Re: evmpl license

From: David Mertz <voting-project_at_gnosis_dot_cx>
Date: Thu Sep 11 2003 - 00:50:55 CDT

Clay Lenhart <clay@lenharts.net> wrote:
|This is hinting at a downside to the audit trail. It seems that it is
|easy to circumvent. Just have a text file with the general ideas of
|what changed.

There -is- an issue here. I think Clay has the false sense that CVS
cures it; it does not. But in general, requiring a revision log in the
license isn't the same thing as requiring a *good* and *accurate*
revision log.

Then again, I cannot imagine how we could write a license that made
enforceable demands about the -quality- of the revision log.

|I am in agreement with David, that there will be a Live CD for a
|computer without a harddrive.

There MAY be a CD. Just because we recommend something, that does not
make it so in the actual voting booths.

|I don't see a company taking the software and making major changes to
|sell it. That would require them to pay for development when people
|like me will do it for free.

It's easy to come into the project, and imagine that we, the EVM2003
project, are lawmakers and elections boards. We're not! The political
landscape is far more complex than are the technical issues.

In particular, if our project flies, companies most certainly WILL make
major changes to EVM, or its descendent code (earlier in the discussion,
it was clear that different project members have different hunches about
how much continuity there will be between the demo and the production
system).

There are two financial issues that basically mean that developers like
Clay really won't do it all for free. One is that it cost a bunch of
money to get a voting system certified for official use. I frankly
think that the existing certifications of commercial systems are a sham
(I think Doug Jones agrees, although he'd put it in more measured
terms). But even if certifying companies don't do a good job, they
still charge a lot more than volunteer developers have as pocket change.

The other part is that a lot of money is spent by government to conduct
voting, including a lot paid for the machines. What these two facts add
up to, in practice, is that companies with some resources will sell or
lease EVM-based voting machines to elections authorities. As much as I
might prefer the whole thing was done by government itself, that's not
how thing work in the USA.

Moreover, these companies with costs and profits would not solely play
shell games. There really are different requirements in different
elections. Just because a system is certified in California (or maybe
only for a certain county), that doesn't mean Texas will use the system
as-is. Ballot requirements may well differ, which means customization.
And requirements will evolve over time.

Please take another look at my prior note describing Acme and the like;
that really is the scenario to have in mind.

|Using the software for elections for public office or votes on public
|measures is considered "distribution of software" under this agreement.
|All rights and restrictions must be obeyed as if the software were
|actually distributed.

In terms of this clause, the second sentence is superfluous; or
actually, it even weakens the first. As I wrote before, I believe that
distributing the software installed on machines -already- constitutes
distribution under the GPL (or EVMPLv1.0). Clarifying this intention
with the first sentence is good; but claiming that it is only an "as-if"
distribution is not good. The EVMPL v.1.0.1 should include only the
first sentence, IMO.

Yours, David...

--
 mertz@  _/_/_/_/ THIS MESSAGE WAS BROUGHT TO YOU BY: \_\_\_\_    n o
gnosis  _/_/             Postmodern Enterprises            \_\_
.cx    _/_/                                                 \_\_  d o
      _/_/_/ IN A WORLD W/O WALLS, THERE WOULD BE NO GATES \_\_\_ z e
==================================================================
= The content of this message, with the exception of any external 
= quotations under fair use, are released to the Public Domain    
==================================================================
Received on Tue Sep 30 23:17:02 2003

This archive was generated by hypermail 2.1.8 : Tue Sep 30 2003 - 23:17:09 CDT