Live from the Hubble Space Telescope


PART 1: Challenge Questions: last week's answer and a new puzzle

PART 2: Today's WebChat: the folks who operate the spacecraft

PART 3: Passport to Knowledge at NSTA in St.Louis

PART 4: Converting our data from round to rectangular

PART 5: Seeing Hyakutake from Kitt Peak (Arizona)

PART 6: Overloading the database server


Last week we asked:
You are packing for your vacation on Pluto. Of course, you want to bring along your camera to take pictures of the fantastic scenery. What would happen to your photography plans if you forgot to bring you camera's tripod?

ANSWER from Marc Buie:
The basic answer comes from knowing that Pluto is much further from the Sun than we are here on Earth. That greater distance means that sunlight will be fainter. How much? Pluto is currently at 30 AU (Astronomical Units) which means it is 30 times further from the Sun than the Earth (that's right Earth is at 1 AU from the Sun). There is a law in physics known as the "inverse square law" that means if you move twice as far away from a light bulb (or star) that the light bulb will appear four times fainter. So, you take the distance, square it and that is how much fainter. For Pluto at 30 AU, this means that the Sun appears to be 30 * 30 or 900 times fainter than what we see here on Earth.

To fully answer this question, it helps to know a little about photography. First, a tripod is needed to hold a camera steady when you using really long (slow) exposures. I find that for a normal 50mm camera lens (the most common) you can take hand held pictures if the exposure time is 1/60 sec or faster.

Now, when taking a picture, there are three things you can control about your camera to adjust for different light conditions.

Exposure time(how long you let light hit the film)
Film speed(This is the ISO or ASA number like 100 or 200)
Lens aperture(This makes then lens collect more or less light)

A typical sunlight day with no clouds at noon would require these settings:

So how would you adjust your camera settings for Pluto? If you expose the film for twice as long, then twice as much light can hit it. If you open the aperture wider you let more light in. This is a little trickier to figure out but the smaller the number, the more light you collect. An f/11 aperture lets 4 times as much light in as an f/22 aperture (factor of 2 on the number is a factor of 4 for light hitting the film). Finally, if you double the ASA rating, then the film is twice as sensitive to light.

So we need to adjust these settings by a factor of 900 to get a good picture on Pluto. We only have one notch to go on exposure time. If we set it to 1/60 second then we've adjusted by only a factor of 2, not nearly enough. Next, lets open the lens aperture to let more light in. My camera lens goes to f/1.8. Compared to f/11, that would let in 11/1.8 * 11/1.8 more light in, or a factor of 38 more light. Combined with the change in exposure time, we now have adjusted by a factor of 76, still not enough. If we don't change our film, we would definitely need a tripod since we'd have to use a 1/4 second exposure to record the scenery.

We could have planned ahead though and brought more sensitive film. You can get pretty good stuff that is 1000 ASA or 10 times more sensitive. That would do the job nicely. So, for Pluto, you'd set your camera to

and you wouldn't need a tripod. But, if you had a telephoto lens, you would definitely need a tripod because they need faster exposure times. Don't worry too much if all these adjustment factors don't come out to exactly 900. Film is pretty forgiving and you can still take a good picture even if thing are either a little too bright or a little too dark. The actual factor is 760 but that is close enough to 900 to work.

If you have access to a good camera, you might try a little experiment. Good 35mm cameras have a light meter built in that tells you how bright the scenery is that you point your camera at. You can put in these camera settings and adjust the brightness in a room by turning lights on or off until the camera tells you these settings will work to take a picture. When you match these settings and the light is right then you have just simulated how bright the light will be on Pluto.

I found that a room with the drapes drawn and no direct sunlight hitting the drapes on a sunny day is just about right. You could still read a book see colors, and there would be noticeable shadows. Similarly dark lighting would be during a strong thunderstorm here on Earth though that level of lighting would be in full Sun on Pluto (no clouds).

Challenge Question for this week:

This Challenge Question is from our Neptune Planet Advocate, Heidi Hammel.

If your spaceship needed some emergency repairs, and you were in the vicinity of Neptune, where would you decide to land? Why there?


If you are able, join us today (Tuesday, March 26) in the Live from HST WebChat room. We'll be discussing the Hubble Space Telescope as a spacecraft and all of the very detailed engineering parameters that need to be monitored to keep things running smooth.

The Chats are scheduled for 9-10am and 11am-noon Pacific (noon-1pm and 2-3pm Eastern). A basic Web browser is all the technology you need to join us. Alternate access is available from the main LHST homepage by choosing Teachers' Lounge.

