From abriggs at westnet.com.au Fri Jun 3 09:28:45 2005
From: abriggs at westnet.com.au (Anthony Briggs)
Date: Fri, 03 Jun 2005 17:28:45 +1000
Subject: [melbourne-pug] Livejournal update
Message-ID: <42A006AD.3010907@westnet.com.au>
Hi All,
I've just updated the livejournal page
with some meeting
details. Let me know if you've got any thoughts about topics for
presentation or the meeting format.
Thanks,
Anthony
From abriggs at westnet.com.au Thu Jun 16 04:50:41 2005
From: abriggs at westnet.com.au (Anthony Briggs)
Date: Thu, 16 Jun 2005 12:50:41 +1000
Subject: [melbourne-pug] Meeting Tonight
Message-ID: <42B0E901.2090200@westnet.com.au>
Hi All,
Just to let you all know that the meeting tonight is definitely going
ahead (I've had a couple of last minute enquires). I'll see you at 6.30:
Intrepid Travel
12 Spring St,
Fitzroy
16th June, 6.30pm
It's pretty close to the 96 and 112 tram, and only a short hop from
Parliament station. Not too sure about parking, but there's a fair bit
of street parking around, particularly after hours. There are also a
large number of cafes and restaurants along Brunswick street.
Thanks,
Anthony
From richardjones at optushome.com.au Thu Jun 16 05:05:23 2005
From: richardjones at optushome.com.au (Richard Jones)
Date: Thu, 16 Jun 2005 13:05:23 +1000
Subject: [melbourne-pug] Meeting Tonight
In-Reply-To: <42B0E901.2090200@westnet.com.au>
References: <42B0E901.2090200@westnet.com.au>
Message-ID: <200506161305.23765.richardjones@optushome.com.au>
On Thu, 16 Jun 2005 12:50 pm, Anthony Briggs wrote:
> Just to let you all know that the meeting tonight is definitely going
> ahead (I've had a couple of last minute enquires). I'll see you at 6.30:
Sorry, I'm not going to be able to make it :(
Richard (who feels quite the piker for not making the inaugural meeting)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050616/7159cf69/attachment.pgp
From abriggs at westnet.com.au Thu Jun 16 05:38:48 2005
From: abriggs at westnet.com.au (Anthony Briggs)
Date: Thu, 16 Jun 2005 13:38:48 +1000
Subject: [melbourne-pug] Meeting Tonight
In-Reply-To: <200506161305.23765.richardjones@optushome.com.au>
References: <42B0E901.2090200@westnet.com.au>
<200506161305.23765.richardjones@optushome.com.au>
Message-ID: <42B0F448.1030305@westnet.com.au>
Richard Jones wrote:
> On Thu, 16 Jun 2005 12:50 pm, Anthony Briggs wrote:
>
>>Just to let you all know that the meeting tonight is definitely going
>>ahead (I've had a couple of last minute enquires). I'll see you at 6.30:
>
> Sorry, I'm not going to be able to make it :(
>
> Richard (who feels quite the piker for not making the inaugural meeting)
You piker!
No drama - I've still got 15 people listed as coming, plus 6 (including
you) who are interested but can't turn up to the first meeting for one
reason or another. That's also not including another 6 people who
haven't RSVPed and may still turn up...
Anthony
From byte at bytebot.net Thu Jun 16 06:16:57 2005
From: byte at bytebot.net (Colin Charles)
Date: Thu, 16 Jun 2005 14:16:57 +1000
Subject: [melbourne-pug] Meeting Tonight
In-Reply-To: <42B0F448.1030305@westnet.com.au>
References: <42B0E901.2090200@westnet.com.au>
<200506161305.23765.richardjones@optushome.com.au>
<42B0F448.1030305@westnet.com.au>
Message-ID: <1118895417.3559.32.camel@arena.soho.bytebot.net>
On Thu, 2005-06-16 at 13:38 +1000, Anthony Briggs wrote:
>
> No drama - I've still got 15 people listed as coming, plus 6
> (including
> you) who are interested but can't turn up to the first meeting for one
> reason or another. That's also not including another 6 people who
> haven't RSVPed and may still turn up...
Can't turn up, though I'd definitely like to. Sigh! Today's just not
conducive for me
--
Colin Charles, http://www.bytebot.net/
FUDCon II @ LinuxTag
June 24-25, 2005 in Karlsruhe, Germany
http://fedoraproject.com/fudcon/
From daedalus at eigenmagic.com Thu Jun 16 06:32:38 2005
From: daedalus at eigenmagic.com (Justin Warren)
Date: Thu, 16 Jun 2005 14:32:38 +1000
Subject: [melbourne-pug] Meeting Tonight
In-Reply-To: <42B0F448.1030305@westnet.com.au>
References: <42B0E901.2090200@westnet.com.au>
<200506161305.23765.richardjones@optushome.com.au>
<42B0F448.1030305@westnet.com.au>
Message-ID: <1118896358.3982.5.camel@norbert>
On Thu, 2005-06-16 at 13:38 +1000, Anthony Briggs wrote:
> Richard Jones wrote:
> > On Thu, 16 Jun 2005 12:50 pm, Anthony Briggs wrote:
> >
> >>Just to let you all know that the meeting tonight is definitely going
> >>ahead (I've had a couple of last minute enquires). I'll see you at 6.30:
> >
> > Sorry, I'm not going to be able to make it :(
> >
> > Richard (who feels quite the piker for not making the inaugural meeting)
>
> You piker!
>
> No drama - I've still got 15 people listed as coming, plus 6 (including
> you) who are interested but can't turn up to the first meeting for one
> reason or another. That's also not including another 6 people who
> haven't RSVPed and may still turn up...
Hey Anthony,
Is there somewhere I could park a bicycle securely at the venue? I've
got panniers full of laptop and stuff. It'd save having to come back to
the office after the meeting.
--
Justin Warren
From abriggs at westnet.com.au Thu Jun 16 08:35:12 2005
From: abriggs at westnet.com.au (Anthony Briggs)
Date: Thu, 16 Jun 2005 16:35:12 +1000
Subject: [melbourne-pug] Bicycle Parking
In-Reply-To: <1118896358.3982.5.camel@norbert>
References: <42B0E901.2090200@westnet.com.au> <200506161305.23765.richardjones@optushome.com.au> <42B0F448.1030305@westnet.com.au>
<1118896358.3982.5.camel@norbert>
Message-ID: <42B11DA0.6060205@westnet.com.au>
Justin Warren wrote:
> Hey Anthony,
>
> Is there somewhere I could park a bicycle securely at the venue? I've
> got panniers full of laptop and stuff. It'd save having to come back to
> the office after the meeting.
Sure - we've got a courtyard out the back - there are lots of people at
Intrepid who cycle to work.
Anthony
From daedalus at eigenmagic.com Fri Jun 17 02:04:04 2005
From: daedalus at eigenmagic.com (Justin Warren)
Date: Fri, 17 Jun 2005 10:04:04 +1000
Subject: [melbourne-pug] libsnmp
Message-ID: <1118966644.2810.3.camel@norbert>
A few people expressed an interest in libsnmp last night, so I thought
I'd point to the code: http://seafelt.unicity.com.au/libsnmp/
Source downloads available from:
http://seafelt.unicity.com.au/downloads/
Cheers,
--
Justin Warren
From djc at object-craft.com.au Fri Jun 17 02:52:32 2005
From: djc at object-craft.com.au (Dave Cole)
Date: Fri, 17 Jun 2005 10:52:32 +1000
Subject: [melbourne-pug] libsnmp
In-Reply-To: <1118966644.2810.3.camel@norbert>
References: <1118966644.2810.3.camel@norbert>
Message-ID: <42B21ED0.8070204@object-craft.com.au>
Justin Warren wrote:
> A few people expressed an interest in libsnmp last night, so I thought
> I'd point to the code: http://seafelt.unicity.com.au/libsnmp/
>
> Source downloads available from:
> http://seafelt.unicity.com.au/downloads/
Already had a look at that. After a quick traversal over the unicity
site I think I would like to learn a bit more about BEEPy at a future
meeting (if possible).
- Dave
--
http://www.object-craft.com.au
From daedalus at eigenmagic.com Fri Jun 17 02:59:56 2005
From: daedalus at eigenmagic.com (Justin Warren)
Date: Fri, 17 Jun 2005 10:59:56 +1000
Subject: [melbourne-pug] libsnmp
In-Reply-To: <42B21ED0.8070204@object-craft.com.au>
References: <1118966644.2810.3.camel@norbert>
<42B21ED0.8070204@object-craft.com.au>
Message-ID: <1118969996.2810.7.camel@norbert>
On Fri, 2005-06-17 at 10:52 +1000, Dave Cole wrote:
> Justin Warren wrote:
> > A few people expressed an interest in libsnmp last night, so I thought
> > I'd point to the code: http://seafelt.unicity.com.au/libsnmp/
> >
> > Source downloads available from:
> > http://seafelt.unicity.com.au/downloads/
>
> Already had a look at that. After a quick traversal over the unicity
> site I think I would like to learn a bit more about BEEPy at a future
> meeting (if possible).
