The Python Papers Anthology

MPD WebAMP

Embedding a Python interpreter into your program

Designing Semiconductor Heterostructures with Python

The Python Papers, Volume 2, Issue 4 79
Upcoming Events
The  following  events,  taken  from  the  python.org  events  wiki20,  are  being  held  between
December this year and March 2008.
December 12, 2007: Cologne, Germany, monthly pyCologne meeting 
December 13, 2007:  IRC, Internet,  Kubuntu Tutorials Day features "Get programming with
PyKDE 4" 
March 15, 2008: Portland, Oregon Python workshop SIGCSE'08 (workshop 32) led by Atanas
Radenski 
March 13, 2008: Chicago, Illinois, PyCon 2008 Tutorial Day 
March 14-16, 2008: Chicago, Illinois, PyCon 2008 
March 17-20, 2008: Chicago, Illinois, PyCon 2008 Sprints
To include your  event  in our  next issue,  or  to include expanded event information,  please
contact  us directly  to  ensure that  your  event  is  represented as you would like.  All  events
available from the python.org events wiki will be included with a basic reference.
20  http://wiki.python.org/moin/PythonEvents


Upcoming events

The Python Papers, Volume 2, Issue 4 6
GrrlCamp: An Open Source Python Project for Women
Gloria W. Jacobs
The evolution of Free/Open Source Software was as much fun to watch as it was to help
push  along.  When  I  started  designing  and  developing  software  professionally  in  1989,
compilers cost thousands of dollars and freeware was just a tiny dream held by a small
group of UNIX developers and college professors. 
I worked for a company which ported their software to the AT&T 3B series, the Pyramid
symmetric multiprocessor and the IBM XT. The XT version sold for $200 while the versions
for monster-sized machines sold for tens of thousands of dollars. When I asked why, the
response I got was essentially that companies which can afford these monster machines
can also afford monstrous software and licensing fees. Still  no fair,  I  thought, especially
since the features on all versions were the same. But I was green and scooped up into this
very good job six months before I graduated, so I did not protest. I was happy to have a
company  want  to  hire  and  train  me,  as  well  as  allowing  me  to  finish  my  studies
simultaneously. 
There  simply  were  no  other  options  in  1989.  People  paid  what  companies  wanted  for
software, even down to the compiler and it's supplemental components. The hobbyist in this
field was either rich, or moonlighting on university equipment to design and develop the
software  of  their  dreams (often  with  great  difficulty).  Cross-platform testing  was  nearly
impossible. Lab time had to be negotiated. It was not fun, to say the least.
In  the  early  90's,  I  began  to  see  some small  embedded-systems  companies  play  with
software from the Free Software Foundation. The “gas” (GNU Assembler) was often tweaked
to work with proprietary hardware. The GNU C compiler (gcc) was maintained privately by a
tiny group, as well as professionally by companies charging $7000 a year for support. The
professional bug fixes were not shared with the public and were not rolled back into the free
code base, thus defeating the purpose of the free GNU tools in the first place. We were far
from a solution, but knew we would get there eventually.
As  hardware dropped in  price,  more enthusiasts  and hobbyists  sprung forth,  and more
freeware was developed. Now moonlighting could be done in the comfort  of  one's own
basement, on a better PC, or even on a $700 Motorola ethernet-ready prototype board. The
quality got progressively better, and the projects bigger, to the point where we are today.
Free/Open Source Software, or F/OSS,  dominates the application development world (does
it?) and has infiltrated, but not yet consumed, the embedded systems world. In both cases,
software design and development is easier, cheaper and takes less time to ramp up. We
have a plethora of  tools  to draw upon and start  with,  so that project  time and cost  is
incredibly less than it was 15, 10, or even 5 years ago. F/OSS has exploded all over the
world, enabling everyone with a PC and an idea to explore and create. 
I saw this evolution and helped push it along wherever I could. I helped embedded systems
companies hack the GNU assembler and use other GNU tools to automate building and
testing. I helped set up SAMBA servers for companies, showing them how Linux provides
affordable and reliable file  servers  (and authentication services?)  for  Windows clients.  I
taught people how to use GNU make, wrote Perl wrappers, web interfaces and automation
tools for various companies over many years. 
It was great fun, but lacking one aspect: women. I was often the only female for miles in
many of these environments. I didn't mind; I learned how to handle the bad, enjoy the good
and  focus  on  my  ultimate  goals  of  learning  and  getting  the  work  done.  My  hobbyist
tendencies also flourished as I could afford to buy better hardware, run free software on it
and have a few side projects at any moment in time. It was and still is great fun for me. 
The Python Papers, Volume 2, Issue 4 7
Today,  from where  I  stand as  an  experienced software  developer,  I  survey  the  current
landscape and see a similar picture to that of ten years ago. Only this time it strikes me as
sad. Ten years ago, I considered it more to be happenstance, but that excuse is no longer