Most of the following NASA folks promise to be present today: Chris Wilkinson, Eric Isaac, Kathy Southall, Tony Cruz and Angela Manifold. They represent various parts of spacecraft operations and if we get lucky, they'll actually be chatting from the SMOR, next to the main HST operations area.

If you do join us, prepare yourself for a bit of chaos. This is our first expert WebChat and the format is unmoderated. This means that we will all have to cooperate to have the WebChat operate smoothly. Three basic rules to keep in mind:

  1. Please identify yourself with at least school grade and location

  2. Don't be a chatter hog; ie, ask your share of questions, but try make sure that others get a chance as well

  3. Keep to the topic; we don't want to see a bunch of "what's your favorite band" chatting between kids

But with our typical Passport to Knowledge explore-and-grow attitude, it should be educational and big fun. Tally-ho!


The Passport to Knowledge team will be all over the NSTA Convention in St. Louis later this week. Please set aside some time to join us at any of the following events:

Refreshments and informal conversation on Friday, March 29th from 8PM to 10PM at the Marriott Pavilion in Pavilion B. The Marriott is located on One Broadway, downtown St. Louis. We took you to the South Pole, the Stratosphere, Neptune and Pluto....won't you please join us for dessert!

A formal presentation on Saturday, March 30 from 5:00-5:30PM. The title of the talk is "Passport to Knowledge: Real Science, Real Scientists, Real Locations, Real Time". Join us in Room 261 of the Cervantes Convention Center for this fun

Also, another presentation on Friday March 29 from 2:00-3:00PM is scheduled for the Hawthorne 1 Room in the Marriott Pavilion. We'll be presenting a detailed overview of the Passport to Knowledge system of exciting students about science learning, including future opportunities.

Passport to Knowledge Team members will also be found hanging at the IITA booth in the NASA space in the Exhibit Hall. Look for the cheery folks with the six high-speed Interrnet computers and dualing VCRs. We'd love for you to visit us and drop in for a turbocharged Internet surf session.

A special event will be our CU-SeeMe session from the exhibit floor of the NSTA convention. If you are with us physically, join us in the IITA booth between 11:30AM-1:30PM Central (9:30-11:30AM Pacific (12:30-2:30PM Eastern). Else, join us virtually by connecting to the reflectors at: and


Heidi B. Hammel

March 22, 1996:
During the past week, Wes and Heidi have worked very hard. Much of our effort has been focused on analyzing the 1994 and 1995 data - since those are the largest and most complete data sets.

But the excellent 1996 LHST data have not been neglected! We have been able to set up and run the navigation programs on the new data. That took quite a bit of effort on Heidi's part, but it finally all came together just this afternoon.

For the LHST data, we can now convert the round images of Neptune into rectangular maps of latitude versus longitude, just like we did last year for the earlier images. It is in these maps that we will be able to tell if there is still a Great Dark Spot (or not). Lucy Lim, an MIT undergrad who has been working as our student assistant, has been converting the new images into maps. We will work more on that this weekend, and try to put some of the maps on the web site by next week.

On Wednesday, a producer and cameraman showed up here at MIT with several suitcases full of video equipment: lights, tripods, a camera, lots of cables, extension cords, and microphones. They wanted to film us actually working on this data at the computer in Heidi's office, and across the hall at Heidi's main computer station. We showed them (you) how the images of Neptune and the maps look on the computer screen where we make measurements. We also talked about the different things we do during analysis, and about what we still have left to do. They filmed for about 5 hours, but in the final show next April 23rd, they will probably use about 5 minutes of that (that's show biz!).

Wes has been working hard on writing more text for the paper, which will now definitely include the LHST data (along with a reference to the school kids of America who chose Neptune as a target!). This paper is tentatively titled: "Atmospheric Structure of Neptune in 1994, 1995, and 1996: HST Imaging at Multiple Wavelengths" and we will probably be submitting it to a journal called Icarus, which is all about planets. So these observations will make a real contribution to our knowledge of Neptune.


Trisha Borgman

March 25, 1996
Well, right now I'm in the console room for the 0.9-meter telescope at Kitt Peak. We actually arrived here last night (tonight is the first night of our run). My boss is also here for a 3-night run on the 4-meter telescope, so last night I went up to the 4-meter and tagged along with him for most of the night.

