Transterrestrial Musings

Comments

It's because the flight rate was far too low.

Q: How do we get flight rates up? A: Get customers besides the government to start paying for flights.

CSI had a great quote in their July 2006 Lunar Express presentation: "Commercial doesn't mean contractors [it] means companies with private capital at risk"

And since ESAS seems to be even more secure after collapse of the DIRECT initiative NewSpace especailly needs to find a pile of money (not taxpayer sourced) to accomplish their mission.

Sam Dinkin may now be blazing an essential trail by beginning to tap into the trillion dollars spent annually on advertising. Which NewSpace venture will get the job done? SpaceX? Kistler? Benson's Dreamchaser? t/Space? I dunno and no one knows but with Sam's approach it does not matter.

Posted by Bill White at January 17, 2007 12:28 PM

When the wool industry in England was in recession hundreds of years ago, decrees were passed that when you died, you had to be buried in a woollen shroud. A similar thing was done in France with linen shrouds.

Perhaps we can do the same with payload shrouds - ie space burial. It's not like it would raise the price of a funeral these days.

Posted by Roger Strong at January 17, 2007 01:35 PM

That's why average costs are so misleading. The average cost was high because they flew so few times, and they flew so few times not because either the average or marginal cost was high, but because they had no need for any further flights after they won the prize.

I was kind of surprised that Burt didn't fly SS1 at least one more time with him in one of the passenger seats. However, it's possible they didn't because SS1 was quite difficult to fly and had some very narrow margins for stability as witnessed by Mike Melville's spiraling flight caused by excessive dihedral effect. While it would've been great to see it fly some more, it's probably best that SS1 is in it's place of honor in the Smithsonian.

You're right that increasing the flight rate is crucial to achieving lower launch costs regardless of the technology. Any time you combine the need to amortize R&D costs and add in the essential fixed costs with a low flight rate, then your launches are going to be expensive whether you're talking about reusable vehicles or "Big Dumb Boosters."

Posted by Larry J at January 17, 2007 01:56 PM

the question here is .. why didnt SS1 fly more ? Wasn't the market there at marginal cost + nice profit margin, i.e. around $200K ?

Posted by kert at January 17, 2007 01:58 PM

"why didnt SS1 fly more ? Wasn't the market there at marginal cost + nice profit margin, i.e. around $200K ?"

Scaled never intended SS1 to be a commercial venture. It was a three-seater to fulfill the requirements of the X-Prize. The $200k seat price is on the SS2, not the SS1.

Rutan himself has said that, while he wanted to fly the SS1 a few more times, it was better off going to the Smithsonian in one piece rather than as a pile of pieces (I'm paraphrasing, but you get the point).

Posted by John Breen III at January 17, 2007 02:14 PM

Noting the inadequacies of the X Prize I have long been wondering what the next prize should be. My latest thoughts are say $50 million each to the first two non-government groups to perform twenty separate personed orbital flights.

This structure creates ongoing competition – a real race, necessitates a high flight rate with sustainable operation and reliability to match, favours a small low crew number vehicle, and necessitates some space personed commercial venture, (tourism, science, satellite delivery, orbital assembly, space stations, etc.), to close the business case. It is still expensive though, and of course I do not see NASA paying for it.

What would be the minimum value for such a type of prize that companies would still go for? Can it be made more symbolic so that companies are merely encouraged and given some external justification for doing that which they commercially wanted to do anyway? Could one invoke advertising to pay for it?

Posted by pete at January 17, 2007 04:48 PM

> I would guess that the marginal cost of an SS1 flight was less than a hundred
> thousand dollars, even with their expensive choice of propellants.

I heard it was around $150,000, but they were essentially custom-building each motor by hand. Production motors would certainly be less expensive.

Another problem with Dempsey's math: Burt did not spend $20-30M for just 3 flights. Those three flights were just the final phase of the flight test program. The complete test program included 66 powered flights of White Knight and 17 powered and unpowered flights of SpaceShip One.

The entire test program was accomplished with just two vehicles: one booster and one upper stage.

To carry out an equivalent flight test program with an expendable rocket would have required building 66 boosters and 17 upper stages -- a total of 83 stages in all.

It should be obvious that building 40 times as much hardware would greatly increase the cost.

I anticipate that someone will say, "You wouldn't have to that many test flights with an ELV." Which is true, if you don't care about reliability, but that's a false economy.

