Some day…

It was some day. And someday I’ll get a clue. I had a major confrontation at work today, though it was with someone who never did like me all that much. Everyone who’s heard the story says she was being unreasonable. But I just can’t help but notice one thing: Every major confrontation I’ve ever had in the workplace during my professional career has been with an older woman. By “older,” I mean 20+ years my senior.
I don’t like that pattern.

On a brighter note… I was quoted on CNET! It’s Linux’s 10th birthday, so CNET solicited some opinions. A lot of people said Linux can overtake Microsoft, an equal number said no way, but I don’t think anyone said what would have to take place for it to happen.

Essentially, I said someone with an anti-Microsoft chip on its shoulder would have to bundle Linux and StarOffice, already configured and ready to go (meaning it boots straight to a desktop when you turn it on–no setup questions or license agreements whatsoever), price it at $349, and make it available in places people normally shop.

That’s not the only scenario that I see working, but it’s the one that’d work best. History states people will sacrifice the status quo if the price is right–Commodore and Atari mopped up the floor of the home market with Apple and IBM for most of the 1980s, because they gave you twice the computer for half the money. It’d be impossible to do that today, but if someone with name recognition (say, Oracle or Sun) stamped its name on Taiwanese-made clones (made by, say, Acer or FIC) and got into the distribution channel, pricing it below an eMachine and using an ad campaign like, “We made performance computing affordable for big businesses. Now we’re making it affordable for you,” they’d stand a chance. They’d probably need to go outside the company to run the operation. Maybe Jack Tramiel, a veteran of both Commodore and Atari, could be coaxed out of retirement.

What about applications? An awful lot of home users live with Microsoft Works. StarOffice is better. Internet access? Take a cue from the iMac and stick an icon on the desktop that signs you up for Earthlink. Games? There are tons of open-source games available for Linux. Include any and every game that doesn’t crash XFree86. Cut a deal with Loki to include demo versions of all their games, and maybe the full version of an older title. Loki needs the exposure anyway. Digital imaging? Include The Gimp, along with drivers that talk with a certain type of digital camera. Include a coupon for a decent-sized discount off that camera.

It won’t dominate the market, but I can see it grabbing a decent-sized chunk. It’d do everything a small percentage of the population needs to do, and it would do it cheaply and reliably and quickly.

Will it happen? I doubt it. It’s a risk. For a company to be able to pull this off, this operation has to have little or nothing to do with the company’s core business. Shareholders don’t like ventures that have nothing to do with your core business. As much as Scott McNealy and Larry Ellison hate Microsoft, I don’t think they’re willing to risk hundreds of millions of dollars just to try to steal a couple million sales from Microsoft each year. The company that does it would have to have name recognition, but it’d be best if the general public didn’t know exactly what they sell. A company like IBM or HP couldn’t do it, because they can’t afford to offend Microsoft, and the general public expects an IBM or HP computer to run Windows apps.

04/07/2001

Caveat emptor. Dan Bowman wrote in asking about a barebones Duron system he found. $399, just add an HD. He was nervous about it but wanted confirmation. It sounded OK until I found the word “Amptron.” You should translate the word “Amptron” as “No,” or, “Don’t buy it.”

Amptron is part of the PC Chips group, and the running joke about PC Chips was that PC Chips made parts so bad that even Packard Bell wouldn’t touch them. Typically a PC Chips board will sell for $65 or less and have integrated everything, so you can build a really cheap system in it. Mom-and-Pop stores will use systems built around a PC Chips board to undercut the consumer electronics chains, because, let’s face it, an integrated motherboard selling in the mid two figures will allow you to undercut even the entry-level $399 eMachine.

My personal experience with PC Chips boards has been horrendous. Defects abounded. I had one system that worked fine until you tried to access the floppy drive. Then the PC would bluescreen. Every single time. I had two other PC Chips boards that didn’t work at all. That was when I swore the company off for good. Some of my coworkers have bought cheap PCs at Mom-and-Pop shops, had problems, and brought them in to work for me to look at. Wouldn’t you know it? PC Chips. In both cases the problem turned out to be really, really dumb statements in config.sys and autoexec.bat so the motherboards weren’t at fault, that time. Honestly, I was surprised.

So… Friends don’t let friends buy Amptron, or any of the other members of the PC Chips group. Eurone and Bondwell are the only others I can think of off the top of my head.

