Crawford's Corner

Library Hi Tech News

ISSN: 0741-9058

Publication date: 1 August 2000


Crawford, W. (2000), "Crawford's Corner", Library Hi Tech News, Vol. 17 No. 8.



Emerald Group Publishing Limited

Copyright © 2000, MCB UP Limited

Crawford's Corner


Ghost Towns and Public Libraries

"You begin to see a future in which the Web brings us nearly everything almost instantly. And while we're glued to our keyboards, today's bustling malls could start to resemble ghost towns-quaint relics of an age in which people left their homes to buy stuff."

That's Harry McCracken in PC World, closing the May 2000 installment of his new "Web Savvy" column. Supposedly, McCracken "looks critically at a new or developing Web site genre" each month. "Always in the back of his mind will be the question 'Is this good for consumers?'" according to a late introduction to the column. If that paragraph and the column that precedes it are indicators, he's either asking the wrong questions or he's the wrong person to be asking them. The May column is about and similar rapid-delivery services: order an item online and get it within an hour.

McCracken certainly thinks Kozmo is "good for consumers." He lives in one of the six urban areas covered by the service, and his orders have always arrived on time. offers free delivery and McCracken claims that "its prices aren't set artificially high to compensate." How can Kozmo make money doing house-to-house delivery of orders as small as two pints of Ben & Jerry's, at competitive prices, without delivery charges? Go ask Kozmo's venture capitalists.

Suppose it works. Is it good for consumers? If the consumers are housebound, certainly. If they're the cocooning telecommuting paranoid geeks that Faith Popcorn had us all becoming by now, sure: we're all working from home and afraid to go out in any case. Otherwise the question may be too narrow. Some other questions that I think McCracken should be asking:

Does the service make economic sense in the long run?

Is the service good for society-for the polis, for neighborhoods, communities, towns, and the like-or does it tend to undermine community interaction and ability to fund local services?

Does the service explicitly favor one group at the long-term expense of other groups?

Does the service make sense for people who lead reasonably rounded lives?

In other words, would sensible people be enthusiastic about this service?

The second question could be reworded in special-interest terms: Does this service make public libraries and other community services more effective and feasible, or does it undermine community services?

I like good local stores. I also like good chain stores that employ local people. I particularly like interesting local restaurants-not necessarily big-deal destination places, but little delights like Mountain View's El Paso Café and Yakko.

I like well-maintained roads, police forces good enough that you never know they're there, well-maintained interesting parks, responsive fire departments, and community social services that work. We don't have children, but I appreciate good public schools and am willing to pay for them. We do use public libraries-and I treasure them.

I won't claim to answer all of those questions for Kozmo. But I doubt the soundness of their business model, and I'm not thrilled about anything that encourages people to stay out of their communities and in their homes. Fortunately, I think the paragraph that begins this article is nonsense. Most of us have no particular interest in staying "glued to our keyboards" all day every day; most of us are not so loath to leave our homes to buy stuff. We may not recognize the direct links among local business, local employment, and local services-but we function as parts of a community. As a gimmick, Kozmo doesn't bother me; as a universal model, I am only comforted by its sheer improbability. Ghost towns don't have public libraries.

Getting It

San Francisco Chronicle reporters have moved beyond the "Internet as life" view; they've covered it for too long to buy into all the hype. I suspect that's also true for the San Jose Mercury-News. I've seen recent stories that include the line "The Web doesn't change everything," going on to note where Internet systems can reduce friction and otherwise improve life.

Other people are getting it, too, little by little. An Industry Standard review of The Social Life of Information by John Seely Brown and Paul Duguid (Harvard Business School Press) suggests that Brown and Duguid get it. "The way forward is paradoxically not to look ahead but to look around." These authors care about context and are aware of the rich mythologies that have grown up around information.

Consider universities. People who should know better predicted that virtual universities would replace physical plants-but it's not happening, partly because real universities do a lot more than just deliver information. The same can be said for many other massive paradigm shifts: by leaving out issues of community and society, they leave out reality.

Is PC World getting it? As with most enlightenment, it comes one person at a time. Brad Grimes (PC World's executive editor for features) offers a guest editorial for the June 2000 issue entitled "The bloom is off the Web's rose." In this editorial, he notes that being able to do "everything on the Web" doesn't automatically mean that you will-or should. "Frankly, the Web's gee-whiz period is over." One can only hope!

Grimes notes that services like Webvan may make sense for a couple with twin two-year-olds (Grimes and his wife), "but my bachelor brother up the street correctly shuns the siren call of Web delivery with a big 'So what?'" As do my wife and I-and, yes, we do the grocery shopping together. Finally, Grimes says that "If Web sites are to sell the average consumer on their virtues, they have to be as good as or better than their alternatives. Otherwise, few people will want to 'do that' online."

That's a little too simple. I'd argue that "better than" should be the test-otherwise, why bother? But that's a decision that different people will make differently. I buy CDs on the Web, mostly because local record stores make it clear that they don't want to deal with us old geezers who still respect what's left of our hearing. I buy books locally if it's possible, because local bookstores (independent and chain) are pleasant places to shop. And so it goes.

