Re: looks like it's happening everywhere but the US.

On 27 Oct 2003, Jan C. Depner wrote:

> 	Read slashdot for a week and see how many times you hear someone
> whining about "there are no software jobs because they're moving
> everything offshore".  Not that I agree with that.

  Not to be insulting, but SlashDot isn't what I call a scollarly journal 
to learn about law, economics or good public policy ;-)

  People who are looking only to be surfs and looking for "jobs" may not
be a fan of OS.  The entrapraneurs on the other hand can recognize that
one way to keep jobs from going "offshore" is to more closely meet the
demands of your local customer.  This is yet another case where FLOSS
demand-side methodologies have huge advantages over manufacturing jobs
which are easier to move elsewhere.


BTW: Off topic, but I am a supporter of LETS (Local Employment and Trade
Systems) <http://www.flora.org/lets/>.  FLOSS is another one of those
transformative economic changes which has allowed software creation to be
included in LETS systems. The non-LETS currency costs of entry into
software creation are greatly reduced.  Getting payed in a local currency
for incremental value-add to FLOSS software is possible where the 
larger returns necessary for manufacturing methodologies wouldn't work.

>  I definitely agree but that's not what I was saying - once again, all I
> said was that you weren't going to see a change in the foreign policy
> anytime soon because Micro$oft is a cash cow at the moment.

  What I am asking is whether that is in fact true: is Microsoft a cash
cow for the US government.  The comparison is not to having nothing, but
to having a vibrant demand-side FLOSS economy domestically in the USA. I
don't see the evidence to back up Microsoft being considered a cash cow
rather than an economic drain on the US economy.

  When talking to Canadian policy makers I talk the other angle which is
to suggest that if Canada adopted FLOSS-favoring policies that this could
put the Canadian FLOSS economy at a competitive advantage to the legacy
"software manufacturing" economy of the USA.  In this case what I suggest
is good for Canada is of course bad for the USA.

> You do have to understand that not all systems can or will be built by
> OS developers.

  I suspect most people understand this, but also recognize that this is
the exception and not the rule.  Not to always seem to pick on them
specifically but Microsoft is a prime example of the problem.  The only
divisions of Microsoft that are turning a profit at the moment are
operating systems and office suites, two product lines that should have
been commoditized and been part of the infrastructure years ago.  
Commodity infrastructure software is one of the things that open
collaborative models can always do better than manufacturing
methodologies.

> Some are too expensive to have the equipment to work with and others are
> too boring.

  Not sure how these fit in.

  One example that I often bring up is the Disney support to upgrade WINE
to allow them to run Photoshop.  Not only was this demand-side customer
making use of the FLOSS WINE project to bypassing limitations imposed by
Adobe (the supply-side supplier of Photoshop), but Disney got together
with two competitors to do the work:

 http://www.eweek.com/article2/0,3959,1210083,00.asp
   Rather than wait, Disney, along with two other motion
   picture animation studios (which declined to be named for this
   article), decided to jointly fund the development of a
   Windows-to-Linux porting solution. The idea: develop technology using
   the Wine emulator to run Adobe Photoshop on Linux.

   While animation studios compete fiercely for ticket sales and are not
   known as team players, all three agreed that a project that would
   benefit the entire open-source community--while delivering a
   technology they needed--was worth their cooperation, said Jack Brooks,
   director of technology at Walt Disney Feature Animation.


  I also don't understand the "too boring" reference.  If you are paid to 
do the work, does it matter if it is too boring?  FLOSS software does not 
only come from the voluntary sector, but all other parts of the economy.  
Whether a customer will be making money with the software or saving money 
with the software, there is money to pay software developers to do the 
"boring" stuff.

  Please don't confuse the Free/Libre software movement with some sort of
mythical Free/Gratis software movement.  While some FLOSS software is
volunteer work, much of it is paid work as well -- having the software
project involve many types of contributions from all sectors is one of the
huge advantages of FLOSS over the uni-sectoral alternatives.

