Friday, October 31, 2008

CPSRENEWAL.CA Weekly: IM/IT Open Source of Off-the-Shelf?

A couple of days ago, we pointed to this article. The main premise of the article is that economic uncertainties are leading our country into a federal deficit and that Parliament's new budget officer has taken to reviewing one of government's biggest and riskiest capital costs -- large information technology projects.

To be honest I have zero experience with large Information Management / Information Technology (IM/IT) projects on a government-wide scale, so I can't really speak to it with any degree of certainty. However, what I can speak to with greater certainty is the opportunity to reduce costs of IM/IT at a lower – say the program – level. If you want evidence that the sum of small contributions aggregate quickly, I ask you to look into the success of the Barack Obama fundraising campaign.

Perhaps this is a by-product of my own experience with technology (being Gen Y, and all), but my gut reaction is to ask why we aren’t using more open source technologies in government? They are easy to use and do not require licensing. I am told that software licensing costs the Government of Canada approximately $750 per machine, I was also told that that particular network has approximately 65,000 users… and that is a lot of licensing fees.

Conversely, it should be mentioned that procuring new hardware doesn’t mean additional licensing fees for new software if the new hardware replaces the old hardware (e.g. licenses are transferable). The point is that the majority of licensing fees have already been paid and have already been paid over the preceding fiscal years. It should also be mentioned that not all Public Servants are attached to computers, but then again there are usually number of empty workspaces in any given office building with active licenses attached to them.

The only conclusion I can draw here is that determining the potential savings from open source is an extremely daunting task. Therefore, while I am drawn to the idea of moving to open source, I’m unsure as to whether or no it’s the right move.

Lacking the requisite background knowledge, I did what anyone in my shoes would do: turn to the internet. In my search, I stumbled upon a paper out of Bond University entitled Open Source or Off-the-Shelf? Establishing an institutional repository for a small institution. It provides the rationale for the university’s decision to go with an Off-the-Shelf product instead of an open source product and in so doing, lays out a comparative analysis that perhaps we can draw some lessons from.

The paper explains the costs and benefits of the two options, precisely what I’d set out to explore in this column, but what really struck me was a single question in the comparative analysis chart: How does open source sit in the culture of the organization?

Great Question

It is probably the first question that should be asked. Why bother conducting a cost-benefit analysis if the culture doesn’t jive with the change? Sadly, the paper doesn’t explore the issue beyond posing the initial question. This leaves me in the position of simply having to make an experience-based guess. My estimation is that the corporate culture within the GoC has one foot firmly placed in favour of off-the-shelf solutions, while carefully looking for the right place(s) to slowly bring down its other foot on the open source side.

Enter GCPEDIA, the GoC’s new wiki – one of those ‘right places’ that has been carefully being identified as a ‘right place’ to step (Aside: We are planning a more comprehensive response for our next weekly, and to be completely honest, I am quite excited about the opportunity to contribute to GCPEDIA and have already started brainstorming around how exactly to do that in both my official and unofficial duties as a public servant).

At least in the context of GCPEDIA, I think that open source and open culture are poised to be mutually reinforcing. Not only that, but I think that these types of opportunities are set to grow in number, and I hope that those who have advocated for their implementation take full advantage of them. For some reason, the old adage, use it or lose it, comes to mind.


  1. I can tell you from experience that IT managers won't implement open source unless they are backed into a corner without any other resort. Open source is seen as unreliable, not because of bugs, but because you don't have a company that you can hold responsible if and when something goes wrong - so the onus falls to you to implement the product rather than blaming it on the company that made it - as is so often done in government.

    Another main reason is just basic price. In a decision between a 0 dollar solution and a million dollar one, government managers will nearly always take the million dollar solution (and I'm speaking from experience and not exaggerating the costs on this one) cause a million dollar solution means that your project is much more important, needs more staff, funding etc... while a free solution is exactly that. Free, free of funding, staff, importance, and priority to senior management above you. So by choosing the cheaper solution you automatically de-prioritize your project in the eyes of the rest of the department.

    Anyways, this is something I can go on at extreme length about, but thought I'd drop some of the basic points down.

  2. In the case of Bond University the decision was always
    straightforward: First you create a requirements document:

    must be able to upload and download files

    , and then you purchase an expensive product (BlackBoard)
    and then you invest the same amount of money for implanting
    it into the organisation. Student fees well spent....