PC Values: June 2000

The standard configuration for June includes 128MB RAM, 16x or faster CD-ROM, AGP graphics adapter with 16MB dedicated display RAM, V.90 fax/modem or 10/100 Ethernet adapter, wavetable sound card, speakers, and a 15.6-16" viewable display. "Pluses" and "Minuses" are shown where applicable, along with hard disk size and software. Top systems are taken from company Web sites as of 8 June 2000.

Three items are noteworthy for June. First, top systems show sharp increases in value at all prices, thanks to significantly improved configurations. Second, the oddity of the top power system having a better value ratio than the top midrange system is readily explained by the notation "75GB HD": Dell has equipped the XPS B866r with an absurdly large hard disk, and my current point scheme overstates the worth and cost of large hard disks. Third-and possibly most interesting-"other" systems represented worse values than top-vendor systems in both budget and power categories, so those systems don't appear here.

Top, Budget: Dell Dimension XPS T700r: Pentium III-700, 20GB HD. Extras: MS Works Suite 2000, Altec Lansing speakers with subwoofer. $1,289, VR 13.41 (+30% since 3/2000 and 12/99).

Top, Midrange: Gateway Select 900: Athlon-900, 30GB HD. Pluses: 32MB display RAM, DVD-ROM drive. Extras: MS Works Suite 2000, Boston Acoustics speakers with subwoofer. $1,999, VR 11.04 (+18% since 3/2000, +27% since 12/99).

Top, Power: Dell Dimension XPS B866r: Pentium III-866, 75GB HD. Pluses: 18" display with 64MB display RAM, DVD-ROM. Extras: MS Works Suite 2000, Harmon Kardon surround speakers with subwoofer. $3,089, VR 11.33 (+28% since 3/2000, +42% since 12/99).

Other, Midrange: Quantex SM800s: Pentium III-800, 30GB HD. Pluses: 32MB graphics RAM, DVD-ROM. Extras: MS Office 2000 SBE, Altec Lansing speakers with subwoofer, Zip drive, Ethernet. $1,899, VR 11.70 (+20% since 3/2000, +38% since 12/99).

Product Watch

Special-Purpose Printers

PC World for May 2000 reviews two new printers that can produce photo prints that "look, feel, and last as if they came from a photo lab," according to Harry McCracken. Sony's $389 UP-DP10 is compact, sits upright, uses dye-sublimation printing, and prints nothing but photos: 4" x 6" at most, in about 90 seconds, costing $0.60 each. You can choose glossy, matte, or textured finish. McCracken says that $0.60 is comparable to the cost of film and prints at a midrange photo finisher. That depends on where you are. At our local Safeway, using Kodak's high-end Picture Processing (with its new DuraLife paper and index print), we pay $0.36 each, combining the cost of Kodak Gold 200 film, developing, and 4" x 6" prints; regular processing would be significantly cheaper. But then, we live in Silicon Valley: maybe film printing is cheaper out here?

The other printer apparently produces even better quality photos if you use its special photo paper and inks, and the manufacturer claims that the prints will last as long as traditional photos. Surprisingly, this printer is cheaper and isn't a one-trick pony: it's Epson's $299 Stylus Photo 870, a general-purpose inkjet. Figure $0.50 per 4" x 6" print, but you can also produce bigger prints when you need them.

PowerBook 2000: The Vagaries of Mac Reviewing

Macworld for June 2000 loves the PowerBook G3/400 and G3/500, and Andrew Gore's review (love song is more like it) makes no pretense of objectivity: "A powerbook is more than a computer. It's a loyal companion, an indispensable assistant, a home entertainment system, and a gateway to the world." Both units get 4.5 mice, even though there are clear deficiencies. The deficiencies might matter if these were (boo, hiss) Windows notebooks-but they're PowerBooks, and immune to serious criticism.

What I find most remarkable is what's missing from a lengthy two-page review that covers two essentially similar models. We get the CPU speed, graphics controller name and graphics RAM, system RAM, hard disk size, and notice of the DVD-ROM drive.

There's no indication of screen size, screen resolution, existence (or speed) of a modem, or bundled software. I can't imagine printing a notebook review that doesn't mention screen size or resolution-but then, I'm not a Mac person.

DataPlay: Another Writable Disc?

PC Magazine for May 23, 2000 discusses this proposed new product with this charming lead sentence: "In fact, the world does need another miniature mass-storage device, contends" The startup company thinks there will be a big market for little write-once optical discs that store 250MB on each side and cost $5 to $10 each. That's ten times as expensive as 650MB write-once CD-Rs, and these discs won't be all that fast. They could also be pressed as read-only devices for perhaps $1 each, with individual numbers to provide for locked content. Drives should cost $100 or so.

What's the benefit? These are tiny discs: roughly an inch square. That could make them ideal for digital photography or (possibly) portable music players. For some uses, the lack of rewritability can also be a feature-although the article's suggestion that "rewritability aids piracy" strikes me as a bit loony. The ability to copy digitally makes casual piracy easier; whether or not you can write over a medium makes no difference at all.

Will this one make a difference? It seems improbable, but very little is impossible. Incidentally, people use a lot of CD-Rs, more than you might guess. The May 2000 EMedia includes a 1999 market summary showing 1.583 billion CD-R discs sold in 1999, up from 698 million in 1998.

