Well, drones ain't drones.

The government has cancelled a plan to buy unmanned aircraft just as Ukraine is showing how useful drones can be. But what we've cancelled was hardly like what Ukraine has been using.

The decision, revealed quietly just a week before the election campaign began, was the right one - but came far too late.

This program to buy armed propeller-driven drones has been a disgraceful example of the inability of the Defence Department and armed forces to change direction quickly.

Criticised for the decision to scrap it, the department says exactly the right thing: the government must "prioritise resources in response to the complex and challenging strategic environment we face", meaning the rising threat from China.

Yes, but how did the program last until 2022? It should have been put down almost as soon as the Royal Australian Air Force first mentioned the idea in 2014.

In those days the air force could imagine using armed propeller drones against the Taliban, the only sort of targets such weak and vulnerable aircraft can attack. But even then the RAAF should have seen that serious fighting power close to home was a higher and rising priority.

The specific type of drone we would choose was always likely to be General Atomics MQ-9 Reaper, because the US Air Force was using it, and Australia confirmed that choice in 2018.

Pretty late in the program, with our military presence in Afghanistan and Syria much reduced, the requirement changed from watching and hitting easy land targets to something completely different: watching easy sea targets around Australia. Now the equipment would be MQ-9s in a suitably modified version.

So we got the impression that any requirement would do, providing the answer was "MQ-9s".

That may sound juvenile, but sometimes defence procurement really is like that.

Once there is momentum in our defence organisation to buy some kind of kit, it becomes hard to stop.

Defence ministers might suspect that their department or the services are pulling a swifty in arguing for a purchase, but it's hard to resist advice from experts - especially when they have the advantage of military secrecy in avoiding independent criticism.

Sometimes it also seems that planning just runs on rails, with no serious review of programs once they're set in motion.

The excuse to use MQ-9s for maritime surveillance was weak.

We already had, or were getting, three types of aircraft for that role: the Boeing P-8 Poseidon (based on the 737 airliner), the big Northrop Grumman MQ-4C Triton jet drone, and, with the Australian Border Force, the Dash 8 propeller plane.

Why would we need a fourth type, with all the cost and complexities of new infrastructure and training? Why would we need one carrying little ground-attack weapons and therefore fit for going back to the Middle East?

If we needed more ocean-surveillance capacity (and in fact there's hardly a limit to how much we could use), the answer would have been buying more aircraft of a type that was already in the plan, probably Tritons.

The precise minimum price for the MQ-9s was public - not because our government told us, but because the US government did. It was $2.1 billion for only 12 aircraft: $175 million each, before paying for a few extras.

So MQ-9s are nothing like those little Turkish Bayraktar drones, which have been giving hell to the Russian army in Ukraine and which, costing only a few million dollars each, are almost expendable.

It may seem unfair to heap criticism on the government and department after they've done the right thing, finally canning the MQ-9 program.

But everyone who watches defence policy knows we have a problem with sclerotic decision-making. Even officials in the department know it.

And, while we'd like to think that common sense has won the day, we can't help noticing that the cancellation can be explained by two other factors that drive decisions in the Australian armed forces, notably in the RAAF.

One is that our services strongly desire to have equipment like that of bigger armed forces that they admire, especially those of the US. That equipment is not always what we really need.

As it happens, the US Air Force is moving to get rid of its MQ-9s, even trying to give some away. So it's not surprising that we're suddenly dropping our MQ-9 ambitions like a hot scone.

Second, all air forces love the best technology, preferably future technology. And it so happens that the RAAF and Boeing are developing a very snazzy jet drone in Australia, the MQ-28A Ghost Bat.

So something much more tempting than the MQ-9 is on the RAAF's horizon.

By the way, the Ghost Bat, a kind of aircraft called a loyal wingman, so far looks outstandingly useful.

Something almost as important as the drone cancellation also happened before the election - or, rather, didn't happen. The government didn't announce a supplier and manufacturing location for up to 450 infantry fighting vehicles that the army wants.

Since these shockingly costly IFVs are supposed to be made in Australia (jobs! votes!), this program was perfectly poised for a pre-election go-ahead. But the election has already been called and the government's in caretaker mode, so there will be no announcement.

Why didn't the government take this opportunity? Probably because it wouldn't be politically helpful to say we're making a massive purchase of equipment just like the stuff cheap Turkish drones have been blowing up in droves.