I'll probably use BEEPy for examples of twisted as part of my talk next
month, since that's how I came to use twisted.
Also, you can check out the rudimentary website for BEEPy:
http://beepy.sourceforge.net/
--
Justin Warren
From miked at dewhirst.com.au Fri Jun 17 10:37:01 2005
From: miked at dewhirst.com.au (Mike Dewhirst)
Date: Fri, 17 Jun 2005 08:37:01 GMT
Subject: [melbourne-pug] Thank you Anthony and Intrepid
Message-ID: <20050617083701.20086.qmail@webmastery.com.au>
By way of saying thanks, I thought I'd make a proposal.
We need an idea-path so newcomers understand where we are. What are the
shared objectives and hopes regarding the group?
Personally, I want to learn to use Python effectively. For me that means ...
- interface to Apache on Linux
- interface to SQL92 standard RDBMS
- python application development
- python "standard" libraries
- python special purpose libraries
- rich client front ends via wxPython and (probably) Boa Constructor
- thin client front ends via W3C standard browsers
- any other standards-based development infrastructure which is python
related
- agile methodologies applied to python development
My prime motivation is to use open source AND cross-platform development
infrastructure for the rest of my life. I won't willingly use closed source
OR single-platform tools ever again. I have done so in the past and I admit
I was wrong.
I understand the Windows / Linux divide and how religious wars can destroy
communities. I don't like it but I have no choice but to work with my
customers all of which have Windows desktops - exclusively. Most have
Windows Terminal Servers and Novell file servers. The rest are 100% Windows
so I can never get away from Windows.
YMMV so please add to the list.
I would like to propose that in future the group focuses on two things per
meeting ...
1. A presentation by anyone competent in an item from the above list
2. Actually developing an application to be owned by the group and with
potential to become a useful open-source project.
The idea for the group project is to put best-practices into context.
Hopefully, best practice in any area is likely to be contentious. I say,
bring it on :) I want to learn.
I can hear some questions already so I'll try to pre-empt a couple ...
The list should be held in safe-keeping for presenters to look at before
deciding what to present.
We should keep score so we know what is popular and when something is due
for regurgitation.
The list can be added to and subtracted from by group members by consensus
at any meeting.
The group project needs lively discussion and may not get off the ground
quickly. It needs precursors such as the best practices in setting up any
project but especially a Python project which IMO should be cross-platform
and totally reliant on open source. We would need to obtain certain
resources. I have a 24/7 Apache + Subversion server available on a Linux box
with a fixed IP address but there may be more we decide that we need.
If the project is ambitious enough I would expect it will provide
presentation material aplenty for years to come. Especially if we document
why we did what we did.
If the project is good enough the group may decide to licence it and
generate revenue upon which we can all retire :)
If the project is useful enough to other Python developers the group may
decide to open-source it.
My own view (developing as I write) is that group discussion of best
practices for a real project will have far more value for everyone than
hearing general good advice. I would also be very keen to see working code
constructively criticised in very specific context. The idea of an ongoing
context will also have drawing power at meetings.
Enough. I'm off my horse now.
Thank you Anthony and Intrepid for the inaugural meeting
I enjoyed it very much.
Regards
Mike
From huw.davies at kerberos.davies.net.au Fri Jun 17 12:46:43 2005
From: huw.davies at kerberos.davies.net.au (Huw Davies)
Date: Fri, 17 Jun 2005 20:46:43 +1000
Subject: [melbourne-pug] Thank you Anthony and Intrepid
In-Reply-To: <20050617083701.20086.qmail@webmastery.com.au>
References: <20050617083701.20086.qmail@webmastery.com.au>
Message-ID:
On 17/06/2005, at 6:37 PM, Mike Dewhirst wrote:
> By way of saying thanks, I thought I'd make a proposal.
>
> We need an idea-path so newcomers understand where we are. What are
> the
> shared objectives and hopes regarding the group?
>
> Personally, I want to learn to use Python effectively. For me that
> means ...
>
> - interface to Apache on Linux
> - interface to SQL92 standard RDBMS
> - python application development
> - python "standard" libraries
> - python special purpose libraries
> - rich client front ends via wxPython and (probably) Boa Constructor
> - thin client front ends via W3C standard browsers
> - any other standards-based development infrastructure which is
> python
> related
> - agile methodologies applied to python development
Roughly what I want/need as well although I'm not currently
interested in developing for the web.
> I would like to propose that in future the group focuses on two
> things per
> meeting ...
>
> 1. A presentation by anyone competent in an item from the above list
>
> 2. Actually developing an application to be owned by the group and
> with
> potential to become a useful open-source project.
>
> The idea for the group project is to put best-practices into context.
> Hopefully, best practice in any area is likely to be contentious. I
> say,
> bring it on :) I want to learn.
Agree with 1 - I'm happy to listen to anyone talk about Python.
I think a group project is a great idea but perhaps we should look at
developing smaller things and doing a "sprint". My thoughts would be
to have several small projects and try to complete them in no more
than a day. In small groups with one or more experienced Python
people helping those who may well have lots of programming knowledge,
just not in Python.
I've done lots of programming and am slowly learning the Python way -
as I've said before it's easy to write Pascal programs in any
language. To be accurate, substitute Algol-68 for Pascal but I doubt
too many people know about Algol-68 :-)
Finally, sorry I couldn't make the meeting, it's an entertaining
story but requires an NDA before it makes sense....
Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au
Melbourne | "If soccer was meant to be played in the
Australia | air, the sky would be painted green"
From mauriceling at acm.org Fri Jun 17 16:10:22 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sat, 18 Jun 2005 00:10:22 +1000
Subject: [melbourne-pug] developing something as a group
In-Reply-To:
References: <20050617083701.20086.qmail@webmastery.com.au>
Message-ID: <42B2D9CE.6010801@acm.org>
>>2. Actually developing an application to be owned by the group and
>>with potential to become a useful open-source project.
>>
>>
>I think a group project is a great idea but perhaps we should look at
>developing smaller things and doing a "sprint". My thoughts would be
>to have several small projects and try to complete them in no more
>than a day. In small groups with one or more experienced Python
>people helping those who may well have lots of programming knowledge,
>just not in Python.
>
>I've done lots of programming and am slowly learning the Python way -
>as I've said before it's easy to write Pascal programs in any
>language. To be accurate, substitute Algol-68 for Pascal but I doubt
>too many people know about Algol-68 :-)
>
>
>
Well, I agree that this might be a good idea as there is some form of
ownership as a group and continualty, given the number of people we have
seen yesterday.
Personally, I have the tendency to start off projects faster than I can
ever complete them. Currently, I am working on something with 5% of my
time, which I think affects all of us. The problem I face and foresee is
this, when we try to install a Python program or sort, there are usually
dependencies to fulfill prior actual installation, and each of these
dependencies may have other dependencies.
This is made worse by the fact that we can have serveral Python
installations within a system. I am a Mac user, so I have
Apple-installed Python and Fink-installed Python, and each maintains
their own site-packages. And my question boils down to 2 things (I think):
1. Is there a way or system to satisfy the dependencies if I choose to
install a package?
2. When I upgrade my Python installation, is there a way to re-install
all the packages of my older python version into my newer python version?
I am using ideas from CPAN, Darwinports, Fink and other package
management systems.
What are the uses of this proposed system?
1. Imagine a system admin have installed 50 different packages and
dependencies and now have to upgrade to a newer version of python.
Re-installing all those packages again is a pain.
2. How many times we hope that when we try to install a program with
"python setup.py install" that we wish for a system that can just help
us handle all the dependencies?
I did make an announcement in catelog-sig and python mailing list that
this work I've just described, is intended as an academic project. So, I
hope I can solicit some interests in this group. If there are some
interests, then I do think it make sense to attempt it as a group effort.
Please tell me what you think about this and we will take it from there.
Sorry, this is a bit long for Friday night read...
