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:
The Mailbag

Jerry Pournelle jerryp@jerrypournelle.com
Copyright 2007 Jerry E. Pournelle, Ph.D.

October 29, 2007

We begin with two corrections to last week's column:

Jared Tracy of Alexander Innovation Wizard comments on the review of the ECO LED Flashlight Pull-To-Charge:

Only problem is that alexiw.com does not give customers an option to make a purchase. It should be gadgetuniverse.com. I knew the confusion would start taking place with the AIW brand.

And indeed it is confusing. I had the products and the press release sheets, and still I found the wrong web site. It remains an excellent flashlight for emergency preparedness. Recommended.

Subject: Plantronics link

Hi Jerry,

First let me say I love your books and columns, have for some time.

I thought you'd like to know, the link which takes us to the Plantronics DSP500 headset from your October 11th column requests a compatibility check before showing any purchase info. I put in Windows and USB and it said that my system was NOT compatible and then showed me a bunch of other headsets. Is the link in error or is Plantronics making you look bad? :-)

Thanks and keep up the great work,


Not sure what happened to you. I found this link works just fine. I usually link to the product manufacturer, not to a vendor. I figure that readers can get the specifications: you're sort on your own about finding where to buy it, although I'll tell you if I know about a great buy or a very reliable source.

I really like the Plantronics DSP 500.

We have both mail and advisor comments on Net Neutrality, and there will be a Comcast discussion in the last segment of the October column.

Hello Mr. Pournelle,

You wrote, "I remain opposed to Congressional intervention in a free market". I agree with that principle, however is that the case with cable? Sure, you can go with DSL, but that's usually slower, and fiber isn't available in many places yet. I happen to live in an area where there's an alternative cable provider (Wideopenwest in the Detroit area) but as I understand, that's still pretty uncommon. I'm not familiar with all the details, but local municipalities have given the cable companies government sanctioned monopolies for a given area. Most people don't have a realistic alternative to their single cable provider for true high-speed internet access.

My solution, rather than regulating network neutrality would be to remove the monopoly and require the cable companies to allow competitor access to their networks. They could even trade, Comcast lets Cox sign up customers in Detroit in exchange for access to the Atlanta (no idea if that's Cox, but you get the point) customers.

Of course, this is probably technically infeasible, or at least the cablecos will say it is, but we really need a true open market.

Scott Bailey

Competition doesn't have to be direct. One of our problems has been local monopolies. In some cases those were granted to induce the cable companies to string communications - in those days generally coax - and of course any time you have monopolies granted by politicians and political bodies, you will get lobbyists skating closer and closer to the edge of bribery.

Technology has moved us past those necessities. Today the new "cables" whether for telephone companies or "cable" companies tend to be fiber and can carry any kind of signal. The Telcos have a lot of copper out there and some keep hoping to get a few more years of profit out of that sunk cost by using various new technologies, but the world is going over to fiber everywhere including that last 100 feet to individual households.

In general, the less regulation the more competition. Of course there are exceptions. Adam Smith said it well: when two capitalists get together they conspire to bring in the government to restrict entry into their business. This often takes the form of regulations that requires considerable effort and experience for compliance. The result is that startup companies face great initial expenses.

I would not myself be in favor of adding regulations and price controls, which would be the effect of what you suggest. I'd rather see fewer regulations. We are entirely in agreement on the need for more competition. I suspect that absent a lot of government interference, technology will force that.

Dan Spisak said:

Some information regarding the ComCast BitTorrent throttling story, with some technical information that looks to confirm exactly what Comcast is doing to Bittorrent users. The short version seems to be that Comcast allows Bittorrent traffic to come into its network from non-Comcast users but blocks Comcast users from seeding torrents to users outside of Comcasts network.

You can read the info here:

How To Bypass Comcast's BitTorrent Throttling

Comcast Wrongfully Denies Interfering with BitTorrent

So apparently Comcast uses a Sandvine packetshaping appliance which you can read about at this link.

-Dan S.

Robert Bruce Thompson said

Subject: More on Comcast blocking

They've apparently added Lotus Notes to their blocking list.

Ars Technica article...