Posted by Edward Wright at January 17, 2007 04:52 PM

Rand,
OK, I will admit ignorance. After reading your post I realize that I do not understand the term "marginal cost". I had it confused with average cost which I do understand. Would you please explain marginal cost to me. Given that it took several million dollars to design, develop, build and test the SS1, how can the cost of the actual flights be reckoned at only $100K? I'm obviously missing a key concept here.

Posted by Michael at January 17, 2007 09:24 PM

Rand, you're right to emphasize marginal costs, but I take issue with what I see as your implication that they necessarily represent actually achievable launch costs. There's the big economic question of whether demand for flights will rise rapidly enough as the price falls to make the higher flight rates possible.

To illustrate with an (extreme, unrealistic) example: if it takes a flight rate of 1000 a day to push the price of a ticket down to $50,000, then it'll never happen. There's no chance 5000 people a day will fork out $50,000 for five minutes of weightlessness 100 km up. If it takes a flight rate of 1 day, then maybe it will.

The jury's still out on this key question, I'd say.

Posted by Carl Pham at January 18, 2007 02:32 AM

The jury's still out on this key question, I'd say.

The jury is out on that question because you deliberately loaded it. It's an absolutely ludicrous to assume that to meet your completely arbitrary price level you need 1000 flights per day. Even Emirates doesn't make 1000 flights per day.

Posted by Adrasteia at January 18, 2007 03:20 AM

Marginal costs are the additional expenses of flying another mission. In the case of SS1, that would be the cost of the hybrid engine and laughing gas, the fuel for White Knight, and the labor costs for preparing the crew and vehicles for the flight.

Average costs involve extra expenses like amortizing the R&D expenses, depreciation of the vehicle (only good for a finite number of missions), and the fixed costs that will be there regardless of whether you fly or not (such as facilities infrastructure, insurance if applicable, etc.)

These points are often argued when talking about the cost of operating the Space Shuttle. NASA advocates have long claimed that the marginal cost of flying an additional mission is about $200 million, although this figure hasn't changed in a long time so it may be unrealistically low. Others point out that if you divide the multi-billion dollar allocation that NASA gets each year by the number of Shuttle missions flown, the average cost is over $1 billion per flight (and that doesn't factor in depreciation and other factors).

It's like when looking at how much it costs me to fly my Piper Cherokee. The marginal cost is the fuel, oil, and an engine overhaul allowance. The average cost has to include fixed costs like hangar space, annual inspection, insurance, IFR recertification, and other expenses. If I total the fixed costs and divide that by the number of hours flown each year, then add the marginal costs, I come up with a much more accurate true cost of owning and operating my plane. That's a number I hope my wife never sees...

Posted by Larry J at January 18, 2007 06:36 AM

Even Emirates doesn't make 1000 flights per day.

Emirates? How about a different example:

According to the BoT, U.S. carriers operated 8.8 million domestic and international flights during the first 10 months of 2006.

Let's not set the bar so low that we consider 1000 daily flights to be "absolutely ludicrous"

Posted by Stephen Kohls at January 18, 2007 06:47 AM

I knew it wouldnt take long for self proclaimed mathematician Edward Wright to make it to this Kool-Aid kegger. The stated goal of SS1 was attained in three flights. It cost ~30 mil. End of story. Adding up development flight tests, unpowered drop flights and solo white knight flights and counting them as "features" is ridiculous. If he doesnt already work there, he should consider a career at Microsoft.

As for Rand's if-you-flew-it-300-times-it'd-be-cheap argument, that's the alt.space equivalent of if-your-aunt-had-balls-she'd-be-your-uncle. Both scenarios would be true if they happened, but they didnt.

Posted by Greg at January 18, 2007 09:24 AM

Carl makes a good point. Supply must be ample enough for to meet the price point necessary for demand. It is one reason that many believe very high flight rates (1 a day) won't occur until spaceflight becomes point to point. Unfortunately, point to point raises GSE costs, which increases the average costs per flight, if not the marginal costs as Rand defines it. This is the primary driving force for building rockets that can either take-off anywhere or being air-launched like SS1-SS2. At least you avoid the costs of fixed based pads. However, you still have to deal with propellant costs and storage.

