Given how similar certain intersections both sound and spell quite similar (e.g. in the south of France), if given a direct, you can enter it but confirming the spelling can be awkward - so commonsense and situational awareness to avoid a Gi-Go error is, as you rightly say, of paramount importance. Presumably airlines have SOPs, but if directed to an unfamiliar intersection, it would seem prudent to use HDG rather than NAV until the new point has been confirmed?
Yes. When flying along, one has the route printout and should be aware of the intersections along the route. They should be loaded into the GPS anyway. ATC tends to issue shortcuts, and one tends to ask for extra ones all the time anyway. If ATC issue a DCT to some intersection which is not immediately recognisable then I normally immediately ask them to spell it.
What happens when it is entered as a DCT depends on the autopilot mode at the time, and the type of HSI. If one is in NAV mode, and the HSI is a mechanical one, the AP will not do anything (much) until you manually turn the course pointer (CP) to the new track as indicated on the GPS. If the HSI is an electronic one (EHSI, or one of the "glass" products) then the CP will flip round by itself and the AP will turn onto the new track automatically, and yes it could be embarrassing if you entered the wrong waypoint
But the track will be obvious on the moving map anyway, where it will appear about the time the AP is turning onto it.
It is obviously good practice to always have the HDG bug set to the current heading, even when flying in NAV mode, so you can revert to HDG mode, or even fly manually, immediately, at any time.
In some cases switching over to HDG is highly advisable during track changes. One case if where one is intercepting a track. Most GA autopilots do not do intelligent intercepts. They use the CP as the initial "heading" and then tweak the heading according to the lateral track error, so a lot of intercepts happen slowly, and slowly enough to irritate ATC. The Honeywell APs in particular will not do a clean intercept unless the track error is at least 3 divisions (about 60% HSI full scale) or greater when you press NAV.
So one has to know how to play one's system
I flew up to Coventry the other day and spent about 80% of the flight in HDG, with ATC vectors... The rest of it in HDG without ATC vectors
I also note m_j's point about 'head down' button pecking
I don't see that an issue if one knows one's equipment.
As I understand the TSOs, 'IFR' GPS approval applies to the entire system, including antenna and external feeds, but applies only to en-route IFR cruising and terminal approaches (subject to additional TSO requirements).
That depends on the approved flight manual supplement (AFMS) detailing the GPS. In the USA, the GPS AFMS normally permits all IFR i.e. including GPS approaches. In Europe, those I have seen tended to allow only enroute and perhaps not even sids/stars. This can now be corrected, under both EASA and FAA regimes, but it can be nontrivial especially under FAA (N-reg) if based in Europe. It is just a paper exercise but one needs it to be legit.
So if you're flying under VFR and navigating above cloud, you don't actually need an 'IFR' GPS. But what about the IMCR pilot climbing up through cloud to reach a VFR level, or descending below it at the other end? "No, I just note the heading and ignore the GPS until I become visual" Oh really?
As they do with their own SatNavs in hire cars, many pilots are going to prefer to use their portable GPS units in clubs' rental wreckage and the CAA will have to accept that 'VFR' GPS
is going to be used more and more often as a primary navigation tool.
VFR GPS has been used for GA nav for longer than I have been flying, and it doesn't need to be an IFR approved installation. The ANO (etc) specifies equipment to be
carried, not equipment to be
used. I think too many people like to read between the lines on that one, but one is either debating legal requirements, or not. There is no middle ground. Legally, any handheld is legit for primary or even sole navigation (regardless of the wisdom of it, etc) provided you
carry the prescribed equipment, and this is true even in high altitude IFR (airways) flight (non AOC, obviously).
"Fly and navigate visually, only use the GPS once you have verified its accuracy against something else, and cross-check regularly", as SSL 25 puts it, is going to be observed less and less often in future.
Of course, but that is the way the world is moving.
And I support that (with the usual caveats, like backing up with VOR/DME) because GPS is far more reliable than visual nav, dead reckoning, whatever.
I fly 100% on GPS, with VOR/DME backup where possible.