Cheers
Maurice
--
Maurice Han Tong LING, BSc(Hons)(Biochem), AdvDipComp, CPT, SSN, FIFA,
MASBMB, MAMBIS, MACM
Doctor of Philosophy (Science) Candidate, The University of Melbourne
mobile: +61 4 22781753, +65 96669233
mailing address: Department of Zoology, The University of Melbourne
Royal Parade, Parkville, Victoria 3010, Australia
residence: 9/41 Dover Street, Flemington, Victoria 3031, Australia
resume: http://maurice.vodien.com/maurice_resume.pdf
www: http://www.geocities.com/beldin79/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050618/ed83c433/mauriceling.vcf
From richardjones at optushome.com.au Sat Jun 18 00:25:50 2005
From: richardjones at optushome.com.au (Richard Jones)
Date: Sat, 18 Jun 2005 08:25:50 +1000
Subject: [melbourne-pug] developing something as a group
In-Reply-To: <42B2D9CE.6010801@acm.org>
References: <20050617083701.20086.qmail@webmastery.com.au>
<42B2D9CE.6010801@acm.org>
Message-ID: <200506180825.54932.richardjones@optushome.com.au>
On Sat, 18 Jun 2005 12:10 am, Maurice Ling wrote:
> 1. Is there a way or system to satisfy the dependencies if I choose to
> install a package?
> 2. When I upgrade my Python installation, is there a way to re-install
> all the packages of my older python version into my newer python version?
For the information of others on this list, there's a number of projects alive
at present attempting to satisfy these concerns. See:
http://wiki.python.org/moin/DistutilsProjects
Richard
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050618/83e6cf66/attachment.pgp
From mauriceling at acm.org Sat Jun 18 01:11:55 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sat, 18 Jun 2005 09:11:55 +1000
Subject: [melbourne-pug] developing something as a group
In-Reply-To: <200506180825.54932.richardjones@optushome.com.au>
References: <20050617083701.20086.qmail@webmastery.com.au>
<42B2D9CE.6010801@acm.org>
<200506180825.54932.richardjones@optushome.com.au>
Message-ID: <42B358BB.4050404@acm.org>
Richard Jones wrote:
>On Sat, 18 Jun 2005 12:10 am, Maurice Ling wrote:
>
>
>>1. Is there a way or system to satisfy the dependencies if I choose to
>>install a package?
>>2. When I upgrade my Python installation, is there a way to re-install
>>all the packages of my older python version into my newer python version?
>>
>>
>
>For the information of others on this list, there's a number of projects alive
>at present attempting to satisfy these concerns. See:
>
> http://wiki.python.org/moin/DistutilsProjects
>
>
> Richard
>
>
Uraga seems to be the closest to what I have in mind.
I chose to put this up for development in the group because it affects
everyone and although I intend to start it as an academic project, I may
not have all the time it needs to see the daylight. Having said that, if
the group is interested to develop it to production, I am willing to
take the lead.
I have a very raw prototype in sourceforge, under ib-dwb/centipyde. Just
go to the CVS and take a look.
maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050618/8dd58761/mauriceling.vcf
From cropleyb at yahoo.com.au Sat Jun 18 02:46:08 2005
From: cropleyb at yahoo.com.au (Bruce Cropley)
Date: Sat, 18 Jun 2005 10:46:08 +1000 (EST)
Subject: [melbourne-pug] developing something as a group
In-Reply-To: <42B2D9CE.6010801@acm.org>
Message-ID: <20050618004608.78747.qmail@web60216.mail.yahoo.com>
> >I think a group project is a great idea but perhaps
> we should look at
> >developing smaller things and doing a "sprint". My
> thoughts would be
> >to have several small projects and try to complete
> them in no more
> >than a day. In small groups with one or more
> experienced Python
> >people helping those who may well have lots of
> programming knowledge,
> >just not in Python.
If we choose any particular project and work on it for
more than one session as the official Melb PUG
meeting, the danger is that the people who are
interested in python but not in that project will not
come along. I believe that a sprint for a particular
open source project would be a more effective way to
achieve something while sharing knowledge.
If anyone is interested practicing python and Test
Driven Development, let me know. I'm running a little
group that meets every two or three weeks. We usually
start from scratch on a new mini-project so new people
can join at any time. The next session is on this
Thursday at Zanzibar in Collins Place starting around
6:15. We are going to do some refactoring of a project
that we started last time (unique id server), so I can
mail you the source as it stands if you are
interested.
As someone else mentioned, small projects and
presentations don't necessarily show up the recurring
issues that you get with large projects. Also,
presentations of experiences with large projects may
not be as useful to people who haven't experienced the
same problems on similar projects.
Bruce
Send instant messages to your online friends http://au.messenger.yahoo.com
From mauriceling at acm.org Sat Jun 18 03:01:25 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sat, 18 Jun 2005 11:01:25 +1000
Subject: [melbourne-pug] developing something as a group
In-Reply-To: <20050618004608.78747.qmail@web60216.mail.yahoo.com>
References: <20050618004608.78747.qmail@web60216.mail.yahoo.com>
Message-ID: <42B37265.2040406@acm.org>
>If we choose any particular project and work on it for
>more than one session as the official Melb PUG
>meeting, the danger is that the people who are
>interested in python but not in that project will not
>come along. I believe that a sprint for a particular
>open source project would be a more effective way to
>achieve something while sharing knowledge.
>
>
>
Sprints may be good as an active spur of development. But it may not be
the only way we can do. We can also work on a project (open source) as a
group. If I am not wrong, this idea of developing something as user
group had been done before. And I think that a user group does provide
the environment for it. Own mailing list, periodic meetings, geographic
vicinity makes it easier to have coffee...
maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050618/f40b88d1/mauriceling.vcf
From abriggs at westnet.com.au Sat Jun 18 05:07:37 2005
From: abriggs at westnet.com.au (Anthony Briggs)
Date: Sat, 18 Jun 2005 13:07:37 +1000
Subject: [melbourne-pug] developing something as a group
In-Reply-To: <42B358BB.4050404@acm.org>
References: <20050617083701.20086.qmail@webmastery.com.au> <42B2D9CE.6010801@acm.org> <200506180825.54932.richardjones@optushome.com.au>
<42B358BB.4050404@acm.org>
Message-ID: <42B38FF9.2070108@westnet.com.au>
Maurice Ling wrote:
> Richard Jones wrote:
>
>> On Sat, 18 Jun 2005 12:10 am, Maurice Ling wrote:
>>
>>> 1. Is there a way or system to satisfy the dependencies if I choose to
>>> install a package?
>>> 2. When I upgrade my Python installation, is there a way to re-install
>>> all the packages of my older python version into my newer python
>>> version?
>>
>> For the information of others on this list, there's a number of
>> projects alive at present attempting to satisfy these concerns. See:
>>
>> http://wiki.python.org/moin/DistutilsProjects
>>
> Uraga seems to be the closest to what I have in mind.
> ...
> I have a very raw prototype in sourceforge, under ib-dwb/centipyde. Just
> go to the CVS and take a look.
Just as a minor aside, there's often a tendency amongst programmers to
reinvent the wheel - I've been guilty of it myself on a few occasions. A
better way to go might be to get a few interested developers together
and add the functionality that you're missing to Uraga.
It shouldn't take too long to get up to speed, and the main developer of
whichever package you choose will probably be glad of the help :)
Just my 2c worth,
Ant
From abriggs at westnet.com.au Sat Jun 18 06:24:36 2005
From: abriggs at westnet.com.au (Anthony Briggs)
Date: Sat, 18 Jun 2005 14:24:36 +1000
Subject: [melbourne-pug] developing something as a group
In-Reply-To: <20050618004608.78747.qmail@web60216.mail.yahoo.com>
References: <20050618004608.78747.qmail@web60216.mail.yahoo.com>
Message-ID: <42B3A204.4020709@westnet.com.au>
Bruce Cropley wrote:
>>Huw Davies wrote:
>>
>>> I think a group project is a great idea but perhaps we should
>>> look at developing smaller things and doing a "sprint". My
>>> thoughts would be to have several small projects and try to
>>> complete them in no more than a day. In small groups with one or
>>> more experienced Python people helping those who may well have
>>> lots of programming knowledge, just not in Python.
>
> If we choose any particular project and work on it for
> more than one session as the official Melb PUG
> meeting, the danger is that the people who are
> interested in python but not in that project will not
> come along. I believe that a sprint for a particular
> open source project would be a more effective way to
> achieve something while sharing knowledge.
To my mind the better way to go (unless you're really sure most of the
group will be interested) is to have a mini-meeting or sprint at some
point, but don't do it at the meetings. There are some pretty key
advantages to doing it externally:
* Coordinating 14-15 people on a project is fairly daunting - I find
that development goes much faster with a smaller group of 4-5.
* You'd generally want to be pretty sure what you're working on -
thrashing around trying to figure out what you're going to be doing .
Getting agreement from 15 people would be similarly hard.
* The smaller group could present their 'findings' to the larger groups,
and you might get more interest by introducing the project gradually, or
feedback from people who may not be interested in development, but can
still contribute valuable ideas.
* If you're developing during a meeting, it's likely to be after work
for a lot of people, so you'll be tired and won't be as productive.
Meeting for a day during the weekend might be a better option for some.
* Development normally involves working at a computer, and not everyone
may have a laptop they could bring. For a smaller group it would be
easier to borrow one or two.
