Dr. Jerry Pournelle

Email Me

Why not subscribe now?

Chaos Manor Subscribe Now

Useful Link(s)...


Hosting by

Powered by Apache

Computing At Chaos Manor:
September 23, 2008

The User's Column, September, 2008
Column 338, Part 2
Jerry Pournelle jerryp@jerrypournelle.com
Copyright 2008 Jerry E. Pournelle, Ph.D.

In the June issue of the column I recommended The Black Swan to anyone interested in statistics and financial security. I've mentioned the book and the concept many times over the years, latest at this link. If you haven't paid attention, you may want to do so now. The financial crisis isn't over.

In the past month, investment banks managed to lose more money than all the banks in history cumulatively have made as profit on risky investments. Lehman Brothers, which survived the Civil War, the turbulence of the late 19th Century, World War I, the Crash of '29 and the Great Depression, has vanished. Major stockbroker houses, which used to know that the first rule was never to gamble with house money, collapsed. Note that the crisis was easily predicted by anyone except the smartest guys in the room who controlled America's financial centers. They, apparently, couldn't see it coming.

It is time to lay out, in very plain language, precisely what happened; and while I do not usually make this kind of comment here (as opposed to my comments in The View From Chaos Manor), this crisis greatly shrinks our pool of investment capital and thus affects all our high tech industries; and the smartest people in the room — and this is common to the leaders of both major political parties — either do not understand that what happened was predictable, or pretend not to.

As often (but not always) happens when government interferes with economics, the origin of the collapse was due to good intentions. Aristotle tells us that democracy is rule by the middle class, and the middle class are those who possess the goods of fortune in moderation. Most political scientists, economists, and intellectuals in general are agreed that ownership of one's home is a key element in defining the middle class in America. It's neither necessary nor sufficient, but it's still important, and the more people who own homes, the more people with a stake in America and the existing social order. People who own houses work hard to keep them. Home ownership is good for the owners, and it's good for America.

It was decided that the usual market forces weren't sufficient to spread the joys and benefits of home ownership, and government help would be needed. There are several ways that might be done including taxing the rich, buying houses, and giving them to those who couldn't afford them, and that was actually proposed at one time; but sanity prevailed. After all, if you could get a free house by being poor, why work hard to save a down payment and take out a mortgage? Better to spend yourself poor...

Eventually a solution was found. Private corporations called Fannie Mae and Freddie Mac were established with government backing, and by implication, government guarantees. These firms, managed by executives paid on the modern scale (millions to hundreds of millions in 'incentives' and bonuses), were sent forth to make it easier to own a house by making it easier to get a mortgage loan on that house.

This meant that people who couldn't convince bank loan officers that loaning them money would be profitable were able to borrow money because the government guaranteed the repayment. Fannie Mae and Freddie Mac could obtain money at low interest from the Federal Reserve System. Of course the Feds don't have any money — the budget runs a deficit — but it's easy to print some and borrow more by selling Treasury bonds, so Fannie Mae and Freddie Mac were well financed at low interest. Given those loan guarantees to lenders, people who otherwise would not be able to buy a house were enabled to do so. And of course as more people were able to enter the housing market, more bids were made on houses, and housing prices rose. As prices rose, the construction industry was stimulated to build more houses. Employment went up. For a while everything worked about as hoped.

It couldn't last, of course. Whatever the intentions of the people at the top, the loan salesmen — you could hardly call them loan officers — got commissions for pushing loans, and of course some of them began pushing loans to people who had no chance of paying them off. That didn't matter because as housing prices rose the buyer could borrow more, or sell the house for a profit.

As long as prices went steadily up, all was well. Of course the housing inflation kept many people out of the housing market, and forced others to buy in locations far from where they worked, which was not precisely what Fannie Mae and Freddie Mac were intended to do; and the inflation got out of hand. In my community middle class houses that sold for $30,000 in 1968 were suddenly "worth" a million dollars and more for essentially the same house.

