I think PN makes a very good and important point about the decisions that have to be made in the MoD, and which attract a lot of criticism in this and other forums.
He's absolutely right that the staffs have to make decisions a long time before the systems that arise from those decisions enter service. They have no better crystal balls than anyone else, but have always been forced to justify their decisions using a range of assumptions about how the systems will be used, the expected costs, and how efficient they will be. A good many of those assumptions will not be correct, but the reasons for that can vary.
In some cases, the 'errors' the staffs make are simply due to unforeseen changes in circumstances - again, it's easy (with hindsight) to say that the changes should have been foreseen, harder to get it right.
However, in some cases the assumptions made by the staffs are wrong because they simply don't know enough about the systems that they are developing requirements for. This is an area that I've worked with for a long time, and I have to say that requirements development is far and away the weakest area of the air staffs (all three services) that i've worked with.
Too often, requirements are driven by personal operational experience, or worse, the last sexy gadget they saw at an air show. Sorry if that sounds flip, but it's just too common. It's been especially true for UORs.
Turning to the subject of this thread, I don't think that the RAF have been 'anti-cannon' - but I do think that successive air staffs have been unable to develop a proper set of requirements for cannon (or gun) systems, mainly due to an almost total lack of the required R&D and analysis within industry or MoD to determine how requirements should be set and how far they should try to reach. Take a basic one - what is the required lethality of projectiles? Has anyone used the standard STANAG target set?
The best one I was involved with was the Lynx cabin mounted .50 calibre gun, which was precisely defined in terms of 'we want a gun that does this (classified) to this target (classified) at this range (classified)'. Surprisingly, it was an SF requirement.
The worst example I came across was for the attack helicopter, which merely specified that the gun system should 'be capable of engaging a target'. That was the entire (verbatim) SRD entry for a large, costly and complex system, which has become the most frequently used weapon system on the Apache. In a similar vein, Typhoon's requirements set specified a Mauser 27mm cannon with existing shells, again with no serious requirements for effect or lethality.
Yes, it was procured with a gun system, not just a bay. The gun system was taken out as a savings measure in the late 90s, at about the time the Aden 25 for the GR7 was cancelled. At that time, the Air Staffs were definitely assuming that no cannon would be required in the future. The ensuing saga of the Typhoon gun might be funny if the taxpayer hadn't had to pay so much to play 'Mauser hokey-cokey'.
So, that would be my plea - get the staffs informed and resourced to develop and write meaningful requirements for gun systems.
Best regards to those who do their best,
Engines