With all due respect to the respective posters, I must take exception with one statement, and comment on another..
>>>Dispatch should stick to what the word says.
It must first be recognized that the functions and responsibilities that an airline's dispatch office and an individual dispatcher have are quite non-standardized across the globe, since various countries have various rules. In some places, a "dispatcher" could be the one that waves the aircraft in/out of the parking area with the pretty flashlights. In some places, the "dispatcher" is the local person doing the weight and balance shipside. In some other places, the "dispatcher" is the lad/lass working in a centralized (usually) airline office that initiates the computer flight plans and ATC filings, amongst other things.
In the USA, the FAA Part 121 regs take that third "dispatcher" definition above and expand on it by making the dispatcher and PIC jointly responsible for "operational control" and the safe conduct of the flight. The US FARs (Part 1.1) state: "Operational control, with respect to a flight, means the exercise of authority over initiating, conducting or terminating a flight.
Everyone seems to understand the "initiating" function (pulling flight plans and filing ATC strips), but the previous statement "Dispatch should stick to what the word says" fails to acknowledge the "conducting" and (potential) "terminating" aspects of flight operations. While the vast majority of an airline's flights are routine and communications between PIC dispatcher will be minimal, that changes rapidly when weather-related or mechanical-related issues arise.
A key benefit of US FAR Part 121-type operational control is that the dispatcher is in a position to "backstop" the PIC when it comes to some critical decisionmaking as far as additional information that the PIC might not possess. Additionally, in of the event of a PICs ill-advised decision to "press-on" into an unsafe situation, the US FAR Part 121 dispatcher has the ability to -independently- initiate the declaration of an emergency condition.
Back to the Hapag-Lloyd 3378 accident, in the absence of an official report, I cannot say whether Hapag-Lloyd had the same operational control set-up as a US airline (as some non-US airlines do), or whether their "dispatcher" just pulled flight plans and filed ATC strips, and devoid of any supporting regulatory language, was forced to leave the critical safety-of-flight decisionmaking to management personnel. If it had been a Part 121-style operation, the flight would have had a different outcome.
I can personally attest that, in my 25 years as a dispatcher for 3 different US airlines, I have seen numerous situations where dispatcher intervention (mine, or that of colleagues) has been the "voice of reason" as far as countering a PIC's desire to do something that really wasn't safe, and PIC and dispatcher jointly (there's that word again) agreed on a more conservative course of action, which while not pleasing some of the management types, was truly in the airline's best -overall- interest, since it helped them to potentially stay off the evening news.
There may be some who read this and assume that I'm saying that "pilots are stupid" or some other nonsense, and I don't believe that for a second. What I do believe is that pilots, as well as dispatchers, are all human beings, and thus capable of some well-intended but poor decisionmaking at times. The duality afforded by US FAR Part 121-type operational control can (properly supported and exercised) serve in a similar "crosscheck" capacity as does the "two key" philosophy on a nuclear missile submarine. That is not to say that I expect a "Mother, may I?" call for every PIC deviation around weather, but I do expect one on obvious safety-of-flight items.
While operational control certainly can't prevent all accidents, it can sure prevent some. Anyone so inclined can see this in various accidents such as Air Illinois flight 710 (US, HS-748, 1985), Avianca 52 (US, B-707, 1990), and Valujet 558 (US, DC-9, 1996, and no, this was a different one than the Everglades crash). There are also the ones you -don't- hear about, for obvious
PR reasons. I know of two potential fuel-starvations that were averted, one involving a MLG door stuck down, sucking fuel, and that one necessitated the dispatcher declaring an emergency and diverting the flight to an offline airport. The other was a MLG itself stuck down, sucking fuel, and the PIC wanting to continue to his desired XYZ (instead of a much closer ABC) due to his FMC telling him "he could do it." The dispatcher got the flight diverted to ABC without declaring the emergency, but his finger was on the proverbial trigger had the PIC not agreed. Those who do not remember the past are condemned to repeat it, as the man once said...
>>>Dispatch and maintenance put a lot of pressure on the crew to make it to Munich or Vienna (both have Airbus maintenance)
Irrespective of whatever kind of dispatch operation an airline has, this kind of pressure (assuming it occurred, again in the absence of an official report, and the former Hapag-Lloyd F/O's comments notwithstanding), is simply unacceptable. One has to wonder if the pressure was coming from the "dispatcher" who did the flight plan and filed the ATC strip, or some management person (or two).
That said, another benefit of the US FAR Part 121-type operational control is that the dispatcher is licensed by the FAA, just as the flightcrew and mechanics are. It's not always a comfy place to be, this little no-man's land between company pressure and FAA accountability, and some outfits don't want/support that objectivity.
I'm quite fortunate in that my airline does support it. A PIC once called me on the last flight of the night (to a MX base) with a grounding item. It was a full flight, and he felt the pressure, but wasn't happy with the aircraft. I conferenced MX control in, and it was obvious that they wanted the bird back at the MX base for the night for the easiest and most convenient repair. The PIC asked me what I thought, and I opined that I wanted the aircraft back at the MX base just as much as the PIC and MX did, but that it didn't matter to me whether we MX-ferried it empty or taxied it 500 miles home on the interstate highway. The MX guy actually told me "Well, this aircraft is in no condition to MX-ferry", so I asked the inevitable question that nobody else seemed to be willing to ask "Then why are you trying to con us into carrying revenue pax on it?" No answer from the MX guy, but the PIC then chimed in "Yeah, I'm not going to take it." And we didn't. After an overnight light-twin charter with parts and personnel, they had it fixed for the morning.
Had my airline not supported me, my union and I were each quite content to have taken the issue directly to the FAA. Sadly, at some places, that's all too often an occurrence, and it's largely a matter of an airline's corporate culture.
In closing (sorry for the length), there are some kinds of "dispatch" that do need improvement, but there are some types of operational control that work quite well if properly understood, supported, and exercised...
Cheers...