Dumb Ideas Revisited

Not to make a value judgment here, but I think eMedia's title for an article on SpectraDisc is right on the money: "Divx without the dial-ins?" Or consider Ted Pine from InfoTech Research: "It seems that the one-rental concept is the Count Dracula of the home video industry. No sooner does the consumer stick a stake in its heart than it comes back demanding new blood."

Okay, so Pine needs to study up on his vampire lore: we seem to have missed the heart with Divx. SpectraDisc, a new Rhode Island company, has this spiffy new idea for DVDs (or CDs or CD-ROMs, for that matter): "environmentally safe chemistry" that causes the information layer to start decaying as soon as you open the packaging. Once opened, the disc will work only for a programmed amount of time-say 24 hours, 48 hours, a week.

The hotshot behind this notion thinks it's a sure winner. Stores would sell these DVDs for $2.99. We all hunger for them because they eliminate that awful trip back to the rental store. Once you watch the movie, "the disc can be tossed into the plastics recycling bin." I wasn't aware that the kind of plastic-metal combination used in DVDs was readily recyclable, but that's a detail. As is convincing retail stores that they should use twice as much floor space and that the road to big bucks is paved with $2.99 items (with a hefty chunk going back to the studios and SpectraDisc, presumably). Sigh.

Location Without GPS

It's nice to follow a dumb idea with a good idea, particularly a "Who woulda thunk it?" idea. U.S. Wireless is building a nationwide "location information network" that will pinpoint the location of cell phone users-without GPS technology. How? By measuring the pattern of reflections in the phone signal as it bounces off buildings, hills, and so on. Take enough reflections as captured by enough base stations, apply loads of processing, and couple that with a database-and it should be possible to locate the caller.

This is an important development if only because 911 does not work very well from cell phones: it's hard to dispatch a fire truck when the dispatcher can't tell where the call's coming from. Real-time navigation assistance may be another important use, while wonderful stuff like e-coupons for your nearest store will get more attention than it deserves.

What about privacy? Will this technology mean that you can never get lost even if you want to? Not as long as your cell phone has an Off button-or you're able to leave home without one.

Maps Where They Matter

I hate it when Microsoft gets it right: it makes demonizing them harder. So it is with Microsoft Streets & Trips 2001, the latest version of their trip planner. It finally includes GPS support so you can use it while on the road-but, more significantly, it's designed so that you can load the whole CD-ROM on your notebook computer. That can take up to 835MB of hard disk space, but it means that you'd have exactly what you might need for a traveling companion-and it now includes Canadian cities.

Free WordPerfect!

What have we come to? Corel has a new WordPerfect Suite: WordPerfect Suite 8 OEM. It's only available preinstalled on select PCs-and it's free. I'm not sure what that means, since bundled software on any PC is "free" from the buyer's perspective. What is clear, however, is what you get along with the free software. It's not hard to guess: it's what you get with most freebies, on the Web or elsewhere.

That's right! Advertising along with your word processing! So far, it's minor: ads pop up when you launch a program but disappear when you start working on a project. Will it stay that way?

Sheet Music Online

Here's another one that makes a fair amount of sense. Musicnotes (you can figure out the URL) sells sheet music, some 200,000 pages in all. That's nice if you're not close to a big music store, and the selection beats most local stores-but that's not the kicker. Six thousand titles (so far) are in downloadable digital format: you can see the sheet music, but you can also hear it performed (presumably using MIDI).

Years ago (three of them, roughly), back in the Dark Ages, quite a few enhanced CDs offered synchronized viewable scores as accompaniments to the sound recordings. One company produced a few CDs composed primarily of printable and playable scores, including the complete works of Scott Joplin. The idea still has merit-at least for the tens of millions who can read musical notation.

Reality Check: The Death of CD-ROM?

By now, DVD-ROM should be pushing CD-ROM out of the marketplace, at least in terms of new drives. That's what market forecasters predicted, and (blush) I tended to think they were probably right. DVD-ROM drives don't cost much more than CD-ROM drives and they can do so much more.

The CD/DVD-ROM sales index chart for the last half of 1999, appearing in the May 2000 EMedia, shows what's actually happening. In the third quarter, 3.7 million DVD-ROM drives shipped worldwide-but so did 25.5 CD-ROM drives. For the fourth quarter, DVD-ROM shipments were up to 4.6 million-along with 28 million CD-ROM drives. The good news here is that more than eight million DVD-ROM drives shipped in the second half of 1999. The tricky news, if you're looking for clean transitions from one technology to another, is that the share of market grew from 12.7 to 14.1 percent. At the end of 1999, people were still buying seven times as many CD-ROM drives as DVD-ROM drives; it will be a few years before the installed base of DVD-ROM surpasses CD-ROM.

Press Watch

Seymour, J. (2000), "Interactive TV: set to take off?" PC Magazine, Vol. 19 No. 10, p. 99.

Speaking of failed ideas that just keep coming back... This article has the clichés, most if not all of them. "The one-way experience of watching traditional television has gotten, in this Interactive Age, very old." (Capitalization and emphasis Seymour's.) "The Web has changed everything." "We'll finally be able to shop over TV, buy tickets for sports and entertainment events...and jump from one weather report to another." In my simple brain, "buy tickets" and "shop" are pretty much identical, and I'm not sure that "jumping from one weather report to another" is that compelling as a TV activity-but never mind.