Posted by Leland at January 18, 2007 09:25 AM

Kert and John make valid points. The reason SS1 didn’t fly for profit was because it couldn’t maintain controlled flight and eventually would have killed the pilot and all passengers.

BTW I don’t think the spin was caused by aerodynamic forces I think it was caused by axial torque due to large scale vorticity in the flow. There are several papers written on this topic and it is a common problem for Hybrid rocket engines (although I freely admit that I didn’t follow the “relive the X-15 program” much). Another 10 seconds of powered flight and the rates would have killed the pilot. Anyway all that’s irrelevant, it didn’t fly because it was unsafe.

My original point stands. Rutan demonstrated that he could fly 3 suborbital flights at $5-10M a pop. Had he continued to fly, the costs would have gone up due to the lawsuits brought by the grieving relatives of those who died.

Posted by brian d at January 18, 2007 11:22 AM

No, ignoring the ignorance of your second sentence, that wasn't your original point. Your original point was that SS1 didn't reduce per-flight costs as claimed, because you used the innumerate legerdemain of average rather than marginal costs. You remain wrong, for reasons already stated.

Posted by Rand Simberg at January 18, 2007 11:29 AM

>that wasn't your original point.

Hu? This is what you said I said, so I guess it was my original point.

“SS1 cost between $20-30M for 3 flights”

I guess we’ll just have to agree to disagree, I live in the world of what is ($5M to $10M), and you live in the world of what could be.

Posted by brian d at January 18, 2007 01:58 PM

No, Brian, I live in a world in which we make honest arguments, and talk about actual costs per flight, not disingenuous ones. And even by your foolish accounting, they made many more flights than three, as Ed pointed out.

Posted by Rand Simberg at January 18, 2007 02:01 PM

Well. SS1 really didnt fly more than a bare minimum.
Here is a question. You have a vehicle that has accomplished its primary goal. Assuming there is an ample demand for $200K a pop suborbital spaceflights, and your marginal flight costs are around $100K.
You have these options:
1) put the vehicle in Smithsonian. Start work on version 2.
2) fly it more, work out the kinks, spin it off to a company for operating it and keep flying it, recouping your investment. Getting all sorts of first-of-a-kind deals, you are the first commercial suborbital operator after all. All while working on version 2, with lessons learned.

In reality, version 1 happened. Question is, why ? Was the vehicle really that FUBAR so that fixing and operating it would be out of the question ? How are they going to make sure that version 2 is not going to be FUBAR, then ?

Posted by kert at January 18, 2007 02:02 PM

Was the vehicle really that FUBAR so that fixing and operating it would be out of the question?

Probably, for both cost and safety reasons. Plus, he wanted to get it safely into a museum.

How are they going to make sure that version 2 is not going to be FUBAR, then?

Presumably by incorporating into its design lessons learned from SS1, not to mention operating at a different scale.

Posted by Rand Simberg at January 18, 2007 02:07 PM

> Another 10 seconds of powered flight and the rates would have killed
> the pilot.

No, the pilot could have stopped the roll at any time. He chose not to, because he judged the roll rate to be tolerable* and did not want to use extra RCS propellant. As Mike Melvill said, it was a dynamic roll, not an aerodynamic roll. Far from being about to die, Melvill actually let the motor run long to gain some extra altitude. That's how concerned he was.

The reason the roll rate was no problem should be obvious to anyone with an aviation background. Since the vehicle was going almost straight up, roll did not effect its heading and did not need to be controlled. And since it was a dynamic rather than aerodynamic roll, the rate was not increasing. The statement it "would have killed the pilot" is sheer ignorance. Far from being concerned, Mike Melvill actually let the motor run long to gain some extra altitude.

> Anyway all that’s irrelevant, it didn’t fly because it was unsafe.

Reality check. SpaceShip One DID fly again after the corkscrew. It was not grounded for safety reasons, even temporarily. The next flight occured less than a week later, with some changes to pilot procedures but no significant changes to the vehicle. All of your allegations are proveably false.

> Had he continued to fly, the costs would have gone up due to the lawsuits
> brought by the grieving relatives of those who died.

Do you know the meaning of the word "slander," Brian?

Does Lockheed have any ethical guidelines about employees making false statements about competitors in a public forum?

You are the Brian Dempsey who works for Lockheed, aren't you? I found a paper with your name on the web. Despite your boasting about "real rockets," it looks an awful lot like a paper rocket to me.

