I think that it would be nice to start a Common Lisp Janitors project,
much like the Linux Janitors project http://janitor.kernelnewbies.org/
Basically the spirit of CLJ would be to polish up areas of CL that are
currently messy. When I say CL, I really mean Common Lisp libraries
and Common Lisp Implementations. CLJ should be pitched at newbies
wanting to get started with CL, get us newbs to do the work, have an
expert mentor us & at the end of the day some CL libraries get
improved.
Off the top of my head, some things that could be done for libraries
that are already mostly useful:
- documentation
- testsuites
- portibility between implementations
- asdf installable
Though since I am new, I don't know exactly what libraries need help &
where.
I figure that to get this project moving we need
1) An administrator - setup the mailing lists, wiki pages etc
2) A few experts willing to create, and rate for difficulty messes,
that need cleaning up
3) A bunch of newbies that want to learn and contribute at the same
time.
I personally fall in category #3.
Thoughts?
Cheers
Brad
bradb wrote:
> I think that it would be nice to start a Common Lisp Janitors project,
> much like the Linux Janitors project http://janitor.kernelnewbies.org/
>
> Basically the spirit of CLJ would be to polish up areas of CL that are
> currently messy. When I say CL, I really mean Common Lisp libraries
> and Common Lisp Implementations. CLJ should be pitched at newbies
> wanting to get started with CL, get us newbs to do the work, have an
> expert mentor us & at the end of the day some CL libraries get
> improved.
Good idea, although finding enough willing mentors to shepherd a horde
of newbs into lisp might be difficult. (not to discourage anyone :-)
>
> Off the top of my head, some things that could be done for libraries
> that are already mostly useful:
> - documentation
> - testsuites
> - portibility between implementations
> - asdf installable
All good ideas.
>
> Though since I am new, I don't know exactly what libraries need help &
> where.
> I figure that to get this project moving we need
> 1) An administrator - setup the mailing lists, wiki pages etc
> 2) A few experts willing to create, and rate for difficulty messes,
> that need cleaning up
> 3) A bunch of newbies that want to learn and contribute at the same
> time.
>
> I personally fall in category #3.
Number 3 here also, although I am interested in helping out with it.
> Thoughts?
>
> Cheers
> Brad
On Tue, 6 Dec 2005, bradb wrote:
> I think that it would be nice to start a Common Lisp Janitors project,
> much like the Linux Janitors project http://janitor.kernelnewbies.org/
>
> Basically the spirit of CLJ would be to polish up areas of CL that are
> currently messy. When I say CL, I really mean Common Lisp libraries
> and Common Lisp Implementations. CLJ should be pitched at newbies
> wanting to get started with CL, get us newbs to do the work, have an
> expert mentor us & at the end of the day some CL libraries get
> improved.
>
> Off the top of my head, some things that could be done for libraries
> that are already mostly useful:
> - documentation
> - testsuites
> - portibility between implementations
> - asdf installable
>
> Though since I am new, I don't know exactly what libraries need help &
> where.
> I figure that to get this project moving we need
> 1) An administrator - setup the mailing lists, wiki pages etc
> 2) A few experts willing to create, and rate for difficulty messes,
> that need cleaning up
> 3) A bunch of newbies that want to learn and contribute at the same
> time.
>
> I personally fall in category #3.
> Thoughts?
Hi,
Been a lurker here for awhile,
this sounds like a pretty good idea really.
I'd fall into category #3 as well, but if this got off the
ground I'd like to help out.
Creighton Hogg <······@login01.hep.wisc.edu> writes:
> On Tue, 6 Dec 2005, bradb wrote:
>
>> I think that it would be nice to start a Common Lisp Janitors
>> project, much like the Linux Janitors project
>> http://janitor.kernelnewbies.org/
>>
>> Basically the spirit of CLJ would be to polish up areas of CL that
>> are currently messy. When I say CL, I really mean Common Lisp
>> libraries and Common Lisp Implementations. CLJ should be pitched
>> at newbies wanting to get started with CL, get us newbs to do the
>> work, have an expert mentor us & at the end of the day some CL
>> libraries get improved.