Maybe the Web has changed everything, but what Seymour is talking about is not the Web. It's an effort built around DVB, Digital Video Broadcasting, a proposed standard for data within the interstices of TV broadcasting. It's data over TV; it's not the Web.

"Cynics used to say that we were going to sit at our PCs, treating them like TVs. The new cynics will say we'll be sitting at our TVs, treating them like PCs. And they won't be far wrong." I consider myself a skeptic but not a cynic-and I say that most of us have exceedingly little interest in treating TVs like PCs or vice versa. Maybe that remote control isn't interaction enough for Seymour-but for me and apparently most other people, the TV is fundamentally a passive medium. I work with my PC; I watch TV. In different rooms.

Loftus, T. (2000), "The business of community," FamilyPC, Vol. 7 No. 6, p. 28.

"Back in the Web's early days, Netizens embraced the idea of virtual communities." But that was four or five years ago (the "early days" of the Web don't go back much before 1995), back when we used quaint terms like "Netizen" and didn't understand The Rule:

The Web is about Business. Period. If it isn't Business, it isn't worth discussing.

So now we have faux communities, with a "new crop of sites revitalizing an old-fashioned sense of community"-as long as there's a buck to be made in the process. "Make no mistake, these third-generation sites are not the free-wheeling, grassroots Web communities of old." These new "communities" include Epinions,, and Best of all, these commercial sites are "slick clicks that incorporate a retro-Web sense of community and creativity." So isn't some ragtag band of protesters: it's a site that sells organic products.

How much are you charging for library cards these days? Or do you print ads on the due-date slips instead? After all, isn't that what community is all about: selling and buying stuff?

Ledbetter, J. (2000), "The new paper chase," The Industry Standard, Vol. 3 No. 14, p. 117.

Pay attention to the signs. Real life has a hold on people. In this case, as the teaser says: "Online publishers are eyeing a new source of revenue: Old-fashioned print magazines.", a remarkably well done adult-oriented (OK, erotic) online magazine, launched Nerve in April 2000: a bimonthly print magazine with content similar to the online magazine. is apparently thinking about a print publication. Yahoo! and eBay both have related print magazines; Yahoo Internet Life has a paid circulation approaching 750,000. The item says, "Call it reverse convergence." Maybe. I like the comment from Rufus Griscom,'s CEO: "The idea of the pure Internet play is going to sound ridiculous in 18 months."

Block, D. G. (2000), "DVD title return rates: not the return of CD-ROM," eMedia, Vol. 13 No. 4, pp. 22-23.

This little story is innocuous enough other than some remarkable mathematics. Apparently CD-ROM publishers used to see astonishingly high return rates. That may have been because there were loads of defective discs, but it's just as likely that people found discs difficult to install or operate or just returned them because they didn't like them.

I was bemused by one sentence halfway through the story: "DVD yields are nowhere near as good as current CD yields, which are in the high 90th percentile, or in some cases even above the 100th percentile."

Wow! A yield above the 100th percentile means that there are more perfect discs coming out of the process than there are raw materials going in! I'm impressed.

I can't come up with an innocent explanation-unlike a similar case in the cruise industry. The biggest names in mass-market cruising frequently cite occupancy rates of 105 percent or higher. That's legitimate, because "full occupancy" is defined as two people in each cabin, but many cabins will actually hold three or four-perhaps not comfortably, but it will hold them. But I don't see how a production line can get more than 100 percent yield.

Tynan, D. (2000), "Privacy 2000: in Web we trust?" PC World, Vol. 18 No. 6, pp. 102-16.

Have Webheads woken up to privacy concerns in time, or is it already too late? If you ever wanted a reason to avoid Sun computers, Scott McNealy gave you one with his snide little comment "You have zero privacy anyway. Get over it." I've used that last sentence myself, and I apologize: "Get over it" is another trigger wording like "inevitable." This PC World article is one of two major June 2000 articles concerning privacy abuses and what (if anything) you can do about them.

It gets particularly interesting for California natives (like me), given that so much of e-commerce is based in these parts. The US Constitution does not, in fact, have any specific protection for privacy. The California constitution, on the other hand, does-so when Web hotshots in California (like McNealy) say that privacy is a moot point, they potentially have the state arguing against them. I believe California's constitution is likely to come into play repeatedly over the next few years-or at least I hope it does.

This article is well worth reading, despite some grotesque overstatements. "As the Net gradually becomes the medium most Americans use to get news, buy groceries, rent movies, obtain medical advice, and possibly vote for presidential candidates, what little personal privacy they once had may soon disappear." Can we raise the privacy alarms without the absurd claim that most of us will be buying our groceries and getting all of our news online?

Some of the sidebars are horrifying. Some are amusing. It's useful to consider how many e-commerce business models rely on in-depth awareness of individual buying habits-and on selling that information to others.

The article after this one discusses products and services that can give you more privacy. It's also worth reading.

Glass, B. (2000), "Keeping your private information private," PC Magazine, Vol. 19 No. 11, pp. 117-36.

