Diffstat (limited to 'contributing/discussion.mdwn')
1 files changed, 67 insertions, 1 deletions
diff --git a/contributing/discussion.mdwn b/contributing/discussion.mdwn
index 5a6bfd7c..11e8ac0e 100644
@@ -1,4 +1,4 @@
-[[!meta copyright="Copyright © 2011 Free Software Foundation, Inc."]]
+[[!meta copyright="Copyright © 2011, 2013 Free Software Foundation, Inc."]]
[[!meta license="""[[!toggle id="license" text="GFDL 1.2+"]][[!toggleable
id="license" text="Permission is granted to copy, distribute and/or modify this
@@ -19,3 +19,69 @@ Invent something.
# Mailing Lists
Add link to [[mailing_lists]] to page, and suggest following these.
+# IRC, freenode, #hurd, 2013-08-05
+ <nalaginrut> hi guys, I'm new here. I'm a developer from Guile community,
+ and I think maybe it's a proper time to do some work to make GNU stuff
+ use Guile increasingly, but I found the wiki and docs seems a bit old,
+ and I can't find an entry from Hurd source, since there're too many
+ things. Anyone point me out?
+ <nalaginrut> thanks
+ <nlightnfotis> nalaginrut what exactly is it that you need help with?
+ <nalaginrut> I've no idea, I saw MIG and I think if it's a language I can
+ write a front-end on Guile platform. But someone suggest me write hurd
+ binding will be a good start
+ <nalaginrut> I cloned incubator which is cl-binding for hurd, but I've no
+ idea too, since there's nothing in master branch
+ <pinotree> well, fixing guile on the hurd would be a start:
+ <braunr> i won't talk about this, as my personal opinion on the matter is
+ that it's not a proper time to do it
+ <braunr> but at the same time, people should do what they're interested in
+ <braunr> so feel free to do it
+ <nalaginrut> braunr: is there any reason why it's not a proper time?
+ <braunr> nalaginrut: two words: mig sucks
+ <nalaginrut> so it'll be replaced by a new stuff?
+ <teythoon> any more reasons to have alternatives, no?
+ <braunr> sure, please do it :)
+ <braunr> actually it's more than just mig
+ <braunr> the low level internals of the hurd are almost fine, but not good
+ enough to reliably develop over it
+ <braunr> gccgo is currently proving it
+ <braunr> and such projects are good opportunities to identify and fix such
+ <braunr> but the, if you want to work on guile, be prepared to work on a
+ lot more than just guile
+ <nalaginrut> I'm afraid I have to collect the reasons and evaluate when is
+ proper to do that, if Hurd has to be redesigned, it is not a proper time
+ <braunr> it also happened with openjdk, jeremie had to fix signals (!)
+ <nalaginrut> anyway, I just want a suggestion how to start
+ <pinotree> <pinotree> well, fixing guile on the hurd would be a start:
+ <nalaginrut> ok, I'll try
+ <antrik> nalaginrut: "incubator" is a somewhat strange beast. every branch
+ in there is a completely different project. you have to find the right
+ branch for the CL bindings...
+ <nalaginrut> antrik: thanks for reply, I guess it's clisp branch?
+ <pinotree> nalaginrut:
+ <antrik> nalaginrut: sounds like it :-)
+ <antrik> braunr: I'm believe it's important to encourage work on as many
+ different levels as possible. there is no motivation for fixing low-level
+ issues unless there are some interesting high-level things relying on
+ <braunr> antrik: i agree
+ <braunr> 11:50 < braunr> but at the same time, people should do what
+ they're interested in
+ <antrik> in fact, it's pretty much impossible to identify what we really
+ need at the lower levels unless working on high-level stuff as well...
+ <braunr> yes
+ <braunr> 11:57 < braunr> but the, if you want to work on guile, be prepared
+ to work on a lot more than just guile
+ <nalaginrut> I prepare to work on Hurd, is that an fair answer?
+ <antrik> nalaginrut: perfect! ;-)
+ <nalaginrut> ;-) well, easy to say, but I'll try what I can do
+ <antrik> yeah, just see how far you get. might be an interesting ride :-)