Previous Article
Next Article
Table of Contents

Lurking About at the 2.1-m Telescope (1Dec92)
(from KPNO, NOAO Newsletter No. 32, 1 December 1992)

Bill Keel recently completed a run at the 2.1-m telescope in which he
never left home (probably could not find his American Express card...).
Instead, he "observed" in what has become known as "eavesdropping" or
"lurking" mode, in which he logged into the 2.1-m data computer over
Internet to "watch" the data come in and to keep abreast of the data
quality while a collaborator was present at the telescope. Bill
provided us with the following account of his experience.

"Remote observing has not been my favorite notion in the past, but once
you have a family and faculty position time pressures can make it an
attractive option. In this case, it went very successfully.
Communication with my collaborator at the telescope was straightforward
via the Unix talk utility, and direct use of the CCD computer (lapis)
was reasonably fast. In no case was the network connection lost. The
only holdup was in looking at fully two-dimensional data. The network
connection never goes much faster than 5 kbytes/second, so bringing the
image to a local machine by IRAF networking is more sensible than
trying to display directly from the remote node and running the risk of
having to do it twice to see the full dynamic range of the data. For
typical long-slit spectra, it took 3-5 minutes to copy the relevant
range of rows. This is no problem for routine observing (our exposures
were 10-40 minutes), but when checking calibration and focus tests, I
had to rely on doing row and column plots. The only aggravating problem
was that when plotting on the remote machine, it is possible to send
IRAF plotting tasks into limbo if certain keys are struck before the
remote signal to enable cursor readout has been received.

"There were some positive advantages to having a remote observer.
First, with three people on the project, we would do little more than
bump elbows in the control room, especially since all the data
interaction takes place via a single workstation. Having a second
person free to look at the data full-time gives a real addition in
capability, as well as something to do. The distant partner also may
have access to extra information that becomes useful during the run
(say the complete dataset from an earlier observing run) that is not
practical from the mountaintop.

"I also kept notes on how close this capability currently is to genuine
remote observing, since that is a capability that will be desirable for
dynamic scheduling in the future. I set up several monitors with useful
information, including a PC running a planetarium program set to show
the sky over KPNO so I could easily track hour angles and air mass.
(For interested parties, there is a $20 piece of shareware called
Skyglobe that did the job nicely). I also have an X-windows tool to
show weather satellite photos, but admittedly I did not get hold of it
until after the run. Fully routine observing is basically possible now,
in the sense that if the operator has no trouble identifying the object
or its placement in the field, one can exercise full control over data
taking by network. It would be very useful to experiment with the frame
grabber at the 4-m and see whether a remote user can access an
acquisition TV image quickly enough for real-time object identification.

"Fully remote observing is a close possibility, but doing it now would
take an extra measure of responsibility from the setup people and
telescope operators. I realize that these are both delicate issues, but
this is a direction that many new instrumental developments are driving
at several observatories, and it may be gratifying to see how close the
existing setup is to being able to work in this way."

For those interested in lurking about, keep in mind some of the
consequences of not being at the telescope. For example, while you may
be close to your family, it is unlikely that they will encounter you in
a conscious state; and, if you have a noisy daytime household (dogs,
cats, kids), sleep may not come easy. In a more scientific vein, most
lurkers find it very important to have a collaborator at the telescope
to be in harmony with the conditions and observations ("what was that
loud bang I just heard when the comparison came on?", or "did not the
shutter sound like it opened kinda slowly that time?"). Also, many
"remote" observers find the distractions of their home institution
interfere with the more important act of observing, whereas when one is
sequestered on Kitt Peak, the cross-section for interruptions is much
lower.

Nevertheless, eavesdropping provides an attractive alternative under
certain conditions as Bill noted, and we are interested in seeing
others participate.

George Jacoby

Previous Article
Next Article
Table of Contents