It's a strong season for privacy concerns, as this second lengthy feature on software privacy should indicate. While certainly not flawless, this is worth reading, including sidebars that review cookie managers (Cookie Crusher and Cookie Pal get the highest ratings) and anonymous browsing services (Anonymizer, IDSecure, and Freedom all rate five dots out of five). The last few pages are actually a separate article on designing privacy for e-business, probably worth reading for librarians as well-except that the privacy standards for libraries should be much more stringent than for businesses.

Furger, R., and McEvoy, A. (2000), "The buying game," PC World, Vol. 18 No. 6, pp. 160-70.

What's the best way to buy a PC? What's the worst? PC World tries to answer that question every couple of years through surveys and undercover shopping. Most shoppers would buy from the same seller no matter what route they took-even at CompUSA, the worst place in the survey, 79 percent would buy again.

The article is interesting. The conclusions are peculiar if only because they describe one "retail" chain oddly-namely, the only retail chain that scored above average overall. Gateway Country Stores are physical establishments, but they're showrooms, not really stores: you can't go to one, buy your computer, and walk out with it. You can see what you like and place an order-but that order is handled the same as it would be if you called Gateway or connected to (Gateway scored above average for Web and phone sales, too, as did Dell, Micron, and Quantex.)

Overall, the conclusion seems to be that phone sales may be your best bet if you're a newbie, while Web configuration systems work just fine for people buying their second or third PCs. Retail stores just don't do all that well-and I'd argue that the best direct-order brands are better computers in any case. For what it's worth, Circuit City and Office Depot scored average; Best Buy and Staples joined CompUSA in the below-average category.

Stewart, S. (2000), "Anxiety disorder," The Industry Standard, Vol. 3 No. 19, pp. 156-80.

Go read this lengthy article. It considers the panic about attacks on the Internet, whether that panic makes any sense, and what it says about the press, the Net, and the assumptions we make.

The article is well researched and well written, and includes some cautionary notes about e-commerce and the extent to which commercial concerns can undermine noncommercial uses of the Internet. That isn't a perspective you'd expect in "The Newsmagazine of the Internet Economy," but it's the kind of nuanced thinking I expect from The Industry Standard.

Slater, M. (2000), "Surf's up for info appliances," Computer Shopper, Vol. 20 No. 6, p. 85.

Yet another "post PC" article in which an expert tells us that hot new "information appliances" may replace PCs, or at least make them irrelevant. It's interesting to see how these experts trivialize the use of PCs" we use them for "Web browsing, e-mailing, game playing, and managing personal information." Genealogical research? Writing? Household budgets? Tax preparation? Forget it.

Maybe people really will browse the Web on their cell phones or put dedicated appliances all over their houses. I tend to agree that tens of millions of home PCs really aren't being used as full-fledged computers.

I was struck by the closing paragraph. "Let's not forget how mainframe-computer makers viewed the PC in its infancy. They derided it as hopelessly inadequate, right up to the time the PC put them out of business." Oddly enough, that isn't the history I remember. I seem to remember that the biggest mainframe computer maker was also, for many years, the biggest PC maker, doing more to legitimize PCs than any other company. Then again, IBM sold $9 billion dollars worth of mainframe-class computers in 1999: an odd definition of "out of business" in my book. But then, Slater is an Industry Expert, where I'm just an ignorant observer who works at a place that still uses mainframe computing-just as thousands of other places do. What do I know?

Starrett, R. (2000), "The speed of sound," EMedia, Vol. 13 No. 5, pp. 28-39.

As CD-RW drives get faster and CD-R media get cheaper, some concerns arise about the tradeoffs in recording a CD-R at 8x or higher. For audio recording in particular (probably the predominant use for CD-R), will discs recorded in six minutes (12x) sound as good as discs recorded in real time?

This article can't entirely answer that question, particularly given the ideas that arise among the high-end audiophile community, but it does provide some test results. Starrett took a bunch of blank CD-Rs from the major manufacturers and a good-quality computer CD-R drive capable of 12x writing; he then did disc-to-disc copies at each supported speed and tested the resulting CD-Rs using professional test equipment.

The results are interesting, but give little comfort to slowpokes. For most name-brand media, if higher-speed recording didn't improve the error levels (which it frequently did), neither did it make them significant. In every case, 12x recording resulted in discs that had much lower error rates than typical prerecorded audio CDs. Recording from WAV files on the computer, rather than disc-to-disc recording, didn't make much difference.

The general message here is that, if your recorder specifies a maximum speed and the media claim support for that speed, there's no compelling reason to run slower. Your audio compilations will almost certainly be more error-free than commercial CDs, even if you only take six minutes to record them.

Review Watch

Desktop Computers

Metz, C. (2000), "Fast PCs: who needs them?" PC Magazine, Vol. 19 No. 11, pp. 146-64.

You probably know the answer to this review's title question: very few normal PC users, except those doing high-end graphics, image-editing, or data-mining work. (The real answer isn't appropriate for the all-business PC Magazine: fast PCs are for gamers.) The article concludes that business professionals can get by just fine with a mere 866MHz PC. Most of us will be more than happy with 400MHz PCs, to be sure, but that's real life. The editors clearly hungered for a chance to review this year's barn-burner PCs and built a worthwhile article around those reviews.