I’m not sure why anyone buys these boards anyway, seeing as you can always get an FIC board for $10 more, and FIC generally makes good stuff. But anyway…

If you’re looking for a cheap Duron upgrade, look for a Gigabyte GA-7IXE4. It sells for about $85, and it’s a bare board. No video, no audio, no nothing. If you’ve got an existing PC, it’s perfect. Pair it up with a Duron-750 at $60, and salvage all your other components from the old machine, and you’ve just hot-rodded your aging PC for $145.

Soyo’s not my favorite motherboard company, but they’re miles ahead of PC Chips, and if you need an integrated solution they’ve got one for a good price. The K7VLM-B uses a VIA KL133 chipset and includes audio and video. Its 2D performance isn’t great, but its gaming performance is outstanding for an integrated chipset. The K7VLM-B sells for about $95, so you can build a low-end Duron gaming rig very inexpensively.

The other problem with barebones systems is the difficulty of knowing what you get. No one cares about anything but the motherboard maker. But what about the case? Will you sever a finger trying to open it? What about the power supply? Will it struggle to light the IDE activity LED? It’s impossible to know. Yeah, it’s a little hassle to spec out a case and power supply, and it’s a little hassle to mount the board instead of having it all done for you, but in the long run it’s more than worth it.

This particular barebones PC also included 256 MB of RAM. That really scared me, especially considering everything else. No one puts an Amptron board in a system and then puts Kingston memory on it. People buy PC Chips boards because they’re cheapest, so they’ll buy whatever commodity RAM is cheapest. A cheap motherboard plus cheap RAM is a recipe for disaster.

You can at least take comfort that they aren’t charging you any extra for the problems. They’re included, free.

03/06/2001

Inexpensive PCs. I want to post something useful before I handle the issue of the day. So here goes.

A longtime friend wrote in wanting some advice on buying an affordable PC. Yesterday’s build-your-own route using closeout components and a Saturday afternoon isn’t realistic for the person he was asking for. (Undoubtedly the issue came up in conversation because he’s pretty computer-savvy, and he said, “Well, my son knows an awful lot about this stuff, and a friend of his writes computer books, so let me see what they have to say.”)

So, what to do when it’s not practical to build your own?

Do the same thing my mom and aunt do and my grandmother used to do when they’re buying anything. Shop around. I still have nightmares in which my aunt exclaims, 10 hours into a shopping marathon, “I’m looking for bargains!” Some of it rubbed off on me. Some. NOT A LOT! (I know, I know, denial’s the first symptom.)

Anyway. Shop around. Not at Best Bait-n-Switch and Circuit Vulgar-Adjective-that-Rhymes-with-City. You don’t want consumer-grade HP Pavilion and Compaq Presario junk. And you definitely don’t want slimy salespeople who know more about sales commissions than they know about computers. (I know, Best Bait-n-Switch salespeople aren’t commissioned, but their managers get monthly bonuses based on sales, so you’d better believe they’re putting the screws to the salespeople.) Avoid the salespeople. Use the Internet. I routinely find great deals at a wide variety of places: www.insight.com , www.onvia.com , www.outpost.com , www.pczone.com , www.cdw.com are all reputable. And a lot of times you even get free shipping from Onvia and Outpost.

Just to prove my point, I went to Insight, the first place I thought of. I found a Compaq Deskpro Celeron-500 with decent specs (64 megs of RAM, I forget what sized hard drive, and Windows 98) for $450. So you get the quality Compaq reserves for big corporate clients for an eMachine price. Very nice. That leaves a monitor. I know this person’s working on a tight budget and $450 may already be pushing it. I find some bottom-feeder 15″ monitors for $119. Then I spy a 15″ NEC monitor for $150. I’ve got a six-year-old NEC monitor that’s still going strong. And last year I threw away an NEC monitor built in 1988 that finally died after about 12 years, much of it hard use. NEC quality is definitely worth $30 extra.

Now, does NEC give you the same quality on their entry-level monitors today as on my older monitors that were top-drawer when they were built? Probably not. But they’re certainly better than Proview, and better than most monitors PC makers relabel. And while Compaq does some things that drive me up the wall, I’ll take a Deskpro over absolutely anything I can buy at a consumer electronics or office supply store. Heck, I’ll take a used Deskpro over any new consumer-grade PC.

So there you go: quality worthy of corporate use, for 600 lousy bucks.