Housing prices rose steadily. Fortunes were made selling kits on how to get rich speculating in real estate. People took out loans on their new "equity". As long as there was easy mortgage money, the bubble expanded. It couldn't last, and many saw that coming: but Fannie Mae and Freddie Mac were private companies, and as private companies were permitted to form Political Action Committees to make campaign donations; and they were able to lobby. Lobby they did: they threw parties for Congressional staffers, held "education sessions" indistinguishable from parties, and did all the other things that lobbyists do to woo Congress, and what they lobbied against was any restriction on either their lobbying or their money lending practices. In fact, Congress repeatedly gave Fan and Fred new loan powers, and encouraged them to create even more home owners. This was done with the best of intentions — everyone was careful to say so — but the result was that the bubble kept growing.

Note that this was not deregulation. The Community Reinvestment Act of 1977 was amended in 1995 to increase regulatory supervision to guarantee that CRA loans were not confined to high and middle income borrowers, and that more sub-prime loans would be made. For the history of the Act and its modifications see this Wikipedia link or use Google to find your own sources. And the bubble grew and grew.

So far we have a foreseeable crisis in housing and construction, serious enough, but still, nothing that could bring down the whole financial system; but at that point the smartest people in the room got really creative. They realized they had a lot of assets in real estate. As the bubble expanded these assets were valued higher than the mortgages held on them. They were, in a word, really first class securities. Perhaps some of them were a bit shaky — after all, when you loan $500,000 to a $40,000 a year gardener so that he can buy a house with a thousand dollars down and interest only payments, there's a chance he's going to default when it comes time to pay the principal — but many of them were solid as a rock. The obvious remedy was to bundle the gardener's loan with a nearly paid off loan by a college professor whose house had appreciated like mad over the years, and sell that package as a "mortgage guaranteed" loan to an investment bank. Now the loan company has even more money to loan out on even more questionable sub-prime deals.

At some point the investment banks began to realize that they'd bought a pig in a poke, and got creative by selling bundled packages of "mortgage guaranteed loans" to retirement funds, other investment banks, and anyone else they could find. After all, a mortgage guaranteed loan was as safe as a Treasury bond, and paid higher interest. Great investment, with no risk.

They sold a lot of those bundled mortgage guaranteed loans; and that is how the entire financial system was endangered. To learn how many bankers were caught by surprise because they didn't look very hard or didn't want to look, see "How Wall Street Lied to Its Computers" by Saul Hansell. Note that some of the lies were deliberately ordered by management who wanted the bubble to continue (and who departed their ruined companies with enormous bonuses). Given incorrect models to work with, the computers continued to forecast profits right up to the crash. For a cartoon summary of the above, see this Sheldon Comics strip.

So: that's where we are. As to what can be done, it may not matter. That is, it's important what we do, but the chance that it will be done sanely and rationally is very small. What will be done must be decided by the most unpopular Administration in nearly a century in connection with the most unpopular Congress in history; and everyone involved in finding a remedy was in one way or another a part of creating the mess. By everyone, I mean everyone: the Administration, the Treasury, the Congress under Carter and Clinton, Congress under Reagan and Bush, Congress controlled by both Democrats and Republicans, the regulatory agencies, and the "experts" now out of jobs who will be hired to manage the new institutions that will be set up to buy bad debts: every one of them. What will be done will be settled by politics, not by economics.

I say this because those who did foresee this disaster tried repeatedly to rein in Freddie Mac and Fannie Mae, but the Fred and Fan lobbyists were easily able to defeat those efforts. Moreover the leaders of Fred and Fan were fired, but left with multi-million dollar bonuses, as did the leaders of various firms ruined in the disaster. The remedies being proposed aren't going to do much more than create a bureaucracy. Once that happens, Pournelle's Iron Law of Bureaucracy will take over, and whatever is required to keep that bureaucracy healthy will be done. One thing is certain: the people who must pay for this debacle will largely be those who took out sensible loans and have kept up their mortgage payments; those who did nothing wrong, but will be handed the bill. Depend on it. But however it comes out, and whatever we do, there are some inevitable consequences of this debacle, and we need to be concerned.