The Editors' Choice for content creators isn't one of the gigaHertz units; it's Dell's $3,479 Dimension XPS B866r, a well-equipped unit that also performed well. Micron's $2,899 Millennia Max 866 gets an honorable mention in this category. An Editors' Choice for business users goes to Systemax's $2,360 Ascent PVO-850A, based on an Athlon 850: it scored well on business benchmarks and offers great bang for the buck. An honorable mention goes to another cheap system from a little-known maker: ABS' $2,299 Supreme 3D Alpha.

If you want a machine this fast ("need" is rarely the right word), read the article and more recent reviews. It's probably worth looking at any four-dot or five-dot system, and in addition to the Editors' Choices and honorable mentions those include Compaq's $2,800 Prosignia Desktop 330, Dell's $4,289 XPS B1000r, and Gateway's $2,799 GP7-800 and Select sb (same price, similar configurations, but one has a Pentium III-800 while the other has an Athlon-1000). Any of those systems should keep you happy for a month or two.

Digital Cameras

Grotta, S.W. (2000), "3-megapixel cameras," PC Magazine, Vol. 19 No. 10, pp. 32-40.

The magic number is six megapixels: that's what a 35mm negative yields scanned on typical drum scanners. Effectively, that's as much raw resolution as most experts believe can be gained from regular film (although "regular film" covers a huge range of formulations). Consumer-priced digital cameras aren't there yet, but at three megapixels these five cameras come awfully close.

The summary of this review says that "we were hardly surprised to discover that all five cameras take gorgeous images that are virtually indistinguishable from those of film-based cameras." Oddly, that's not what the image quality comparison shows. They photographed a still life using a 35mm Nikon (film not stated), then took the same image with one of two Editors' Choices (the Olympus C-3030 Zoom) at its highest-quality setting and at a significantly lower-quality setting. The differences between the two Olympus versions are subtle-but the differences between both digital prints and the film print are striking! It has the snap, color range, and vividness you'd expect from a Nikon; both digital versions are dark, dull, and muddy by comparison.

That quirk aside, the review offers individual writeups of each camera so that you can see which (if any) of these units might suit your needs. If you've read good reviews for Nikon's Coolpix 950, Macworld's favorite digital camera for some time, you won't be surprised that the newer Nikon Coolpix 990 ($1,000) wins one Editors' Choice. The Nikon is a little strange looking but it has great image quality, lots of features, and simple operation. Olympus' C-3030 Zoom costs the same and has a steeper learning curve, but it offers "just about every conceivable feature" and produces the best images in the roundup.

Displays and Graphics

O'Brien, B. (2000), "Graphics at warp speed," Computer Shopper, Vol. 20 No. 6, pp. 168-77.

Give Bill O'Brien credit for telling the truth: today's hot new graphics cards are designed for PC gamers. "After all, how fast do you need your spreadsheets to scroll across the screen?" It's not quite that simple, but there are very few non-game applications that strain last year's good graphics accelerators.

If you're one of those who needs the hottest graphics, this article provides a careful review of six cards costing $172 to $284, all with 32MB RAM, all using the AGP bus (standard for graphics on almost all new PCs). Most of the cards use nVidia's GeForce 256 graphics accelerator and double data rate (DDR) memory. That includes the Best Buy, Hercules' $254 3D Prophet DDR-DVI, which includes video outputs (S-Video and composite) and a DVI output for digital LCD displays. Honorable mention goes to Creative Labs' $250 3D Blaster Annihilator Pro (don't you just love that name!), which lacks the video outputs but is just about as fast and includes some games.

Poor, A. (2000), "LCD monitors go to work," PC Magazine, Vol. 19 No. 10, pp. 182-201.

Surely you know the reasons that LCD displays overthrew CRTs years ago: compact size, high resolution, no flicker. Then there's the little glitch that keeps real-world purchasers going back to CRTs in 97 percent of cases: price. The fourteen 15" (viewable) LCD displays in this roundup cost $1,000 to $1,580; name-brand 16"-viewable CRTs (that is, so-called 17" monitors) cost $220 to $450. You can get a superb 18"-viewable CRT for considerably less money than these 15"-viewable LCDs: that continues to be a compelling argument.

Still, LCDs do keep getting better. This roundup provides detailed discussion of each unit, including objective performance tests. Editors' Choices go to ViewSonic's $1,580 VP151 at the high end and Philips' $1,000 150P for bargain hunters.

Mac Upgrades

Read, D. (2000), "G4 processor slot upgrades," Macworld, June, pp. 34-5.

You don't hear much about processor upgrades for PCs, but they continue to be significant for Macs. That may reflect the difference in system pricing or the relative simplicity of upgrading a Pentium; it may have something to do with huge recent changes in Apple's peripheral buses.

This brief review covers eight upgrade cards costing $600 to $850, carrying 350MHz and 400MHz G4 CPUs. The upgrades won't make an old PowerPC Mac quite as fast as a new G4 Mac, but they come close, even though the upgrades aren't painless.