I can’t always find a steal of a deal at Insight, but I can usually find something good at one of those five places.

So shop around, and when you find a price you’re willing to pay, ask a knowledgeable friend about it just to make sure. It’s not just people who build PCs in their basements for fun who get all the good deals.

Subscriptions. I was going to say something about this trend then I decided I’d wait until someone brought it up. Yesterday, two people did. I’m not too surprised; ever since Gun-Bob Sweatpants went the subscriber route, I’ve noticed a dramatic increase in my traffic.

I currently have no plans to adopt a subscription model.

For me, it’s an easy question. If I were to go to that model, 20 people would subscribe. Then, for less than I get for publishing a typical magazine article, I’d be committed to produce quality content on a regular basis for a year. Bad move.

But let’s say 600 people subscribed. At, say, $25 a year, that amounts to 15 grand. That’s a decent chunk of change. But once again, I’d be committed to produce quality content on a regular basis, I wouldn’t feel as free to experiment, and if I had to take a week off–or, heaven forbid, a few months off like last year–it would be very, very bad.

And besides, when people pay money, you feel worse about hacking them off because they have ownership. Take last month’s gun debate as an example. I killed off that topic because I found it boring and I suspected other people did too. After I started writing about computers again, readership climbed back to pre-gun levels. But what if that discussion had involved my five or six longest-time subscribers? Would I have been able to kill off the topic? I don’t know. I can hack off those five and risk losing five subscribers, or humor them and risk losing a few hundred subscribers. Somehow, when there’s no money involved, it’s easier to take those kinds of risks. The best way for a writer to grow is to take risks, and no publisher’s going to pay me to take risks.

Now let’s look at this another way. When I started out, I had about 25 readers. Soon I had 200. Soon after that I had to take my extended sabbatical to let my wrists heal. Within a month or so of coming back, I had about 175 readers and I stayed flat. I started working hard to promote the site, and pushed that to 400. I stopped promoting, and it dropped slightly. Now, it looks like I have about 600 readers.

Now, 600 subscribers, at $25 apiece, would net me 15 Gs. But if I keep my content free, I have every reason to believe that within six months I’d have 1,000 readers or maybe even 1,500. That gives me more leverage when I  write professionally. The first question any editor asks is, “Can you write?” I can say yes. But what if I say, “Well, I write a daily essay on whatever I feel like, and 1,500 people around the world read it.” Isn’t that a much better answer?

Can I (or my agent) use my daily readership as leverage when negotiating writing deals? You bet. Can I get more than $15,000 worth of benefit from that? Not this year. But if I get one magazine article or book chapter because of this site’s impact, I make more than I’d make under a subscription model. (Which will net me even more readers, and more clout. See the vicious circle?) But that’s a moot point anyway, because there’s no way anyone’s gonna pay 15 grand to listen to me spout off every day. Not this year. And not next year either.

I also realize that people’s needs change. Yes, I paid to subscribe to Jerry Pournelle’s site, because for a while, his site was absolutely invaluable to me. But over time his content changed, and my needs changed. I visit his site a couple times a month now. But I used to visit a couple times a day. When I visit, I don’t learn nearly as much as I used to, at least not about computers. Part of that’s because my needs have changed. And part of it’s because in the time since then, I’ve spent some 6,500 hours being paid to work on computers and I’ve written a computer book, a half-dozen articles for publication, half a book manuscript no one will ever see, done a tech review for someone else’s book, and reviewed three chapters of still another person’s book before it was published. I guess you could say I outgrew him.

And I fully expect a good percentage of my readers to outgrow me. Hopefully they’ll still find some reason to visit anyway, if my content remains free.

Besides, I get my Web hosting and most of my promotion for free. Is it right for me to charge people under that model? I don’t really think it is. I can remedy that, but frankly, for a while I even felt a little guilty using this site to promote my book. Then I decided that me promoting the book gives me credibility, which lends credence to Dave Winer’s saying a large number of professional journalists and authors use Manilla, which lets him sell more Manilla and make money, so he benefits and I benefit. So I no longer have a problem with that.

If my content really is worth something to you, there are a couple of things you can do. You can buy my book. I benefit monetarily from that, and book sales give me clout just like Web readership does. Some people even buy my book and give it to people for Christmas and birthday presents. That may be going a bit overboard but I appreciate it. And if you buy the book from my link on this site, I get a kickback from Amazon in addition to my royalty from O’Reilly.

