Thoughts on Services development (was Re: [IRCServices] Can't even get it to build on OSX 10.3.8)

[Andrew Church]

[This is a copy of the message I sent to David M in response to his private
 message to me.  I will not post that message, as doing so would violate
 etiquette, but I feel it important to make my thoughts on Services
 development known, for the benefit of others who may have similar
 concerns.  I have elided portions of my message irrelevant to that topic.]

     [...]  From my viewpoint (subjective, certainly, but everyone's
viewpoint is subjective by definition), you have accused me of carelessness
without taking the time to fully understand the issues involved.  Despite
the fact that IRC Services is merely a free-time project of mine, I have
taken considerable care to ensure that it is compliant with modern
standards, as described in section 2-1 of the manual, and I try to ensure
that it will compile and run on commonly used operating systems.  However,
I lack both the time and the resources to check its behavior in all
environments--in fact, all I have at my disposal is one computer which can
run either Linux, OpenBSD, FreeBSD, or Windows.  As a result, I have to
rely on reports from users to resolve any problems on other systems.

     I had hoped my previous message had demonstrated that I was quite
willing to admit my error, but it regrettably seems that that is not the
case.  Nonetheless, I am still open to receiving any information you wish
to provide regarding your problem.

     It is true that I have deliberately removed support for older
compilation tools from IRC Services, and I will continue to do so in the
future as those tools evolve.  However, this is not due to carelessness; on
the contrary, it is the result of efforts to clean up the source code and
bring it in line with a single standard.  As I mentioned earlier, Services
is only a free-time project of mine, and I simply cannot spare the effort
to support multiple versions of development tools.  If perhaps I am
adopting new standards faster than commercial OS vendors do, it is
certainly not out of malice or spite, but as I am distributing this program
(which is the result of nearly nine years of work--perhaps you are not
aware, but Anope itself is a second-generation derivative of IRC Services
4.3) free of monetary charge, I do feel entitled to ask that users of the
program make some effort of their own in order to use it.

[...]

  --Andrew Church
    achurch at achurch.org
    http://achurch.org/