If I were a Comcast customer, I'd be [unhappy]. If I were one of the cities that has franchise agreements with Comcast, I might be talking to them about pulling the franchise agreement if they don't stop this horsematite.

-- Robert Bruce Thompson

My view remains that we don't need more regulation, but I will applaud the debate: it may scare the service providers into doing that they know they ought to be doing anyway. And I still believe that truth in advertising will be entirely sufficient - and a great deal easier to enforce.

On Digital Rights Management. Last week I said

"The goal here is not protection from casual copying and sharing, but from commercial exploitation."

Bob Thompson commented:

As I've said repeatedly, DRM is doomed to fail for either purpose. As far as commercial exploitation, DRM has to be completely and utterly uncrackable, or it will be cracked by those seeking commercial advantage.

Those who talk about plugging the "analog hole" are whistling past the graveyard. Human eyes and ears are analog, and there's no way to plug that hole. At some point, regardless of the DRM scheme, the data must be presented in analog form. When that happens, that material becomes freely copyable, and there's nothing any DRM scheme can do to prevent that.

I don't believe uncrackable DRM is possible, even in theory, but even if we assume that it is possible, how Draconian will it have to be to achieve its goal? Enough so to make the products unusable for ordinary people, which by Santayana's redoubling rule entirely defeats the original purpose of the DRM.

And as to casual copying and sharing, remember that DRM has to win every time while crackers have to win only once. If you protect a movie with DRM that's 99.9999999% perfect, someone somewhere is going to crack the DRM on that movie, even if only by re-recording the audio and video. The de-DRM'd movie ends up on the Internet, where anyone can download it.

Robert Bruce Thompson

I do not think DRM has to be perfect. It does need to be invisible to individual users. Given that, the goal is to put in a primary hamper for pirates: DRM needs to cause them enough effort that it's obvious that they went to some trouble to circumvent it. In law, a conspiracy requires that you collude to commit some unlawful act, then at least one of you commits at least one overt act in furtherance of that illegal goal.

And as I have said often, what's important to me is that if someone seeks an electronic copy of a copyrighted work, the first hit from the search engines ought to be a legitimate copy offered for sale, not a pirate edition for free.

Dr. Mark Huth comments on ebooks:

Subject: Ebook Readers


Saw one of these the other day and think it's the reader you've been talking about. I'd buy it and I'm addicted to books. I'm seeing what's available in digital books and it looks as though that may hold me back. This connects via USB and does not use a closed format.

Sony product link

Peter Glaskowsky had the older Sony Reader, and now has the new one. Here is his report:

I did get a new Sony Reader, and my review is here:


and here:


In summary, the PRS-505 is a small but significant improvement over the older PRS-500. I would not upgrade if I had a PRS-500, but anyone who doesn't have an ebook reader and wants one should look at the PRS-505.

There are a couple other ebook readers to consider, though. The Bookeen Cybook Gen3 is much like the Sony Reader:


The iRex iLiad is a larger device using the same kind of electronic paper display (8.1" vs. 6"), has a touch screen that allows the user to make notes, drawings, etc. on existing documents (or blank "paper"), and supports WiFi... but it's $699 vs. $299 for the Sony.


There's also the forthcoming Amazon ebook reader code-named Kindle. According to photos in an Amazon FCC filing, Kindle is (or was) huge and ugly. I suspect that's why we didn't see it earlier this year; Amazon is probably re-doing the industrial design.

I think the PRS-505 is the best overall ebook reader on the market today, especially when considering price/performance and software support, but I'm sure it'll look primitive and awkward compared with ebook readers we'll see in just two or three more years... and by that time, ebook distribution should be greatly improved as well. For those who can afford $300 to try out an ebook reader, the Sony is the best choice, but most people should probably wait.

. png

Ebook readers still cost too much, and they're one more thing to carry. I believe that the real break in electronic book publishing will come when millions of people routinely carry a telephone that is Good Enough as an ebook reader. Another iteration of the iPhone may do that.