No Editors' Choice, but five cards earn four mice each: Newer Technology's Maxpowr 350 and 400MHz cards ($599 and $799, respectively), Sonnet Technologies' Crescendo 350 and 400MHz cards ($500 and $650), and PowerLogix' $599 PowerForce G4 350.

Portable Digital Assistants

Brown, B., and Brown, M. (2000), "A pocketful of PC," PC Magazine, Vol. 19 No. 11, pp. 32-6.

There's an old saying that Microsoft doesn't get it right until the third try (or various permutations of that theme). It's based on good evidence-Windows wasn't usable until Windows 3; Word 3 was the first really workable Word; does anyone remember Internet Explorer 1 or 2? For those who aren't wedded to their Palm or Visor handhelds, this First Looks mini-roundup may be interesting, as it covers two early Pocket PCs. Pocket PCs are PDAs that run the Microsoft Pocket PC OS-which is the third iteration of Windows for handheld devices. It does appear to be a huge improvement over the earlier Windows CE and the devices themselves are interesting, but perhaps not a direct challenge to Palm OS devices.

Pocket PCs cost $500 to $600, weigh 6 to 9 ounces, run 6 to 12 hours of continuous use on a set of standard rechargeable batteries, and display quarter-screen VGA: that is, 320 x 240 pixels with either 12-bit (4,000-color) or 16-bit (64,000-color) color. While the screen size and color depth are both superior to any Palm, these devices cost more, weigh more, and don't last as long on a set of batteries.

These aren't minimalist PDAs. Pocket PCs come standard with ROM-based Pocket Word, Pocket Excel, Pocket Internet Explorer, Pocket Media Player, Pocket Money, Pocket Outlook, Microsoft Reader with Cleartype-and (tada) Solitaire. They have microphones, speakers, and stereo headphone jacks. They can function as MP3 players, voice recorders, ebook readers, and limited PCs-and they can handle the appointment and task-list functions of Palms, including synchronizing with desktop units.

These are miniature PCs without keyboards more than they are Palm-style PDAs. The CPUs are many times faster than Palm CPUs, but they're handling a lot more overhead. The editors give both the Casio Cassiopeia E-115 ($600) and HP Jornada 545 ($500) a five-dot rating (the highest possible); many competitors will emerge.

If I had to guess, I'd guess there's a significant market for Pocket PCs-but that Palms will continue to outsell them. Sometimes simpler is better.


Howard, B. (2000), "Choosing the ideal network printer," PC Magazine, Vol. 19 No. 11, pp. 186-210.

The 16 printers reviewed here all claim to produce at least 12 pages per minute-and their manufacturers had to have a measurable share of the market (one percent or more). These are all monochrome laser-equivalent units; networkable color lasers appear in another review (see Stone, M. below). As always, the reviews merit careful reading, as do the performance charts; as is typical these days, there are no print samples.

One Editors' Choice appears for each of three major use categories, but it makes sense to consider other five-dot (out of five) and possibly four-dot printers as well. Lexmark's $6,080 Optra W810dn garnered the top spot for departmental printing with its speed, ease of setup and use, and good management software; the $2,180 Optra T614n won the Editors' Choice for large workgroups. Xerox' $1,780 DocuPrint N2125 gets the nod for small workgroups, with HP's $1,100 4050 TN getting an honorable mention.

Other five-dot printers include IBM's $1,180 Infoprint 21 and Xerox' $2,860 N2825 and $5,670 N4025; there are also four-dot printers from Brother, HP, Kyocera, Lexmark, and Minolta.

Labriola, D. (2000), "All the news that's fit on printers," FamilyPC, Vol. 7 No. 6, pp. 130-2.

This family-oriented review includes an odd mix of inkjet and laser printers costing from $99 to $499. Top-rated in the group is HP's $299 DeskJet 952C, although it's slow when producing photo-quality prints. Two other printers earned Outstanding ratings, either one or two points behind the HP. Closest is Epson's $499 Stylus Photo 1270, which can produce oversize projects (up to 13 x 44") and offers Epson's new "permanent" photo inks and which is the fastest of the group for printing full-page photos (3.5 minutes compared to 8 minutes for the HP). Third place goes to the lone laser in the roundup, Epson's $349 EPL-5700i; it's light, cheap (consumables run two cents a page), and fast-but it's a monochrome printer.

One interesting sidebar evaluates actual costs of producing full-page photographic output and the costs estimated by manufacturers. Uniquely, Hewlett-Packard's estimate is not too far from the truth: estimated cost $1.02, actual cost $1.45. In every other case, the actual cost was at least twice the estimated cost, with Lexmark representing the extreme: $0.63 estimated, $2.84 actual.

Stone, M. (2000), "Color on the network," PC Magazine, Vol. 19 No. 11, pp. 64-8.

The network-printer roundup in this issue of PC Magazine concentrates on monochrome lasers, still the workhorse printers for business. This First Looks piece reviews five color laser printers designed for network use. That's a specialized market, but the cost per page for these printers may be tolerable-and the output from today's color lasers is remarkably good even on ordinary copier paper.