What Happens Next?

Some immediate effects are obvious. First, there will be inflation as the government creates money to buy up bad assets (which will eventually be sold at fire sale prices, in many cases to those who bought the properties with inadequate resources, but mostly to speculators). Inflation is a tax on fixed incomes, savings accounts, and conservative investors holding what used to be called blue chip assets. Those on pensions will be hit twice: the money they get won't go as far because of rising prices, while the amounts they get will fall as the capital assets generating their pensions lose value.

Second, as government deficits rise, investment in science and research will be cut. That's been happening over the past few years anyway. Only about 15% of grant proposals are funded now. Of course that's not entirely a bad thing; back in the days of "As long as you're up, get me a grant" we funded some very silly projects. Alas, there are some good projects not getting funds either. Basic research is risky, and has little direct payoff; but investment in technologies implied by new technologies changes the world.

I am aware that some economists including my friend David Friedman argue that government ought not invest in any scientific research, but that's another discussion. David and I aren't agreed on government investment. Alas, there's no doubt that the effect of cutting research investments will reverberate through our high tech industries for some time to come. We will all be affected.

Third, there's going to be considerably less money for both external and internal investment in the high tech industries. Fortunately we've just been through a huge growth in hardware capabilities, so there are developments hidden away in back rooms, and our hardware capabilities way outstrip our software's ability to make use of them. One might or might not make some useful inferences about the future from that.

All of which is to say there are hard times ahead. Not as hard as they might have been, and particularly not for the kind of people who read this column. We all of us have capabilities few had during the Great Depression. We all of us have some familiarity with applying high tech to everyday problems, and the ability to tool up to do more of that. Most of us once learned how to write programs, and most of us have a lot of high tech equipment with far more power than earlier software developers ever had.

I once said that the computer revolution would really take hold when it was more important to know how to do something that a computer might do than to be able to teach the computer to do it. I suspect that's even more true now.

The world won't come to an end, but now would be a good time to take stock of one's resources and decide which ones ought to be developed. This inventory should look at everything: from vegetable gardens to software development. I was once an editor of Survive Magazine; this was back in the days when there was a small but real probability that civilization would end with a bang. We now have a small but non-zero probability that it will end with a whimper. We have a much larger probability that it won't end, but there may well be frightening dramatic changes.

Look out for Black Swans. And you might go read The Gods of the Copybook Headings.

Emily Is On Line

Emily is our latest system built on the Intel Core 2 Extreme Quad chip, and Intel Extreme DX48BT2 motherboard. The cooling system is a Zalman CNPS9700 Ultra Quiet CPU Cooler. The case is an Antec Titan550. Emily uses a WD one terabyte hard drive, and 4 GB of Kingston KHX12800D3K2/4G Hyper memory. The DVD drive is a Plextor 800A DVD R/W.

Many Intel Extreme systems are sold to gamers, but Emily is intended as an enterprise system. The other frequent users of Extreme boards are high end users who have heavy video performance needs. The DX48BT2 board has dual PCI-e slots to allow dual video boards. Video editing performance on the Extreme won't be up to a full XEON workstation, but for many it will be good enough at far lower costs (particularly memory costs).

In our case we didn't install the very latest and greatest in video boards. Instead, Emily's video board is a Diamond All-In-Wonder HD Premium (as in High Definition) based on the ATI HD3650 video chip. The board has built in on-board HD TV tuner and a number of other video editing features. For a more technical review see this link.

I wasn't feeling quite up to par, so I had Eric Pobirs build the system. Eric's original name for the system was Bones, but when he installed Vista Ultimate 32-bit he used the placeholder name of BonetrailBetty. Note that the Intel portions of this system fall under the Intel Bonetrail platform spec. When I transferred the system from a workgroup to my Chaosmanor Domain, she was renamed Emily after the actress Emily Deschanel, AKA Dr. Temperance Brenner of the Bones TV show. It seemed appropriate for a beautiful machine that's the smartest and fastest system in the house.

