrees on
that view, it's easier to allow yourself to push an improved version
of someone else's patch.

There may also be legal issues involved, as it occurs to me now.
Patches do not generally have to be licensed as GPL until they are
merged with OpenOCD, thereby creating a derived work. So it's not safe
to assume that anyone has a right to use patches in gerrit as they see
fit. The author has pushed a work to gerrit for the purpose of merging
it into OpenOCD, and has thereby implicitly licensed it for inclusion
into mainline OpenOCD, where that derived work will be licensed as
GPL. If the patch is not accepted, nothing says that it's legal to use
the patch for anything else, including creating new versions (derived
works) of the patch and merging that with OpenOCD, or using it in
another project.

Of course, IANAL and all that garbage.

/Andreas

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
OpenOCD-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openocd-devel

Reply via email to