Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/ops/stories/5.text Дата изменения: Tue Jul 13 17:17:41 1999 Дата индексирования: Sun Apr 10 14:27:02 2016 Кодировка: Поисковые слова: http astrokuban.info astrokuban

- "US" vs. "THEM" - - Then vs. Now -

O Organization

o The major difference in the way "we" used to do things was that there
was a "we", but there was also a "they". And it was always they who
failed to get us the data we needed, or didn't understand what we sent
them, or had any idea what we did or refused to acknowledge our existence.
"We" were whoever was working on our system and "they" were everyone else.
Of course each system thought of themselves as we and everyone else as they.
Now we have become all us where we know just what web page to access as
the need arises.

o Continual seemingly random re-organization of the dizzying byzantine variety
of Institute "Divisions" into a multitude of acronyms that had meaning
mostly only to those serving within each or to those that set them up.
But as long as the front door remained open and the power kept running
I didn't pay them no nevermind. Connecting systems that passed vital
information had interfaces defined by meticulously defined "ICD's",
but in the spirit of "We/Them", I never had much idea who or what was
on the "other side" of the interfaces. Now stability reigns and everyone
knows where everyone belongs.

o The obvious result of this was a total disconnect between what we
prepared for the telescope and meaningful input from the PIs of what
they wanted and when. Getting a response as to whether we could trim
an exposure to make it fit on a calendar required sending a message to
someone who would contact someone else, to who knows how many levels
of interference, to reach the PI, to return through similar layers.
The iterative "black box" shuffle. It wasn't until soon before Presto
came about that I discovered the identities of some people who were vital
in this chain. A nice test now would be a quiz to each division where
people are asked what Presto is, what role it plays in the project and
who is in it. And the same for us on their roles.

o Hidden away in the 'crypt' ("The vault under the main floor of a church")
- Room 112. This work area was two floors away from most of our offices,
necessitating moving all our supporting papers and notebooks (and coffee
cups for those dependent on such) up and down stairs at least daily.

o Two separate sites for operations - Science Support Center (SSC) at Goddard
Space Flight Center and SCience Institute Facilty (SCIF) here. This would
have necessitated split work shifts with travel up and down and up and down
and up and down the Baltimore Washington Parkway on a rotating basis.
The Washington Post once mentioned the relative risks of Space Shuttle
flight compared to commuting from DC to Baltimore. Commute is riskier.
At least we didn't use cable cars.
Now telecommuting reduces these risks. For some.

o Projected staffing levels for SSC and SCIF.
I have a copy of a sobering video made in late 1985 summarizing the
presentation of our readiness to operate SOGS in support of the upcoming
launch (then) expected a few months later. In spite of SOGS SPSS version 1.0
still pending delivery, and a sub-skeletal crew still being trained, we
declared our readiness to go for it. I suppose if we didn't have chutzpah
we wouldn't be in this business. But it's amazing to view it.
The design expected SOGS people to swap back and forth from the Science
Planning and Scheduling System ([Short-Term] Scheduling portion)
and the Observation Support System (OSS) for real-time contacts.
For about three years, I actually did alternate between OSS and SPSS.
(Even did one week on PODPS.)
Now when I come in I know that I'll either be working on building a calendar,
manually processing routine ephemerides, resolving contact conflicts,
generating ground support and spacecraft products, testing new installations,
learning new ways of doing things, or dealing with the unexpected, weird
or unique. And I sleep at night. Sometimes.

o Midnight shift work (at an astronomical observatory??? - how retro!)
Round the clock and through the weekend shift work, with corresponding
shifts of time of day or length of week. Maybe once a week or month
could everyone be in the same room for a common meeting. Some people
were only seen at shift handover.
For an example of the conflicts of what such a schedule can do, just
how many opinions are there for the meaning of "Monday midnight"?
Suppose your shift were to start then. Would you show up Sunday night/
Monday morning between 11:59 p.m. Sunday and 12:01 a.m. Monday?
Or would you wait until Monday reached its end boundary and arrive between
Monday 11:59 p.m. and Tuesday 12:01 a.m.? Hmmmm???? A short story (by
Odgen Nash?) was written about the tragic consequences when two folk assumed
different interpretations of this ambiguity. (I could also mention the
ambiguity of 12 a.m./p.m. when m. means meridian and a. and p. at noon and
midnight are by definition undefined, but assumed nonetheless. To make it
worse, some might say m. is meridian OR midnight. And n. might mean noon
OR nadir. Insurance companies use 12:01 to avoid the financial consequences
when people try to take advantage of various interpretations.) And don't
even ask about time cards when working over weekly boundaries. Or to which
day the over-night shift report belongs.