* If you were going to do some sort of sprint at a meeting, you could
work on multiple projects (eg. 3 projects x 5 peoople)
Not that I've averse to some sort of group development - in fact, quite
the opposite, but I think we should start small and work up from there.
The meetings are a perfect place to recruit new developers, though :)
> As someone else mentioned, small projects and
> presentations don't necessarily show up the recurring
> issues that you get with large projects. Also,
> presentations of experiences with large projects may
> not be as useful to people who haven't experienced the
> same problems on similar projects.
You could sidestep these issues by adding functionality to a larger
project - provided you chose something that was relatively easy to do.
Perhaps we should be considering projects that would be fun to add to,
and then someone could give us all an overview of how the code fits
together, so that the barriers for getting up to speed on a new project
are much lower?
Anthony
From mauriceling at acm.org Sat Jun 18 09:27:37 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sat, 18 Jun 2005 17:27:37 +1000
Subject: [melbourne-pug] developing something as a group
In-Reply-To: <42B38FF9.2070108@westnet.com.au>
References: <20050617083701.20086.qmail@webmastery.com.au> <42B2D9CE.6010801@acm.org> <200506180825.54932.richardjones@optushome.com.au> <42B358BB.4050404@acm.org>
<42B38FF9.2070108@westnet.com.au>
Message-ID: <42B3CCE9.9090600@acm.org>
>>>>1. Is there a way or system to satisfy the dependencies if I choose to
>>>>install a package?
>>>>2. When I upgrade my Python installation, is there a way to re-install
>>>>all the packages of my older python version into my newer python
>>>>version?
>>>>
>>>>
>>>For the information of others on this list, there's a number of
>>>projects alive at present attempting to satisfy these concerns. See:
>>>
>>> http://wiki.python.org/moin/DistutilsProjects
>>>
>>>
>>>
>>Uraga seems to be the closest to what I have in mind.
>>...
>>I have a very raw prototype in sourceforge, under ib-dwb/centipyde. Just
>>go to the CVS and take a look.
>>
>>
>
>Just as a minor aside, there's often a tendency amongst programmers to
>reinvent the wheel - I've been guilty of it myself on a few occasions. A
>better way to go might be to get a few interested developers together
>and add the functionality that you're missing to Uraga.
>
>It shouldn't take too long to get up to speed, and the main developer of
>whichever package you choose will probably be glad of the help :)
>
>
>
I've looked more closely at Uraga (sf.net/projects/uraga), especially at
the codes in CVS. From what I can see, neither Uraga nor Centipyde (my
attempt) is off the ground yet. As far as Uraga is concerned, they have
place Centipyde on the radar. Essentially, both projects are trying to
implement PEP 262.
Anthony, I think all of us are guilty of reinventing the wheel some time
or another. I've not known of anyone that hadn't programmed something
that they knew had existed before, some time in their development
career. What we can do is reduce that. In this case, Uraga or Centipyde,
it is hard to day which way to go, and I suppose this is the case with
many package managers for Linux (rpm, apt, yum, emerge, etc etc)...
maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050618/f598a368/mauriceling.vcf
From daedalus at eigenmagic.com Sat Jun 18 10:12:39 2005
From: daedalus at eigenmagic.com (Justin Warren)
Date: Sat, 18 Jun 2005 18:12:39 +1000
Subject: [melbourne-pug] Thank you Anthony and Intrepid
In-Reply-To: <20050617083701.20086.qmail@webmastery.com.au>
References: <20050617083701.20086.qmail@webmastery.com.au>
Message-ID: <1119082359.6006.30.camel@localhost.localdomain>
Interesting discussion stemming from Mike's original post.
On Fri, 2005-06-17 at 08:37 +0000, Mike Dewhirst wrote:
> By way of saying thanks, I thought I'd make a proposal.
>
> We need an idea-path so newcomers understand where we are. What are the
> shared objectives and hopes regarding the group?
>
> Personally, I want to learn to use Python effectively. For me that means ...
>
> - interface to Apache on Linux
> - interface to SQL92 standard RDBMS
> - python application development
> - python "standard" libraries
> - python special purpose libraries
> - rich client front ends via wxPython and (probably) Boa Constructor
> - thin client front ends via W3C standard browsers
> - any other standards-based development infrastructure which is python
> related
> - agile methodologies applied to python development
I have to admit I haven't really thought about my reasons for coming
along to mel-pug meetings. I think I figured I'd just turn up a few
times and see how things went. And then I somehow ended up volunteering
to give a talk. :)
> I would like to propose that in future the group focuses on two things per
> meeting ...
>
> 1. A presentation by anyone competent in an item from the above list
I think sharing knowledge is one of the reasons I'd want to turn up;
listen to other's experiences and learn from what they've done, while
also sharing my own experiences, hopefully to someone else's benefit.
> 2. Actually developing an application to be owned by the group and with
> potential to become a useful open-source project.
>
> The idea for the group project is to put best-practices into context.
> Hopefully, best practice in any area is likely to be contentious. I say,
> bring it on :) I want to learn.
Hmm. I agree with others' suggestions that this be treated as a Special
Interest Group kind of thing that happens as an adjunct to the group.
Like many of us, I already have multiple projects of my own that I don't
get to spend enough time on, as well as earning a living. I'm not really
interested in being part of another project as a major focus for the
group.
I'd rather learn things from the group that I can apply to my own
projects and work. If there are lessons learned from the group project,
then by all means, share them with the whole group. Perhaps a lively
discussion on a particular problem that is encountered on anyone's
project, including the group one, would be a good way to learn? "What do
people think is the best way to do N?" kind of thing.
> My own view (developing as I write) is that group discussion of best
> practices for a real project will have far more value for everyone than
> hearing general good advice. I would also be very keen to see working code
> constructively criticised in very specific context. The idea of an ongoing
> context will also have drawing power at meetings.
I can see the appeal of this approach with those new to Python, and I
agree that the best way to learn is to get your hands dirty. I'd say
that joining the group project would be a good way for Python newbies to
get involved with a Python project and learn by doing, and would
certainly be a drawcard for them. However, I'd prefer that the group not
become centred around this one project as that could limit the scope of
discussion topics.
I think that by discussing the variety of projects people have been
involved with we will all have the opportunity of learning something
new, or perhaps gain another perspective on how we're doing things with
our own projects. A group project would allow those without their own
projects to put some of these lessons into practice.
Oh, and on reinventing the wheel, or duplicating another project; I
don't think it's necessarily a bad thing, for a couple of reasons:
- Solving a problem is a good way to learn. You usually start off
learning a new language by reimplementing "Hello World!" for the 472nd
time, after all.
- If it's been done before, you can learn from others' mistakes and
hopefully do it better. If you're trying to learn something new, you can
get an Aha! moment of "Oh right, so that's how you do that..." by
referring to someone else's work.
- Competing projects enable the best one to win out by virtue of being
better, hopefully.
It all depends on what you're trying to achieve. Why build a house when
you can buy one ready-made?
--
Justin Warren
From mauriceling at acm.org Sat Jun 18 10:37:43 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sat, 18 Jun 2005 18:37:43 +1000
Subject: [melbourne-pug] Thank you Anthony and Intrepid
In-Reply-To: <1119082359.6006.30.camel@localhost.localdomain>
References: <20050617083701.20086.qmail@webmastery.com.au>
<1119082359.6006.30.camel@localhost.localdomain>
Message-ID: <42B3DD57.8040307@acm.org>
>Hmm. I agree with others' suggestions that this be treated as a Special
>Interest Group kind of thing that happens as an adjunct to the group.
>Like many of us, I already have multiple projects of my own that I don't
>get to spend enough time on, as well as earning a living. I'm not really
>interested in being part of another project as a major focus for the
>group.
>
>
I agree that a group project shouldn't be central or even near central
to the group. It has to be a SIG kinda thing so that members who aren't
interested in the project can still participate in the group meetings.
>I think that by discussing the variety of projects people have been
>involved with we will all have the opportunity of learning something
>new, or perhaps gain another perspective on how we're doing things with
>our own projects. A group project would allow those without their own
>projects to put some of these lessons into practice.
>
>
I agree with this.
>Oh, and on reinventing the wheel, or duplicating another project; I
>don't think it's necessarily a bad thing, for a couple of reasons:
>
>- Solving a problem is a good way to learn. You usually start off
>learning a new language by reimplementing "Hello World!" for the 472nd
>time, after all.
>- If it's been done before, you can learn from others' mistakes and
>hopefully do it better. If you're trying to learn something new, you can
>get an Aha! moment of "Oh right, so that's how you do that..." by
>referring to someone else's work.
>- Competing projects enable the best one to win out by virtue of being
>better, hopefully.
>
>It all depends on what you're trying to achieve. Why build a house when
>you can buy one ready-made?