>>
>> Off the top of my head, some things that could be done for libraries
>> that are already mostly useful:
>> - documentation
>> - testsuites
>> - portibility between implementations
>> - asdf installable
>>
>> Though since I am new, I don't know exactly what libraries need help &
>> where.
>> I figure that to get this project moving we need
>> 1) An administrator - setup the mailing lists, wiki pages etc
>> 2) A few experts willing to create, and rate for difficulty messes,
>> that need cleaning up
>> 3) A bunch of newbies that want to learn and contribute at the same
>> time.
>>
>> I personally fall in category #3.
>> Thoughts?
>
> Hi,
> Been a lurker here for awhile,
> this sounds like a pretty good idea really.
> I'd fall into category #3 as well, but if this got off the
> ground I'd like to help out.
Okay, I don't know if this is #1 or #2 or what but I'll volunteer to
be a human clearinghouse--if anyone Lisp library/resource authors out
there can identify some tasks that might be doable by relative Lisp
newbies send a brief description to me. If you're a Lisp newibe who
wants to do some work that will help out the Lisp community, send me a
short note about what kinds of things you'd like to work on. I'll try
and match up tasks with newbies and keep things rolling along.
At the moment I'm working on moving my long-ago registered
lispniks.com domain somewhere that I can actually do stuff with it; if
I can actually pull that off I may even volunteer to do #1, setting up
mailing lists and so forth. (But in the meantime, let's see if there
are people actually ready to do some actual work--we don't need any
fancy infrostructure for that.)
-Peter
--
Peter Seibel * ·····@gigamonkeys.com
Gigamonkeys Consulting * http://www.gigamonkeys.com/
Practical Common Lisp * http://www.gigamonkeys.com/book/
Thank you very much Peter, I'm sure that with your name in the mix this
project has a much better chance of getting up to flight speed.
I'll email you offline with my interest areas.
Thanks
Brad
Hello,
> At the moment I'm working on moving my long-ago registered
> lispniks.com domain somewhere that I can actually do stuff with it; if
> I can actually pull that off I may even volunteer to do #1, setting up
> mailing lists and so forth. (But in the meantime, let's see if there
> are people actually ready to do some actual work--we don't need any
> fancy infrostructure for that.)
I'm game for it, I am personally interested in these things
materialising.
1. XML/Feed Libraries (RSS/Atom hacking). (I can actually churn out
some code for these).
2. Portable User Mode threads i.e Across OS as well as Implementations.
(I can probably work on sbcl/linux, but I don't have enough knowledge
about internals of SBCL. I am basically a lisp newbie, willing to work
on this, maybe we can get other lisp newbies to with sufficient
experience in other things (multitasking, threads) to implement it.
(I think this is a nice place to start for people who want to do some
FFI learning..).
3. Making SBCL's C runtime ".so", so you can link it to programs and
make ELF binaries out of lisp code. (No idea about how to implement
these).
4. Focal Point Lisp distro. This is something that I think thats very
crucial to make inroads into the "new blood" thats out there, and by
that I mean, a single install (download) which comes with
"batteries-included" so they can start writing code that actually works
and see results.
LispBox, is a good start, but it needs more libraries (I would first
put LTK there), and probably a small editor, like IDLE, not Emacs, but
something thats very newbie friendly. Once they are ok with lisp's
complexity, its only a matter of time before a newbie moves to Emacs +
SLIME.
Personally, I can contribute to #1 and maybe #2, but these are things
that can be done.
cheers
Vishnu Vyas.
vishnuvyas wrote:
>
> 2. Portable User Mode threads i.e Across OS as well as Implementations.
> (I can probably work on sbcl/linux, but I don't have enough knowledge
> about internals of SBCL. I am basically a lisp newbie, willing to work
> on this, maybe we can get other lisp newbies to with sufficient
> experience in other things (multitasking, threads) to implement it.
> (I think this is a nice place to start for people who want to do some
> FFI learning..).