o Twice-Daily shift handover meetings run by 'SMURFS' in the 'alcove' (Room
103) next to the 'crypt'. Progress reports given for every one of the seven
weeks being built.

o Proposal prep being done - OPR fixes, Guide star requests, GS_ACQ, updates,
etc. - as the week being built was started. Oftimes being changed as
a visit would drop from one week to the next.

o Build a calendar, putting observations down in the best workable order.
Then tear it apart to minimize the solar array slews (bad SADE!)
And avoid the solar system folk's favorite observing time - opposition -
to avoid overheating the magnetometer sensor.
Then Dr. Musgrave handled them tiny non-captive screws for us.
Now we only worry about ORIENTs. And LINKS. And DECONs. And TOOs.
And Gyros.

o Knowing what's coming next.
We could count on the "Hubble Constant: Two Years till Launch".
Now all that's left is waiting to "Go To 'L Too".


O Software

o Case sensitivity in OSS - upon initial delivery SOGS suppliers mentioned
they routinely hit the SHIFT LOCK key to avoid problems with it interpreting
upper and lower case differently. Thank heaven we won't ever have _tHaT_
problem ever again.

o Parity reversals and confusion in spacecraft coordinates.
"Are we moving the spacecraft _or_ are we moving the stars?"
"Are we rolling the detector _or_ are we rolling the sky?"
"Is that U3 or V3?"

o Scientific Instrument aperture list, FOVs, coordinate systems with
inconsistent names, sizes, positions, orientations, parities with no
central list or understanding. "Banana Mode". And "breathing."

o Some obvious horrors that others may report:

COMET, MENU_MODE, and on-line "HELP".
"Which is a qualifier and which is a parameter?" Some were required to
have values and others were not. Some values were allowed and some
were not and sometimes the software would check if they were legitimate
and sometimes it would not. Some would assign default values if none
were entered, some would not. Some would allow you to change the
default values in menu-mode, but they might have to be re-entered
each time you used them. Some parameters/qualifiers had similar
inputs (time was the most common example), but the spelling of the
parameter or qualifier name would be inconsistent from one command to
another or the format of the time string would be different from one
another. Now we only need to remember the difference between /tstart
and /begin and "place" and the like. And remember which systems use
UT and which use local EDT/EST.

"CHILD" capability - to allow a command to be started in one area
and then "CONNECT" to another to do something else. However, if the
first CHILD needed to output a message to the screen, it would suspend
if you were in another CHILD. And connecting back would prevent you
from doing another connect change. Only way out would be to kill the
CHILD, or execute a BackSlash ABORT. Nasty business.

No scrollback of the screen capability. If you didn't snap it (on a
thermal printer) or scribble it down, you'd have to have a good memory.
No recall of previous command capability.
Long command strings would have to be typed and re-typed over and over.
With very limited ability to check back to see what had just been used.
Command line input logging. Everything went to a log file that could
(sort of) be accessed, but not by just typing it out. Another convoluted
command would be needed to report on just certain types of contents.
You almost needed to know exactly what you were looking for to access it.
(Kinda like manpage, but it could be indexed.)

o PFORMS - others will probably mention these. At birth, these were intended
to be the means of entering all proposal data. We actually did manually
enter quite a number of proposals this way until the alternate system was
shown to work. Continued to use it to make OPR related corrections or
adjustments. AGD, APD, IAC are about the only residue of this still in
active use.

o Software installations that would take an entire day or two. If they
didn't have to be backed out. Although I think they were 'patched'
rather than ever once rolled back.

o Delivery deadlines in May that were met on May 42nd.

o Order Forms - I'll let others describe _this_ instrument of torture.

o "This page intentionally left blank."

O Hardware

o Obvious - VT-100 for interface, DeAnza for graphical displays, no scrollback
capability, no recall of previous command capability. (See above.)
Mouse? What's a mouse? They came with the MACs. Emulating a VT-100.
But they would allow scrollback.

o Chilly air from the underfloor air flow (Room 112) to keep the computers cool.
Comes from being hidden away from the rest of the Institute and the public
in the 'crypt'. Some allegedly kept their beer cold under the floor.

o Classes on what to do if the Halon dumped. Fending off demands of outside
safety officers to do a real Halon dump to show that the valves would work
and that the entire area would be instantly and extensively disabled by
the extensive dust that would be blown around. This might have also
disrupted the mouse ecology under the floors.

o Emergency power shut-off buttons that were loose enough to be engaged by
accidentally leaning against them.

o No parking garage - parking on University "Parkway" or nearby residential
streets, then parking garage offices, then no garage offices. North and
South wings added to the central core.

o Data plugs that had to be pulled from operational system to test system
(pre-launch.) Done so often (daily or oftener) that the pins eventually
wouldn't seat right and would need to be replaced.

