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:
November 9, 2009

The User's Column, November, 2009
Column 352
Jerry Pournelle jerryp@jerrypournelle.com
Copyright 2009 Jerry E. Pournelle, Ph.D.

Net Neutrality

The debate on net neutrality continues, but it's not obvious to me just what is being debated: everyone seems to have a different idea of what net neutrality ought to be, and which regulations are needed to accomplish it. Meanwhile, I haven't seen much in the way of specific claims to practices that ought to be stopped; most of the debate I see is speculative.

Much of the debate is summarized in a pair of op ed columns in the Wall Street Journal on October 30. The first, Net Neutrality: Spur to Entrepreneurship . . . by Mitchell Baker and John Lilly of Mozilla says that FCC intervention is needed, although the example they give, the success of Mozilla, would indicate that the problem they describe can't be all that serious. Mozilla has grown and thrived without government action. Of course that was the wired Internet, and Baker and Lilley are arguing for regulations of the wireless network. The case they make is a mostly theoretical defense of keeping the burdens low for new firms entering the market. I certainly believe that ease of entry into the market is a good thing, but I don't believe they made the case that government action is required to achieve that. I point out that Adam Smith warned that capitalists usually conspire to persuade government to regulate their industry in order to prevent newcomers from entering...

The counter case was made on the same page of the Journal by Republican Senators Hatch and DeMint in . . . Or Barrier to Broadband Investment? Their argument is that things are going splendidly now. They cite as evidence the success of iPhone and "there's an app for that," which is in my judgment pretty convincing. It ain't broke, and government attempts to fix it aren't needed.

When I look for specific complaints that do need fixing, the most significant one I find is the charge that some Internet Service Providers advertise unlimited high bandwidth, but if a customer uses too much bandwidth, they cut back the access rate. I don't know how widespread this practice is, but surely we don't need new Rules and a new bureaucracy to prevent it? It's a plain case of untruth in advertising, and we already have means for dealing with that.

Of course some advocates of "net neutrality" mean that users cannot - or should not - be charged more for higher bandwidth use. Many "consumer advocates" are hoping for some kind of regulation that will let them have all the bandwidth they want for the same flat fee that Aunt Minnie pays for access to her email and pictures of her nieces. Of course that kind of regulation is not fair, and I would not have thought that any serious regulators have anything like that in mind when they consider "net neutrality", but it's often the motivation for people who are "for net neutrality". FCC Chairman Julius Genachowski has made ambiguous remarks on the subject. I'd feel better if he just plain ruled out that "rule." There's no questioning that unlimited access at a low flat rate would stimulate more Internet use, but that coupled with "packet equality" would soon swamp net capacity.

Another set of advocates suppose competitive discrimination. A recent letter to the editor of the Wall Street Journal decries "the reality of bureaucrats at broadcast companies enforcing their contracts with, say, Microsoft to slow downloads of the next revolutionary browser or email application. Imagine a world in which Internet Explorer was still the users' only option for exploring the Web." No examples of such contracts were given, and I find it hard to imagine them, but presumably the letter writer was serious.

Another threat that net neutrality advocates want dealt with is the "use my pipes" argument, which was an ill considered notion that since most ISP users connect to Google, then Google ought to pay for use of those channels. This neglects the fact that the user has already paid for the bandwidth use. The issue was made nugatory by a consent agreement by AT&T in their last major acquisition.

The last major argument for regulative "net neutrality" is the principle of competition: that government ought to dictate the rate at which the owners of the "pipes" must charge those who want to compete with the pipe owner. If some small local ISP wants to provide access service to a list of subscribers but can only use AT&T or the local cable company for its own access to the Internet, the pipe owner must rent the access at some "reasonable" rate. This one is a bit more complex. There are principles involved, including the expertise of the rate setter: who knows what that rate "ought" to be? Is it set to help the users or to provide incentive for the pipe owner to build new pipes? Is there a monopoly? Are there wireless alternatives? And so forth.

