While I agree with this to some extent, I also think you are being very unfair on the people who bring these capabilities into service. A lot of the issues you highlight is simply the situation changing after the decision is made. I'd be a lot more critical if anyone else in a similar situation was doing much better.
For example, look at the Tiger data-links you highlighted. What was the alternative, noting the decision was made in 2001? Were the people making the decisions to digitise the Army with LAND 200 supposed to constrain every other platform in defence just to make sure they could talk to the Tiger? Or were the people making the decision on Tiger in 2001 supposed to predict what system would be chosen for LAND 200 in ten years time? Even if the Apache was chosen, as hindsight would dictate, they still wouldn't be able to talk to the BMS that was eventually chosen for the rest of Army.
Ditto the Abrams. What BMS system should have been fitted to the vehicles in 2004 besides FBCB2/BFT? Were they supposed to predict what was chosen for LAND 200 five+ years in the future? Or should the LAND 200 decision been constrained to what was fitted in the Abrams, noting it was a very old system already? As it was, the FBCB2 system delivered provided excellent service, and was used successfully for many years on the deployed ASLAV and Bushmaster fleet.
There was no problem with lack of foresight in regards interoperability with the LLCs - they simply aren't able to deliver the capability promised by the manufacturer (kind of). There is also the not insignificant issue that when the decision on JP2048 was made, they were basically guessing what sort of amphib capability would be required. Its hard to fault them for not predicting the results of 10+ years of changing strategic direction and policy (for example, the fact that the LCHs wouldn't be replaced), and five+ years of experimentation with the amphib capability (who knew what an ARE/ARU/ARG would look like in 2004?)
Of course, a significant problem is that for everything to be interoperable, particularly for comms, you are more or less forced to customise new platforms, thus adding to cost, risk, and schedule. Imagine the commentary on the internet if we, for example, chose to fit a different BMS to the Abrams fleet when we bought it. The editorials write themselves. The only alternative to customising new platforms is to simply buy everything off the shelf from the US, which, while attractive in a lot of ways, is hardly a solution to everything.
It is very easy to be critical in hindsight, not so easy to make a better decision with the information available at the time. Which isn't to say that much better outcomes couldn't have been achieved, just that usually failed programs aren't the result of everyone on the internet magically being smarter than the people running the programs.