>
>
>
I think 2 popular reasons are:
1. there is no house that fits my design
2. it is the kinda ego-like thing about living in the house you designed
and built.
maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050618/3d3c7261/mauriceling.vcf
From daedalus at eigenmagic.com Sat Jun 18 10:56:34 2005
From: daedalus at eigenmagic.com (Justin Warren)
Date: Sat, 18 Jun 2005 18:56:34 +1000
Subject: [melbourne-pug] Thank you Anthony and Intrepid
In-Reply-To: <42B3DD57.8040307@acm.org>
References: <20050617083701.20086.qmail@webmastery.com.au>
<1119082359.6006.30.camel@localhost.localdomain>
<42B3DD57.8040307@acm.org>
Message-ID: <1119084994.6006.45.camel@localhost.localdomain>
On Sat, 2005-06-18 at 18:37 +1000, Maurice Ling wrote:
> >It all depends on what you're trying to achieve. Why build a house when
> >you can buy one ready-made?
> >
> >
> >
> I think 2 popular reasons are:
> 1. there is no house that fits my design
> 2. it is the kinda ego-like thing about living in the house you designed
> and built.
Exactly. ;)
--
Justin Warren
From abriggs at westnet.com.au Sat Jun 18 11:05:17 2005
From: abriggs at westnet.com.au (Anthony Briggs)
Date: Sat, 18 Jun 2005 19:05:17 +1000
Subject: [melbourne-pug] Thank you Anthony and Intrepid
In-Reply-To: <1119082359.6006.30.camel@localhost.localdomain>
References: <20050617083701.20086.qmail@webmastery.com.au>
<1119082359.6006.30.camel@localhost.localdomain>
Message-ID: <42B3E3CD.3030608@westnet.com.au>
Justin Warren wrote:
> I have to admit I haven't really thought about my reasons for coming
> along to mel-pug meetings. I think I figured I'd just turn up a few
> times and see how things went. And then I somehow ended up volunteering
> to give a talk. :)
I think that should be "volunteering" :)
> Hmm. I agree with others' suggestions that this be treated as a Special
> Interest Group kind of thing that happens as an adjunct to the group.
> Like many of us, I already have multiple projects of my own that I don't
> get to spend enough time on, as well as earning a living. I'm not really
> interested in being part of another project as a major focus for the
> group.
Sounds like you're at the other end of the spectrum - too many projects,
not enough time. Perhaps you should be looking for a couple of people to
help out? I'm not sure about how much specialist knowledge would be
needed for beepy, but if you're like me there are probably another
couple of projects eating up your time...
> I'd rather learn things from the group that I can apply to my own
> projects and work. If there are lessons learned from the group project,
> then by all means, share them with the whole group. Perhaps a lively
> discussion on a particular problem that is encountered on anyone's
> project, including the group one, would be a good way to learn? "What do
> people think is the best way to do N?" kind of thing.
> ... I'd prefer that the group not
> become centred around this one project as that could limit the scope of
> discussion topics.
I definitely agree with both of these points.
> Oh, and on reinventing the wheel, or duplicating another project; I
> don't think it's necessarily a bad thing, for a couple of reasons:
As long as you have a definite reason to duplicate a project, that's
fair enough, but I've seen quite a few people produce duplicates that
aren't necessarily better, when they could have built upon the existing one.
> - Competing projects enable the best one to win out by virtue of being
> better, hopefully.
Sure, but cooperation also produces good results :)
Anthony
From mauriceling at acm.org Sat Jun 18 14:17:31 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sat, 18 Jun 2005 22:17:31 +1000
Subject: [melbourne-pug] members' OSS efforts
In-Reply-To: <42B3E3CD.3030608@westnet.com.au>
References: <20050617083701.20086.qmail@webmastery.com.au> <1119082359.6006.30.camel@localhost.localdomain>
<42B3E3CD.3030608@westnet.com.au>
Message-ID: <42B410DB.3000709@acm.org>
>>Hmm. I agree with others' suggestions that this be treated as a Special
>>Interest Group kind of thing that happens as an adjunct to the group.
>>Like many of us, I already have multiple projects of my own that I don't
>>get to spend enough time on, as well as earning a living. I'm not really
>>interested in being part of another project as a major focus for the
>>group.
>>
>>
>
>Sounds like you're at the other end of the spectrum - too many projects,
>not enough time. Perhaps you should be looking for a couple of people to
>help out? I'm not sure about how much specialist knowledge would be
>needed for beepy, but if you're like me there are probably another
>couple of projects eating up your time...
>
>
>
Kinda same story here actually. That is why I am thinking of throwing
Centipyde back for community development, even though I thought of doing
it as an academic project (which is my statement made in catelog-sig and
python mailing list). Just trying to solicit some interests here...
Nevertheless I think our group here is pretty active in OSS development,
and this is exciting (for me at least). Personally, I think at least
this means that we can look at each other's OSS work and provide
constructive criticisms...
Maybe we should declare our open-source development efforts (python or
not)...... For me, I am involved in the following developments:
1. For my thesis, I am developing a biological modeling and simulation
platform (named "Mosirium"). At the moment, I am trying to get IP
clearance from my industrial sponsor to release it in SF.
2. For my thesis again, I am developing a pipeline system for text
analysis and mining for protein-protein and gene-protein interactions,
named "Muscorian". It is more of a meta-tool that incorporates existing
text-mining and analysis engines for synergistic use. The 1st engine
incorporated is MontyLingua, written by Hugo Liu in MIT Media Labs.
MontyLingua is in Python and is the core of both OpenMinds project and
ConceptNet project. I've written a paper on Muscorian for submission to
a conference next year, and getting IP clearance for release at this moment.
3. For my AdvDip in Computing, I've prototyped a data warehouse builder
(IB-DWB) based on Borland Interbase 6 open edition (currently developed
as Firebird) and published a conference paper
(http://arxiv.org/pdf/cs.DB/0307015). IB-DWB (sf.net/projects/ib-dwb) is
developed in Borland Delphi. Currently, my housemate and myself is
re-designing it as OpenDWS (sf.net/projects/opendws). OpenDWS will be
developed in Java/Jython/Python combination.
4. Centipyde as described in previous mails. It is almost a spur of the
moment thing but I do think it is worth it. Personally, I'll work on it
with the little time I have for handling the Python dependencies I may
have in Mosirium, Muscorian, and OpenDWS in future.
5. This is embarassing. I started JMaths (sf.net/projects/jmaths) as a
library of mathematical support for Java. But sadly had lost momentum.
Currently, contemplating of making the "J" from Java to Jython......
So far that's all with my open-source development and blunders. Of all,
I can certainly say that only Mosirium and Muscorian will see the
daylight in near future. They are the foundation stones for my PhD.
That's all from me....
Cheers
Maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050618/b7e12f6a/mauriceling.vcf
From richardjones at optushome.com.au Sun Jun 19 02:19:14 2005
From: richardjones at optushome.com.au (Richard Jones)
Date: Sun, 19 Jun 2005 10:19:14 +1000
Subject: [melbourne-pug] Another project idea (oldie but a goodie)
Message-ID: <200506191019.18552.richardjones@optushome.com.au>
Maurice has already mentioned his efforts to develop a download-and-install
tool. I've promoted the link to the distutils projects page. Something people
here might not be aware of though is that PyPI is one of my projects (apart
from the sprinting at PyCon this year, I've been the sole developer). It
needs work, in a number of areas. I welcome volunteers, especially those in
my own timezone, to work on the code with me.
And if you work on PyPI, you'll get a nice warm fuzzy feeling that you're
really helping Python the language in a big way :)
I concur with others on this list though - this sort of thing is definitely
adjunct to the PUG itself. It should impact as little as possible on the
regular meetings.
Richard
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050619/a726b04e/attachment.pgp
From mauriceling at acm.org Sun Jun 19 02:30:05 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sun, 19 Jun 2005 10:30:05 +1000
Subject: [melbourne-pug] Another project idea (oldie but a goodie)
In-Reply-To: <200506191019.18552.richardjones@optushome.com.au>
References: <200506191019.18552.richardjones@optushome.com.au>
Message-ID: <42B4BC8D.9050207@acm.org>
Richard Jones wrote:
>Maurice has already mentioned his efforts to develop a download-and-install
>tool. I've promoted the link to the distutils projects page. Something people
>here might not be aware of though is that PyPI is one of my projects (apart
>from the sprinting at PyCon this year, I've been the sole developer). It
>needs work, in a number of areas. I welcome volunteers, especially those in
>my own timezone, to work on the code with me.
>
>And if you work on PyPI, you'll get a nice warm fuzzy feeling that you're
>really helping Python the language in a big way :)
>
>
>
>I concur with others on this list though - this sort of thing is definitely
>adjunct to the PUG itself. It should impact as little as possible on the
>regular meetings.