This latter might be worth discussion. My proclivity is against Federal regulation unless there's a clearly established need, but I'll listen to arguments, particularly in the cases where the pipe owner has a monopoly. Do note, though, that this isn't what's usually meant by the advocates of net neutrality.

My conclusion is that the phrase "net neutrality" no longer has much meaning, and in any event those who favor regulations haven't made the case for their necessity. Leaving service providers free to try different pricing schemes should be the default; if real problems arise, it's a lot easier to start regulating than to deregulate once there's a dependent bureaucracy.

Windows Security Essentials

I now have several weeks of experience with Microsoft Windows Security Essentials. It is running on four Windows 7 (formerly Vista) systems including Roberta's, and two XP systems. So far it has worked just fine to protect those systems.

There is a problem. The Windows Security Essentials controls allow you to schedule updates, but there is no "Update Now" command. Instead you give it a scheduled update time. That works just fine - unless the computer is a laptop and isn't turned on at the scheduled time. In that case, the next time you turn the system on, you get warnings about your computer at risk because your virus protection program isn't up to date, and you can go nuts looking for a way to update it. So far as I know, you won't find one. Fortunately it's nothing to worry about; the story has a happy ending.

When I recently turned on Orlando, the t42p ThinkPad running XP, I got the warning. I was also notified that there were operating system updates available, and I ought to install them. I did the installation and a reset. When the machine came back up, Windows Security Essentials was telling me that I hadn't scanned the system in a while. That was true enough: although the system was set for automatic scans, it's a laptop and was asleep at the scheduled time. This was easily fixed, so I told it to do a Quick Scan. That took about 15 minutes, after which that warning turned off.

WSE was still giving me the warning about my virus detection software not necessarily being up to date. Searches on both Bing and Google found no place where I could simply update WSE, nor could I find it in any control I could access. I did have the alternative to reschedule the updates, and I seriously thought of simply telling it to do it at the next hour (the choices of times are always on the hour). Just for the heck of it, I went to the general Microsoft Update site, and told it to scan my system. It took it longer than I thought it ought to, but eventually it told me there were no critical updates available. Now, though, Windows Security Essentials was no longer warning me about my virus detection software needing an update. Apparently the active-x program that does the scanning reset that particular flag in WSE.

Alas, WSE was now telling me that I hadn't scanned my system in some time - although I had certainly just done that. The simplest solution to that problem was to tell it to scan again, and I did, at which point all the warnings were gone and Windows Security Essentials was happily telling me all was well.

All these are annoyances, not fatal flaws. There are glitches in the internal communications in the programs - if you can get the Windows Security Essentials control screen it has an "update" button, but the warning comes up on a system security screen which does not. Annoying, but only that, and as Microsoft becomes aware of these problems they'll be fixed, probably without announcement, in one of the automatic updates. For example, if you miss your automatic scheduled update of your operating system because your system was turned off, it doesn't notify you of that, it just goes out and looks for the updates, then offers to install them. Clearly Windows Security Essentials needs to be set up the same way, and I make no doubt that it soon will be. Until then the important thing is to be sure your laptop is updated before you take it on the road.

Security expert Rick Hellewell tells me there is a very long discussion thread at this link, and that Microsoft is aware of the problems. In theory, if the machine is turned off at the time that Windows Security Essentials is supposed to update, an update should happen in the first quarter hour after the system is turned on, and on most systems it will do that. It doesn't seem to have done so on mine. The important fact is that Microsoft is fixing this, and it wasn't that serious to begin with.

My conclusion remains: Windows Security Essentials works, it's easy to use, and it mostly runs invisibly. I know from experience that it works with Windows 7 and Windows XP, and I used the beta version with Vista before I converted the last Vista system at Chaos Manor to Windows 7. WSE is free, and it's my understanding that it would be built into Windows 7 if Microsoft weren't fearful of monopoly accusations, particularly in Europe. It doesn't take more than fifteen minutes to download and install WSE, and the first Quick Scan won't take much more time than that - at least it hasn't on my system with 200 GB of files. (Full scan takes hours; I schedule one at 3 AM each Sunday.) WSE replaces previous Microsoft anti-virus scanners such as OneCare - it also runs faster. All told WSE is a lot more than good enough, and the price is right. Recommended.