Eric Pobirs sends these notes on the video board:

The All-In-Wonder (AIW, henceforth) comes with its own extensive suite of apps to exploit its capabilities. Those of us who recall the original AIW board in the 90s might regard ATI as the true inventor of the DVR. They let you put a TV tuner in your PC and record TV shows using a channel guide a good year or more before TiVo and RePlayTV (then known as the Panasonic ShowStopper) hit retail shelves. But I never got around to installing the AIW's application suite. I was still intent on getting all of the drivers in order when I became curious about the USB Infra-Red receiver dongle in the accessory pack. Would Vista know how to use it without the ATI software?

Vista did indeed know. The bundled IR receiver is supported by a native Vista driver and the remote control from Diamond included in the package is fully compliant with the spec for Media Center, allowing both that and the ATI suite to be used. I didn't realize at first how far the support went. I pressed the power button on the remote and Emily immediately went into hibernation. After rousing her, I found I could easily invoke Media Center by pressing the logo button on the remote and play a video test file I'd already copied to the Video folder. Some parties whine about having all sorts of latent functions in an OS but the way this just worked was really a pleasure.

For those desiring extensive TV manipulation capabilities and fairly good gaming 3D in a single slot (the optional input bracket uses an external slot panel but not an actual bus slot) the Diamond All-In-Wonder HD looks to be my first recommendation. This kills many birds with a single stone.

I do note that Hauppauge Computer Works has WinTV, which works; we reviewed it here. I have used it with my laptop, but that computer isn't fast enough to do HD TV even with strong signal strength. Since it works in HD with Roxanne, a Core 2 Duo with Vista 32, it certainly would work with Emily.

One more note: when Emily was first assembled, she ran hot: by hot, I mean HOT, as in the case was uncomfortably warm to the touch. This was in the early stages after installing Vista and doing the Microsoft Updates. Fortunately Windows Problem Solver knew about some Intel update drivers. We then went to the Intel site and obtained the updates for the Intel Extreme board. Once those were installed the system cooled down, and doesn't get hot even when doing a highly complex battle in Medieval Total War at highest resolution.

Emily has been running for days, but she became fully functional when she was added to the Chaosmanor domain only a few hours ago, so this is a very preliminary report; we'll have a lot more detail next month.

The preliminary report is: Wow! She does everything beautifully and very very fast. As usual I have been fighting Vista to get the security features properly configured so that the system is safe without pestering me to death, but that hasn't really been a problem; and other than that I have to say she has performed beautifully. She plays movies, including movies stored on the hard drive, and an HD movie I picked up off the air with a TV antenna: it played well enough even though Chaos Manor is in a nearly dead TV reception zone so there was frequent interference from everything including traffic outside. Emily plays Medieval Total War II in highest detail and very fast: it's beautiful. Disk operations are extremely speedy. If I sound enthusiastic, I am: this really is the fastest machine in the house.

More next month, but I'm really happy with this system.

iStore, iPhone, and Mac Apps

There are now new third party applications for the iPhone. They're available from the Apple iStore (actually the App Store within the iTunes Store on the Apple site). My first experience with buying and installing those applications wasn't pleasant, but that's largely because I did that at 0400 after a night of insomnia, and I forgot the "Mac Way" of doing things.

That is, use the iMac 20 to download an application. In the process of doing that I had to give my mac.com user name and password, so I presumed that iTunes Store would now know those. Then I connected the iPhone to the iMac. Synchronization began — and up popped a message on the iMac 20 saying that the application "could not be installed on iPhone "iPhone" because you are not authorized for it on this computer." This was frustrating in the extreme, and I said so angrily in my day book. A number of readers took me to task for that, because the solution is easy enough, and not difficult to find with Google, although of course I also got the remedy from just a lot of users, along with some commiseration about the opacity of Apple error messages.

The remedy is to go to the menu at the top of the screen, select Store, and on that drop down menu select "authorize computer". Then enter the mac.com user name and password (yeah, I'd already done that, but that was just to set up an account, not to authorize I guess). When I did that, all my iPhone apps instantly installed on my iPhone.


