Re: posible bug in daemontools

On 15 March 2010 03:25, Mark Johnson <[email protected]> wrote:
> On Sun, Mar 14, 2010 at 9:25 PM, Chris Pugh <[email protected]> wrote:
>
>> Suggestions:
>> ------------------
>>
>> Youthful enthusiasm or not, you might wish to lose the somewhat glib
>> and smart-ar*e style comment, state license information correctly, and
>> also maybe follow Dr Nerstein's mildly put recommendations?
>>
>>    http://cr.yp.to/distributors.html
>
> There is no license.  The djb derived bits are not copyrighted.

They are in the 'Public Domain'.  That covers a wide area.  Being in
the Public Domain could even be considered a form of Licensing!

> You're the first to complain about the attempt at humor.  If I get
> further complaints, I will consider removing it.

Poor attenpt if I may say so. ;o)   It wasn't really meant as a
complaint, I was merelky expressing an opinion.   You are at liberty
to add or remove, as you so please.

> As to recommendations, I take it you're referring to quote below?

>  'Please take time to ensure that your distributions of my software
> support exactly the same interface as everyone else's
>..
> frivolously  break scripts that work everywhere else.'

> Up to this point, I believe I have failed to honor that request only in the case of >daemontools (since my qmail fork is not yet operational).  I'm not a fan of slashpackage.

Again your choice.  I'm not sure I'm a die-hard fan either, especially
of the more extreme Package Managers out there, but, it is still an
interesting, logical and useful way of doing things.  To be expected
of a Mathmatician, eh?

> If you are, feel free to fork my fork and do it Dan's way.  I have no interest in doing so
> or arguing about it.

Neither have I, and I could indeed fork it Dan's way.  Might be a
useful learning exercise. ;o)

>> Why bite the hand that feeds you?
>
> I don't believe I am.

Maybe just a tad more respect is due then, for the guy who wrote the
original, and gave you the materail to fork with ( feel free to morph
the penultimate word! )? ;o)

>> If you are patching where you think appropriate, you might also wish
>> to give the user the option to sequentially apply ( or not ), rather
>> than simply embedding them in the source.
>
> The patches don't all play nice with each other.  I only want to
> integrate them once.  Not once for every combination somebody might be
> interested in.  Note that some of the stuff I've added can be enabled
> (or disabled) at run time.  Most (but not all) of the rest is stuff
> where I (and others) consider the stock behavior incorrect or broken.

I'm aware of that.  All the more reason to make 'em!  I have yet to
compile your fork btw.

As to 'incorrect' or 'broken' and/or 'non-functional', there's enough
flaming on this list as to that!

Cheers,


Chris.