Personally, portable threads is the thing that I need the most. Maybe,
what we need is a layer that maps to the threads support provided by
the lisp implementation where available and provides a simple (probably
inefficient) implementation of threads elsewhere (eg: CLISP in
Windows). I volunteer to work on such a project.
Vishwas.
"vishnuvyas" <··········@gmail.com> writes:
> 3. Making SBCL's C runtime ".so", so you can link it to programs and
> make ELF binaries out of lisp code. (No idea about how to implement
> these).
Some time ago, Fred Gilham made CMUCL for xBSD generate ELF
executables. You might check that work for ideas.
Paolo
--
Why Lisp? http://wiki.alu.org/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools:
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- CFFI: Foreign Function Interface
Hello,
>
> > 3. Making SBCL's C runtime ".so", so you can link it to programs and
> > make ELF binaries out of lisp code. (No idea about how to implement
> > these).
>
> Some time ago, Fred Gilham made CMUCL for xBSD generate ELF
> executables. You might check that work for ideas.
>
Is there any link documenting this? (web-link).
Cheers
Vishnu Vyas.
"vishnuvyas" <··········@gmail.com> writes:
> Hello,
>
>>
>> > 3. Making SBCL's C runtime ".so", so you can link it to programs and
>> > make ELF binaries out of lisp code. (No idea about how to implement
>> > these).
>>
>> Some time ago, Fred Gilham made CMUCL for xBSD generate ELF
>> executables. You might check that work for ideas.
>>
> Is there any link documenting this? (web-link).
Try things referenced from here:
<http://www.justfuckinggoogleit.com/search?q=Fred+Gilham+CMUCL+BSD+ELF+executables>.
Christophe
On Tue, 06 Dec 2005 20:30:13 -0800, vishnuvyas wrote:
> Hello,
> 2. Portable User Mode threads i.e Across OS as well as Implementations.
> (I can probably work on sbcl/linux, but I don't have enough knowledge
> about internals of SBCL. I am basically a lisp newbie, willing to work
> on this, maybe we can get other lisp newbies to with sufficient
> experience in other things (multitasking, threads) to implement it.
> (I think this is a nice place to start for people who want to do some
> FFI learning..).
>
Yes! I would be eager to help out with something like this.
(See previous rants, fumblings:
http://groups.google.com/group/comp.lang.lisp/browse_thread/thread/9d4a9427d122930a
http://groups.google.com/group/comp.lang.lisp/browse_thread/thread/8e85c4b976190ecb
)
--
"You do not really understand something unless you can
explain it to your grandmother." — Albert Einstein.
"Peter Seibel" <·····@gigamonkeys.com> wrote in message
···················@gigamonkeys.com...
> Okay, I don't know if this is #1 or #2 or what but I'll volunteer to
> be a human clearinghouse--if anyone Lisp library/resource authors out
> there can identify some tasks that might be doable by relative Lisp
> newbies send a brief description to me.
Suggestion:
Remove the requirement that the "authors identify some
task...DOABLE BY..NEWBIES" [emphasis added].
The newbies (or maybe intermediates, i.e., those not in the #2 expert
category) intend to improve and might not remain newbies if challenges
are offered.
Crummy code can be criticised by the "experts" or just ignored until
and unless it is improved.
Having the suggestions be accompanied by (optional) suggestions for
level of difficulty but leaving out something important that might
interest a newbie (or intermediate) won't help and might hurt.
BTW, the suggested categories were all VERY USEFUL, but not
likely to be of INTEREST to many people who wish to learn to CODE:
(Testsuites and portability are coding so might be interesting
choices in some/many cases though):
- documentation
- testsuites
- portability between implementations
- asdf installable
I would also add: Code reviews of existing code but this may be
worse than the others, and might be part of "testsuites" but perhaps
less ambitious.