Now maybe you’ve already bought my book and you’ve given copies to everyone you know and a few people you don’t (thanks). Or maybe you’ve read the reviews, read the sample chapters, decided my book’s not for you, but you like the things I write about here (thanks). If you want to give something, then the next time you’re planning to buy a book or a CD or software, click on my Amazon link and buy it. I get a small kickback.

So when people do that, I get a few bucks every quarter. And it doesn’t cost my readers much of anything–they’re just buying stuff they probably would have bought anyway. And it’s anonymous, so I get most, if not all the benefits of subscription without any imagined pressure. Everyone benefits.

Now, what about the subscriptions and members that Manilla speaks of? That’s just signing up for e-mail updates and the right to directly post responses to the things I post here. There’s no cost involved, and what gets e-mailed out are my daily posts. Granted, they differ slightly sometimes from what stays here because sometimes I go back and edit the content slightly. Subscribers and non-subscribers basically get the same benefits.

01/29/2001

Slim pickings. There isn’t a whole lot of new content going up over the weekend, so I’ll hit two useful, if slightly flawed hardware articles. And I guess I’ve gotta come up with some of my own stuff.

Shopping. I went computer shopping Saturday for the first time in, oh, years probably. I build my own and have been doing so since 1996, so I normally take no interest in retail PCs. But a friend of a friend is in the market for a PC, and I didn’t want her to get ripped off, so I offered to go shopping with her to keep such a thing from happening. Note I didn’t offer to build her one–I’m trying to get out of the build-a-PC-for-a-friend business for the most part.

So we hit Office Depot. The superdupercheap trend seems to be abating a little; there’s a lot for under $1,000 still, but the $499-and-under market is waning. eMachines is playing there, but Compaq and HP seem to have retreated. I noticed plenty of Durons, which ought to make AMD happy.

We also hit Computer Renaissance. I’ve heard horror stories about the place, but figured I could probably handle the slimy salespeople. I can talk way over their heads when I want to (“I don’t care what anyone says, compared to Microchannel, PCI is rubbish. At least with Microchannel, I knew where the resources were going and I knew they’d stay put!”), and I can play intimidation by dropping OS/2 and Linux compatibility questions. They left us alone though, which was nice. I saw P200-based Compaq Deskpros for $199, including 15″ monitor. I wanted more power than that for her. HP Vectra PII266s were $399; PII233s were $379. Both included monitors. What caught my eye was a $299 Compaq Deskpro. It had a Pentium Pro-200, which was about as fast as the PII-233 due to its on-die, full-speed cache, and it had a SCSI hard drive. For productivity use, this Deskpro is fine. Its 32 MB RAM is awfully low, but that’s curable. DIMMs are cheap. The SCSI will be nice. And it’s hard to find a better-built machine than a Compaq Deskpro. Life expectancy of this machine will be much higher than that of a new Compaq Presario, eMachine, or HP Pavilion.

But if mail-order had been an option, I probably would have pushed her in the direction of a Compaq iPAQ. For about $399 (without monitor), you get Windows 2000 and a corporate-quality (as opposed to consumer-quality) PC. Expandability is nil, except for the memory, but for word processing and Internet use, it’s great, and that’s what she wants. And it’s got USB and Ethernet built-in. If I had to equip a small business with a fleet of PCs quickly, that’s probably the direction I’d go. And I like them for home use too.

On to the reviews.

ATA-66 vs. ATA-100 (Real World Tech)

Good methodology, at least as far as hardware selection goes, and he explains his methodology as well. Full disclosure is always good, as it shows confidence you have nothing to hide.

The testing is a little suspect though. Using three trials and taking the highest number isn’t the accepted method. It’s better to take 9 or 10 trials, discard the highest and lowest, and average the remaining scores. I say this because in my own tests, I sometimes get a string of two or three weird scores that are awfully high or low. Running more than that, then discarding the outliers gives scores more likely to reflect the real world.

For these purposes, the flawed method probably suffices; it shows the slight advantage of ATA-66 and ATA-100 over ATA-33, though it may be exaggerated. The tests show little or no advantage to using ATA-100 over ATA-66.

This isn’t the best online test I’ve seen, but it’s definitely not an atrocity either. There is carefully-planned research here, by someone whose experience shows.