Please tell me how many rockets you have built that completed their first 66 test flights without a single catastrophic failure.

Then tell me how many of those flight test programs were completed for no more than "30-50%" more than the cost of SpaceShip One.

If it's more than zero, I'll start to take your statements seriously.

Posted by Edward Wright at January 18, 2007 03:14 PM

Note also that even with the infamous 'space shuttle', the 'marginal cost' was surprisingly reasonable. We know that because when that Spacelab mission had to be flown over again -- same flight plan, same crew, same trajectory -- it cost an additional $80 million. Hand-crafting each mission with unique payloads and need-to-be-trained-almost-from-scratch astronauts etc multiplies that cost by a factor of 5 to 10.

Posted by JimO at January 18, 2007 03:18 PM

kert: All of the speculation about why Burt decided to put SpaceShip One in the Smithsonian instead of doing something else is irrelevant.

Burt didn't decide. He wasn't doing SpaceShip One for himself, but for a sponsor (Paul Allen). The sponsor wanted it in the Smithsonian, and the sponsor got what he paid for. End of story.

Posted by Edward Wright at January 18, 2007 03:23 PM

Let us not forget that the SS1 and its backers got $10 million in cash, a lot of publicity, and orders for their next generation vehicle. A great return on a $30-35 million investment. Rutan certainly did very well even with only 3 full blown flights.

Posted by Karl Hallowell at January 18, 2007 07:34 PM

..which still didnt get us closer to validating a data point in the suborbital tourism market.

Regardless of all the studies and refundable deposits, we still dont know how much real demand at $200K a pop there would be for suborbital flights.
SS1 would have had a chance to make a mark there, they didnt, it wasnt their (investors) mission.
I guess we'll have to wait for a few more years.
regardless, i remain moderately optimistic.

I guess mr. Allen was just too rich investor, for him a place in Smitshonian mattered more than trying to make that profit operating that thing.

Posted by kert at January 19, 2007 02:23 AM

You aren't being fair to Mr. Allen. The SS1 doesn't have the proper space for a legitimate tourist vehicle. All the designs I've seen have room for at least six people. And it was inherently dangerous due to its experimental nature. Remember too that any loss of life would provide considerable ammunition to the anti-space crowd.

Posted by Karl Hallowell at January 19, 2007 04:43 AM

You also cannot discount the fact that Burt's total development cost was split between WhiteKnight and SpaceShipOne. While SpaceShipOne is in the Smithsonian, White Knight is still flying commercial services for NASA contracts. So some portion of that $30 million gets amortized over all of those White Knight flights as well. I suspect that once you look at those flights the marginal starts getting reasonably close to the average.

I run numbers like this about once a week. One of the major components of marginal cost is insurance and range fees. If we can work to get those down then fuel and salaries really do become the overriding cost.

Posted by Michael Mealling at January 19, 2007 07:34 AM

You aren't being fair to Mr. Allen. The SS1 doesn't have the proper space for a legitimate tourist vehicle. All the designs I've seen have room for at least six people.
Well, maybe im not fair but thats the situation. The six people argument doesnt tell anything actually. Like John Carmack has said for instance, if i had the cash to go up, i'd much prefer to have a privilege of going up with just the pilot and my wife perhaps, but not with a bunch of random other people.
Its going to remain a once in a lifetime experience at that price point, and i really wouldnt like some random bozos along on that journey.
Yes, a craft capable of carrying six still can go up with two on a special "im filthy rich so book the entire ship" occassion, but saying that three-person suborbital isnt practical isnt immediately and obviously true.

As for its safety .. im still left puzzled why didnt they take and improve it and gather some real operational data with it. One possibility is that they really gained everything they needed to know from those few flights, and SS2 design derived from that data will be your ultimate suborbital ship hands down.

Other people take more incremental approaches, learning and readjusting their ideas and designs along the way. ( Masten and Armadillo for example )

Posted by kert at January 19, 2007 07:53 AM

> One possibility is that they really gained everything they needed to know from those few flights, and SS2 design derived from that data will be your ultimate suborbital ship hands down.

A more reasonable possibility is that they learned what they could from SS1 to build SS2 and plan to use what they learn from SS2 together with what they learned from SS1 to build SS3, and so on.