  Royalty payments are a single (and legacy) business model, not the only
way to make money with software.

> There are people begging for developers for the boring financial
> software.  Plenty of developers available for digital audio/video though
> ;-)

  Are these people begging for handouts, or are they providing resources
for a legitimate demand they have?  This is one of the things that most
gets people confused, and sometimes angry with me for my response. When a
third-party user of some software makes demands of open source projects as
if they were a supplier of technology, and still want everything for
$free/gratis, I have huge problems with that.

  These people need to make themselves into involved participants in the
projects they have a demand for.  If they are like the majority of people
and are not software developers this does not mean they can't be
participants.  Funding software developers, technical writers (for
documentation, developer or user), or other such contributions are
important as well.  If people have money to pay royalty fees for "software
manufacturing", where does this money go when they find a FLOSS project
that "almost" meets their demand?

> There is one other reason that OS developers do a better job.  It's peer
> review.  I know when I work on a piece of software that everybody and
> their brother gets to look at it so I'd better do it right.  At work I
> am a government serf but I still program the same way ;-)

  I write submissions to the government the same as I write software
(except I do policy work primarily as a volunteer, software only when
paid).  I release what I am doing early (before a version 1), advertise
it, and encourage feedback.

  I suspect people are sick of the number of times I've referenced
http://www.flora.ca/copyright2003/ and http://www.flora.ca/patent2003/ in
this forum, but I have received very useful feedback from doing this.


  Where Lawrence Lessig says that "Code is Law"  
<http://cyberlaw.stanford.edu/code/>, I go further and suggest that "Law
is Code"  
<http://www.linuxsymposium.org/2003/view_bio.php?login=russellmcormond>.  

  I believe that in a democracy that both code and law should be open and
collaboratively created, and distributed/executed in a way that is both
transparant and accountable.  Government mandating FLOSS in all but a few
exceptions should be no more controversial than existing access to
information laws (which also have narrow exceptions that must be justified
when used).

> 	Less taxes doesn't mean no taxes.  What would you rather have 2% of 1
> billion or (the Canadian way) 18% of 1 million?

  Nice attempt at a snipe at the Canadian tax system, but your analogy
doesn't work.  What you are comparing is .01% (if you are lucky) of a
billion (Microsoft Way) or 2% of hundreds-of-billions (The FLOSS way).

Note: I am not a fan of income/profit tax at all, but that would be
another longer conversation that is outside the scope of this thread.
  http://www.flora.ca/russell/align-left.shtml

> 	I tend to agree with that but our methodologies would differ.  I
> think that the government should mandate open standards thus taking away
> Micro$oft's advantage.

  I think that this is missing the forest for focusing on a single tree.  
Anti-competitive vendor dependence in software procurement is only one
problem among many.  While properly enforced requirements for open
standards would deal with that problem, it doesn't address the other
problems.

  How about voting machines?  Is it any better to have 10 competing 
secretive companies rather than only one?  The problem is that voting 
should not be under the control of technology vendors.

  Suggesting that a technology vendor is an uninterested third party in an
election is about as credible as suggesting that the candidates themselves
are.  Electoral fraud isn't just something that private citizens do, but 
that the private sector has a huge interest in as well.  Just look at the 
last US presidential election:  Assuming you believe that there were more 
than 2 presidential candidates, out of the top 3 you had:

  a) Someone who said he would ignore anti-trust law and let Microsoft 
     and other such criminals off the hook.
  b) Someone who said he would stay the course with the anti-trust suits.  
     Then again, he also said he invented the Internet ;-)
  c) Someone who helped hosting the Appraising Microsoft  conferences.
     http://www.appraising-microsoft.org/

  Would Microsoft be considered trustworthy enough to be involved in
voting machines?  I believe there was even a trial of some Microsoft
machines in Florida, but don't have the reference for that handy.