Gigabyte GA-7ZXR review (BX Boards)

All I can say is this is a very unremarkable hardware review. They didn’t disclose the testbed setup other than the CPUs used. Benchmarks were limited to Winstone 99 and Quake 2, and then they didn’t list any competing boards, so you’ve got a bunch of numbers but nothing to directly compare them with!

Lots of pictures and a list of features, but frankly there’s nothing here that probably wouldn’t be on the manufacturer’s Web site.

The quality of writing is better than average, but this review’s usefulness is limited to introducing you to a board you may not be familiar with yet. Unfortunately to learn much of anything meaningful about it other than what it looks like, you’ll have to wait for one of the other sites to get their mitts on it.

01/11/2001

Mailbag:

My docs; Apple; Lost cd rom drive

It’s that time of year again. MacWorld time. I work with Macs way too much, so of course I have opinions. If you expect me to withhold them, you don’t know me very well.

Let’s face it: Apple’s in serious trouble. Serious trouble. They can’t move inventory. The Cube is a bust–unexpandable, defect-ridden, and overpriced. The low-end G4 tower costs less than the Cube but offers better expandability.  Buying a Cube is like marrying a gorgeous airhead. After the looks fade in a few years, you’re permanently attached to an airhead. So people buy a G4 tower, which has better expandability, or they get an iMac, which costs less.

Unfortunately, that gorgeous airhead metaphor goes a long way with Apple. The Mac’s current product line is more about aesthetics than anything else. So they’ve got glitzy, glamorous cases (not everyone’s cup of tea, but hey, I hear some people lust after Britney Spears too), but they’re saddled with underpowered processors dragged down by an operating system less sophisticated under the hood than the OS Commodore shipped with the first Amiga in 1985. I don’t care if your PowerPC is more efficient than an equivalently-clocked Pentium IV (so’s a VIA Cyrix III but no one’s talking about it), because if your OS can’t keep that CPU fed with a steady stream of tasks, it just lost its real-world advantage.

But let’s set technical merit aside. Let’s just look at pure practicalities. You can buy an iMac for $799. Or, if you’re content with a low-end computer, for the same amount of money you can buy a low-end eMachine and pair it up with a 19-inch NEC monitor and still have a hundred bucks left over to put towards your printer. Yeah, so the eMachine doesn’t have the iMac’s glitzy looks. I’ll trade glitz for a 19-inch monitor. Try working with a 19-inch and then switch to a 15-inch like the iMac has. You’ll notice a difference.

So the eMachine will be obsolete in a year? So will the iMac. You can spend $399 for an accelerator board for your iMac. Or you can spend $399 for a replacement eMachine (the 19-inch monitor will still be nice for several years) and get a hard drive and memory upgrade while you’re at it.

On the high end, you’ve got the PowerMac G4 tower. For $3499, you get a 733 MHz CPU, 256 MB RAM, 60 GB HD, a DVD-R/CD-R combo drive, internal 56K modem, gigabit Ethernet you won’t use, and an nVidia GeForce 2 MX card. And no monitor. Software? Just the OS and iMovie, which is a fun toy. You can order one of these glitzy new Macs today, but Apple won’t ship it for a couple of months.

Still, nice specs. For thirty-five hundred bucks they’d better be nice! Gimme thirty-five hundred smackers and I can build you something fantabulous.

But I’m not in the PC biz, so let’s see what Micron might give me for $3500. For $3514, I configured a Micron ClientPro DX5000. It has dual 800 MHz Pentium III CPUs (and an operating system that actually uses both CPUs!), 256 MB of RDRAM, a 7200 RPM 60 GB hard drive, a DVD-ROM and CD-RW (Micron doesn’t offer DVD-R, but you can get it third-party if you must have one), a fabulous Sound Blaster Live! card, a 64 MB nVidia GeForce 2 MX, and in keeping with Apple tradition, no monitor. I skipped the modem because Micron lets me do that. If you must have a modem and stay under budget, you can throttle back to dual 766 MHz CPUs and add a 56K modem for $79. The computer also includes Intel 10/100 Ethernet, Windows 2000, and Office 2000.

And you can have it next week, if not sooner.

I went back to try to configure a 1.2 GHz AMD Athlon-based system, and I couldn’t get it over $2500. So just figure you can get a machine with about the same specs, plus a 19-inch monitor and a bunch more memory.

