[evlatests] Anagrams. also, Archive averaging broken?

[Vivek Dhawan]

Thing 1. Rick was quoted today, as saying "the monster awakens" re WIDAR.

         So here, IMHO, are the top 5 anagrams for "WIDAR correlator":
         (how _did_ they waste time before the internet?)

         5. Radical Terror OW!
         4. Torrid War Oracle
         3. Erratic Raw Drool
         2. Erotic Lard Arrow
         1. Radio Terror Claw

Thing 2. Some pointing data were taken Thursday evening, and I fetched it
    via the archive tool with spectral avg=2 and time avg.=2s just to try
    it out.

    Estimated time was 80 min, it took 3 hours, but the data did appear.
    (UVfits output, ~6G after averaging, 23GB before).

    The delays were high as reported by Rick yesterday, but not disastrous.

    Antenna flags were clearly not present.

    The fringes went very weak on all antennas after about 1hr into the
    2-hour data, and just about then, many many zeros appeared in the
    visibilities, in different patches on different antennas.

    I quickly accused the CBE of making the zeros, but Martin could find
    nothing amiss in his log. So I re-fetched the same data from the archive
    with no time or spectral averaging, and now it looks just fine for the
    entire 2 hours.

    So now I am accusing the archive averaging (which is done in CASA,
    not in the filler, right?). I have both versions in AIPS. Can someone
    investigate the CASA end of it?

The filename in the archive:  syspt2hc.55231.974717060184.uvfits