First Looks articles don't normally feature Editors' Choices, and this one is no exception. Five-dot ratings (the highest possible) go to Lexmark's $3,400 Optra C710dn and Tektronix' $4,100 Phaser 750 DX by Xerox, both limited to legal-size prints. If you want tabloid printing, all three reviewed units garnered four-dot ratings, still very good: HP's $11,800 Color LaserJet 8550 MFP, HP's $8,500 Color LaserJet 8550 GN, and Minolta's $3,700 QMS magicolor 6100. In all cases, the price cited is the price as reviewed, including appropriate paper-handling and networking extras; base configurations run as low as $1,910 (Lexmark) to $4,500 (both HP printers).


Machrone, B. (2000), "Protect & defend," PC Magazine, Vol. 19 No. 12, pp. 168-200.

This article is actually the introduction to two group reviews for software (and in some cases hardware) to help guard PCs and networks from various forms of attack. Most of these programs and devices are firewalls; some offer more specific or more generalized capabilities. It's a good set of articles to consider if you don't already have firewall protection and have become sensitized to the dangers out there.

Norton Internet Security 2000 2.0 ($60) gets the Editors' Choice to secure a single desktop; it combines a firewall with Norton AntiVirus and optional filtering software. By the time you read this, you should be able to buy the firewall by itself; in any case, the software integrates with Norton SystemWorks if you own that.

For corporate firewalls, the Editors' Choice is WatchGuard LiveSecurity System 4.1, chosen for its ease of installation, superior reporting, and general good performance. It costs $4,990: LAN solutions don't come cheap. Small-office firewalls don't get the full treatment, but the highest rated of the group are Watchguard SOHO ($350 for 10 users, $500 for 25) and Ramp Networks WebRamp 700s ($480 for five users, $850 for 25).

Web Tools and Platforms

Clyman, J. (2000), "Know your site," PC Magazine, Vol. 19 No. 11, pp. 169-82.

You've built a great Web site for your library, consultancy, or business. How is it being used? This review provides excellent background on what you might be looking for and covers seven tools for analyzing Web log files. All five products scored well, but the editors found one clear favorite: the Editors' Choice, WebTrends' $499 Log Analyzer 5.0. An honorable mention goes to Hitbox Pro, a Web-based service that costs $19.95 per month (and up, depending on page activity measured). Sidebars discuss other aspects of Web site monitoring and tools to help with those aspects.

Dragan, R., and Derfler, F. J. Jr. (2000), "Build the right site for your business," PC Magazine, Vol. 19 No. 10, pp. 154-70.

This thoughtful article suggests the kinds of sites that an organization might want to build and examines three popular Web server platforms (and one promising Novell platform).

The background and discussions here are more interesting than the conclusions. For the record, two platforms offering high scores on every measure both get Editors' Choices; surprisingly, neither Linux nor Apache is involved. One choice goes to Microsoft Windows 2000 Advanced Server and MS Internet Information Server 5.0; the other to Sun's Solaris 8 using iPlanet Web Server Enterprise Edition 4.1.

Freund, J. (2000), "Creating the perfect site," Computer Shopper, Vol. 20 No. 6, pp. 178-88.

As another data point, this article covers five Web creation programs-but the reviewers here differ from PC Magazine (see Mendelson's article below) as to what matters most. Editors' Choice here goes to HomeSite 4.5 because it gives you the most control over your site; while the article doesn't cite runners-up, both Dreamweaver 3 and Microsoft FrontPage get four stars (to HomeSite's five). (Computer Shopper and PC Magazine both come from Ziff-Davis, but there's never been much doubt of their editorial independence.)

Grotta, S. (2000), "State of the art," PC Magazine, Vol. 19 No. 12, pp. 201-9.

This group review could as readily go into Graphics, but these days the measure of a graphics program is how well it supports the Web. As you'd expect, the article offers worthwhile background information and good individual reviews of the six programs reviewed, which include old reliables such as Photoshop and CorelDraw. The Editors' Choice goes to Macromedia Fireworks 3 ($200); an honorable mention goes to Ulead's $80 PhotoImpact 5.

Mendelson, E. (2000), "Design a great site," PC Magazine, Vol. 19 No. 10, pp. 134-52.

Still handcoding HTML-or are you just saving Office 2000 files as Web documents and working from there? Today's Web authoring products offer more flexibility, whether you want to tweak your own code or work at a higher level. This review provides worthwhile discussions of eight current products and yields two Editors' Choices for two different user types.

If you need to put together a good basic site, consider Microsoft's FrontPage 2000 ($150). It provides an "intuitive solution for business users who want to design simple sites" with wizards, a wide selection of templates, and a range of advanced features and automated functions.

For advanced site design, the Editors' Choice (and a package with loads of buzz in the field) is Macromedia's Dreamweaver 3 Fireworks 3 Studio ($399). Besides an elegant interface with powerful palettes, it offers some of the best advanced functions-but it's not for novices.

Honorable mention goes to Allaire's HomeSite 4.5.1 ($99 on CD-ROM), primarily useful for people who want complete control over HTML.

The Details

Crawford's Corner is written by Walt Crawford, an information architect at the Research Libraries Group, Inc. (RLG). Opinions herein do not reflect those of RLG or MCB University Press. Comments should be sent to CD-ROMs and DVD-ROMs for review should be sent to Walt Crawford, c/o RLG, 1200 Villa Street, Mountain View, CA 94041-1100; Windows only. Visit my Web site: