After reading the changes to open-source policy released by the government today, I couldn't help thinking that getting a greater percentage of open-source software into government is going to be just as difficult as getting more Australian-grown smaller scale companies supplying government.
Both are generally viable answers to government needs and both benefit a burgeoning community that has more flavour and less clout than big companies.
However, as we've seen with small business, there's a hesitation when deciding on a product that not many other organisations are using. There's a tendency to look around and see what the norm is and think, "well, if I choose this and it goes pear shaped, at least I'm not the only one who'll suffer".
This makes it difficult for open source to really get a foot in the door.
Then there's the stability issue, with government generally wanting a company that isn't about to disappear off the map at the first sign of hard times, or software that has a defined support path.
And — I'm not going to say this always happens, but it can — there are the discounts. Big business can often afford to offer a very low price up front, with hopes that later they can recoup their losses when they're firmly entrenched. Changing vendors is sometimes just too hard. Organisations just go the easy route, as we've seen in the past with Westpac, which stuck with IBM, for example, or the Australian Taxation Office again choosing HP Enterprise Services (formerly known as EDS) for its centralised computing needs.
Then there's the skills issue. If a government agency has lots of people trained in a particular proprietary software, it makes it difficult to move.
Of course, this will be factored into the "value for money" consideration, which is very much present in the government's mind. If people have to be retrained completely to implement a new product, that product sits at a disadvantage.
So what do we do to make sure Australian companies and those firms championing open source get their fair go?
This policy is certainly a good start, moving government support for open software further than its former "informed neutrality" to a request for active consideration. However, I'm not sure that it will be enough to balance out the problematic factors I've laid out above. To steal the words from former Electronic Frontiers Australia vice-chair Geordie Guy on Twitter, "You can lead a horse to water, but you can't make it risk its career recommending unsupported software..."
Another option for open source could be a government open-source advocate like the government's small business advocate Don Easter. Certainly in enterprise, open source seems to get more of a run if there's a passionate advocate in the company (think De Bortoli CIO Bill Robertson or CIO for the Sydney Anglican diocese George Lymbers).
Yet, this might not work the same way in government, as it's not one company but a sprawling mass of agencies. Plus, I'm not sure how much the SMB advocate has done for our Aussie battlers.
Another option is to lean on suppliers, which the government seems to be doing in its policy, asking government agencies to phrase their tenders so that suppliers are encouraged to use their open brethren. This to me has great potential, as long as partnerships don't take the "open" out of the equation.
I'm sure there are more possibilities out there.
How do you think the government could increase the use of open source? Do you think the policy will change things? Do you think it is worth pushing for open-source usage in government?