Then again, I'm assuming that they're actually somewhat competent, which is clearly unreasonable given that they've actually built a suborbital craft while kert is, umm, I need help here. Why should I believe kert instead of Rutan?

Posted by Andy Freeman at January 19, 2007 08:46 AM

Well, maybe im not fair but thats the situation. The six people argument doesnt tell anything actually. Like John Carmack has said for instance, if i had the cash to go up, i'd much prefer to have a privilege of going up with just the pilot and my wife perhaps, but not with a bunch of random other people.
Its going to remain a once in a lifetime experience at that price point, and i really wouldnt like some random bozos along on that journey.
Yes, a craft capable of carrying six still can go up with two on a special "im filthy rich so book the entire ship" occassion, but saying that three-person suborbital isnt practical isnt immediately and obviously true.

I have a feeling that, once you get up to "black sky", there isn't going to be much talking going on in the cabin. And, it's not like they're going to put 4 passengers in between you and your wife to keep you from holding hands or anything.

What about the people that would like to take their kids up with them instead of just their spouses? Or a couple that wants to take their last remaining parent up with them for their parent's dream journey? Pilot+2 is very inflexible. Besides, if it weren't for the "Capable of carrying three people" rule for the X-Prize, the SS1 would probably have been a single-seater.

Personally, I wouldn't want to take a paid ride on SS1 without a re-design of the entry/exit door. a) it swings inwards, which Apollo 1 should have taught us is a BAD idea, and 2) it's tiny, which doesn't make things easy for people who aren't used to crawling in and out of things.

Other people take more incremental approaches, learning and readjusting their ideas and designs along the way. ( Masten and Armadillo for example )

It's pretty apparent from a lot of the comments on this topic that almost nobody followed Scaled during SS1 development, or that they seem to have forgotten history. As Ed Wright stated, there were 66 flights of the White Knight, and 17 powered and unpowered flights of SS1 (also detailed here). There WAS incremental testing, which was kind of the point of the whole process. If you don't think that they learned anything from the process, then you obviously didn't watch the second X-Prize flight, in which the vehicle did NOT roll as it did in the Sept. 29th flight. That, to me, says that they learned something from their approach.

Maybe I'm missing something as an "outsider" to the space industry. Can someone tell me whether or not comments like Kert's (chastising a successful program launched by rich guys, and saying it's not as good as incomplete "grass roots" programs) pretty much par for the course in the space community? Is there really THAT MUCH jealous in-fighting? And is really just jealousy disguised as ignorance, or is it true ignorance?

Posted by John Breen III at January 19, 2007 08:57 AM

Maybe im not expressing myself clearly, but im not critizising Rutan for what he does and how he does it. Especially is there no need to "believe me" as i havent made any extraordinary claims here or anything.

I am just wondering why they left an opportunity go unused. They had a craft that flew, it could possibly have seen a lot of different uses, including testing new designs, possibly flying paying experiments and/or passengers or whatever.

They didnt do it for their own reasons, whatever those are. I am trying ( maybe shortsightedly ) guess what those reasons were.

Note that the flights that they did were basically all development flights. I would guess that gaining operational experience, even flying dummy payloads is valuable in its own right. Or maybe it isnt, if the vehicle you are flying is really that experimental and nothing like the operational version.

Posted by kert at January 19, 2007 09:25 AM

>Do you know the meaning of the word "slander," Brian?
Why yes I do Ed. It means to SPEAK malicious lies meant to damage someone’s reputation. I think the word you’re looking for is libel.

And yes I’ve worked on dozens of real rockets. Mostly RCS thrusters but also rocket engines that are used for creating lasers (do you know how that’s done?), but not for Lockmart. In fact I’m holding a part of a real rocket engine in my hand right now. Alas LM doesn’t MAKE anything, they just buy stuff from people that do real engineering and science and ASSEMBLE it.

Posted by brian d at January 19, 2007 12:22 PM

Guess I’ll post my response here since I wasn’t able to post at CONTINUING OBFUSCATION

>if Burt had decided to fly SS1 one more time, do you claim that he therefore would have had to come up with five to ten million dollars more to do so?

No. Then I would claim (correctly) that it cost $20M (if that’s what was spent developing it) + whatever it costs to get it ready to fly again divided by 4 (I think he made 3 powered flights to space, or was it just 2?), anyway you get the idea by now I hope.