What Kept Them?

It's a bit odd that it took Microsoft a decade before the company brought out an anti-virus system. Surely Microsoft had the capability to do that long ago, and it isn't unreasonable to expect to find something like that built into the operating system itself. After all, the OS knows how to find and examine files better than any third party program could.

It's only a guess, but I think I have an explanation. Bill Gates famously said "In 1989 I went to all the software developers and asked them to write applications programs for Microsoft Windows. But they wouldn't do it. So I went to the Microsoft Applications Group, and they didn't have that option."

That's a fairly accurate account of what happened, but it leaves out the exceptions. Back when Commander Gordon Eubanks ran Symantec, he put a lot of effort into writing applications for Microsoft Windows, and after Symantec acquired Peter Norton's company in 1990, Norton Software switched to writing anti-virus applications in Windows (from DOS where Norton programs had a well deserved high reputation). Norton Anti-Virus was an early and effective Windows application. I recommended it at the time.

Bill Gates was fairly careful not to harm the early third party Windows software developers. He was also friends with Eubanks. That, I think, is why Microsoft didn't develop or buy its own anti-virus software until rather late in the game.

Others would say that Microsoft didn't get security conscious until just before the millennium, and not doing anti-virus software was just a symptom of that.

Whatever the reason, it took Microsoft quite a long time to get into the virus prevention business, but now that they've done so it will be interesting to see what happens to the third party security software developers. I don't predict a very bright future for them.

Windows 7 and Office 2007 at Chaos Manor

I have not installed Windows 7 on my laptops. Orlando, the t42P ThinkPad, is sort of at the margin when it comes to hardware capability, and since I am still in the process of moving to the Apple MacBook Pro as my road machine, I am taking no chances with Orlando; he'll stay with XP until or unless I hear reliable reports of Windows 7 on a t42P.

The other system still running XP is LisaBetta, the HP Compaq TabletPC. She's far too old and slow for Windows 7. I am happy to report that she still runs just fine, and it is still my opinion that a TabletPC with OneNote is one of the best research tools you can ever have. One day I suppose I will have to replace her, probably with a ThinkPad Tablet, but we'll see. I'm still wondering what Apple will do for a Tablet - it's all just rumor so far - and I am in no real hurry. It's not the Tablet alone that's important for research, it's the fortunate combination of wireless connection, Tablet capability, and OneNote that makes it all so effective. A Tablet with wireless is also pretty good at rapidly dealing with a lot of your email, the only glitch being that if you use Outlook and the Tablet is not your usual email machine, you have to transfer the outlook.pst file back and forth between machines; it's a big file and that can take a while. That's the main reason I don't regularly use LisaBetta to sort and clear mail at the breakfast table.

But with those exceptions, all the Windows systems at Chaos Manor are either running Windows 7 or shortly will be. They were also all updated to Microsoft Office 2007. I have had no reason to regret either decision. Fortunately, Outlook in Office 2007 uses the same file structure and .pst files as Office 2003 did, so no conversions are needed when I use one of the laptops which continue to have Office 2003. I suppose I ought to update the laptops, but I never have. Note that Outlook 2003 has several differences in displays and controls from Office 2007, and in fact some 2007 features and flags just don't exist in Outlook 2003. Obviously you can't see or set those flags if you're using a 2003 system, but when you copy the .pst file back over to a Windows 7 Office 2007 system, they'll be visible again.

