Sent via BlackBerry by AT&T
Well, the reason why we were talking about gitit and ikiwiki the other
day was because they are about the only two options for wikis that use
"real" revision control systems under the hood, not something customly
integrated or whatever. What this means is that this would easily
expand to the open source hardware projects in the future while
simultaneously not being too much of a pain in the ass at the present,
since it could both do code hosting, documentation hosting, and be a
fancy web interface like everyone wants. I'm not sure what instiki
uses for its revision control .. it does say something about ruby, so
maybe there's a ruby module we could throw in to give it access to git
or something, but honestly I think we should take the path of least
resistance.
http://www.instiki.org/show/WhatIsInstiki
Just for the record, it isn't really "my" content. Yes, I contributed
some of the code. Anyway, if you want to look at who has been
contributing to it, you can look at the logs (run "git help log" I
guess). On the other hand it's possible that there are a number of
other anonymous people who have at least looked at it (which I guess
is who you might get responding to you).
> Secondly, why can't we keep using openwetware as the main wiki?
I'm starting to feel like a broken record.
Also, most wikis are not good for revision control of anything more
than just text on a page somewhere.. in the case of diybio projects,
the toolchain stuff in the git repo for instance can be used for
community infrastructure purposes, etc. etc.
I agree. I am not saying that we should all stop using OWW or
something "or else". Remember, I'm still the one making most of the
edits on the diybio FAQ page on OWW. I don't think a wiki is a good
place to host code, for instance, or the deeper projects that we're
working on.
> That thread was mainly discussing the fact that on OWW you need to use
> a real name. Doesn't seem like that is the argument you're making
> below. Nor does that seem to be the best reason to choose to
> use/not-use OWW.
The last few posts in the thread are what I should have linked to, I
admit. In particular the comment about how it's more like ignoring the
years and years of experience that are actually out there about how
online collaboration actually works, the comment about Hawaii, etc.
>> Also, most wikis are not good for revision control of anything more
>> than just text on a page somewhere.. in the case of diybio projects,
>> the toolchain stuff in the git repo for instance can be used for
>> community infrastructure purposes, etc. etc.
>
> I don't know what the best option for DIYBio is, I'll leave it up to
> the folks who are bigger contributors then me to decide -- however my
> 2 cents is that I would optimize on whatever enables the largest
> number of people to contribute.
There are certain tools for certain jobs, but that doesn't mean we
should just throw away all of the hacker tools because most people on
diybio don't know how to use hacker tools. I think it should be ok if
we use tools to advance our mission.
> So even if something isn't perfect, but has a really easy to use UI
> (Mediawiki benefits from all the training that Wikipedia has done on
> people, though there are other good ones) or a pre-existing community
Maybe you could make a mediawiki CSS layout thingy for gitit or
ikiwiki if you think it's such a huge problem. But I'm not about to
start pushing code commits to a mediawiki RCS.
> that can be tapped (OpenWetWare) it might be a better choice then
> something that has a technical advantage that might matter further
why can't diybio have its own toolchain instead of depending on OWW?
> down the road or only to a small fraction of the community (I don't
> know if this characterizes instiki, don't know much about it). Good
> luck, it's not easy to choose the right platform to inspire
> contribution as anyone who has tried to get an online community off
> the ground knows.
Maybe you should use tools that contributors are using.
OWW doesn't really provide the infrastructure that diybio is
interested in developing. Development tools are development tools, and
a wiki is a wiki. These other tools provide the best of both worlds.
Jason, I am wondering whether or not you do any programming, and if
so, have you ever collaborated with others by pasting code on to a
wiki? And how did that work out for you?
Really? I have never heard of an engineering company using svn before.
That's neat. I haven't ever configured twiki, although I have used it
on various different servers, is the backend somewhere that it would
be easy enough to just dump files and incorporate that into a make
system?
I wouldn't mind trying out twiki either, then. Maybe we should set up
/sandbox/wikis/ on the server for a few testbeds to see if we can
commit and checkout between each other on the diybio.org server, or
something, and try out at least two or three different wikis (I'd be
willing to set them up).