Ok, see that is one of the things that I am talking about (well, trying to poorly, rather). Yes, the satellite did a lot of damage, but it was an offscreen thing. I would probably have let the player do this by spending a fate point on the "Blackstaff" aspect and be done with it. He is the blackstaff, he can do things like that simply by being the blackstaff. Especially, since Ortega had already been defeated, it was nothing more than a finishing move, which helped the story along.
I would look at a lot of thaumaturgy rituals this way, meaning having a look at the end rather than the means. For example Harry calling ab the wikipedia demon, you could go and make it a full fledged "summoning a demon and trapping him inside a circle" spell (means), or you can treat it as a contacts replacement ritual to gather some information (ends). They will be the same on the narrative end, but the mechanics would be entirely different.
And I see nothing wrong with the field of giants resurrection. It was a well written ritual from prep to finish, I see nothing to object. Plus she left her blood on every last one of them, and I would love to have that turn around on her. Yes, you could set it up to be a hundred shifts or more, but how much is the actual accomplishment worth? Would it bring her into debt with her sponsor? In trouble even, if the giants do something wrong?
Did it hurt the story? That would be my only objection. I haven't read the whole story, only a short section before and after the ritual, and it seemed to fit into it rather nicely.
If you look at it from a "how many shifts do you need to raise a thousand dead giants", then I agree, the preparation alone should have probably taken weeks. But if you look at it from another perspective and treat it as a presence replacement to convince them to join the fight, it is going to go down in numbers rather drastically. The need for a ritual would still be the fact that they are dead, so you wouldn't be able to talk to them, and they wouldn't be able to help otherwise. But raising them is not the interesting part here, at least in my eyes, so it should be alright to just gloss it over with a good description.
I stressed the earlking ritual as an example of a high complexity ritual cast in a few moments. But this, too, could have easily been done without any numbers involved by Harry accepting a compel. The GM and Jim agree to let him do the ritual successful, but the Earlking will still get loose and roam through town, due to Cowl and Kumori interfering.
With all that in mind, it should be possible to limit any actual thaumaturgy to sane numbers and treat the big rituals as plot devices. That's kind of what I had in mind the whole time, it just took a while to shake it loose.
It can ease a whole number of things, actually. For example I think it would be highly annoying to roll and roll and roll on the zombies throwing themselves against Harry's wards to get in. First of all, you'd need to determine a strength value for Harry's wards, which would probably be rather high, but how high exactly? Doesn't really matter, they are going to get through eventually, so you can just say something like "At the rate the zombies are throwing themselves at your wards, you have 5 exchanges to come up with a plan, that is equal to your complexity for wards. The clock is ticking."
And this is sort of what I meant by "unopposed", too. If a ritual is unopposed, the only reason you are doing it is its importance to the story for one reason or another. And if the ritual isn't even important to the story, then it can most often than not fall under the "mundane effects" category. I don't roll on background rituals as well. The big bad summoning a demon, killing someone remotely, etc., all of that happens, because the story demands it to. So if the story demands something of the PC, why not let them do it similarly? Now I'm not proposing to throw the rules overboard completely, they are still plenty valuable. I'm just proposing to solve this problem with an integral part of the fate rules, narrative power.