Cut-throat competition in PC land means you get a whole lot more bang for your buck with a PC. And PC upgrades are cheap. A Mac upgrade typically costs $400. With PCs you can often just replace a CPU for one or two hundred bucks down the road. And switching out a motherboard is no ordeal–they’re pretty much standardized at this point, and PC motherboards are cheap. No matter what you want, you’re looking at $100-$150. Apple makes it really hard to get motherboard upgrades before the machines are obsolete.

It’s no surprise at all to me that the Mac OS is now the third most-common OS on the desktop (fourth if you count Windows 9x and Windows NT/2000 as separate platforms), behind Microsoft’s offerings and Linux. The hardware is more powerful (don’t talk to me about the Pentium 4–we all know it’s a dog, that’s why only one percent of us are buying it), if only by brute force, and it’s cheaper to buy and far cheaper to maintain.

Apple’s just gonna have to abandon the glitz and get their prices down. Or go back to multiple product lines–one glitzy line for people who like that kind of thing, and one back-to-basics line that uses standard ATX cases and costs $100 less off the top just because of it. Apple will never get its motherboard price down to Intel’s range, unless they can get Motorola to license the Alpha processor bus so they can use the same chipsets AMD uses. I seriously doubt they’ll do any of those things.

OS X will finally start to address the technical deficiencies, but an awful lot of Mac veterans aren’t happy with X.

Frankly, it’s going to take a lot to turn Apple around and make it the force it once was. I don’t think Steve Jobs has it in him, and I’m not sure the rest of the company does either, even if they were to get new leadership overnight. (There’s pressure to bring back the legendary Steve Wozniak, the mastermind behind the Apple II who made Apple great in the 1970s and 1980s.)

I don’t think they’ll turn around because I don’t think they care. They’ll probably always exist as a niche player, selling high-priced overdesigned machines to people who like that sort of thing, just as Jaguar exists as a niche player, selling high-priced swanky cars to people who like that sort of thing. And I think the company as a whole realizes that and is content with it. But Jaguar’s not an independent company anymore, nor is it a dominant force in the auto industry. I think the same fate is waiting for Apple.

Mailbag:

My docs; Apple; Lost cd rom drive

Mac emulation and insights

I’m scaring myself. I’ve been playing around with Mac emulation on my PC at home (I can get an old Quadra or something from work for nothing or virtually nothing, but finding space to set it up properly in these cramped quarters would be an issue, especially since I’d have to give it its own keyboard and mouse and possibly its own monitor). My Celeron-400 certainly feels faster than the last 68040 I used, and I greatly prefer my clackety IBM keyboard and my Logitech mouse over anything Apple ever made, so this emulation setup isn’t bad. I’ve got MacOS 8.0 running on my Celeron 400, though on an 040 (especially an emulated 040), 7.6.1 would be much better if I can track down an installation CD for it by some chance.
Of course, there’s the issue of software. A lot of the ancient 68K Mac software is freely available (legally) these days, and it raises the old “Are we better off now than we were 10 years ago?” question. I don’t know. I still think the software of yesterday was much leaner and meaner and less buggy. By the same token, programs didn’t necessarily work together like they do today, and the bundles of today were virtually unheard of. Software ran anywhere from $99 to $999, and it typically did one thing. More, an outliner from Symantec (not to be confused with the Unix paging utility), made charts and outlines. That was it. And it cost around $100. The functionality that’s in MS Office today would have cost many thousands of dollars in 1990. Of course, the very same argument could be made for hardware. You couldn’t get the functionality available in a $399 eMachine for any price in 1990–there were very high-end machines in 1990 with that kind of CPU power, of course, but the applications weren’t there because you don’t buy a supercomputer to run word processing.

Messing around with this old Mac software gave me some insights into the machine. One of the freely available packages is Think Pascal. In high school, we did computer applications on Macs and programming (at least the advanced programming classes I was taking) on IBM PCs. So I know Pascal, but this was my first exposure to it on the Mac. Reading some of the preliminary documentation on programming a Mac in Think Pascal gave me some insight into why the Mac has (and always had) such a rabid following. I don’t really find the Mac any easier to use than Windows (and there are some things I have to do that are far easier in Windows) but I won’t deny the Mac is a whole lot easier to program. Implementing “Hello, World!” in Think Pascal on a Mac is much easier than implementing it in C on Windows, and the Think Pascal version of “Hello, World!” makes more sense to me than even the Visual Basic version of “Hello, World!” on Windows. It’s more complicated than the main() { printf(“Hello, World!\n”); } you would use in DOS or Unix, but if you use all available tools and put the dialog boxes and buttons in resources it’s not much more complex, and programmers can rough in GUI elements and get on with the code while they shove the GUI elements off to artsy people, then it’s easy to use ResEdit or another resource editor to put the final GUI elements in.