valid.  There  really  should  be  more  female  F/OSS  enthusiasts,  hobbyists,  tinkerers  and
hackers. I have heard many women echo my concerns, and many more ask where they can
find such projects, so that they may become those enthusiasts, hobbyists, tinkerers and
hackers.
This inspired me to start something called GrrlCamp. It is an Open Source project which will
be written mostly in Python, my favorite language to date. The intuitive nature of Python
syntax,  plus it's  diversity,  robustness and overall  speed makes it  the perfect  choice for
every project I've done since 2001. Python is ideal for  GrrlCamp also because it is fairly
easy to teach online. Since I run my own publicly accessible servers from my basement,host
commercial software from these servers and have taught online before, I figured it was time
to dedicate one server (and much of my time and energy) into starting this project.
This project is unique in several ways. It is entirely voluntary, even to the point of turning
down offers for corporate sponsorship to ensure that it is guided only by volunteers and
contributors. It is only for women, but it does not matter how novice or experienced they
are. The purpose of the project is for training as well as fun, so a few GrrlCamp members
offer online classes to members for free. It is entirely hosted from my home on my own
server and all participants take part in everything, including system administration. I started
by opening up services on my server, advertising for GrrlCamp participants on the net, and
preparing and organizing online group material. 
I was fortunate to have a volunteer come forward and offer her outstanding organizational
skills to our group. This helps greatly. I have quickly learned that having a “kick-ass” group
manger on volunteer projects of this size is as essential as a right arm. 
Within several weeks, I had a total of 31 participants from all over the world and in so many
time zones that we need to hold weekly repeat chat sessions ten hours after our first ones,
simply to ensure that we reach out to everyone and everyone can reach back. I started by
passing out system admin assignments and the work began. 
Everything from the definition of the groups' operating rules and responsibilities to the build
and  installation  of  all  of  the  necessary  Open  Source  packages  is  done  by  the  group.
GrrlCamp has a mostly flat management structure with the exception of one small “bump”,
being  the  “kick-ass”  group  manager's  absolute  veto  power  over  anything  absurd.
Everything is decided by consensus and discussion. Needless to say, this is verbose and
time consuming, but so is any volunteer project where people contribute great time, energy
and effort.
We are currently at the end of the system admin and group principle definition phases,
finishing  up  the  bash  shell  training  and  beginning  OO  design  and  development  basic
training, as well as Python basic training. We now also begin the design phase of the project
source code and the content evaluation phase of how we document the process as well as
the project. Our plan is to continue online until May 2008, when we meet in person in New
York City for a week, wrap up the project and release a Beta version to the public under a
freeware license. 
My hopes with GrrlCamp are multifaceted. Firstly I want to help a group of women learn and
experience  Open  Source  development  from the  ground  up,  just  as  it  is  done  by  most
startups. This is an experience many women in IT want, but have not yet had for various
reasons. Luckily it is an experience I have enjoyed twenty times or more. I design, develop
and administer projects for startups from scratch, all the way to production, which has, in
hindsight, made me wonder why I did not start GrrlCamp sooner  
With GrrlCamp, I also have a bigger goal of documenting the process as well as the project,
to try and preserve the learning experience and simplify it for future groups. The system
admin tasks, the discussions on participant principles, the design and development process
as well as the online classes, the code and the project, are being categorized, summarized
The Python Papers, Volume 2, Issue 4 8
and  “wiki-fied.”  This  is  only  possible  by  having  31  people  and  an  outstanding  project
manager, and it is so far very successful. 
Within  GrrlCamp, all information is shared, all thoughts are shared and all questions are
aired. General consensus provides the answers and direction. For this reason it is much
more  than  just  another  Open  Source  development  project.  It  is  an  evolving  social
experiment, a project becoming more and more cohesive and substantial by the day. It has
only been three weeks since GrrlCamp started, and we have come a long way. 
As time progresses I'll talk more about the tools and the technical aspects of the project. For
now I'll leave off by saying that we'll be using CherryPy and SqlAlchemy, two of my favorite
development  tools,  on  the  back  end  portion  of  this  project.  More  information  about
GrrlCamp will follow in subsequent issues of The Python Papers. 


Not a member yet

MPD WebAMP

Embedding a Python interpreter into your program

Designing Semiconductor Heterostructures with Python

Upcoming events

GrrlCamp: An Open Source Python Project for Women

Frontmatter, Volume 1 Issue 1

PyGTK, PyQT, Tkinter and wxPython comparison

172

207