“What’s 50 million for voting machine even Mugabe wouldn’t buy?”

Eamonn Gilmore’s performance in the Dail’s vote of confidence vote last night is well worth watching (as is last night’s Oireachtas Report when they update), for his calm tone, tempo and unremitting focus on the government benches… The quote from above is from an impressive delivery from another front bencher, Ruari Quinn…


  • Wan


    The man I hope to see as the next finance minister . I would prefer him to Gilmore as leader of the Labour Party.

  • Mack

    Surely the e-voting ‘scandal’ encapsulates the core problem of statism.

    Let’s think of this rationally.

    1/ Why should each country create it’s own e-voting system?
    Once created it could be re-used elsewhere.

    2/ Why did Ireland take it upon itself to create it’s own e-voting system? Surely all the government had to do was state that they were in the market for one and what there requirements were. Once other states made similar announcements the market should take care of the rest.

    3/ Normally there are multiple tech startups working in the same area. The vast majority fail. If the government is going to get involved in software development – why do we expect the state to have a 100% success rate? That is madness.

    The state should be seen as an economic actor like any other. We shouldn’t expect success rates beyond those achieved by private companies. The advantage of having multiple economic actors competing in a given sphere is that one or more will succeed. Government spending on the creation of their own favoured projects crowds out private investment as no entreprenuer is going to seek investment to build a system that stands in direct competition to a system the government is building (via contractors) when the government is the customer!

  • Mack

    Or to put it another way – if way back in the late 90’s – if instead of building an e-voting system the government decided it wanted to have it’s very own search engine – does anyone think the contractors would have built a Google-like search engine utilising a similar algorithm to Backrub? Bollix. They’d most likely have got a copy of Northern Light (manually selected links) or a poor Alta Vista clone. Why? Because anyone smart enough to think of the Backrub algorithm (let’s call them Sergei and Larry) – would keep it for themselves!

    The state can get involved in development projects where it makes sense (i.e. where the market couldn’t deliver) but we should also have rational expectations!

  • Paddy Matthews

    My understanding is that Nedap (a private company, Mack), who provided election systems elsewhere in Holland (and I think Germany) won the contract – i.e. the Government actually followed step 2 as outlined by you above. It’s not actually the case that a bunch of public servants (boo! hiss! spit!) did the coding or engineering.

    A customised system of some sort was going to be necessary because our electoral system is different to that of other European countries where a voter indicates a single preference for either a party list or individual candidate – handling multiple preferences are not a functional requirement in any mainland European jurisdiction.

    Whether we actually needed an electronic voting system of the type proposed is another matter entirely.

  • Paddy Matthews

    There’s a copy of the tender at:


    Amazing what a little elementary fact-checking will discover.

  • Mack

    Paddy –

    I was aware it was built by contractors – I even mentioned them in my post above. In so far as it appears largely to have been custom built – it’s still statism. (see below)

    I don’t think the additional requirement of having to support STV would justify the creation of an entirely new system just for Ireland. If that’s what they paid €50 million for – they were had! Customising an existing solution to meet #2 above should not cost that much money (although this was the peak of the dot-com bubble).

    If you are attempting to build a new product like that – you can expect that the probability of failure is pretty high. In my experience integration projects and internal tools often score some degree of success, but new products fail one way or the other very regularly.

    E-voting has been abandoned in the Netherlands, although some Nedap machines have been used in Germany.

    Had we waited until all these issues had been sorted out, the price should be limited to some customisation / installation work, hardware costs & licensing. Not €50 million..

  • Mack

    My understanding is that Nedap (a private company, Mack)

    Just to highlight again – you did notice where I said most tech startups fail?

    If the state is going to get involved in ventures like this – effectively becoming an early stage VC fund – admittedly a pretty stupid one as they are effectively funding the creation of this product by Nedap – by the looks of things along with the Germans and Dutch – with no prospect of sharing in any upside – in fact just more fees – d’oh!!).

  • Mack

    Link for the Dutch rejection of Nedap’s solution in 2008.


    Leave the venture capital funding to Seqioua lads!

  • Paddy Matthews

    I don’t think the additional requirement of having to support STV would justify the creation of an entirely new system just for Ireland.

    If you’re going with electronic voting kiosks, then they’re going to require dealing with a different type of user input to other e-voting systems because of the fact that sequential preferences can and will be indicated by the voter. Continental systems either have a straight vote for a party (e.g. Germany) or the choice of a single candidate from a party list (e.g. Holland). Simply taking a system where you push one button and your vote gets recorded won’t work in an Irish system without significant adaptation – this also makes it difficult to offload our unfit-for-purpose voting machines onto anyone else.

    Apart from Malta, Australia and Scotland (at local election level), no-one else outside Ireland uses preferential voting, so you’re going to have a pretty limited market for it to be worth the while of possible entrepreneurs developing such a product without governmental prompting.

    E-voting in the Republic was a botched job, but lack of ideological free-market purity was not the reason for it.

  • Mack

    Paddy –

    Let me assure (10+ years building large scale software systems) you that you do not need a completely different system to handle the kind of minor UI changes you are talking about, or even to handle the results processing on their back-end servers. The vast bulk of the system would consist of a resusable core as all states should have similarly high expectations with respect to security, data storage, redundancy, retention, traceability etc. Nedap failed to meet these for both Ireland and the Netherlands.

    This is hardly free-market idealogy by the way. Two points

    #1 The government was effectively funding an early stage venture.

    #2 Early stage ventures often fail

    So either we decide that governments shouldn’t fund early stage ventures (mostly I think this is true). But where it does, we should be tolerant of failure. The public sector is just another economic actor and liable to be as successful (if that) and no more than the private sector.

  • Mack

    Just in case the hardware requirements are causing some confusion – a commodity pc attached to a touch screen monitor running e-voting front end software would be reusuable no matter which type of voting was in practice.

    That model has been in use for a long time – way back in the 80’s a British firm had a successful line in business card generation based on an Atari ST hooked up to a touch screen and printer, deposited in a nice looking box and distributed to shopping centres throughout the UK and probably also Ireland.