<URL: http://bugs.freeciv.org/Ticket/Display.html?id=16811 >
I think this should be in the ticket, so I'm sending it to RT.
On Sun, Jan 4, 2009 at 3:47 PM, Kevin Benton <kben...@yitr.com> wrote:
> Brandon J. Van Every wrote:
>> I don't see any big deal
>> with RT because it's not my first instinct to look for someone else's
>> bug to work on. My first instincts are to:
> Okay - I'm glad you don't care. I do care, however, because I think RT is a
> piece of garbage when it comes to tracking software development projects. I
> also hate the way it's currently configured to prevent others from being
> able to view bugs and look for duplicates. It's discouraging to those of us
> who actively want to participate in other parts of development like QA,
> Reviews, etc. It's also discouraging for those of us on the "outside"
> looking in trying to get a sense of where things are in the development
> process because things are pretty hidden unless you actively try to look at
> bugs that wouldn't show up in "normal" searches.
My experience in open source is you have to communicate with other
developers somehow. I've been heavily involved in a couple of smaller
projects. They communicated by active mailing lists, and that worked
fine. I've perused the Mozilla bug tracker. Culturally, they use it
for communication instead of mailing lists. They're a huge project so
perhaps they are justified in using bug trackers and newsgroups to
communicate. It is quite amusing, however, to see vast threads of bug
tracker discussion that never go anywhere. Especially when the
discussion goes for years, with big lulls between. It's like the
collective gestalt of such a system is extremely limited. The bug
tracker is only as good as the people and the work they actually do.
>> 2) evaluate the readability of the source code. So far, I find it
> Again - great, but why not let the issue tracking system help in the review
> process to make sure it's better still?
Because I don't personally need or benefit from that much process. I
think "swimming through muck" is a core skill in open source
development. If a person can't do it, then they just don't get
anywhere, and probably find something else to go do. Also, all the
small projects I've seen, that have actually survived for a few years,
have had decently readable code. The ones that don't just die,
because nobody can figure out how to contribute. Freeciv is long past
>> 4) evaluate the political tone and management of the project. What do
>> people really spend their time on? What do they stew and get to
>> loggerheads about? What do they fail to get done? What do they
>> actively obstruct?
> Again - what does this have to do with the issue tracking system?
Because it's a mistake to advance a technology to solve a cultural problem.
> If all
> you want to do is write code, great and bring it on. The problem I see is
> at some point, it needs to be scrutinized by others, integration tested, and
> issues need to be tracked and easy to look at for those who want to write
> code themselves. This is where a good issue tracking system comes in.
Can you point to any open source project that's about the size of
Freeciv, that actually has a lot of programmers doing QA testbeds
instead of core development of new features? Most people find this
quite boring, and I say that as a build engineer who has tried to goad
people into more QA. My experience is that on small projects, it
doesn't happen. So there is a danger of overengineering a bug
tracker, like expecting Mozilla-size development, when in fact it will
> As I said earlier in this ticket, I'm willing to provide the elbow grease
> necessary, however, I am not willing to host it personally because I can't
> guarantee that I won't get hit by a bus and therefore, I am not able to
> guarantee it to be up.
Ok how about providing the elbow grease to solve that problem? I
Googled for a few minutes and found there are definitely open source
services that provide bug trackers. Show us one that's gonna stay
alive independent of you.
> I don't take lightly the commitment I'd be making if I choose to become a
> contributor to the Freeciv (or any other) project. I choose carefully
> because I have a life and I need to make sure I don't give up too much of
> what's really important to me. Freeciv is a way for me to relax. I would
> like to contribute to it in certain areas like the CMA because I really
> think it could be improved a lot. [...and several other game design ideas.]
Most volunteer developers care mainly about relaxing. At least that's
what I see, in practice, as far as what they actually put their energy
into. I think you can see why bug trackers and QA end up at the back
of the queue. If QA isn't the #1 thing you want to do, are you sure
you want to walk down this road? Maybe you should just dig into the
CMA code and whatever else. Possibly accept RT for what it is: an
adequate, not best of breed, but functional and working bug tracker.
Brandon Van Every
Freeciv-dev mailing list