--
Herb Martin
> BTW, the suggested categories were all VERY USEFUL, but not
> likely to be of INTEREST to many people who wish to learn to CODE:
> (Testsuites and portability are coding so might be interesting
> choices in some/many cases though):
>
I agree, documentation is not the most fun job in the world. But if we
have a wiki, maybe each Janitor can keep a notes page of what they are
doing & then distill the notes into documents occasionally. The wiki
could also have a page for every project so that there is low effort
overlap. I think that it isn't too hard to generate notes as you go.
Cheers
Brad
"bradb" <··············@gmail.com> writes:
>> BTW, the suggested categories were all VERY USEFUL, but not
>> likely to be of INTEREST to many people who wish to learn to CODE:
>> (Testsuites and portability are coding so might be interesting
>> choices in some/many cases though):
>>
>
> I agree, documentation is not the most fun job in the world. But if we
> have a wiki, maybe each Janitor can keep a notes page of what they are
> doing & then distill the notes into documents occasionally. The wiki
> could also have a page for every project so that there is low effort
> overlap. I think that it isn't too hard to generate notes as you go.
>
I think cliki, is a good place(wiki) for doing all these. One just needs to add
pages where projects seeking help can advertise themselves, and newbies
can go look around, etc.
Cheers.
--
Surendra Singhi
http://www.public.asu.edu/~sksinghi/index.html
,----
| By all means marry; if you get a good wife, you'll be happy. If you
| get a bad one, you'll become a philosopher.
| -- Socrates
`----
> I think cliki, is a good place(wiki) for doing all these. One just needs to add
> pages where projects seeking help can advertise themselves, and newbies
> can go look around, etc.
I think that Cliki is a good place to host the actual wiki pages. But
I also think that there needs to be one focal point to start from, and
it needs to be a toplevel domain so that people on c.l.l can say "go to
lispnewbie.org", instead of "go to cliki, search for 'newbie' & hope
you get the page I mean, and it is still relevant". This kind of
project needs active management at the beginning I think, otherwise it
won't generate enough momentum to get started.
Cheers
Brad
"bradb" <··············@gmail.com> writes:
>> I think cliki, is a good place(wiki) for doing all these. One just needs to add
>> pages where projects seeking help can advertise themselves, and newbies
>> can go look around, etc.
>
> I think that Cliki is a good place to host the actual wiki pages. But
> I also think that there needs to be one focal point to start from, and
> it needs to be a toplevel domain so that people on c.l.l can say "go to
> lispnewbie.org", instead of "go to cliki, search for 'newbie' & hope
> you get the page I mean, and it is still relevant".
What about having this domain, redirect to some page on cliki?
Will it be approprate to call it www.lispnewbie.org, as some experienced
lispers might also want to help?
Cheers.
--
Surendra Singhi
http://www.public.asu.edu/~sksinghi/index.html
,----
| By all means marry; if you get a good wife, you'll be happy. If you
| get a bad one, you'll become a philosopher.
| -- Socrates
`----
Am 07.12.2005 7:29 Uhr schrieb "Surendra Singhi" unter
<·········@netscape.net> in ············@netscape.net:
> "bradb" <··············@gmail.com> writes:
>
>>> I think cliki, is a good place(wiki) for doing all these. One just needs to
>>> add
>>> pages where projects seeking help can advertise themselves, and newbies
>>> can go look around, etc.
>>
>> I think that Cliki is a good place to host the actual wiki pages. But
>> I also think that there needs to be one focal point to start from, and
>> it needs to be a toplevel domain so that people on c.l.l can say "go to
>> lispnewbie.org", instead of "go to cliki, search for 'newbie' & hope
>> you get the page I mean, and it is still relevant".
>
> What about having this domain, redirect to some page on cliki?
>
> Will it be approprate to call it www.lispnewbie.org, as some experienced
> lispers might also want to help?
>
> Cheers.
You also need to understand that 'the' cliki is about 'free' Lisp on
Unix.
http://www.cliki.net/index
Quote from the start page: " Links to and resources for free software
implemented in Common Lisp and available on Unix-like systems. Listed
software should satisfy the Debian Free Software Guidelines (DFSG). "
Due to these limitations I mostly don't contribute to the site.
