How are we going to get spectra for nearby SNe? A NEAT type survey would produce roughly 1 SNe per night. Moreover, at any given time I would expect about 15 SNe to be part of the spectroscopic monitoring program (10 spectra over -18 to + 60 days). Thus, the demand would be for about 3 spectra per night (1 screen, 2 monitoring). Case I: Let's consider big telescopes with integral field units. Then the set-up time is the same as for imaging, making the idea tractable. Assume 5 minutes for set-up and 5 minutes to confirm a z = 0.05 Type Ia one magnitude below max. For the above program we would need 0.58 hours per night, giving 213 hrs, or 27 nights (1278 spectra). We might expect 3 nights at the VLT (with the Portugeuse), and 5 nights at Gemini north + south (with Hook) IF WE ARE ONLY DOING A NEARSEARCH. Then we might be able to get the equivalent of 2 HET nights by teaming up with Wheeler. This only provides 1/3 of the nights needed for the baseline program. Everyone will argue that this is inefficient use of these cutting-edge telescopes. Also, when you include calibration, the situation is even worse. Case II: Since I think the big telescope route will be shot down, let's look at the medium telescope route, as before: SOAR, NTT or ESO 3.6-m, Lick 3.5-m, NOT, INT, WHT, CTIO 4-m, centered around an ESO Key Project and maybe an NOAO Survey. SOAR is the only one of these telescopes that is explicitely queue scheduled. If we teamed up with the Italians for an ESO Key Project, they might provide an observer through their fellowship program (now ruled out). NTT/ESO and INT do have service observing, but those options involve high overhead to prep the service observer. Classical scheduling would work for monitoring, which is about half of the overall observing budget. This would involve a lot of training of observers and traveling to telescopes. Even so, what might this budget look like? ESO 3.6-m 25 nights/yr (25 for Ia, 25 for II, w/ Italians) Lick 3.5-m 10 nights/yr classical (would Fillipenko have a fit?) NOT 2.6-m 12 nights/yr Ariel/student CTIO 4.0-m 15 nights/yr classical - NOAO survey KPNO 4.0-m 5 nights/yr classical - NOAO survey APO 3.5-m 4 nights/yr classical SOAR 4.2-m 5 nights/yr queue INT 2.5-m 6 nights/yr service? (only if intermediate search quits) WHT 4.2-m 2 nights/yr service? (only if intermediate search quits) If we assume that these telescopes need an average set-up of 15 minutes, and average integration of 35 minutes, each night translates into 9 spectra. So, the above total about half the required ~ 1100 spectra/year. A lot of this time would be fixed, meaning that it would be hard to have access every night. Note that the use of these various telescopes would have to be very seasonal to avoid heavy weather losses. (Letter of intent for NOAO Survey Program to start in Fall 2000 would be due Jan 29, 1999. See http://www.noao.edu/scope/surveys/). Further Ramblings: Is there any chance that 2-m & 3-m class telescopes in the next decade will become like the 1-m class telescopes of today? If we just bought a piece of SOAR, we would need about 30% of the time (exploiting the better image quality and set-up efficiency (due to IFU). I think this shows that we would have to reduce the number of SNe, or the amount of monitoring in order to make this work. It also shows that NEAT can supply as many SNe as we can handle. One could imagine begging the equivalent of 1 night apiece from VLT, Gemini, and HET, yielding another 144 spectra.