[Developers] Re: patch: CactusEinstein/IDAxiBrillBH cleanup

[Tom Goodale]

On Fri, 20 May 2005, Erik Schnetter wrote:

> On Friday 20 May 2005 14:31, Tom Goodale wrote:
>> It all looks good, please go ahead, although as you note at the
>> bottom it would be good if you could split the commits into smaller
>> units, each addressing an individual thing, e.g.
>>
>>       relabeling latex
>>       adding comments to param.ccl
>>       adding the interpolator parameters
>>       adding new output (I agree that it should really be a GF so we
>> can use standard Cactus IO methods)
>>       add debugging stuff
>>       adding new stuff to the documentation
>>       etc
>>
>> In general, small, specific patches will get quicker attention, and
>> thus lower latency, than big patches which do lots of different
>> things.  If you'd sent a note out to developers that you were doing
>> this, and followed with a sequence of patches as you did things, it
>> would have been easier.
>
> This is not how development works.  The original problem is that the
> physics output is wrong, and then one usually doesn't start to send
> patches until the physics output is right.  At which time many
> different things have accumulated.
>
> Taking things apart takes a lot of time, and is in my opinion not worth
> the effort.  It would be worth the effort only if one wants to go back
> in time.  But since the thorn contained an error, this is quite
> unlikely.  Additionally, many small patches depend on each other, and
> that leads to other problems.
>
> Also, Jonathan probably didn't send out a note to developers because his
> intention was to get correct physics.  Had thorn IDAxiBrillBH turned
> out to be unfeasible for the job, then he would have dropped his
> changed and used / written a different thorn.  "Grabbing" a thorn
> exclusively is against our workflow.
>
> We have an old version and a new version of the thorn.  You can take the
> two versions and create a patch from the difference.  Artificially
> creating new versions in between is a waste of effort.

There is a difference between adding new features/documentation, fixing 
bugs, and fixing documentation, and it would be good if that could be 
maintained in the commits/patches.  I have no desire to make lots of extra 
work, however I would like commits to be as specific as possible.

In this specific case, Jonathon has made no mention of fixing physics 
bugs, and presented it as a general cleanup, so your argument does not 
apply at all.  In fact Jonathon says he has been working on this for a 
month or so, and thus could at any point have sent a message to the list 
that he was working on this, and perhaps have sent intermediate patches 
for easily modularised things.  I've just gone back through the archive 
and don't see any such message.

It is great that he has done this work, however in general it is a good 
idea to send a bit of info out if you are spending considerable time 
changing public code and think you're going to submit a patch, just so we 
can track what's going on, rather than it appearing out of the blue.

Tom