>
>
> Richard
>
>
Richard,
I certainly do not know that you are the sole developer of PyPI, though
I do realise that you are in it. To me, PyPI is a web-based browser of
packages. Pleasse enlighten me on its other functions.
PyPI, to me, lacks that download-and-install tool I've been mentioning.
So, I'll be 150% glad to have my tool to fill that gap.
Cheers
Maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050619/4ba0cb1b/mauriceling.vcf
From richardjones at optushome.com.au Sun Jun 19 02:55:42 2005
From: richardjones at optushome.com.au (Richard Jones)
Date: Sun, 19 Jun 2005 10:55:42 +1000
Subject: [melbourne-pug] Another project idea (oldie but a goodie)
In-Reply-To: <42B4BC8D.9050207@acm.org>
References: <200506191019.18552.richardjones@optushome.com.au>
<42B4BC8D.9050207@acm.org>
Message-ID: <200506191055.42669.richardjones@optushome.com.au>
On Sun, 19 Jun 2005 10:30 am, Maurice Ling wrote:
> I certainly do not know that you are the sole developer of PyPI, though
> I do realise that you are in it.
As I mentioned, I was briefly not the sole developer :) I had considerable
help from other developers during the last PyCon sprints:
http://wiki.python.org/moin/DistutilsSprint
Outside of the sprinting, it's just me (so far ;)
> To me, PyPI is a web-based browser of
> packages. Pleasse enlighten me on its other functions.
PyPI is intended to be a central, authoritative database of packages (both
metadata and distribution files). See its PEP:
http://www.python.org/peps/pep-0301.html
To access that information, it needs an open interface for people to submit,
query and retrieve information. On the web user side of things, it's
currently fine, but on the programmatic side, the interface is quite lacking
(submit is fine, but query and retrieval are practically non-existent).
There is facility in PyPI for XML-RPC, but no real query methods have been
written. Ian Bicking has proposed a set of methods to the catalog-sig. See:
http://mail.python.org/pipermail/catalog-sig/2005-May/000634.html
So far no-one's stepped forward with code. I'm busy with work at the moment.
One of these days I'll get back to PyPI...
> PyPI, to me, lacks that download-and-install tool I've been mentioning.
> So, I'll be 150% glad to have my tool to fill that gap.
You'll be needing support from the web interface to achieve this - see Ian's
proposal above. The beauty is that once the web interface supports queries,
all of the download-related projects benefit (EasyInstall, Centipyde,
Uraga, ...)
Richard
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050619/87ef6a75/attachment.pgp
From mauriceling at acm.org Sun Jun 19 04:13:53 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Sun, 19 Jun 2005 12:13:53 +1000
Subject: [melbourne-pug] Another project idea (oldie but a goodie)
In-Reply-To: <200506191055.42669.richardjones@optushome.com.au>
References: <200506191019.18552.richardjones@optushome.com.au>
<42B4BC8D.9050207@acm.org>
<200506191055.42669.richardjones@optushome.com.au>
Message-ID: <42B4D4E1.30609@acm.org>
>>PyPI, to me, lacks that download-and-install tool I've been mentioning.
>>So, I'll be 150% glad to have my tool to fill that gap.
>>
>>
>
>You'll be needing support from the web interface to achieve this - see Ian's
>proposal above. The beauty is that once the web interface supports queries,
>all of the download-related projects benefit (EasyInstall, Centipyde,
>Uraga, ...)
>
>
I can see where you are coming from. On the assumption that PyPI is consistent (which is a set of assumptions), download-and-install related projects (all can be seen as implementations of PEP262 one way or another) can query PyPI through the proposed XML-RPC interface, download the stuffs and install the package.
In order for that to happen, from the viewpoint of a
download-and-install project, the following set of assumptions must be met in order not to reimplement PyPI in a way or another:
1. each entry in PyPI must have a download URL, otherwise it is useless. (entry for biopython 1.21 has no download url, www.python.org/pypi/biopython/1.21). Entry for book 1.0 (www.python.org/pypi/book/1.0) points to non-existing urls.
2. the download url must be a read downloadable file and a website where you can search for downloads. In this aspect, entry for alo-aes 0.3 (www.python.org/pypi/alo-aes/0.3) meets the criteria while IbPy 0.4 (http://www.python.org/pypi/IbPy/0.4) does not. Although if the download url points to a Sourceforge website, it may be vaguely possible to reconstruct the url to the actual downloadable file (as in the case of matplotlib 0.71, www.python.org/pypi/matplotlib/0.71), it is not possible when the file is hosted out of SF (as in the case of MontyLingua 2.1, www.python.org/pypi/MontyLingua/2.1)
3. there must be a way to tell if any dependencies are needed. Currently, entries in PyPI has no such information.
4. if assumption 3 is met, the names of the dependencies must be consistent with that of the actual package name in PyPI. Say if a package (e.g. Scientific Python 2.4.6, www.python.org/pypi/ScientificPython/2.4.6) requires netCDF 3.6.0, then the name netCDF must be in PyPI (which currently it isn't as www.python.org/pypi/netCDF/3.6.0 is non-existent).
Cheers
Maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050619/807e2266/mauriceling.vcf
From qubero at gmail.com Mon Jun 20 04:40:28 2005
From: qubero at gmail.com (Peter Galaxy)
Date: Mon, 20 Jun 2005 12:40:28 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To: <42B4D4E1.30609@acm.org>
References: <200506191019.18552.richardjones@optushome.com.au>
<42B4BC8D.9050207@acm.org>
<200506191055.42669.richardjones@optushome.com.au>
<42B4D4E1.30609@acm.org>
Message-ID:
I'm between ISPs right now so I don't get to check my email every day, so
I've just been reading through this list's posts. Here's a summary about
what people have said on doing a group project:
There is debate about whether the PUG meetings should be centred around the
group project, or where the group prpoject should be put asside to
mini-meetings.
There is also debate about whether to have sprints on many projects or to
have fewer (one?) long running projects.
And whether to continue an existing project (leveraging peoples existing
projects, but perhaps giving less satisfaction to others) versus a totally
new project (possibly reinventing the wheel)
And how to choose to a project that is inclusive of everyone (and doesn't
scare people away) or whether to specialise (away from the main group).
Maybe i inferred that last one rather than read it is a post. But anyway.
Okay, now for the "how do we solve these issues?" bit.
Firstly, we need to acknowledge that the reasons for joining the Python
Users Group are diverse. People come to learn new skills, learn more about
their own areas of interest, socialise, trade obsolete hardware, yada yada.
So why do we want to start a group project? to get some hands on experience,
work in a group, learn skills, satification of making something, and maybe
to write a driver for that 8 inch floppy drive with a Z80 controller.
So the goals of a group project aren't so radically different from the goals
of the PUG in general. But still people are scared to spend precious time
hacking together some useless code with some know-nothing geeks who don't
understand the first thing about reloading microcode on System/370
Mainframes of 1967.
So what can be done?
Make the group project a part of, but not the central part of, PUG meetings:
- Have a couple of people give lightning talks (5-10 minutes) on their pet
projects or proposals one month (July if people are ready: after the main
presentation. Could go on for a number of months if there are enough
presentations)
- Have a 10 minute attempt to meet consensus on which project(s) we'd like
to take up as a group (August)
- Spend 5-10 minutes organising people, times, places and coding schedules
as well (August or Sept? which can be taken into email if it goes too long)
- Organise a sprint to get people involved, to be run after the main meeting
(October)
- and have a full presentation from the group when there's something to
present. (Aim for March)
If anyone would be scared away by 10-20 minutes of the meeting being taken
up by lightning talks and sceduling, please say so. I don't see any reason
why a group project would need to be the central part of PUG meetings, and
at the same time it doesn't need to be totally seperate. Likewise for
sprints vs long-term, there's no reason we can't do it both ways.
The main issue that I see is finding a project broad enough that enough
people can bring their own skill to it, and interesting enough that people
will be willing to make a commitment. The main ideas proposed so far are
PyPI (Richard Jones) and Centipyde (Maurice). I don't know much about either
so I'd like to see Richard and Maurice "volunteer" to give short (5-10
minute) presentations on these projects in July or August if they're willing
and able, along with anyone else who has ideas. I'd like to see us make the
group project a part of, but not the central part of, PUG meetings.
Peter (Pengo) Galaxy.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/melbourne-pug/attachments/20050620/a1dc3c7a/attachment.htm
From abulka at bigpond.net.au Mon Jun 20 05:14:26 2005
From: abulka at bigpond.net.au (Andy Bulka)
Date: Mon, 20 Jun 2005 13:14:26 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To:
Message-ID: <026101c57546$2a247500$6901a8c0@a30p>
> I'd like to see us make the group project a part of, but not the central
part of, PUG meetings.
