April 17th, 2013

Today’s rant is about computers in general: they’re too damn difficult to use. By this I mean all computers of all kinds: desktops, pads, and smartphones. Here are a few quick examples:

Email setup
Setting up a mail program to get email from an account should be trivially simple: you give the computer the name of the account (, the password, and a label for the mailbox (“Business”). That’s all the information that should be needed to get the job done. It then figures out everything else. But no, that’s not how it works. In the real world of computers, you have to specify the incoming mail server (which is always “”) and the user name, which never seems to be consistent. Most of the time the user name seems to be the same thing as the email address (“”) – so why do we have to re-enter it? But sometimes I’ve been required to enter just the naked name (“user”). What gives?

But we’re still not done burdening the user; the mail program demands to know the outgoing mail server (SMTP). Does anybody know what SMTP means? I sure don’t – so why are they pushing this meaningless acronym onto me? Just to intimidate me?

We all know the answers to these questions: because programmers have set up email systems to be big and complicated so that they can do all sorts of big, complicated things with them. But there are now billions of people using email and they don’t need all that big, complicated crap. But the world of computing is so screwed up that we retain these archaic methods.

Worse, it doesn’t work! Setting up an email account has always been a pain in the neck; I can remember only a few times when everything went smoothly. Recently I spent two hours trying to get my wife’s iPhone to get mail from her standard email account. She has no problem getting it using her desktop computer. But for some reason the iPhone refused to accept exactly the same information that her desktop computer uses to access the email account. I tried every plausible variation on the entry parameters. I even went into my ISP account and changed the password to be absolutely certain that I was using the correct password. I never got the email working. What a skunky!

Saving mail on the server
My wife has a special account for her business email. Recently she received a notice from her ISP that her email box was full; she had too many emails in it. At first I found this hard to believe, because emails take up so little space. But in her work as a tax preparer, she receives lots of scanned documents from her clients, which really gobble up space.

In a sane world, a mail server would delete emails after they’ve been delivered. But no, we want more power, so we’re going to make it more complicated. The mail server will not delete emails until the user instructs it to. But why do we give the user this option – who needs it? Why in the world would a reasonable user ever want to save emails on the server? Once an email has been received by the email program, the user has a copy. If the user is backing up the computer, there’s even a backup of the email. But no, we want another copy, just in case. Just in case of what? The computer is stolen or destroyed? That problem is not restricted to emails; it applies to all the user’s data, and so is addressed with standard backup systems. Why would anybody need two backups?

The counterargument is that a user might want to access the same email from two or three different devices. Nowadays, people have multiple devices with which to access their email, so deleting an email upon delivery would prevent access from multiple devices. I confess that this counterargument has force, but there are better ways to do this. The obvious way is for the user to have a cloud service that can hold whatever data the user wants to share between devices. Let’s face it, most of the emails we get aren’t important enough to require access from multiple devices; we read it and toss it in 30 seconds. What’s wrong with requiring the user to mark selected emails for inclusion in the cloud? Another possibility is to register each device with the email server; it doesn’t delete an email until each of the registered devices has received the email. But no, we have to keep things messy.

These should be standard on all computing devices. Apple has done a pretty good job with its Time Machine, which makes backup and recovery about as simple as can be done. Microsoft, on the other hand, apparently has failed to make backups transparent enough; my wife’s business partner recently lost data when her computer was taken over by malware and had to be wiped. Which brings me to:

Vulnerability to malware
We take this vulnerability for granted, but the only reason why computers are vulnerable to malware is that they’re too complicated. It boils down to the astounding fact that operating system designers don’t understand their own creations. After all, how could OS designers leave a vulnerability in the OS if they actually understood how it works?

The underlying problem here is a combination of greed and hubris. The people who make operating systems want their creations to have the latest, greatest, most powerful feature set in the universe, so they reach further than they can truly grasp. They always overestimate their own programming skills and so end up creating software with holes in it. Wise designers would be conservative enough to confine their efforts to what they know will work. But in the computer world, wisdom always plays second fiddle to technical prowess.

One other thing: there’s absolutely no reason why the Internet should be insecure; it’s just tradition. It would be a simple matter to alter the protocols to insure that every packet can be traced to its true source. Defenders of Internet freedom denounce all such schemes, on the grounds that they would remove anonymity from the Internet, and anonymity is important to its power. That’s true, but anonymity is also necessary for Internet crime. Freedom of expression is a human right that should be protected by governments, not the Internet. I realize that many governments deny their citizens freedom of expression, but the Internet cannot be used as a vehicle for freedom without simultaneously being used as a vehicle for crime. Given the billions of dollars that Internet crime is already costing us, prudence requires us to close off the criminal opportunities. If we don’t, the problem will continue to grow. How serious must it become before we do something about it?

I’m sure that there are lots of other bad aspects of computers that I can’t see because I am inured to them.