o Using FAX machines outside the operational stream to receive vital,
critical data for accessing the spacecraft. Such data hand-written on
paper, then processed through workstation keyboards to reach another
system via convoluted steps in a GUI to reach the final system in
which it would be used. Whoops, sorry - this "story" is about how it
"used to be done". All this used to be done remotely by people we never
saw and rarely heard with turn-around times of the better part of a day.
More of "them". Now it's done by "us".

o Hard-copy paper 'bible' of SPSS procedures. Periodically updated with
new hard-copies after the one 'official' book got too many hand-written
notes to be legible. All now on-line.

o ad hoc variety of desk phones and answering machines in every division
and office. Except no answering machines for people in SOGS.
Now phone-mail for all.

o hard, smooth tile that looked good, but provided excellent slipping
potential when wet. Occasional bullet holes in the windows facing
the back lot. Near by the cafeteria that got converted to a photo
darkroom brewing photo chemicals rather than greasy V***-burgers.

o road access via a street that neither the campus nor the city would
admit it owned or was responsible for. No provision for drainage
at curve at Hopkins crosswalk just north of entrance, hence perpetual
potholes or icy spots. Last to be plowed after snows. Solved that
by not snowing much last winter. Or hardly any rain this summer.

o Unshielded 'security' lights that prevented views of the night sky
and whose glare made it even harder to be aware of surroundings and
nightcrawlers. Yet we could still see the comet crash into Jupiter
from a scope set up on the sidewalk.

o number of keystrokes needed to crash SOGS (seven would do it, but
two if you were at the right line-printer.) For graphics capability,
one touch of the trackball at the DeAnza would freeze it up.
Now it takes an entire GUI to freeze a workstation.

o "PAC-MAN" processes that would 'eat-up' active processes if they
discovered problems. These would hibernate so they wouldn't be seen
with a normal 'show processes', but would stay active unless specifically
killed. If you could logon to get at them.

o Right Ascension displayed on graphs with Zero at the left and increasing
to the right. That is, west was displayed to your left when facing south.
But then, how many know their vesper from their orient anyway?

o A major league baseball team that set new records for futility.


O public perception

o Dick Gregory in DC and Balto

Now this is an amazing story not well known.

In July 1990, a few months after launch and a short time after the
announcement of spherical abberation, the well-known speaker Dick Gregory
appeared on "D" Street, NW in Washington DC at the public open-air funeral
of homeless activist Mitch Snyder. Among the thousands there under the hot
sun, I was probably the only one with a connection to the Institute or the
HST project. Although I was videotaping the event, unfortunately I had the
camera off just at the time of the episode related here. (I was trying to
get closer to get a better picture, but all available vantages were filled
- a case of trying just a tad too hard at the wrong time.) Otherwise, I
would have shown this on video.

Anyway, there were all sorts of speakers, from Jesse Jackson, to Cher,
to one of the Berrigan brothers to the actor who played Snyder in the
movie "Samaritan". And as you can imagine, the circumstances were
rather tense and grim. But Mr. Gregory has the ability to get people's
attention and get his points across with admirable humor.

He got his biggest response of the entire day when he compared
Mitch Snyder's ability to see racism in society with
"Our billion dollar telescope in space that can't see."
The audience response was explosive laughter. The kind of knowing
cathartic laughter that comes from a shared view of not feeling responsible,
but somehow still paying the price. It seemed that everyone know what he
was talking about.

I returned to the Institute for my next shift and we all carried on.

As a followup, some years later, Mr. Gregory returned to speaking
engagements on tour. He had suspended that part of his career because
at the time all the venues were smoke filled restaurants or clubs that
typically served alcohol. Another aspect of his career is diet management
and he could not continue supporting those environments. But with advent
of health sensitivity (albeit with an amazing ignorance of the role of
Polonium-210 in smoke), smoke and booze free theaters and forums became
more prevalent. (Remember when smoking was common inside the Institute?)
So he appeared in such a venue in Baltimore. Out of the many laughs he got,
again the biggest of the whole event came when he compared the then recent
ability of the HST to detect brown dwarfs with the inability of the
government to accurately count the "number of nappy hairs" at the
recently concluded "Million Man March" on the Washington Mall.
I was probably the only one in the audience who worked on the calendar
in which those brown dwarf observations were scheduled.

o Computer Sciences Corporation routinely misspelled in Institute documents.

o Finally, what may be the most amazing story is the tale of the trust of
the powers-that-be in the ability and dedication of a superlative staff to
fix and overcome the horrendous shortcomings of the nearly still-born SOGS.
A testimony to their confidence and the service of the people then here
and yet to come. Or maybe it was just the attitude that we were going to
make this work no matter what. At least _that_ hasn't changed.