At the beginning of last night, it was much too cloudy for his galaxy observations, so we took many images of Comet Hyakutake. Have you been able to see it? I've been watching it for several nights from my backyard in Phoenix, and it's been spectacular. Even from our house on the outskirts of town, we could easily see it with the naked eye. It looked like a little, fuzzy cotton ball, and it moved quickly from night to night. However, nothing I saw at home could have prepared me for the view I saw last night.

Like I said, it was fairly cloudy at dusk yesterday. We could see the comet, but not very well. By 10:00 pm or so, the view was spectacular! I could easily see the entire comet, and the tail stretched almost halfway across the night sky. It was beautifully bright! I remember seeing Halley's comet back in 1985 (I was in 5th grade then), but we could barely see it at all. I've never seen anything like last night's view of Hyakutake. I have a feeling that I will remember everything I saw last night for the rest of my life...

Anyway, I stayed in the 4-meter telescope with my boss for several hours, just watching everything he did. Even though it took 15 minutes for each of his exposures, I had a lot of fun! By the time I went to sleep at 2:30 am, the comet was streaking across the sky with an unbelievable brilliance.

Today I have been learning more about how to operate the 0.9-meter. Dusk here on the mountain is at about 6:45 pm, so we'll finish taking our calibration images (remember the images I talked about earlier--the ones which remove the "signatures" of the telescope itself? Those are the images I'm talking about now). It will take about an hour and a half to complete all of those exposures, so we'll start those soon then go down to the cafeteria for dinner. And as soon as it's dark, we'll start taking our pictures for the catalog!

I've already learned so much, and I am so fortunate to be here (especially while Comet Hyakutake is at its peak!) -- I just keep pinching myself to make sure I'm not just dreaming!

I will write more after our observations tonight!
Trisha :)


Tom Comeau

March 13, 1996:
Aaaaaaaaaaaaarrrrrggggghhhh!!!!!!! :-}

:-}	<--- (That's an "emoticon", which is computer nerd speak for an
	emotion icon.  You'll also hear them called "smileys". This
	particular emoticon means something like "pained grin". 
	The conventional smiley, :-) or ;-) just means "smile". 
	Since people can't see what your face looks like when you
	email them, you have to give them hints, or they might
	think you were mad at them when you're just trying to say
	something funny.

I'm a little peeved right now, because my parallelism has come back to haunt me.

Remember from my first journal, when I wrote about how the software changes I was making would allow DADS to work on more than one thing at a time? We used to have just five programs, and each did one step of processing HST observations in sequence. Our new version has seven programs, and lets us do an arbitrary number of jobs in parallel. I can now run eight or ten copies of each of those programs, and work on eight or ten HST pictures at a time!

I've completed all the changes, and I've been testing the complete new system. One thing each program does is open a "connection" to a database server -- another computer that does nothing but handle our database. The database describes all the HST data we have on Optical Disks in the Archive. Unfortunately, having all those extra copies of the programs means I've now overloaded the server, and I have to get the people who maintain that system to change their setup.

Imagine that you have one person who is real good at looking things up in the phone book. The phone book is a database (a paper database, but a good database nonetheless). The person who looks things up is the database "server". There is one "in box" for requests to look things up, and several "out boxes" for completed requests.

The first thing you do is "Open a connection" by marking one of the "out boxes" as yours, by putting your name on it. Only one "client" can use an "out box" at a time.

You get the "server" person to look numbers up by putting a 3x5 card with a request on the card in a little "in box" in front of the "server."

The "server" takes cards out of the "in box", finds the phone number, writes it on the card, and puts it in an "out box", where you pick up your card. There is a separate "out box" for everybody who uses the server.

You take the card out of your "out box", and it has the phone number you want on it. Then you take your name off the "out box", so somebody else can use it.

If there are only a couple of people putting cards in the "in box", the "server" can keep up. If there are 30 people putting cards in the "in box", the server gets behind and the "in box" fills up.

Also, if there are only five "out boxes", and a sixth person wants to use the "server", the "server" doesn't know what to do with the request once he's finished, because there are no "out boxes" left!

That's what's happened to me: My database server ran out of connections, and got flooded by requests.

So I'm getting the database people to "build a bigger 'in box'", to make sure that there is enough space to handle incoming requests, and make more "out boxes", so that I can have more connections to the server.

Which also reminds me how many people are working on developing, maintaining, and operating the Archive. Meanwhile, it's a nice day. It's in the mid-60s, which is great even if you don't consider that we had two inches of snow last Friday. Another member of the Archive Team has brought a baseball and a couple of gloves, so I think I'll go throw with him for a while. ;-)