I program in python most of the week, so would not be keen to see the group
turn into a externsion of my work day! I am happy for a group project to
exist in some way, and maybe I too would would eventually contribute or
learn something - who knows. But the main focus I think should be
presentation based.
Presentations need a projector of course and I think that you need at
least 30- 45min to get a decent presentation. Presentations should always
include a code component so that people can watch some code being built and
executed.
just my 2c
Andy Bulka
http://www.atug.com/andypatterns
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/melbourne-pug/attachments/20050620/7045588a/attachment.htm
From mauriceling at acm.org Mon Jun 20 05:30:58 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Mon, 20 Jun 2005 13:30:58 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To:
References: <200506191019.18552.richardjones@optushome.com.au> <42B4BC8D.9050207@acm.org> <200506191055.42669.richardjones@optushome.com.au> <42B4D4E1.30609@acm.org>
Message-ID: <42B63872.2010405@acm.org>
> There is debate about whether the PUG meetings should be centred
> around the group project, or where the group prpoject should be put
> asside to mini-meetings.
I think we can safely say that the general "pressure" is not to center
the group meeting around a project. The focus of the group should never
be on the project and the projects (if any) should be adjunct to the group.
>
> There is also debate about whether to have sprints on many projects or
> to have fewer (one?) long running projects.
Maybe either way does not matter, as long as there is a lead in each
project.
>
> And whether to continue an existing project (leveraging peoples
> existing projects, but perhaps giving less satisfaction to others)
> versus a totally new project (possibly reinventing the wheel)
A few existing projects may be a nice thing. I think the general idea I
get from Justin and Richard is that we have more than enough projects
but too few people.
>
> And how to choose to a project that is inclusive of everyone (and
> doesn't scare people away) or whether to specialise (away from the
> main group). Maybe i inferred that last one rather than read it is a
> post. But anyway.
As long as the group isn't focused on projects, we shouldn't be scaring
away people.
>
> Okay, now for the "how do we solve these issues?" bit.
>
>
> Make the group project a part of, but not the central part of, PUG
> meetings:
> - Have a couple of people give lightning talks (5-10 minutes) on
> their pet projects or proposals one month (July if people are ready:
> after the main presentation. Could go on for a number of months if
> there are enough presentations)
> - Have a 10 minute attempt to meet consensus on which project(s) we'd
> like to take up as a group (August)
> - Spend 5-10 minutes organising people, times, places and coding
> schedules as well (August or Sept? which can be taken into email if it
> goes too long)
> - Organise a sprint to get people involved, to be run after the main
> meeting (October)
> - and have a full presentation from the group when there's something
> to present. (Aim for March)
This may be a good idea. After hearing what projects everyone is doing,
I think it may be good to have a slightly deeper view from the horse's
mouth. Listening to what problems they encounter and how they see that
will go is certainly educational.
>
> If anyone would be scared away by 10-20 minutes of the meeting being
> taken up by lightning talks and sceduling, please say so. I don't see
> any reason why a group project would need to be the central part of
> PUG meetings, and at the same time it doesn't need to be totally
> seperate. Likewise for sprints vs long-term, there's no reason we
> can't do it both ways.
I think none of us will want to go to PUG meetings, take out our laptops
and start punching codes. That itself is a preventive force from the
project being the central to the group. We still want variety and a
larger perspective. In my opinions, sprints are just to give an
accelerated spur to a project which may take weeks or months on normal
schedules. For those of us who are on several projects, we know that we
do not develop on each project at the same rate but essentially, we
carry out one-man sprints on each project.
>
> The main issue that I see is finding a project broad enough that
> enough people can bring their own skill to it, and interesting enough
> that people will be willing to make a commitment. The main ideas
> proposed so far are PyPI (Richard Jones) and Centipyde (Maurice). I
> don't know much about either so I'd like to see Richard and Maurice
> "volunteer" to give short (5-10 minute) presentations on these
> projects in July or August if they're willing and able, along with
> anyone else who has ideas. I'd like to see us make the group project a
> part of, but not the central part of, PUG meetings.
Most people can almost smell that I will also like to have a group
project or sort, at least that gives a belonging and ownership of sort.
You can at least tell yourself, what have you contributed to the group.
Taking a step back, even though there are only 2 proposals, PyPI and
Centipyde, it is almost logical that these 2 guys should sit down and
talk to each other. Even after declaring a "conflict of interest", I
still feel that PyPI and Cnetipyde can be synergistic.
As far as short project presentation is concerned, I am happy for it to
happen. So if Anthony Briggs gives the signal, I'll go.
Maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050620/d59b5bbb/mauriceling.vcf
From richardjones at optushome.com.au Mon Jun 20 08:45:26 2005
From: richardjones at optushome.com.au (Richard Jones)
Date: Mon, 20 Jun 2005 16:45:26 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To:
References: <200506191019.18552.richardjones@optushome.com.au>
<42B4D4E1.30609@acm.org>
Message-ID: <200506201645.26619.richardjones@optushome.com.au>
On Mon, 20 Jun 2005 12:40 pm, Peter Galaxy wrote:
> - Have a couple of people give lightning talks (5-10 minutes) on
> their pet projects or proposals one month (July if people are ready: after
> the main presentation. Could go on for a number of months if there are
> enough presentations)
I definitely like this approach.
> - Have a 10 minute attempt to meet consensus on which project(s) we'd like
> to take up as a group (August)
I'd say that once lightning talks are given, all other organisation may be
done in a wiki (on wiki.python.org), with emails to this list if absolutely
necessary (but preferrably not).
> - Organise a sprint to get people involved, to be run after the main
> meeting (October)
A sprint organised around the OSDC would be a good idea, then non-Melbournian
attendees can sprint too. We just do all the hard work of organising it. Of
course, when I say "we" I mean someone other than me ;)
> If anyone would be scared away by 10-20 minutes of the meeting being taken
> up by lightning talks and sceduling, please say so.
Lightning talks are always good value, regardless of whether someone would
want to get involved in a project. It's always good to hear what others are
doing, and a short 5-10 minute lightning talk is a great medium for getting a
bit of publicity. Also, lightning talks are fun as hell. In fact, I'd like to
see a good 20 minutes or so devoted to 4 lightning talks every time we get
together. If we can't make up the 4, then we gang up on each other and get
people to just talk about cool modules they've used recently :)
Could be a good meeting format. Lightning talks followed by an in-depth
presentation by someone?
> I don't know much about
> either so I'd like to see Richard and Maurice "volunteer" to give short
> (5-10 minute) presentations on these projects in July or August if they're
> willing and able, along with anyone else who has ideas.
Definitely.
> I'd like to see us
> make the group project a part of, but not the central part of, PUG
> meetings.
Amen.
Richard
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050620/75dec3af/attachment.pgp
From miked at dewhirst.com.au Mon Jun 20 09:31:31 2005
From: miked at dewhirst.com.au (Mike Dewhirst)
Date: Mon, 20 Jun 2005 17:31:31 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To: <026101c57546$2a247500$6901a8c0@a30p>
References: <026101c57546$2a247500$6901a8c0@a30p>
Message-ID: <42B670D3.5030102@dewhirst.com.au>
Andy Bulka wrote:
> > I'd like to see us make the group project a part of, but not the central
> part of, PUG meetings.
I think that's a good idea.
My original suggestion was to split each meeting into a presentation
taken from a list (which I seeded but to which others should contribute)
plus a group project.
I didn't think of actually doing the project in meeting time but instead
discussing interesting bits of it.
I think someone (Richard or Maurice?) suggested a SIG and I think that
is a very good idea. There could be any number of SIGs which could
discuss their special interests at meetings. I can see some SIGs might
just report on developments in the wider world while others might reveal
project progress from time to time.
The nice thing about SIGs is that they are self-selecting and no-one
needs to be involved if they would find something boring.
As I tried to say in the first place, I guess my real interest is
seeing/hearing group input on best practices - not in theory, not in
general but in specific context. I'd like to see code criticised; design
approaches torn apart and reconstructed; TDD applied in the YAGNI
context; maybe even refactoring to patterns! Unfortunately, I too am
seriously underpaid and overworked so I don't get time to do things
right. Some of that is ignorance, some is inexperience and some is just
cutting corners. I don't (yet) even program in Python unless something
forces me.
So, in agreeing with what most people in this thread have suggested,
here is my summary. I think it would be good to have ...
1. a list of presentation topics (as Anthony has already commenced)
2. a list of SIGs for people to join if they wish
3. time allocated for two 30-40 minute presentations per meeting
4. only one SIG presentation per meeting and
5. a culture of open discussion and constructive criticism
Originally I did suggest a group project but I think the SIGs suggestion
is much better. However, I still think that if any SIGs decide to do a
project we maybe ought to consider IP ownership belonging to the group
with a view to becoming open sourced. That idea doesn't have to be a
showstopper. It could be a presentation topic in its own right.
