On Thu, 12 Dec 2013, Radu Gheorghe wrote:
Regarding documentation:
- good documentation makes people want to use features. That's obvious
- good documentation makes people aware of features. This is actually, IMO,
the big problem rsyslog has. Elasticsearch, for example, is a very popular
destination for logs and, when I tell people (99% having rsyslog installed)
that rsyslog can output to ES they're like "whaaat? how come I didn't know
that?". And omelasticsearch is one of the better documented and blogged
about modules. Don't get me started on the likes of omhiredis or even
impstats
- this leads to the following problem: how do we get better documentation?
LOOOTS of people can contribute documentation. If we make it easy for them
to contribute, it's a big win for everyone. It will attract more people,
more awareness, more serious users, more contributors, etc.
I agree with everything ehre
I, for one, have no idea (off the top of my head) how to deal with raw
HTML, how to make a patch of my changes and so on. It's difficult for me to
contribute. Sure, one might say that I should know this stuff, but the fact
of the matter is that I don't. I don't contribute to projects often enough.
And I bet there are tons of people in my situation.
don't worry about editing html, just write the text that you think should be on
the page and post it to the list, there are lots of people here and at Adiscon
who can reformat the correction into a proper patch and get it submitted.
for most changes, you actually don't need to worry about the html, just open the
.html file in any editor, change the text and submit the result.
I think the point of the whole discussion is how to make it easy for people
to contribute. And how to "reward" them for that. There are some possible
solutions in this thread, some require less effort, some more. In the end,
if it's dead-easy to help, people will help and keep the ball rolling.
I think it's far easier to contribute than people think (another documentation
problem :-)
as for how to "reward" them, the biggest reward is the improved success of
rsyslog
but after that, really the only reward in OpenSource software is getting the
credit for the changes. If you submit a patch directly (through e-mail or a git
pull request), you generally are credited automatically. If you do something
like submitting changed text, the person who creates the patch needs to credit
you. This takes a little more time, and right now it's usually an informal note
in the commit message, but the linux kernel project has developed standards for
putting credits in the commit messages, adopting those would not be a bad idea
(but we really need more contributions before it really matters)
David Lang
2013/12/12 David Lang <[email protected]>
I am not opposed to change in and of itself, but there are issues to
figure out.
There is an underlying assumption that if rsyslog was an Apache projects,
something would be better.
I've seen a lot of projects over the years and a lot of changes, and I
have never seen a case where a healthy project made a major change and saw
any significant benefit from it.
I've seen many cases where a project was not healthy and changes did help
significantly.
This then raises the question of what I mean by a 'healthy' project.
The number of people who have commit rights is not a measure of a healthy
project. The Linux kernel is a healthy project by any measure, and it has
exactly one contributer (Linus)
The number of people contributing code isn't a direct measure, there are
far more significant things, and everyone would like more contributers.
It's actually easier to identify signs of unhealthy projects.
A maintainer who is hard to work with
a maintainer who doesn't have/take time to reveiw and accept patches
a maintainer/corporate owner who isn't trusted.
a codebase that's continually broken
a project that has legal/money problems
a project that's hard to build
a project with no support
There are others that I'm not thinking of now.
The only problems that I see rsyslog having are related to documentation.
for example, because of all the different packages to pull in and compile,
starting from scratch is a bit painful, but an improved document that shows
all the git repos that need to be cloned and what order you need to compile
them in.
What problems do others see rsyslog having? before we start looking at
solutions, let's discuss the problems that we are trying to solve?
David Lang
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad
of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you
DON'T LIKE THAT.
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of
sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE
THAT.
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of
sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE
THAT.