My first iPhone app was a book reader called "Stanza." It's free, it installs easily, and it makes it very easy to read books on an iPhone. One free book is Clausewitz's classic "On War." I downloaded that one and I'm reading it again in small snatches at odd times.

There are a lot of other books offered at various prices. You get them from the App Store, in about the same way that you get books for the Sony Reader; buy through the store (or download for free), then connect the iPhone with its cable. After that it's automatic. For me at least, on the iMac 20 the App Store is pretty slow: it takes about 40 seconds to go from the App Store home page to the Applications section to the "Get more applications" screen. It takes another 20 seconds to get to the page where books are listed, and about that long to go to another page of books. This is a bit tedious compared to getting books on the Kindle, but it's only tedious, not prohibitive.

For those with younger eyes than mine, the iPhone may well be a Good Enough reader. In my case I have to make the text larger, which means fewer words on the page, which means turning pages a lot; and while that's fast, it get tiresome. The upshot is that I really prefer the Kindle, particularly for books with notes (the Kindle handles notes very well). On the other hand, I almost always have the iPhone with me, and I do not always have the Kindle. They won't replace the Kindle, but I expect to be using the iPhone and Stanza a lot in future. Recommended.


This is a preliminary report: I love it. YouNote is an application for iPhone that allows the iPhone to be used as a note taking device: it allows voice recording, notes typed in, and even scribbled handwritten notes.

I've only been using it for a couple of days, so I can't be sure, but I like this enough to recommend that you check it out: it's cheap enough. YouNote isn't a TabletPC with Microsoft OneNote (which I consider the ultimate in research tools for a writer on the go) but it's sure good; enough so that I probably will stop carrying the Olympus WS-100 voice recorder.

More another time, but I'm impressed.

iPhone Notes

At breakfast the other day my friend Joan mentioned that her daughter in law had been a physician in a clinic in Hollister, but now was working in a clinic in Watsonville. There followed a discussion of where Watsonville is. Out came my iPhone.

My iPhone map knew precisely where I was, even though I don't have the G3 with GPS, so it showed me a map focused on the 4'N 20 Pies restaurant in Studio City. I was able to use gestures to zoom out — it takes a bit of learning to do it right, but it's easy enough to learn — and then center on Watsonville (which is on Monterrey Bay about halfway between Monterrey to Santa Cruz).

I find I use the iPhone more and more. It isn't yet the pocket computer I first described in The Mote in God's Eye in 1972, but it's getting there.

The Three Coffins

I was talking with John Dvorak the other day, and he mentioned that he was thinking of doing a fiction novel. I wished him luck (and I mean that), and we talked a bit about markets.

One market that has sort of vanished is the classic mystery detective story that used to be enormously popular. Those have been replaced by the first person female detective/adventure story generally written by women. There are usually a couple of those on the best seller list every week. They're all right, but I rather miss the older kind of intellectual puzzle story; which is how John Dickson Carr's The Three Coffins came up in the conversation.

This is the classic locked room mystery, and it has never been bettered. Moreover, during the story the detective hero Dr. Gideon Fell (modeled on the author G. K. Chesterton) gives a classic lecture on the requirements for a good locked room mystery. I thought about this after John and I finished our telephone conversation, and went out looking for a copy of Carr's book; I was pretty sure I had read it a long time ago, but if so I have forgotten the details.

I got a copy of The Three Coffins through Amazon; it's part of a three book Carr collection called Locked Rooms published by the Mystery Guild, and I think it cost me about eight bucks including shipping and handling. I've now reread The Three Coffins, and it's a hoot. It takes getting used to the style, which is quite different not only from today's detective novels but from the classic detective stories by Chandler and Rex Stout. It takes place in pre-WWII England, when people who had lived in England for decades were still called "foreigners", and Bloomsbury was a well known district near the British Museum. The descriptions are good, and it's very readable. I'm glad I found it. I haven't seen a Kindle edition, though. Recommended.