You seem to want to take credit for flights that didn’t happen but could have. Well maybe they could have and maybe they couldn’t have. Maybe it would have spun up to 3000 RPM on its next flight and created a smoking hole in the desert. I prefer to only work with what I know. We know that it cost $7M per flight. You guess that it could have made 10 million more flights driving the cost down to $100,002 per flight.

Look what happened when Lockmart guessed that companies would be beating down the doors for launch vehicles.

Let me give you a tip. Concede obvious points and it’ll enhance your credibility. Here’s what you should have said: Yes Brian it’s true that SS1 demonstrated three suborbital flights at $7M a pop but SS1 was a developmental vehicle and the lessons learned and the mistakes made (uncontrolled roll, etc.) will help build a much better and safer SS2 that can be used for commercial purposes. To that one can only say we shall see.

Personally I don’t think there will ever be much of a market for suborbital flights. There IS a market for orbital flights, which brings me back to my original, original point. I have reason to believe that we might eventually be able to reduce costs of orbital flights by as much as 50%, but not much more. I’ve seen A LOT of wasted money (but not enough to reduce costs by say 90%), in fact I could have readers rolling on the floor laughing with what I saw just this morning.

Posted by brian d at January 19, 2007 04:34 PM

>>Do you know the meaning of the word "slander," Brian?
> Why yes I do Ed. It means to SPEAK malicious lies

Not quite. According to the Random House Unabridged Dictionary, it means 1) defamation; calumny; 2) a malicious, false, and defamatory statement or report; or 3) defamation by oral utterance.

Since you refused to answer my question about ethics, I looked up the Lockheed Martin Code of Conduct.

It contains an interesting statement about "disparaging, misrepresenting, or harassing a competitor."

> I prefer to only work with what I know.

As Will Rogers said, the problem is not what you don't know, it's all those things you "know" that ain't so.

> And yes I’ve worked on dozens of real rockets. Mostly RCS thrusters but
> also rocket engines that are used for creating lasers (do you know how
> that’s done?),

Yes, snarky, I do. Do you know how to make an argument based on logic or mathematics, rather than insults and namecalling?

You disparaged "Internet rocketeers" like Elon Musk and Jeff Bezos for building "paper rockets."

You boasted that you had more experience with "real" rockets than they do.

So, if you build an RCS thruster, that's a "real" rocket? Even if you haven't built any other part of the vehicle?

But if Musk and Bezos build RCS thrusters, main propulsion systems, guidance systems, and entire flight vehicles, those aren't "real" rockets???

Do you really think your RCS thruster is more of a "real rocket" than Goddard or Falcon?

You're losing credibility fast, Mr. Dempsey.

Posted by Edward Wright at January 19, 2007 08:38 PM

Cain't cure stupidity, Ed.

Posted by Lee Valentine at January 20, 2007 06:19 PM

Personally I don’t think there will ever be much of a market for suborbital flights.

I'd be embarrassed to say such a thing. I'm not going to be a jerk and suppose that you haven't been to an airport. However, you certainly didn't notice the obvious while there. If I can get to point A to point B faster, than I'm willing to pay for it. That's why people pay money to fly, rather than drive (or in the case of trans-oceanic, cruise). I'll agree that SS1/SS2 are not sub-orbital point to point, but the investors are hoping that once space has become a common destination (like one time the sky was a destination unknown to many) then the costs will come down for point to point sub-orbital flight. Indeed, this was the imagined dream of the Space Shuttle that NASA never followed through on, instead they got stuck in the orbital rut.

Posted by Leland at January 21, 2007 05:21 AM

Marginal cost for SS1 is closer to $300k (see my space review X-15 interview of Rutan) hence their bumping the expected price for SS2 up to $200k/seat from previously conventional numbers of $100k/seat. That's still far less than the marginal cost of X-15 (about 3% in constant dollars of $100k per seat vs. 3 million for X-15 in today's dollars). Expect continued price drops with liquid fuel; stronger, lighter materials; more capital market liquidity; and yes, more demand through games.

Posted by Sam Dinkin at January 22, 2007 11:18 AM

> Edward Wright said
> Mr. Dempsey

That's Dr. Dempsey to you.

You assume you know so much more (about me and other things) than you actually do.

Posted by brian d at January 22, 2007 11:34 AM

Post a comment