WTVN Radio • Columbus, Ohio • Sunday morning from 8 until 9

Is this information useful? If so, consider
making a contribution, please.

October 3, 2004

Random thought:

What’s wrong with this @%#!! computer?

I suspect that computers have increased both the quantity and the quality of swearing throughout the world. When something goes wrong with a computer, the explanation is often somewhat unintelligible, particularly if you're not a computer geek. And maybe even if you are.

We've progressed beyond error messages like "00110101" or "Bad!" Yes, at least one application had an error message that consisted of the single word "Bad!" Now we're treated to "This program has unexpectedly quit. Close," (Mac) or "Unable to continue. OK?" (Windows).

I started thinking about computers, how they interact with us carbon-based life forms, and what we expect of our logical devices after an acquaintance, annoyed by all the steps necessary to ensure a reasonably safe upgrade to Windows XP service pack 2 wrote, "We're not being served at all well by the software industry if it has come to this. Why should we have to reload antivirus software? Why should we have to check in with motherboard sites? Why should we have to worry about the BIOS? How can anyone think that this is acceptable?"

The easy, smart-ass answer is that you don't have to do any of those things any more than you have to buy fire insurance for your home, take an umbrella along when there's a 90% chance of rain, or look both ways before crossing the street. It's just that doing those things increase your chance of having a successful experience.

When I hear or read comments like these, I invariably begin to wonder how the speaker or writer would have dealt with life in a covered wagon crossing the prairie. "We shouldn't have to float our wagons across these rivers! Where's the bridge?"

And what about the motorists who drove early model automobiles? You may feel it's a stretch, but I see similarities between early automobiles and today's computers.

I imagine there were people in 1910 who complained about the need to be a mechanic if you wanted to drive more than a mile or two, that cranking an engine the wrong way could result in a broken arm, that some cars had to be driven in reverse up hills because the gas tank was in the back and the gas line was a gravity-feed arrangement that failed if the gas tank was lower than the engine, or that the motorist had to replace tires at the side of a muddy road after running over a twig that punctured the inner tube. And the roads! "Good" roads were paved with gravel. Most roads were dirt.

Today's cars are easier to use.

How does this relate to the PC?

The personal computer is about 20 years old and these devices are a pain in the (choose your body part). Computer technology is moving a lot faster than automobile technology, but it's still no further along than cars were in the 1930s. Back then, motorists had to have their cars lubricated after every rain storm. A careful driver might be able to eke 50,000 miles out of a car before having to replace it. Any head-on collision was almost certainly fatal because the steering column tended to impale the driver.

But 1930s motorists continued to buy cars and, over time, manufacturers found ways to improve them. Brake systems were converted from mechanical to hydraulic and then the hydraulic systems were doubled so that a single failure wouldn't leave the motorist without a way to stop the car. Advancing technology made it possible for an engine in one of today's cars to run with little trouble for 150,000 miles or more. Collapsible steering columns, air bags, ABS brakes, and hundreds of other components make today's cars safer.

But none of this happened overnight. Look at a car from the 1920s and you'll consider it primitive. Now look back at a computer from 1983. Primitive? You bet!

My first computer required me to swap cartridges to run various programs. DOS was a big advancement; I could close a word processor and open a spreadsheet without having to reboot the computer. But I couldn't use both programs at the same time.

Microsoft and Apple developed graphical operating systems based on work done by Xerox in Palo Alto. OS/2 was a stronger version of DOS. Unix (and eventually Linux) came on the scene. Microsoft developed a true multi-tasking operating system with NT and then expanded it with XP. Apple's OS X is powerful, but it's still too hard to use. Microsoft's next generation operating system is in development now and we'll see it by 2007 if the development doesn't get bogged down.

Give hardware and software manufacturers another 20 years and computers will be capable of diagnosing and fixing some of their own problems, just as your automobile's engine can today adjust itself. Until the hardware and software reach that level of complexity, you can choose to take advantage of the benefits a computer brings to your life and live with the annoyances or you can choose to forego the advantages a computer might provide because you prefer not to deal with the distractions.

Making sense of computers

Most of the information people need to operate a computer these days is reasonably approachable by any intelligent person who doesn't choose in advance to be unable to understand what's being said and actively tries to understand. I say that with no intent to offend. I can't tune up a lawnmower engine and I like to tell myself that I just don't "get" engines. But a lawnmower engine isn't a particularly complex device. I could undoubtedly learn to service mine, but I haven't.

It's the same for those who have decided they can't learn how computers work. You don't have to understand the technical manuals; all that's required is a willingness to spend some time with some of the popular computer magazines, books such as Osborne's "Quick Steps" series, or any of the hundreds of websites that explain concepts in plain English.