It is not about Lisp on Windows
(and more important for me 'Lisp machines' ;-) ) and non-free Lisps and
their libraries/usage/... .
Surendra Singhi <·········@netscape.net> writes:
> I think cliki, is a good place(wiki) for doing all these. One just needs to add
> pages where projects seeking help can advertise themselves, and newbies
> can go look around, etc.
See also:
Suggested Programming Projects
http://www.cliki.net/Suggested%20Programming%20Projects
Wish List
http://www.cliki.net/Wish%20List
Paolo
--
Why Lisp? http://wiki.alu.org/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools:
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- CFFI: Foreign Function Interface
"Herb Martin" <····@LearnQuick.com> writes:
> "Peter Seibel" <·····@gigamonkeys.com> wrote in message
> ···················@gigamonkeys.com...
>> Okay, I don't know if this is #1 or #2 or what but I'll volunteer to
>> be a human clearinghouse--if anyone Lisp library/resource authors out
>> there can identify some tasks that might be doable by relative Lisp
>> newbies send a brief description to me.
>
> Suggestion:
>
> Remove the requirement that the "authors identify some
> task...DOABLE BY..NEWBIES" [emphasis added].
>
> The newbies (or maybe intermediates, i.e., those not in the #2 expert
> category) intend to improve and might not remain newbies if challenges
> are offered.
Well, consider "doable by newbies" to be defined sufficiently broadly
to encompass things that might be appropriate challenging.
> Crummy code can be criticised by the "experts" or just ignored until
> and unless it is improved.
Indeed. I even volunteer to do some of that--if someone has
improvements they'd like to see made to something but don't
necessarily want to deal with a novice Lisper's attempts to make the
improvements, send them my way--if I can find an appropriate
volunteer, I'll assign it and then act as a first-pass sanity check on
their work.
> Having the suggestions be accompanied by (optional) suggestions for
> level of difficulty but leaving out something important that might
> interest a newbie (or intermediate) won't help and might hurt.
I'm not sure what you mean by this. Anyway, any info about the tasks
will be appreciated as will any information volunteers care to provide
about their interests and skills--I'm going to try and play matchmaker
so I can certainly use the info.
> BTW, the suggested categories were all VERY USEFUL, but not likely
> to be of INTEREST to many people who wish to learn to CODE:
That's fine. Folks can volunteer for whatever they want. Or if they
know exactly what they want to do, they can just go do it. I'm just
trying to make it easier for folks who don't know where to start. If
folks have coding tasks to be done and other folks want to volunteer
to help out, I'm happy to play matchmaker for that to. That said, I
think documenting undocumented code and testing untested code is an
excellent way to learn a *lot* about how the code works (or
doesn't). And if it's basically good code, that's a great way to learn
a lot about how to write Lisp. And it makes the Lisp world a better
place, which is part of the point.
> (Testsuites and portability are coding so might be interesting
> choices in some/many cases though):
>
> - documentation
> - testsuites
> - portability between implementations
> - asdf installable
>
> I would also add: Code reviews of existing code but this may be
> worse than the others, and might be part of "testsuites" but perhaps
> less ambitious.
Yup.
-Peter
--
Peter Seibel * ·····@gigamonkeys.com
Gigamonkeys Consulting * http://www.gigamonkeys.com/
Practical Common Lisp * http://www.gigamonkeys.com/book/
On a similar note:
Right now, I would (personally) like to see your excellent
Lispbox distribution (especially for Windows) altered to
support newer CLISP (2.36 instead of 2.34) but have hesitated
to bug you by asking that you do that.
If you would give me a rough idea of how to approach that
perhaps I could do it -- my initial (naive) attempts to substitute
2.35 for 2.34 never seemed satisfactory (especially as it
relates to SLIME.)
"Herb Martin" <····@LearnQuick.com> writes:
> On a similar note:
>
> Right now, I would (personally) like to see your excellent
> Lispbox distribution (especially for Windows) altered to
> support newer CLISP (2.36 instead of 2.34) but have hesitated
> to bug you by asking that you do that.