There really isn't any reason not to update my laptops to Office 2007 and the major deterrent has been pure inertia. There's also a bit of superstition: most of my fiction work is done with Orlando, the t42P ThinkPad. I carry him up to the monk's cell. That's my term for the bedroom formerly occupied by the oldest son living here; it has no telephone and the only books not associated with the current writing project are textbooks. I connect the ThinkPad to a Microsoft Wireless Comfort Curve keyboard, and a 20" monitor. I have been doing this for nearly a decade now (at first I kept a desktop up there and transferred files with Zip drives, thumb drives not yet being invented). I'm a bit reluctant to change things that work, and that includes Word 2003.

In fact, though, I am quite satisfied with Office 2007, now that I have learned my way around its radical reorganization of the ribbon. If you need to learn Word 2007, the best way is to get the O'Reilly Word 2007 The Missing Manual. Go through it casually and use it as a handbook. You will also want to know that the keyboard command control-F1 toggles the big Word 2007 ribbon into and out of hiding. That ribbon takes up so much screen real estate that it was a major reason I didn't change to 2007 for fiction; the discovery that it can be made to go away and come back so easily makes Word 2007 much more palatable and actually more useful than Word 2003 - although 2003 remains quite useful for getting words onto the screen. I very much appreciate such things as automatic spelling correction (it turns teh into the without my noticing it) and other such conveniences. Indeed, it's the need for either Word 2003 or 2007 that has kept me from converting any of my fiction operations to the Mac. I know that Word 2008 for the Mac is very similar to Word for Windows, but there is a major difference. For reasons I have never understood, Word 2008 on the Mac has no macros. You're expected to learn Apple Script and write your own. I decline to do that. I'm told that there is coming a new Office for the Mac that remedies this astoundingly silly oversight, and it will be here Real Soon Now. We'll see.

The fact is that Windows 7 and Office 2007 are pretty good tools for professional writers, and despite the very clever PC vs. Mac advertisements, the fact is that Windows 7 moves PC's a lot closer to the ease of use and reliability of Mac OS X.

Whatever Happened to Outlook Express?

One thing missing from Windows 7 is Outlook Express; indeed, Windows 7 doesn't even have Windows Mail, which could be found in Vista although not many people used it. For that matter, not many people used Outlook Express as their major mail reader.

I did use Outlook Express to access newsgroups. Over time newsgroups have become less and less important as a source of news, but the major discussion forum for science fiction writers has been the sff.net newsgroup since GEnie folded; there is a closed member-only area, and most sf writer discussions including professional discussions on legal and marketing matters took place there. Still do, for that matter, although SFWA is trying to get more people to use the newly revised SFWA web site. So far, the sff.net newsgroup is still the successor to the old GEnie discussion groups.

Fortunately, although Outlook Express has gone away, and there is no Windows Mail in Windows 7, the solution to the problem is simple. Microsoft lets you download free Windows Live Mail which has all the features of Outlook Express including newsgroup access. Live Mail downloads and installs quickly and easily, it's easy to set up, and for newsgroup access (the only thing I use it for very much) it works as well as - and about the same way as - Outlook Express. I haven't used Windows Live Mail as a mail handler because I have become accustomed to Outlook and have built a number of rules and contact lists - I keep my subscriber lists in Outlook - using it. If you get Windows 7, you should probably get Windows Live Mail. It's free, it works, and it may come in useful. I presume it would have been built into Windows 7 except for the various accusations of monopoly and unfair competition periodically levied against Microsoft.

Snow Leopard

I have installed the Snow Leopard upgrade to OS X on all my Mac systems. The only problem I have experienced is with networking: my Macs all play nice with each other, and my Windows 7 have no problems seeing the Macs, but connecting to Windows 7 machines is a bit dicey. I can do it, but it's neither regular nor reliable, and I haven't had time to straighten things out. Note that Macs have never been all that easy to connect to PC's although if you are running a virtual PC on a Mac, that will connect to other PC's fairly easily.

The odd networking problem I had connecting Mac systems wirelessly to a wireless network hasn't been a problem here, and it's clearly an odd interaction with specific hardware.