You won't be a computer expert, at least not right away, but you will gain the knowledge you need to fix many things when they break and to understand what a technician is telling you when you encounter a problem that you can't fix. Or you can choose to hire a technician to take care of even the smallest problem.

You may be surprised to learn that I have no technical background, but I've read a lot of books, I've taken classes, I've attended conferences, I've asked questions, and the pay-off is that I'm able to take care of the computers I have to deal with every day. You can, too. Honest.

If you subscribe to my Random Thoughts newsletter and this entire topic seems more than just vaguely familiar to you, it's because I sometimes plagiarize myself.

Semi-useless information
about a totally useless concept

A client asked me recently if 96dpi is a good resolution to use for images that will be used on a website. My response said that dpi doesn't matter with images that will be shown on the screen. DPI is a measure that's used only for print applications.

His response:
J> Not the question.
J> Question is, is 96 dpi a valid resolution for web JPGs?

To which I replied:
The answer remains the same:
DPI has NO effect whatsoever on the screen.
The screen is hard wired.
Screen resolution is whatever it is; it varies from screen to screen.
The ONLY thing that counts is pixels.
A pixel is a pixel is a pixel.

So let's talk about that. This is one of the most perplexing subjects for graphic designers and for people who aren't graphic designers but who must create images for use on the Web anyway.

Consider ...

The monitor on my desktop computer has a viewable area that's 13.75 inches across (not "diagonal" -- that's a whole 'nother story). It runs at a resolution of 1280x1024 -- and that works out to be 93.09 "dpi". My notebook computer runs at 1600x1200 resolution. The screen's viewable width is 11.75", so its "dpi" is 136.17. I can export the video from the notebook computer to a projector and display the 1600-pixel-wide image on a 20-foot-wide screen. That's 240 inches across, so my "dpi" would 6.667 on the projector screen, yet I would be looking at the same image on the notebook's screen. Does the "dpi" change from 136.17 to 6.667 when I glance up from the computer and look at the projection screen?

Yes, it does. And that's one good reason why it doesn't matter.

This is a concept that difficult to comprehend, particularly for people who have any experience with print. In print, dpi matters. It's a lot more complicated than my explanation suggests, but generally an image with a higher dpi will reproduce better than one with a lower dpi. You also have to take into account the size (in pixels) of the original image (or the original scan), whether the reproduced image is larger or smaller than the original, the number of lines-per-inch the output device can resolve, the printing process used, the quality of the press, the kind of paper used, and even the printer's knowledge. But in general, high-quality printing needs a high-dpi image.

On the screen, it doesn't matter and I try to convince people to just ignore the dpi setting in their software. When this fails (and it almost always fails because the assumption is that the software makers put the setting there for some good reason and either I'm too dumb to know what it is or it's some secret that you must be part of a clandestine organization to learn the real explanation) I tell people to just set dpi at 100. That at least makes the math easy.

You've probably heard Windows users say screen resolution is 96dpi or Mac owners claim 72dpi. Maybe those numbers were accurate once -- for one size screen at one resolution. Today it's nonsense.

Careful, this is a trick question:

How big is an image 600 pixels by 800 pixels at 100dpi?
How big is an image 600 pixels by 800 pixels at 300dpi?

They are the same size. Both images contain 480,000 pixels worth of information. If they're the same size dimensionally (600x800) and the same size by content (480,000), then what's the difference? The difference is some information in the header (which in TIFF files* is actually at the bottom of the file) that says "Somebody told me that I should be printed at 100 dpi, but it doesn't matter much to me what you do with that information because I know I'm just a bunch of pixels."
*I consider "TIFF file" not to be redundant even though "file" is included in the acronym. TIFF is an acronym for tagged image file format. Your mileage may vary.

So the whole resolution thing is little more than a canard.

You're still not convinced, are you?

Then look at the images below. One has a "dpi" of 100 and the other has a "dpi" of 1000 -- TEN TIMES MORE, but the images are the same size and clarity. In print (if the application used to print paid attention to the dpi figure, which many do not, and if the image had not been scaled up or down) the images would be different sizes. Which would be larger in print? The one with the LOWER resolution. Which would be clearer in print? (Probably) the one with the higher resolution, although if the resolution is too high, the clarity will actually begin to degrade.

		In print, this image would be 3" wide.*
		In print, this image would be 0.3" wide.*

Note that the dimensions are the same, but the size is slightly different. That slight difference is probably (I'm guessing) the result of the different colors. When GIF images are created, the process throws away some of the color information. The files are similar, but not exactly the same, hence the slight difference in file size.

If you did the math from the earlier example, you may be wondering why this file isn't 9000 bytes (300x30). That's because it's a GIF and some of the information is thrown away to create a smaller file size. But even if you compare the original files (created by Macromedia Fireworks) you'll have two more questions: Why are the files so much larger than 9000 bytes? and Why aren't the files the same size?