>
> If you would give me a rough idea of how to approach that
> perhaps I could do it -- my initial (naive) attempts to substitute
> 2.35 for 2.34 never seemed satisfactory (especially as it
> relates to SLIME.)
Actually I did a bunch of work this past weekend on refreshing all the
Lispboxen since all the Lisps have moved forward. Hopefully I'll have
the new Lispboxen up by next weekend.
And as part of this work I put everything in a Subversion repo, mostly
to make it easier to share between my Mac, Linux, and Windows
boxes. If I can make some progress on some other infrastructure I may
be able to make that repo available on the net and then folks who want
to hack on Lispbox can get the source easily.
-Peter
--
Peter Seibel * ·····@gigamonkeys.com
Gigamonkeys Consulting * http://www.gigamonkeys.com/
Practical Common Lisp * http://www.gigamonkeys.com/book/
> Actually I did a bunch of work this past weekend on refreshing all the
> Lispboxen since all the Lisps have moved forward. Hopefully I'll have
> the new Lispboxen up by next weekend.
>
> And as part of this work I put everything in a Subversion repo, mostly
> to make it easier to share between my Mac, Linux, and Windows
> boxes. If I can make some progress on some other infrastructure I may
> be able to make that repo available on the net and then folks who want
> to hack on Lispbox can get the source easily.
>
> -Peter
You're da Man, Peter.
Thanks again.
Thanks for your book. Thanks for Lispbox. Thanks for your repeated help.
--
Herb Martin
> 1) An administrator - setup the mailing lists, wiki pages etc
> 2) A few experts willing to create, and rate for difficulty messes,
that need cleaning up
> 3) A bunch of newbies that want to learn and contribute at the same
time.
I would also fall into catagory #3. I'm now hoping that this will take
off!
--Andy
On 2005-12-06 20:46:43 +0000, "bradb" <··············@gmail.com> said:
> 1) An administrator - setup the mailing lists, wiki pages etc
> 2) A few experts willing to create, and rate for difficulty messes,
> that need cleaning up
> 3) A bunch of newbies that want to learn and contribute at the same
> time.
Count me in as another #3 provided at least one #2 shows up :)
--
JFB ()
"bradb" <··············@gmail.com> writes:
> Though since I am new, I don't know exactly what libraries need help &
> where.
You may get some ideas by grepping for FIXME or TODO in the SBCL
source tree.
Paolo
--
Why Lisp? http://wiki.alu.org/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools:
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- CFFI: Foreign Function Interface
I quite like the idea. I'd personally fall into category #3.
We could possible subjugate a SourceForge account for our purposes
(mailing list + forums + stuff) until something better is created.
I think for something like this it is important to build a community
out of it, so just the Wishlist and TODO pages from the cliki aren't
going to be sufficient, IMHO.
Regards,
Austin
·······@gmail.com wrote:
> We could possible subjugate a SourceForge account for our purposes
> (mailing list + forums + stuff) until something better is created.
>
> I think for something like this it is important to build a community
> out of it, so just the Wishlist and TODO pages from the cliki aren't
> going to be sufficient, IMHO.
>
I agree. I think that this will be largely an exercise in community
building, a place where newbies can come to learn bits and pieces & at
the same time learn Lisp by actually using it.
Having some sort of souce control is probably a good idea, but I don't
know what structure it should follow, because the Janitors will
probably be submitting (reviewed) patches upstream to the actual
library maintainers. Maybe a repo that we can keep the patches in?
I'm glad we are getting some interest.
Thanks
Brad
·······@gmail.com wrote:
> I quite like the idea. I'd personally fall into category #3.
>
> We could possible subjugate a SourceForge account for our purposes
> (mailing list + forums + stuff) until something better is created.
>
Wouldn't a common-lisp.net account make more sense?
From the main page:
"This site is one among many gateways to Common Lisp. Its goal is to
provide the Common Lisp community with development resources and to work
as a starting point for new programmers."
This seems to be exactly what this project is for too.