I will be taking a more thorough look at Snow Leopard in another column; for the moment suffice it to say that the only problems I have had with Snow Leopard have been in networking the Snow Leopard Macs with a PC network, and those problems can be overcome if you fool around with the system enough. I haven't found a systematic solution to the problem.

If you do convert to Snow Leopard, you will want David Pogue's Snow Leopard, The Missing Manual. Like most of the "missing manual" series, this book lives up to the motto "the book that should have been in the box." Moreover, it's a darned good introduction to using a Mac. If you go through this book with your Mac, you'll go from duffer to power user in a month or so, and you'll get the time back in time you save getting work done. The book is quite systematic, and very well written. Highly recommended.

How Green Is Your IT?

Kermit the Frog used to be fond of saying "It's not easy being green," but nowadays it's very politically correct to be green, and anyone looking for a job in IT had better be aware of that. Indeed, if I were looking for an IT job in the present era, one of my first preparations would be to read John Lamb's The Greening of IT (IBM Press).

This is a practical book for practical people. After all, whatever your convictions on Climate Change or Global Warming, no one wants to waste energy, and much of the energy used in modern companies goes into computers, their support, and their cooling. Lamb goes through energy uses and looks for ways to save energy - and thereby save money. The book is a bit of a hybrid, partly intended to convert CEO's and other non-technical executives to the green philosophy, and partly intended to show IT technical people how to accomplish some savings. I think it does a better job on the former mission than the latter, but there are some meaty sections for the technical people too, particularly for those who haven't thought much about the subject.

Some of the discussions of energy technologies are very introductory. They aren't wrong, it's just there's not a lot there that most people won't know, except for some numbers mostly collected from Wikipedia. I don't consider Wikipedia a reliable single source, but at least the material is collected into one place and presented in an orderly fashion. Those with more familiarity with the basics will find the case studies illuminating, and there are a number of them. These can get down to the nitty-gritty, and they're put in the general context of the book.

Whatever your views on the politically correct subject of Global Warming, Green IT is a subject IT professionals had better be aware of in these perilous times, and CEO's had better be prepared for press - and stockholder - questions on what their company is doing to get green fast. This book is a reasonably smooth introduction to a pretty complex subject. Recommended reading.

Citizen Engineer

Citizen Engineer: A Handbook for Socially Responsible Engineering by David Douglas and Greg Papadopoulos (with John Boutelle) is both an introduction to the title subject and a passionate presentation of a point of view. The view can be stated as a plea for social responsibility, that being more or less defined as a strong preference for sustainability (among many other criteria).

It's pretty hard to be for "unsustainability", just as twenty years ago it was difficult to be for "inappropriate technology" when "appropriate technology" was the big catch phrase. Alas, it's not always clear that "sustainable" sources are the best sources, even on the criterion of energy efficiency; and it's almost never a good idea to try to assign every activity a score to come up with a single figure of merit. In my day in operations research we regularly spoke of "the figure of merit fallacy," and one of the most common mistakes people made was what we called sub-optimization. The best known example of sub-optimization comes from the earliest days of operations analysis: convoy escorts thought their goal was sinking enemy submarines, and thus they designed their attacks to have the maximum chance of sinking the enemy. That turned out to be sub-optimization: the real goal was getting the convoy through, and the best way to do that was to disrupt the enemy submarine wolf packs and prevent their attacks on merchant ships. This required entirely different tactics from the "be sure to sink them" tactics, and when the disruptive tactics were applied, fewer enemy subs were sunk but the tonnage delivered to port went up dramatically.

A more modern example: over the long haul, light water fission reactors are not sustainable: we'll eventually run out of uranium. How soon is arguable, but assume the consensus figure fashionable about ten years ago: fifty years. (I know: it's now easily argued that we have 100 years worth of uranium, and there's indication that there's even more; but let's assume fifty for the argument.) Fifty years is not "sustainable". On the other hand, we know how to build light water fission reactors, and doing so will have dramatic effects on the world energy picture including carbon emissions. Choosing not to exploit nuclear energy on the grounds that it's "not sustainable" is a clear case of sub-optimization. Actually, Citizen Engineer doesn't get deep into specific energy sources, and while it has little to say about nuclear energy, it doesn't argue against it either. I choose this example as an illustration, not an accusation.