More Vista Problems

Jerry, you wrote "Alas, Vista twice blew things up for me. First, it dropped me and flashed a message saying it was all Skype's fault. When I got reconnected, it suddenly popped up a message that my microphone was muted. I could find no setting whatever to un-mute that microphone. Eventually I shut down Vista and restarted, and this time everything worked just fine. I find Vista has this similarity to some of the older versions of Windows: when in doubt, reboot. It often helps. "

I've had similar experience with Vista [it came preloaded on my new laptop, I would never consciously install it on a computer of mine]. I use the laptop with Adobe OnLocation software as a video recording device. Twice now, I have lowered the system sound to 0, in effect muting the built-in speakers. Both times, when I later went to restore the sound level, it not only wouldn't but it indicated that there was no sound driver installed. Opening up the control panel confirmed this. Strangely enough, rebooting brought the "not installed" sound driver back from the dead. So what you have told me three times is certainly so: Vista is not yet ready for primetime, and rebooting often helps.



It's pretty clear that Vista has "memory leaks" - or at least that it has problems indistinguishable from what we called memory leak problems in the old days. Just this week I had a number of windows open in Vista. I closed several of them, opened Word 2003, and watched the machine hang when I tried to bring in a photograph. The problem came when I needed to browse to the photo I wanted. The remedy was to restart the machine: the same operation worked just fine after that.

If you run Vista, restarting daily seems indicated. Didn't we get accustomed to doing that with versions of Windows prior to NT?

On Net Applications and Patents


Pamela Jones over at Groklaw is already in the process of looking into the NetApp patents, and her gang have already surfaced significant quantities of prior art, going back to 1980, at least.

Groklaw article

Experience shows that virtually all software patents are bogus. The "inventions" were described in the public press, many years before the patent application. The process by which patents are granted is substantially the same as that for mail-order diplomas. Or perhaps a more apt comparison might be this: there is a certain type of crooked lawyer who works with a crooked clerk in the county tax office to buy people's houses out from under them for $20. You know the procedure: create a small tax bill, deliberately conceal it from the homeowner, and sell the house for taxes in a "public action" which no one knows about.


The new KSR vs. Teleflex test for patent obviousness is only beginning to be established. The patent office has issued regulations pursuant to the Supreme Court decision, but it remains to be seen whether the patent office will actually follow these regulations in large numbers of patent re-examinations. If it does, then virtually all of the disputed software patents will be canceled. The unavoidable fact is that the great age of invention in computer software was approximately from the 1950's to the 1970's. Someone who claims to have a controlling patent which is still in force invariably got it by much the same means as the crooked lawyer who assumes ownership of Chaos Manor for $20.

Andrew D. Todd

An interesting development worth following. I have always thought that a number of patents were issued for software techniques and "business practices" that look obvious to me. I won't go quite as far as Richard M. Stallman in denying all software patents, but I certainly believe that far too many have been issued. In some cases the companies applied for the patents for self protection: if I hold a patent, even if I have no intention of enforcing a monopoly on the practice, you can't stop me from using the technique.

Unfortunately, some legal advisors are telling corporate boards that they must try to enforce their patents because they have a fiduciary responsibility to be squeezing hands at the grindstone. I am reminded of the lawyers who told the Board of the US Naval Institute that they had a fiduciary obligation to try to steal Tom Clancy's characters and hold them for ransom against his future book sales, because USNI was the first publisher of Hunt for Red October. Many of the naval officers on the Board at USNI considered this dishonorable and several resigned.

We'll keep watching these developments. Thanks.

Peter Glaskowsky, who follows these matters closer than I do, had a somewhat stronger opinion:

This is an astoundingly ignorant thing to say. It amounts to implying that nobody has ever invented anything in software, when we all know that's not true. Todd's statement is the sort of thing people say when they're trying to avoid actually understanding or discussing the situation.

Sure, there are a lot of bogus software patents. There are a lot of bogus patents in every field. But the patents that get litigated are ALWAYS the ones that are most debatable. Patents that are obviously valid are respected; patents that are obviously bogus never get into court.

. png

I agree that there are valid patents. I am also quite certain that many of them should never have been issued. Fortunately, the software lobbyists haven't been nearly as effective as the Disney Corporate Lawyers, and Software Patent Life hasn't been extended to 90 years...

There is additional discussion of this matter in the final part of the October column.