On Saturday night I got involved in a drunken discussion/argument with another astrophysicist over facilities for future astronomy. As I said it was a drunken argument (from my side at least - the other guy could have been completely sober) and I was probably not too coherent, got slightly tetchy and aggressive (sorry) and took it a bit more seriously than I should have, but I thought I'd go over the main thrust of the disagreement as I vaguely remember it.
In general I thought that funding the two major planned astronomical mega-observatories, the optical/infrared European Extremely Large Telescope (E-ELT) and the radio observatory the Square Kilometre Array (SKA), was a very good thing for advancing science. Whereas the other guy (TOG) thought that funding these two projects would drain too much money from smaller current telescope facilities, and they would be produce insufficient new science to be able to justify us losing access to these smaller class of telescopes. This is particularly prevalent in light of the recent funding crisis in which the UK astronomy funding agency (STFC) has been squeezing funding for various projects, in particular saying it maybe withdrawing UK membership/funding from things like UKIRT and Gemini. TOG was also worried (I think) that all ESO resources would be put into the E-ELT at the expense of their other facilities.
Science-wise I think TOG was being overly pessimistic. His basic premise was that these new telescopes wouldn't really be opening up any major new discovery space and they'd just allow us to confirm what we already know, but maybe with a bit more precision. A summary of the science case laid out for the E-ELT can be found here, and that for the SKA can be found here (with each chapter available for free on arXiv, e.g. here). These cases do obviously say a lot about what we can do to expand upon current knowledge (it's far easier to write a case based on what we currently know), and in fact I think their ability to do this gives a sufficient jump in sensitivity that the increase in science in these areas is very worthwhile [I'm going to cop out here and not give any specific examples, but see the above links]. However, I also think that there is a major new discovery space that will be opened up and lots of unknown stuff to find out.
The science case wasn't really the main thrust of TOGs argument thought - it was the funding side of things that was the main concern (you might be interested in telescoper's post about different people's ideas about what should be funded, and I think in this comment he show's similar concerns that TOG has). He said (if I remember correctly) he'd love to have access to an E-ELT/SKA for his research, but not necessarily at the cost of not having access to other facilities. No astronomer is going to have access to unlimited time on the E-ELT/SKA, so they'd obviously like to have access to other telescopes and be able to carry on doing productive science and there's a worry that this may not be possible in the future. I again thought that this was a bit too pessimistic, and maybe I'm being a bit naive, but I can't see how such a decimation of smaller facilities would be allowed - smaller class telescopes may have to find more novel ways of funding though (e.g. the LSST and ATA), or by partnership with smaller countries, or groups of universities. Importantly, I think that developing the E-ELT/SKA (via its pathfinders and precursors) is very important for providing technical innovations and pushing boundaries of observational techniques, which will feed back into making smaller class telescopes cheaper [maybe I have this the wrong way round, so someone can correct me] and able to do better science themselves.
I do have a skewed view of this kind of issue though. As a person working on gravitational wave observations we have a few (relatively expensive) detectors that create one data set for use by everyone (or for the moment at least those inside the collaborations that built and maintain the detectors), and don't have to compete for observing time on individual telescopes. I'm mainly making my decision based on my view that these new mega-observatories will produce far more novel science than the current technology is able to (and in small part that they look so cool). I might have different ideas if I felt I'd be unable to get new data myself due to limited observational chances on fewer telescopes, and consequently produce fewer papers and probably therefore have a diminished competitive edge in the academic jobs market. We do have some vested interest in having smaller class telescopes though in that when we see gravitational waves it's important to do optical follow-ups to get the most information about the sources. We're unlikely to be able to get the E-ELT and SKA to go quickly into a follow-up mode, but the smaller class telescopes will be vital.
Anyway, maybe I'm talking rubbish, what are your views?
[Update - on a related note this paper briefly reviews big versus small science/instrumentation in physics and astronomy.]