And, bite my tongue, it would appear that programming the Mac was easier than programming the Amiga as well. I wrote plenty of command-line tools for the Amiga but I never mastered the GUI on that platform either.

I’m not saying anyone can program a Mac, but having attempted unsuccessfully to learn how to program effectively in Windows, I can say people who wouldn’t program in Windows can (and probably do, or at least did back in the day) program the Mac. My friends Tom Gatermann, Tim Coleman and I stand no chance whatsoever of being able to develop a decent Windows app, but we would have made a decent Mac development team with Tom and Tim handling the GUI and me writing code and all of us contributing ideas.

The next time I need a computer to do something for me that I can’t find a readily made program to do, I’m apt to load up Think Pascal on a Mac emulator and take a crack at it myself. My simple mind can handle programming that platform, and I suspect some of the innovative programs that appeared on the Mac first may have originally been written by people like me who have ideas but don’t think like a traditional programmer.

———-

From: Robert Bruce Thompson

“I can count on one hand the number of people I know who’ve ever built anything from discrete components, myself included…”

You’re hanging out with way too young a crowd. I’m only 47, and I used to build stuff from discrete components, including ham transmitters, receivers, amplifiers, and so on using *tubes*. You probably wouldn’t recognize a tube if it bit you, so I’ll explain that they were glass things kind of like light-bulbs. They were available in hundreds of types, which one used for various purposes–diodes, triodes, and so on. When they were running, they lit up with an orange light. Very pretty. And they did burn out frequently, just like light bulbs.

And I’ll be that if I were pressed hard enough, I could even remember the resistor color codes.

Geez.

———-

Too young and too lazy. But I do know what tubes are–they’re still used in audio equipment, for one, because they give a richer tone than transistors. And I remember when I was really young, there was a drugstore we used to go to that still had a tube tester in back.

But I remember the eyebrows I raised in high school when I was building something that needed a particular logical gate, and I couldn’t quickly locate the appropriate chip. I had a book that told how to build the gate using discrete components, so I did it. Actually I raised eyebrows twice–once for building the thing that required the chip in the first place, and once for making the chip stand-in.

eMachine upgrade advice

I got some mail some time back about eMachine upgrades that I never got around to posting. I’ll just summarize because that’s easier (it keeps me off the mouse).
First off, definitely look into a new hard drive. You can pick up a 7200-rpm drive of decent size (10-15 gig) for under $100 these days. I’ve had trouble getting Western Digital drives to work with older disk controllers, but no problems with Maxtors, and I get better performance and reliability from Maxtors anyway.

Next, eMachines tend to have problems with their power supplies. Get a replacement from PC Power & Cooling. It’s $45. Cheap insurance. And chances are the hard drive will perform better, since the PCP&C box will actually be supplying the wattage it claims to supply (which may or may not be true of the factory box). And remember: low-cost PCs have always had skimpy power supplies. Commodore and Atari made great low-cost computers 20 years ago, but they had horrendous power supplies. Given a properly made third-party power supply, a Commodore or Atari could run for 10-15 years or more (and often did).

Finally, get 128 megs of RAM in the system somehow. If you’ve got 32, just go buy a 128-meg stick. If you’ve got 64, get a 64-meg stick or, if you can afford it, get a 128.

Since eMachines have pretty wimpy integrated video, you might also look into a PCI video card with a Matrox, nVidia or 3Dfx chipset. Matrox gives slightly better 2D display quality, while nVidia and 3Dfx give better speed with 3D games. If you’re into gaming, definitely look into a new card.

That’s the strategy I follow with any upgrade. Get a modern disk in there, then get more memory, and replace anything else that seems underpowered. Do the disk first, then deal with memory, then possibly the video. Then, and only then, do I start looking at CPU upgrades. I’ve turned 200-MHz junkers into very useful machines again just by adding memory and a fast disk. The CPU isn’t the bottleneck in most systems.