  Having voting machine software be FLOSS should be a requirement, but is
still not sufficient for voting.  I am one of those people who simply do
not believe there is a way away from a re-countable paper ballot.  I
believe that ballot should be printed from a touch-screen computer to
avoid the physical ballot problems of the past, but that you can't get rid
of the human-readable ballot.


See my signature for what I am suggesting here:
   Governance software that controls ICT, automates government policy, or
   electronically counts votes, shouldn't be bought any more than 
   politicians should be bought.  -- http://www.flora.ca/russell/


> 	The problem here, as I have pointed out before, is that some projects
> won't have an OS solution or will have a substandard one.

  Public policy development should be aware of these exceptions, but the
general rule should not be dictated by these exceptions.

  I agree that there will always be a place for "software manufacturing" 
methodologies in some small niche markets.  There is obviously a place for 
not distributing the software outside of the organization at all, which 
still (and probably always will be) near 95% of the software authored.

  It is important to remember that while distributed software is a small 
percentage (5%) of software that the 95% would benefit from the 5% being 
created using open collaborative methodologies.
  http://www.flora.ca/osss2002/flora/img1.html

> While I don't want to have to get by Micro$oft's "standard" formats when
> I want to buy something I also don't want to have to prove that the only
> OS solution is junk.  This is a difficult balancing act.

  Why not have to justify exceptions to a general rule?  Procurement would
be far more accountable if any procurement that didn't follow the general
rule was made more publicly visible so that open collaborative feedback
could be provided by citizens.

  Private citizens may even be willing to volunteer to help the government
department meet a given demand that the private sector doesn't yet.  I
know I do a lot of volunteer public policy work currently that would
normally have to be paid for.  I'm neither a government employee or a
special interest lobbiest.

> is the same way that ADA worked when we were paying attention to it.  If
> you asked a company to build a product for you it would cost an order of
> magnitude more if you wanted it in ADA.  A company will do exactly the
> same if you spec OS.

  Why do you believe this? ADA was the opposite problem in that it was 
an environment that required skills and resources that didn't apply 
outside the public sector.   With Open Source you are more likely to see 
the opposite with companies offering solutions at an Open Source discount 
as they know that they can then leverage their experience with that 
software in other environments.

  I am an example of this: My rates are different for different customer 
bases, and I charge *MORE* money (if I offer the service at all) if the 
solution is based on or is intended to be non-FLOSS.
  http://www.flora.ca/rates.shtml


> If you think the government pays enough to keep good programmers you are
> sadly mistaken.  I work here because, for some unknown reason, I think
> it matters.  I can make a lot more money in private industry.

  I don't believe that FLOSS is a magic pill that will solve world hunger 
and avoid war, but I do think that it will help this problem as well.

  I see worry from some in the Canadian civil service that a change away
from Microsoft would risk their jobs.  I see the opposite being true.  
Being a button-pushing surf for black-box technology makes someone both
less skilled and more expendable.  In an environment where the public
sector are expected to participate in cross-sectoral open collaborative
projects to meet departmental needs the public servant becomes much more
skilled and much more valuable to the government.

  I do believe that a movement of governments to FLOSS will actually 
increase wages for public sector programmers.  Wages for these programmers 
would increase and the costs to the government for their software needs 
would still decrease because of resource multiplication.  Sounds like we 
are getting something for nothing, but I believe that "software 
manufacturing" really is that inefficient.


> 	BTW, I'm enjoying this discussion immensely ;-)  I think it's always
> good to bat around different ideas and views.

  I'm Cc'ing the forum and hope that others agree and aren't thinking of 
this thread as another form of SPAM...

---
 Russell McOrmond, Internet Consultant: <http://www.flora.ca/> 
 Governance software that controls ICT, automates government policy, or
 electronically counts votes, shouldn't be bought any more than 
 politicians should be bought.  -- http://www.flora.ca/russell/