Citizen Engineer doesn't confine its discussions to "ecology" and "sustainability". There are long sections on copyright and intellectual property, much of it in favor of creative commons, but with considerable attention to Digital Rights Management as well as Open Source Software. There's a lot about law and legal knowledge.

I have my differences with some of the conclusions in this book, but none at all with their view that these questions are important, and engineers have some obligation to think through the consequences of their actions, and I can agree with Stanford President John Hennessy that this is the sort of book that engineers should read and discuss - provided that it's a genuine discussion. The premise of progressive education is that teachers have nothing to teach: they are supposed to induce their pupils to think for themselves. I have found that many who say that don't act as if they believe it; but the authors of this book do believe it, and they raise many questions that engineers ought to think about.

Learning Python 4th Edition

Python is a free Object-Oriented programming language that has been around for years. The O'Reilly Learning Python series has long been the standard tutorial/handbook for the language. The current fourth edition of Learning Python covers the latest version of Python. It is also greatly expanded, being over a thousand pages long.

Like the previous editions, the book begins with how to get Python installed on your computer, then how to run Program Hello, and goes on from there. Unlike many programming introductions it doesn't spend a lot of time one program structure and architecture. There's no discussion of topics such as the desirability of forcing programmers to declare variables before they can use them, and the optimum top-down programming structure practice. Whether this lack is a bug or a feature depends on how familiar you are with programming principles. Given that the book is over 1,000 pages long, it can certainly be argued that such matters belong in a different book.

Of course the fact that the book is 1,000 pages long is a bit startling: Python was originally developed as a quick and dirty programming language, easily learned yet powerful enough to do most of the work you'd want done. As microcomputers became more powerful, Python became more capable. It has always been an extensible language. Python supports but doesn't require structured programming; in that sense it's similar to LISP. Structured Programming theorists like Niklaus Wirth consider that a defect. To them, the whole point is that the compiler ought to enforce strong typing, mandatory declarations, and other such practices: they shouldn't be voluntary. The goal is that when all the errors are removed and the program compiles - and that might take a long time - it will do pretty well what you intended it to do.

LISP advocates have a different view. I once had this argument with Marvin Minsky, a strong LISP advocate. He explained Wirth's position to Danny Hillis (then his graduate student) thusly: "He thinks you should put on a straight jacket, so you can't make mistakes. That will be faster than just building something." I didn't really find that argument persuasive at the time, and over time I came more and more to appreciate Wirth's view that it's better to catch the errors in the compiler than to compile nonsense and then try to find it in debugging.

Arguments about programming philosophy were once pretty common in computer science, but eventually C and C-derived languages well took over and the discussions were moot and you seldom hear these discussions now. I think that's very much a pity. One reason Pascal-derived languages languished was that our hardware wasn't powerful enough to compile 50,000 line programs quickly and easily. C was much faster both in compiling and in run times, and speed was important. Now, our hardware is so much better than our software that we have the capability to let the computer catch more errors, and it may be time to reconsider Pascal-based languages again. The original C would compile almost anything, and it's successors don't do proper type and range checking; one results has been considerable security vulnerability of programs.

In any event Python is more based on the LISP philosophy. It's extensible, it doesn't enforce structure, it will allow you to create new variables without declaring them, and in general it's very easy to use. The general notion is to build Python programs in small chunks, testing each chunk, and when you're done you run the program. This is great for utilities and filters, and I use it for that. It's also free, there is extensive online support, and it's fairly easy to learn.

If you're going to learn Python you very much need Learning Python; given the book, a decent computer, and some time you can learn at lot at very low cost.

