RolesofTreasuryDepartment (1)It’s a government thing. Releasing a report on the performance of an aspect of government, some years after the fact, or, releasing strategy that relates to years gone past. So it has been that August 10 2015 Scoop reports “Govt agencies spending too much on HR, and staff turnover too high.” That report, covered the 2012 / 2013 year, and we are now into 2015 / 2016, so the value of it is dubious. And of course, ICT is lumped in with the HR bodies as sunk cost rather than a business enabler.

One of the issues that ICT faces is organisations that run by people who think that ICT is a simply a cost centre with little or no other value, certainly not a business enabler. Well, if you ever need to have that debate then switch the data centre off for the day, or at least the Internet connection these days, and you’ll find everyone agrees it is a business enabler rather rapidly.

Of course this is Treasury, who’s wild ideas have included shutting down the regions and moving everyone into either Auckland or Christchurch as well as dumping the National Rail Network.

Bill English on ICT had to say;

“ICT remains a challenge.”

That’s it. You can read into it what you like. He could have said “The sky is often blue”, which would have had about the same impact.

Back then, we spent over a billion on government ICT. $1.103B to be exact. We won’t find out how much we spent last financial year until 2018, which incidentally, is when DIA’s grand strategy for Government ICT will be done and dusted, in fact, its all up for delivery by 2017.

DIA commented on the report, the full version of which you can find here, but there is nothing particularly meaningful in their analysis. They said things like;

  • OPEX vs CAPEX will increase.
  • Continued adoption of IaaS will reduce overall spending in infrastructure.
  • TaaS! TaaS! TaaS! Will save us all.
  • DaaS will reduce end user computing costs.

There are other things they say, but these were the most interesting.

I’m still seeing some huge capital investment going on around town. It’s less about the tin and more about the fact that major programmes and projects are still capitalising their work because no one, yet, has figured out how to actually adjust the financial models to deal with OPEX.

I am not convinced with IaaS as you know. It definitely works for some agencies, generally the smaller ones, however it can be more expensive in some cases and the world is starting to look past IaaS as a first Cloud stop. PaaS is favourable and SaaS is hot. What triple jump your way to SaaS when you can transition there in one step?

TaaS is late. She’s been under construction since 2013. At least. It’s unlikely we’ll see any product set this year, in my opinion. It’s been a really long time coming and the longer it goes on, the less relevant it is likely to be. See OPaaS.

DaaS may actually increase end user costs depending on what set of services you take. That shouldn’t frighten people, but it would just great if we could get away from the “Cloud and DIA Services will save you money.” Because they probably won’t. What they should do is give you more capability than you have today, if they don’t, then don’t buy them.

That’s treating ICT like a sunk cost rather than an enabler, and we really need to stop doing that. Right now.

In fact, when it comes to government ICT, can we stop talking about money? Can we start talking about things like the fact that DIA delivered an excellent, award winning online Passport system? Can we talk about the fact that Statistics provide a rudimentary front end that is accessible to the world with a tonne of data behind it? Can we talk about Pond? Can we talk about IRD’s existing online system? The companies office? And so on.

Can we stop talking about how many houses UFB has supposedly gone past and start talking about the difference that ICT is making all across government?

Can we use plain English when we report on what we are doing instead of statements like this?

A new definition for end user was introduced this year. The definition for ‘internal end user’ is the same as the definition for end user used last year, whereas ‘total end users’ also includes external end users provided with end user devices or services in the agency

Can we stop thinking that we can purchase innovation through procurement “specialists” and just talk to each other instead?

Can we stop marketing Common Capability services that actually aren’t of any value? Particularly up-selling them to Ministers?

Also, let’s really get our teeth into this finding, we wonder why tenders and projects collapse, here’s why people:

The report also found that just 5 percent of public sector employees involved in procuring goods and services had any formal procurement qualifications. Of the 27 agencies covered, 17 had no qualified procurement staff, “including three agencies with annual third party expenditure of $900 million to $1.8 billion.”

It’s absolutely no wonder that we have 97% of software projects over ten millionish that fail with that kind of skill level in procurement.

Can we stop using Treasury to review business cases against a business case format that is more suited to building roads than actual ICT?

Can we have a Ministry that understands, supports, and fosters government ICT?

Can we actually just have our jobs back so we can get on with delivering ICT rather than spending our time filling in forms?

Can we please have more feedback than “ICT remains a challenge.”

That would be great.

9 comments

  1. Ian, another suburb(!) post on this topic. Absolutely reasonable ‘wants’ as well. I must admit I was surprised at the “only 5% of public sector employees” having any formal procurement skills/training – that’s a recipe for a poor control gate of public spending indeed! Absolutely agree IaaS & PaaS are inappropriate consumption choices, but if you’re in an agency with limited or narrow IT skills, or one of the big ‘5’ ICT SI’s providing government services, this is what you will gravitate to to keep your job and the contracts being renewed, because it’s an easy intellectual ‘fit’ (human entropy – replicate OS’s & Apps in the same manner in the ‘cloud’, and hybridise them on the journey while you’re learning ‘how to’ at the same time). Only those SI’s with ISV (e.g. SW development) capabilities can tender to the needs of SaaS integration with API’s etc, and on that note, it should also be pointed out that only a handful of ISV’s are ‘approved’ for government development contract tendering as well.

    Regarding TaaS – this is still years away from accomplishing it’s desired outcome of agility, automation, and cost reduction, because the govt ICT providers today have not yet made the investment step to using Software-Defined-Networking (SDN’s) at their core, so they can only offer TaaS Pricing-as-a-Service on legacy technology platforms, with ‘mantronix’ based provisioning. The DIA is fully aware of this, but insists on pursuing TaaS simply to drive down telecommunications costs through RFP competitive responses. A race to zero for provider margins, but they are currently at healthy margin thresholds still (perhaps excluding mobile), so no panic yet from the providers. That said, IMHO there is a great opportunity here for VNO’s to implement their own SDN based TaaS platforms if they could negotiate reasonable input costs from existing providers.

    There’s one ‘%’ that I would dearly love to uncover though, and that is what percentage of the $1.1 billion ICT spend is on duplication of IaaS services across the agency portfolio? That seems like a glaring ‘first’ fact to understand IMHO.

  2. Just commenting on one small point about ‘award winning on-line passport system’. Actually I was flabbergasted about that award. The passports system was not only late and millions over budget but it failed to deliver its original scope. It was supposed to be a generic application system, not just for passports. ‘Apply’, ‘decide’, ‘fulfil’, is such a common pattern for all sorts of applications, licenses, grants etc, that an adaptable common service really makes sense, else you end up creating passport like systems all over the place. Instead because it was so late and over budget that goal was dropped, which created a knock-on problem for all of the other requirements that were planning to use that functionality.

      1. Yes, despite all the talk about services, it seems that we still find it almost impossible to break away from individual ‘systems’, a passports system, a fines collection system, a consents application system etc developed largely from scratch.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s