I can answer the "so much larger" question: Fireworks retains quite a bit of additional information within the PNG about the original file, the file name it has been output to, the location of the output file, the palette used for output, and such.

I can guess at the "same size" question. It's probably because of the difference in output palette, but nearly everything in the PNG file is binary. The only human-readable content is Macromedia's copyright information, so I can't really say for sure.

Although it's far too late for this, to make a long story short: If you're a graphic designer who's working on a print ad, you'll need to be concerned with dpi. If you're a website designer who's working on something that will be viewed only on the screen, your only concern is pixels.

Nerdly News

TechXNewYork is this week ...

... and I won't be there for the first time in about 20 years. While it's true that the show has been smaller since 2000 and particularly since 2001, that's not the reason that I'm sitting out this year's show. Well, not "sitting out" exactly. Instead of heading east, I'll be flying west to PowerPoint Live.

TechXNY is still an important show. It's the largest East Coast show. But it's a mere shadow of its former self.

In the late 1990s, TechXNY (then called PC Expo) seemed capable of sucking all the taxis in Midtown over to the Javits Center. The show occupied nearly the entire building, and Javits is huge. The main floor was full. The ground floor was full. Most of the meeting rooms were occupied. Even most of the 3rd floor, which also housed the media center, was full. At its height, PC Expo pulled half a million visitors to see more than 1000 vendors.

Since the dot-com implosion and particularly since 9/11, the show has shrunk. Last year's TechXNY was quiet. Sedate. Not what I've been used to since the 1980s. There were maybe 300 vendors and 100 of those represented companies that wanted to outsource jobs from the US.

OutsourceWorld is a big component of this year's show, too. Instead of nearly a week, the show is 3 days. In the past, the website has offered a list of exhibitors. This year, there's nothing. Will 1000 exhibitors be there? 500? 150? 75? There's really no way to tell.

Last year, many were suggesting that this year's TechXNY would be cancelled. It's going forward as scheduled, but with how many visitors from outside the New York City area? With how many exhibitors?

For information (the show begins on Tuesday), visit http://www.techxny.com/.

Real digital cameras finally hit the $100 barrier

You may remember that several years ago the leaders of the digital camera industry were predicting 1-megapixel digital cameras at $100 within a few years. They were right. But they were also wrong.

What you get for $100 is not the latest Canon or Nikon SLR, but it is a THREE (not ONE) megapixel camera with few features.

Priced at $99.99 (which is a dumb marketing trick to make people think that it's under $100), the Concord 3043 is in a case less than 1" thick. There's no zoom lens, but Concord offers "digital" zoom. Forget it. Digital zoom is worthless. It's a canard developed by the marketing department. All digital zoom does is crop the image, render it larger, and make it fuzzy. Just take the picture without using digital zoom and then crop with your photo editing program.

The Concord 3043 comes in red, blue, and silver. It has 16MB of internal memory and a SD card slot. If I were developing a camera, I probably wouldn't use SD because the memory costs quite a bit more than either SmartMedia or CompactFlash. The camera has a 1.8” color TFT LCD. It also provides a "movie mode" at VGA resolution up to 15 frames per second with sound.

Given the price, you probably already know the sensor is a CMOS unit, but that's not the kiss of death it used to be. Canon's low-end SLR also has a CMOS sensor.

For more information, see www.concord-camera.com/. This camera may not yet be on the site, but it's coming soon.

Let us know what you think about this program! Write to:
Bill Blinn --
(wtvn@blinn.com still works)
Joe Bradley --

Joe
(Photo by Sally)

Bill
(Photo by Scampi)

Privacy Guarantee:

I HATE SPAM and will not sell, rent, loan, auction, trade, or do anything else with your e-mail address. Period.

This is the only ad you'll ever see on this site. It's for my website host, BlueHost in Orem, Utah. Over the past several years, they have proven to be honest, reliable, and progressive. If you need to host a website, please click the banner below to see what BlueHost has to offer.

TechByter Worldwide receives a small advertising payment for each new client that signs up with BlueHost but I would make the same recommendation even if the affiliate program didn't exist. (If you don't see a banner ad above and you would like to know more, this link takes you to BlueHost.)

As if you didn't already get enough weather on the radio!

If you do not see a Weather Underground banner above and you use ad-blocking software, please set your application to allow images from "www.wunderground.com" to appear.

Annoying legal disclaimer
My attorney says I really need to say this: The Technology Corner website is for informational purposes only. Neither Joe nor I assume any responsibility for its accuracy, although we do our best. The information is subject to change without notice. Any actions you take based on information from the radio program or from this website are entirely at your own risk. Products and services are mentioned for informational purposes only and their various trademarks and service marks are the property of their respective owners. Technology Corner cannot provide technical support for products or services mentioned on the air or on the website.