[ILUG] E-voting in India

[Colm MacCarthaigh]

On Wed, May 05, 2004 at 01:11:31PM +0100, Bryan O'Donoghue wrote:
> >But I must still recognise that it does not serve our interests to
> >insist on the vendor going open source, something they clearly will not
> >favour, but insisting that they must publish the source and make it
> >available for more general scrutiny.
> 
> The right of the State to redistribute the code is a misnomer.
> 
> The right of the State to change the code to fix it if necessary is a 
> murkier issue though.
> 
> The ability of the State to divest itself of Vendor (x) and maintain the 
> code base of necessary, while not pallatable to the company, *is* in the 
> State's interests.
> 
> That's probably the biggest argument for Open Source Software's usage in 
> the State, but, that's a seperate debate.
> 
> I'd certainly be in favour of an agreement with the Company where the State 
> is granted rights to maintain the code base, on the machines it bought, 
> but, not necessarily make the changes available to the world.
> 
> That's an easy compromise which covers our asses and doesn't leave the 
> company with a forked open source competitor derived from it's IP.

Well as I said, there are good arguments in favour of all
state-procurred software being Open Source, some of which you've outline
above, but as we both realise - it's a seperate debate.

In the context of Evoting, I think you have it the wrong way around. A
good compromise at this juncture would be for the company to accept
that the code must be published, but they can keep it licensed
propeitarily. The state isnt in a position to modify code, it doesn't
have the experience - the only people who can do it under the existing
contract are the vendors, and that isnt going to change. Though the
Dept did insist on an escrow clause so the source is kept in escrow
with a third party and should the vendors fold - they won't be totally
screwed.

But we can trust changes made by the state no more than we can the
vendor, either way they need to be published such that errors have a
better chance of being found.

Btw, in the context of all of this it is interesting to note that the
Commission actually found that;

	 "the need for all software to be open source, to allow the 
	 community to check that it can generate accurate results" 

	...

	"concur broadly with the main themes of the Commission's work"


(pages 17/18 of the Interim report). So they seem to have a moderately 
more pro-Open Source position.


-- 
Colm MacCárthaigh                        Public Key: colm+pgp at stdlib.net