> I think for something like this it is important to build a community
> out of it, so just the Wishlist and TODO pages from the cliki aren't
> going to be sufficient, IMHO.
>
I agree. I think we already have the beginnings of a community in this
thread, so perhaps someone should form a common-lisp.net project and
make a web page.
Just my opinion,
Andy
> Regards,
> Austin
>
Andy Cristina wrote:
> I agree. I think we already have the beginnings of a community in this
> thread, so perhaps someone should form a common-lisp.net project and
> make a web page.
I agree. common-lisp.net also hosts mailing lists so we can create a
list to discuss the project.
You can see here how to create a new project:
http://common-lisp.net/project-intro.shtml#creproj
Pedro Kröger
"Pedro Kr�ger" <············@gmail.com> writes:
> Andy Cristina wrote:
>> I agree. I think we already have the beginnings of a community in this
>> thread, so perhaps someone should form a common-lisp.net project and
>> make a web page.
I'll do it.
-Peter
--
Peter Seibel * ·····@gigamonkeys.com
Gigamonkeys Consulting * http://www.gigamonkeys.com/
Practical Common Lisp * http://www.gigamonkeys.com/book/
There is a lot of interest in the ECL Qt thread, mostly because it is
an example of how to use Qt from Lisp on Windows. Perhaps, one of the
things that CL Gardeners should attempt to do is build bite size
examples that are newbie friendly & portable, GUIs and sockets are two
examples that strike me right away.
Brad
bradb wrote:
> I think that it would be nice to start a Common Lisp Janitors project,
> much like the Linux Janitors project http://janitor.kernelnewbies.org/
>
> Basically the spirit of CLJ would be to polish up areas of CL that are
> currently messy. When I say CL, I really mean Common Lisp libraries
> and Common Lisp Implementations. CLJ should be pitched at newbies
> wanting to get started with CL, get us newbs to do the work, have an
> expert mentor us & at the end of the day some CL libraries get
> improved.
>
I think this is a very good idea and I'd love to see this project take
off!
> Off the top of my head, some things that could be done for libraries
> that are already mostly useful:
> - documentation
> - testsuites
> - portibility between implementations
> - asdf installable
>
> Though since I am new, I don't know exactly what libraries need help &
> where.
> I figure that to get this project moving we need
> 1) An administrator - setup the mailing lists, wiki pages etc
> 2) A few experts willing to create, and rate for difficulty messes,
> that need cleaning up
> 3) A bunch of newbies that want to learn and contribute at the same
> time.
>
> I personally fall in category #3.
> Thoughts?
I'd love to contribute, though I have very little free time. But
anyway, count me in. And I fall in category #3 as well.
Cheers,
Ivan
bradb wrote:
> Off the top of my head, some things that could be done for libraries
> that are already mostly useful:
> - documentation
> - testsuites
> - portibility between implementations
> - asdf installable
Thanks to bradb & peter for getting the common lisp gardeners idea
going. Wanted to throw my two cents in before this moves to a mailing
list.
I think bradb's goals above are great. Understanding code well enough
to document or test it will go a long way towards 'learning to code',
if that's the goal of a volunteer. Personally, there's code I'd like
to _use_ but dont have the (mental?) resources to do so rapidly without
documentation . . . and if that means I need to dig through code and
write docs, so be it.
As for documentation, I'm not sure a wiki is the best way to do that,
or if it is, there needs to be an agreed upon template for people to
start from, i.e.
name:
cl-web-getter
description:
gets web pages so you don't have to.
installation:
(asdf-install:install 'cl-web-getter)
example usage:
(with-cookies
(princ (get "https://some.url.example"))
(princ (get "http://another.url.example")))
details:
....
Any thoughts on boilerplate or how to tie wiki'd docs to distributed
ones?
-cody
> As for documentation, I'm not sure a wiki is the best way to do that,
> or if it is, there needs to be an agreed upon template for people to
> start from, i.e.
>
> name:
> cl-web-getter
>
> description:
> gets web pages so you don't have to.