Winding Down

Two movies that I thought might have potential just came out, but I can't recommend either Fourth Kind or The Men Who Stare at Goats. Both had themes of some interest to science fiction readers, but neither worked very well. That's astonishing for a movie that has both Jeff Bridges and George Clooney in it; you'd think that between them they could use the phone book for a script and still have an entertaining movie. Alas, they weren't able to pull it off this time. The Men Who Stare at Goats takes the true fact that the US Army at one time experimented with "Distance Viewing" (perhaps better known as "Remote Viewing") and other psychic phenomena; and as is often the case with experiments of this kind, the results were just ambiguous enough to justify continuing the experiments for several years. The title comes from the practice of inducing goats to collapse by cursing them, and it's my understanding that there were some actual field tests. Most of the Distance Viewing and Direct Impairment experiments remain highly classified. Whether that's because there's anything that should be concealed or merely to avoid embarrassment to the military I'll leave for you to decide.

The title of Fourth Kind comes from the classification of UFO sightings, which included radar and other instrumental sightings, and Close Encounters: The first kind was visual, second was ground evidence, and Close Encounters of the Third Kind was actual contact. The Fourth Kind would be abduction. There's a considerable literature and lots of discussion of alien abductions of American citizens. Some complaints have resulted in investigations by authorities at various levels of government, but officially there's never been any reliable evidence of the presence of aliens on Earth, much less of kidnappings and abductions. Of course there's no official evidence of the existence of vampires, but vampires inspire better movies.

I can't recommend either of these pictures, so there's no movie of the month.

The books of the month are a pair of light fantasy entertainments, both from Baen Books, and both edited by Esther Friesner creator of Chicks in Chainmail. The two books are Strip Mauled and Witch Way to the Mall. I note that Amazon gives Strip Mauled a subtitle of Supernatural Suburbia, which would in fact be appropriate as a subtitle for either book. Witch Way to the Mall is a collection of stories about real witchcraft in typical Midwestern suburbs, where the local housewife really doesn't expect to find a possessed barbeque grill out by the garage, and bird watchers don't quite understand what they are seeing when an over-eager witch summons a passenger pigeon. Strip Mauled has the same premise, a suburban world much like ours, but sprinkled with people who turn into werewolves come full moon. All great fun, and just the kind of books you want on an airplane.

The Graphic Novel of the month isn't really a novel, because it's a collection of graphic stories on a theme, but I have much enjoyed The Adventures of Chrissie Claus, and now they're collected into a single volume. Chrissie Claus is the granddaughter of Santa Claus. Her father, Santa's son, didn't want to go into the family business, and became an adventurer. During his adventures he encountered Chrissie's mother, an elf princess. Chrissie is being raised more or less as a human in Chicago, and she's now enrolled in Northwestern University. Of course she's always had some magical abilities - for instance she always knows who's naughty and who's nice. Now that she's learned who she is, she has more magical abilities including some inherited from her mother, but she's still a rather naive teen aged girl, fortunately without most of the neuroses of most girls that age. She also has a pair of devoted flying reindeer, and some girl friends who, being raised at the North Pole, are even more naive about boys than she is. If you aren't interested by this time, you may as well forget it. I found her fascinating, and the art work quite good, and I am tentatively set to write my own Chrissie adventure; indeed if I hadn't lost a year to radiation treatments, I'd have done one by now. Marv Wolfman (creator of Blade for Marvel) has given me some tips on how to do that.

The computer book of the month is My New Mac, Snow Leopard Edition. This starts with how to turn on the Mac, and ends with an encryption project. I doubt most regular readers of this column will learn a great deal from this book, but if you're a Windows user converting to the Mac you will find it useful. It would also be a good starting book for first computer users, whether literate school children or Aunt Minnie. It's well written, it doesn't talk down to the reader, and the instructions are pretty complete. I prefer the Missing Manual as more complete, but I'm sure Aunt Minnie and the beginners would rather have this one.