[libvirt] [PATCH] New features implemented in hyperv libvirt driver (Bull)

[Yves Vinter]

Hi Eric,

We initially commented the code with our company name to help us to identify it easily internally.
I totally agree that this comment should not appear in the upstream code; I apologize about that.
No problem; I will make another submission after a code cleaning.

Thanks,
Yves.

-----Message d'origine-----
De : Eric Blake [mailto:eblake at redhat.com] 
Envoyé : vendredi 12 septembre 2014 21:13
À : Yves Vinter; libvir-list at redhat.com
Objet : Re: [libvirt] [PATCH] New features implemented in hyperv libvirt driver (Bull)

On 09/12/2014 12:55 PM, Eric Blake wrote:
> On 09/12/2014 10:36 AM, Yves Vinter wrote:
>> Authors: Simon Rastello (Bull), Adrien Kantcheff (Bull), Yves Vinter 
>> (Bull)
>>  

>> @@ -58,12 +64,97 @@ hypervFreePrivate(hypervPrivate **priv)
>>          wsmc_release((*priv)->client);
>>      }
>>  
>> +	/* Bull */
>> +  	pthread_mutex_destroy(&(*priv)->mutex);
>> +
> 
> What are all the "Bull" comments for?

Maybe I should expand my question.  After a bit of though, I can see that you are using the comment "Bull" to call out sections of code contributed by your company (named Bull), perhaps to make it easier for you to internally identify code that you have touched but which has not yet been merged upstream.  However, such comments add NOTHING to the upstream code itself.  Worse, your company name happens to have an unfortunate negative connotation: at least in colloquial English, if I tell someone a story, and they then respond with a single-word sentence "Bull", they are implying that they think my story is a work of fiction or otherwise sub-par.  So on my first read of your code, that connotation caused me to interpret your single-word comment as "the following code is a joke or otherwise sub-par", even though that is obviously not the intent.  By the way, "Bull" also has a nice connotation of strength and growth, as in a bull market, which I'm sure is more in line with why your company picked that name - it's just that I don't ever hear a single-word sentence tied to that definition.

Let's step back a bit.  We have 'git blame' to say who contributed a piece of code, so if you are worried about finding your contributions, it's easy to use the version control system for that - but that is external to the code.  Meanwhile, comments in the code base are most useful when they explain interface contracts or tricky code constructs, and not when they are just trying to claim attribution.  Therefore, I suggest that when you resubmit, you eliminate the attribution comments, and only leave contract or explanation comments.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org