I know that would be a problem for Maurice's thesis work so I'm not
really sure about all that. I know that my own IP in some areas must
remain a deep secret - partly because it is shared - while in others I'd
be happy to open source it. IP is a tricky thing all by itself and we
can all learn some of the twists and turns by getting involved. That
wouldn't be such a bad thing.
Dismounting
Regards
mike
From daedalus at eigenmagic.com Mon Jun 20 09:41:08 2005
From: daedalus at eigenmagic.com (Justin Warren)
Date: Mon, 20 Jun 2005 17:41:08 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To: <42B670D3.5030102@dewhirst.com.au>
References: <026101c57546$2a247500$6901a8c0@a30p>
<42B670D3.5030102@dewhirst.com.au>
Message-ID: <1119253268.2872.16.camel@norbert>
On Mon, 2005-06-20 at 17:31 +1000, Mike Dewhirst wrote:
> Originally I did suggest a group project but I think the SIGs suggestion
> is much better. However, I still think that if any SIGs decide to do a
> project we maybe ought to consider IP ownership belonging to the group
> with a view to becoming open sourced. That idea doesn't have to be a
> showstopper. It could be a presentation topic in its own right.
>
> I know that would be a problem for Maurice's thesis work so I'm not
> really sure about all that. I know that my own IP in some areas must
> remain a deep secret - partly because it is shared - while in others I'd
> be happy to open source it. IP is a tricky thing all by itself and we
> can all learn some of the twists and turns by getting involved. That
> wouldn't be such a bad thing.
Now there's a topic near and dear to my heart. I'd be really interested
in hearing about other's experiences with Intellectual Property. Perhaps
we can add it to the list of presentation/discussion topics?
--
Justin Warren
From mauriceling at acm.org Mon Jun 20 10:06:19 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Mon, 20 Jun 2005 18:06:19 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To: <42B670D3.5030102@dewhirst.com.au>
References: <026101c57546$2a247500$6901a8c0@a30p>
<42B670D3.5030102@dewhirst.com.au>
Message-ID: <42B678FB.5080405@acm.org>
>I think someone (Richard or Maurice?) suggested a SIG and I think that
>is a very good idea. There could be any number of SIGs which could
>discuss their special interests at meetings. I can see some SIGs might
>just report on developments in the wider world while others might reveal
>project progress from time to time.
>
>
The 1st mention of SIG was by Anthony in the meeting when he gave his
opening "speech" and his words were "SIG is possible as long as it does
not draw too much away from the main group."
>So, in agreeing with what most people in this thread have suggested,
>here is my summary. I think it would be good to have ...
>
>1. a list of presentation topics (as Anthony has already commenced)
>2. a list of SIGs for people to join if they wish
>3. time allocated for two 30-40 minute presentations per meeting
>4. only one SIG presentation per meeting and
>5. a culture of open discussion and constructive criticism
>
>
>
All agreed with both hands.
>Originally I did suggest a group project but I think the SIGs suggestion
>is much better. However, I still think that if any SIGs decide to do a
>project we maybe ought to consider IP ownership belonging to the group
>with a view to becoming open sourced. That idea doesn't have to be a
>showstopper. It could be a presentation topic in its own right.
>
>I know that would be a problem for Maurice's thesis work so I'm not
>really sure about all that. I know that my own IP in some areas must
>remain a deep secret - partly because it is shared - while in others I'd
>be happy to open source it. IP is a tricky thing all by itself and we
>can all learn some of the twists and turns by getting involved. That
>wouldn't be such a bad thing.
>
>
My PhD thesis work (which I did mention in one of the emails) are
currently sealed shut. It is a modeling platform, Mosirium, and a text
analysis platform, Muscorian. I think it is moral in the community that
if someone describes explains or describes this academic work (for the
thesis or what), nobody even thinks of rivaling it.
IP is a tricky issue. It almost requires a schizophreniac to get it well
and right. Having said that, I do have quite a bit of experiences with
IP issues from my former work and graduate school. As mentioned to me by
Prof. Frank Nicholas in Uni of Sydney, it is important to get on top of
IP issues.
As far as Centipyde is concerned, I still have absolute say. I've not
put in my proposal yet.
I think it is fair to say that for any SIG projects, the IP ownership
belongs to the group but entrusted upon the project lead. That is, the
project lead manages and executes the project IP for the group. This is
just to prevent too much red tapes.
Maurice
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050620/da67819f/mauriceling.vcf
From mauriceling at acm.org Mon Jun 20 10:08:21 2005
From: mauriceling at acm.org (Maurice Ling)
Date: Mon, 20 Jun 2005 18:08:21 +1000
Subject: [melbourne-pug] Group Project summary and ideas
In-Reply-To: <1119253268.2872.16.camel@norbert>
References: <026101c57546$2a247500$6901a8c0@a30p> <42B670D3.5030102@dewhirst.com.au>
<1119253268.2872.16.camel@norbert>
Message-ID: <42B67975.3020807@acm.org>
Justin Warren wrote:
>On Mon, 2005-06-20 at 17:31 +1000, Mike Dewhirst wrote:
>
>
>
>>Originally I did suggest a group project but I think the SIGs suggestion
>>is much better. However, I still think that if any SIGs decide to do a
>>project we maybe ought to consider IP ownership belonging to the group
>>with a view to becoming open sourced. That idea doesn't have to be a
>>showstopper. It could be a presentation topic in its own right.
>>
>>I know that would be a problem for Maurice's thesis work so I'm not
>>really sure about all that. I know that my own IP in some areas must
>>remain a deep secret - partly because it is shared - while in others I'd
>>be happy to open source it. IP is a tricky thing all by itself and we
>>can all learn some of the twists and turns by getting involved. That
>>wouldn't be such a bad thing.
>>
>>
>
>Now there's a topic near and dear to my heart. I'd be really interested
>in hearing about other's experiences with Intellectual Property. Perhaps
>we can add it to the list of presentation/discussion topics?
>
>
>
This is really dear to my heart as well. Let's face it, it is a
minefield out there. Sigh......
maurice
--
Maurice Han Tong LING, BSc(Hons)(Biochem), AdvDipComp, CPT, SSN, FIFA,
MASBMB, MAMBIS, MACM
Doctor of Philosophy (Science) Candidate, The University of Melbourne
mobile: +61 4 22781753, +65 96669233
mailing address: Department of Zoology, The University of Melbourne
Royal Parade, Parkville, Victoria 3010, Australia
residence: 9/41 Dover Street, Flemington, Victoria 3031, Australia
resume: http://maurice.vodien.com/maurice_resume.pdf
www: http://www.geocities.com/beldin79/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mauriceling.vcf
Type: text/x-vcard
Size: 334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050620/f62bdfb7/mauriceling.vcf
From wafonso at gmail.com Tue Jun 21 04:52:14 2005
From: wafonso at gmail.com (Wilson Afonso)
Date: Tue, 21 Jun 2005 12:52:14 +1000
Subject: [melbourne-pug] Guido van Rossum's presentation at SDForum
Message-ID:
I'm not sure how many people here follow any type of podcasts, but you
may be interested in this: the presentation given by Guido van Rossum
at the SDForum, titled "Building an Open Source Project and Community"
(but talking a lot about Python and how it, and its comunity, compare
to other languages) is available online from the IT Conversations
website.
The audio files are available from
http://www.itconversations.com/shows/detail545.html
Cheers,
-Wilson
--
Wilson Roberto Afonso wafonso at gmail.com
http://www.netwhatever.com/
From richardjones at optushome.com.au Wed Jun 29 01:13:50 2005
From: richardjones at optushome.com.au (Richard Jones)
Date: Wed, 29 Jun 2005 09:13:50 +1000
Subject: [melbourne-pug] Some Zopeness ad the next Melbourne PUG meeting?
Message-ID: <200506290913.54771.richardjones@optushome.com.au>
Hi all,
Last night's OzZope meeting ended up not going ahead. While we stood around in
the cold I suggested that the Zope users might as well just come along to the
Python users group meeting. OzZope meetings usually attract between 5 and 15
people. Presentations at the OzZope meetings are usually pretty short, and
could be of interest to Python users in general. I'm pretty sure that the
interest goes the other way too :)
So, I'm proposing that last night's OzZope meeting be shifted to the next
Python users group meeting, and we grab a 5-10 minute spot for the
presentation we were going to have last night. Any objections from the PUG
side of the fence?
I believe the next PUG meeting will be on Thursday the 21st of July (third
Thursday of the month). The last meeting's info is at:
http://www.livejournal.com/users/melbournepython/
Richard
ps. http://www.ozzope.org/ for those not in the know
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail/melbourne-pug/attachments/20050629/bf2e7646/attachment.pgp