>
> installation:
> (asdf-install:install 'cl-web-getter)
>
> example usage:
> (with-cookies
> (princ (get "https://some.url.example"))
> (princ (get "http://another.url.example")))
>
> details:
> ....
>
>
> Any thoughts on boilerplate or how to tie wiki'd docs to distributed
> ones?
I actually think the documentation as written should not be in wiki
form. When I started working with python I loved the fact that I could
type help(function-name) at the interactive prompt and get
documentation on that item. It was built from the docstrings of the
code itself, and made things real easy. Worked for just about anything,
and used the docstrings in your own code seamlessly.
That, I think, is something we need. Generating a wiki page from that
is pretty easy as you already have the code + comments on it.
>
> -cody
Good idea. I know CL has (documentation 'function 'funcname) or
something similar, but that is too painful to type - SLIME probably has
that built in but I don't use it.
I think that one of the CLG projects should be a newbies into guide to
Emacs and Slime, and some options if you don't happen to like working
with Emacs.
Cheers
Brad
ajones wrote:
> I actually think the documentation as written should not be in wiki
> form. When I started working with python I loved the fact that I could
> type help(function-name) at the interactive prompt and get
> documentation on that item. It was built from the docstrings of the
> code itself, and made things real easy. Worked for just about anything,
> and used the docstrings in your own code seamlessly.
>
> That, I think, is something we need. Generating a wiki page from that
> is pretty easy as you already have the code + comments on it.
>
I think there are at least two kinds of documentation: for individual
function/variable/class/whatever; and let's call it package
documentation or usage documentation, which describes the purpose and
usage patterns of the whole thing. The former should be put in the
docstrings, but the latter is more like a tutorial and it needs a place
of its own.
I'm not sure if WIKI is the best place to put it but it's a start and
it's very convenient to use it at least while writing the final
deliverable. (damn this RUP jargon).
--
ivant
I like the idea of a Wiki template for starting the documentation. I
was personally thinking that there would be two types of documentation,
explicit documentation and incidental.
Explicit documentation is easy, library foo needs some documentation.
A gardiner takes on the task of writing up a Garden Guide (or some
other cute name) for the library and puts up a relatively complete
document somewhere.
The incidental documentation should be more free form, each Gardiner
just jots down notes about the library as they work on it, useful hints
and tips, problems etc. At some stage there is an effort to formalise
the notes into explicit documentation, and there is a decent set of
existing notes to begin working with.
Cheers
Brad
I've been starting to think of my Lisp image as more of an Operating
System the more I use it. Interacting with it is almost like using a
bash shell on unix. Installing new "software" is just an asdf-install
away. But, I am starting to think that asdf-install could be better,
and I think that the Gardiners could probably help with it.
- maintain an asdf-installable repositry that is NOT wiki based.
Frankly the thought of having asdf-install refer to a page that is on a
publicly accessible wiki gives me the willies. If I am wrong about how
asdf-install works, it is not obvious from the docs. I know that
packages are GPG signed, but really manually checking this won't happen
- it's too much work.
- create an asdf package browser, so that librarys are easy to find
and install
- add more libraries to our repositry
Cheers
Brad
"bradb" <··············@gmail.com> writes:
> - maintain an asdf-installable repositry that is NOT wiki based.
Do you mean something like this?
cCLan - The Comprehensive Common Lisp Archive Network
http://www.cliki.net/cCLan
http://cclan.sourceforge.net/
Paolo
--
Why Lisp? http://wiki.alu.org/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools:
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- CFFI: Foreign Function Interface
Paolo Amoroso wrote:
> "bradb" <··············@gmail.com> writes:
>
> > - maintain an asdf-installable repositry that is NOT wiki based.
>
> Do you mean something like this?
>
> cCLan - The Comprehensive Common Lisp Archive Network
Yep. But from that page it appears that asdf-install has become more
popular. Maybe cClan just needs some water and weeding to grow. If
the only way that asdf-install works is to reference a public wiki page
(I don't actually know this for sure), then that is just asking for
trouble.
Cheers
Brad