Accueil       Draft index

We are developping a system (code named xVote)  for the democratic evoting (i.e. voting by Internet).

In a few words, xVote is the design the first true electronic voting system or by Internet :
You can discover (in French), some technical or others non-technical, some illustrated, documents about xVote. As an example, from the point of view of the citizen* or according to the position of the public administration.
See also (in English) the general schematic and his legend.
    (*) The 8th and final page  -a flyer (in English)- shows that xVote follows the strong criteria of a democratic electronic voting.
Finally you could discover (in English) the presentation of various ways of using its client side, the pattern of its structure on the voter's computer (within a Virutal Machine on an hypervisor), or how to use it remotely.


Firstly, for information, as this is important to value our offer's background :
Switzerland is a semi-direct democracy (with a three state's levels, as in the USA*), and the citizens can monitor closely the government** and vote usualy four times a year (and each time on six to fifteen items). Three fourth of the swiss people are heavy internet users. So, the know-how on voting and e-something is strong here. But, we designed the xVote system to be not swiss-bound, and it could be used in every democracy.
   (*) in fact, the two states' structures and even the two constitutions (and the citizens' spirit or the market's conception) are very similar between the USA and the Switzerland, only the scale is clearly different. The founder fathers of the US came to the Switzerland to see our (old) system, and mid-19e century our new constitution's writers went to the USA to see yours, so the similarities.
   (**) with the initiative right -to add a new law- or the referendum right  -to fight a law proposal of the parliament. Each at the municipal, local state or federal state levels.

And, some words about my own background :
I studied mathematics in the university of Geneva. After years of experience in the industry and banking, in the strategy and in the security field, I had the opportunity to study the three existing pilote Swiss systems of electronic voting and found that it was possible to create a better architecture, which really protects the secrecy and is fully accountable, and to reenforce the security and usability of such a system. I created so a complete concept with the architecture, the protocol and the implementation principles of a new, better and more secured Internet voting system.
English translation of my C.V (html) and original french version (pdf).


A few words about the short document (please, click here to read it), which present the various ways how to use de client part of xVote.
The main ways are using either a local virtual machine (with an hypervisor on the citizen's computer) or a remote virtual machine (virtual desktop) on a full blown server.

The local hypervisor (VirtualBox) use is in the first and third alternatives (full and P2P VM), the second use (VD) is with another hypervisor (KVM) on a robust cluster of servers.
The two illustrations are clickable to enlarge.
Beware, this doc discuss only the very front end part of xVote ! A lot of documents show others parts of the system (but not very internal ones).

Inside the local or remote VM, and on all servers, the software run into Mono.

By the way, the server(s) part -two main clusters and some additional services- uses also intensively virtualization.
Since the exploitation phase, the VD service (four lines above) and the two clusters of servers (last line above) could be run within the other products, with a datacenter automation.

Note : the VD channel is designed also toward netbooks, mobile internet appliances and smartphones; about the last platforms ITU announced that at the end of this year their will be 4 billons of mobile phones, and recently Ballmer said that in five to ten years, all billion mobile phones sold a year will be smartphones.


xVote is fully designed, with all the conception, architecture, protocole, methodology, organization, tools, libraries, infrastructures, services, appended software, systems, etc.
I shall write more on this in a next message.

As a first glimpse of the status of xVote, you could read a short document about the ten principles (or criteria) of a democratic voting system on which xVote was designed. Note that if, in the right column of the upper table, are the links to the swiss law register, these are the same criteria in all democratic countries.
On the opposite of a real democratic voting means, are the three cantonal pilotes which are challenged on those criteria.

As a second glimpse of the status of xVote, you could read a sort of small and rough business plan  for the public market and only the Switzerland one (as said, xVote is also valuable for the private sector and the other democratic* countries). The Swiss public market is simply a test and reference market.
Hints : one swiss franc is 0.926 USD. Outside Switzerland, the vote is less frequent, but the cost of the ballot is about three times higher and the population is largely bigger. In Switzerland, about 80 to 97% of the citizens vote already remotely, by the postal service. The organization of the vote is the burden of the local states (cantons).
 (*)  as a NGO-executive said jokingly to me "as xVote cannot be cheated, it will not interest the authorities in some well known countries.

Those two documents are freshly and quickly translated for you in English -it's only a draft translation, really sorry. But that will be enough to have a good understanding of the qualities and assets of xVote, and to have an insight of the deep design and the high level of its conception.

This schematic thoughts about the rentability (in the last doc above) are based on my extensive experience with the states, know-how from my deep studies on this subject, and close contacts with the authorities.
    A local state's main minister (VS) wrote to us "I can only recommend you and your evoting project"
    A local state's IT manager (FR) said to us that the xVote outsourcing would be a must for them, as the system is so (interestingly) high tech.
    Another local state's votation higher executive (VD) said to us that they appreciate the OSS licence, as a political asset, but they would only take a commercial licence, with the support.
    A French minister said to me "if the Switzerland chooses an eVoting system, as this country is so conservative and perfectionnist, we will take immediatly the same one".
But, I know that the costs and rentability calculation is not my cup of tea, and that you Keith, as a business developpement expert, you would clearly do a much better job than I did !


BTW : At the end of august, we won a local state's political fight (against the main evoting pilote), and the next week we will openly enter the federal politic arena.

I'm looking of much more than the supply of only a single product (eg VBx).

It's much more in quantity, and, it's much more in capacity.

Quantity :

As you could have seen in the last mails and docs, the vision is to give, to the democratic countries, the first real electronic voting (remote by Internet or local* at the voting booth), that follows all the fondamental criteria, that is provable and transparent. A system, which the public administrations could exploit themselves or outsource to local companies (or foundations).
   (*) with mobile internet, that will be with instant remote gathering.

The concerned population is not the one of a high school campus or a corporation's headquarter, it's for entiere countries. Eg, for each one of America, Europa, Brasil or India, even with half participation, and -on this- half electronic voters, the potential users must be counted by tens of  millions.

Capacity :

Such a service, which must serve literally millions of users, on a short peak (but not often), with absolute availability, without any loss or security hole, must be build on robust tools.
  -- nota bene : a failure, a loss or a successful attack, will be the next day in the newscast at prime time ... but a task properly completed will lead to a big market and a huge general reference.

So, we must dispose of a stable and secure OS for the main servers, of two virtualisations technology (desktop and servers -the last with datacenter automation software), of a strong database management system, of good load balancing and high availability cluster management, etc
Moreover, our main partner must have a very strong network awarness, a good image in the technical/Unix/OSS world (supportive opinion leaders), iif possible not only good softwares but also good hardwares, a commitment toward open systems and recognized standards, the know-how for great data centers, etc

And -this is crucial- the main partner must have a global (technicaly and geographically) capacity of support.

This is juste a few notes about our whishes.