Worst inventions in computing
The sheer number of inventions in the field of computing easily surpasses the grand total of the human inventions outside this field. It is no surprise that some of them backfire and some of them quickly age.
Today, I’d like to introduce you to some of them that are still alive and desperately need to die or evolve. I started with five that irritate people around me the most. I am hoping that by the time you see the name of the last, you think: Well, that’s a huge shock and surprise but judging by the other five, this writer is very much sane. And hopefully, by the time you finish reading it, you feel I have a point there.
I fail to find merit for creating screen savers in the first place. It is ludicrous to try to save a screen by keeping it working when it is not needed, wasting energy in process. Powering it down serves the same purpose and saves energy as well. This power-bill–unfriendly hack might have been an acceptable interim solution before hardware vendors added power-saving technologies; however, it must not have survived to see the 21st century. Hopefully, now that Microsoft is planning to make Windows leaner, screen savers may become a deletion candidate.
Auto-dimming the “idle” screen
Some mobile devices temporarily dim their screens if they feel the user is not using them, to save energy. This logic is faulty at core: A screen is either needed or not. An unneeded screen is best turned off, not dimmed. As for a display in use, either the dimming is pleasurable or not. In case of the former, it must become permanent and in case of the latter, it must never be dimmed.
Devices and operating systems with better designs only dim the screen when the device switches to battery power. This form of dimming lasts until the user instructs otherwise. Screens with ambient light sensor can also adjust their brightness automatically.
“Hidden” and “system” files
Back in the dark ages of computing, when everything was very complex, someone though he can simplify things by adding yet another layer of complexity: Hiding files that were deemed more important than others. MSDOS.sys and IO.sys were essential for system startup so, the rationale was that by hiding them from view, they were safer.
However, experience has proven that the “hiding and lying to protect” policy fails in the long run; every competent mother that has raised a toddler would affirm. Eventually, malicious software came along and abused these attributes to evade detection. Naturally, users learned to look for hidden files and folders and delete them. They deleted every hidden file or folder that they did not recognize, including these legitimate files. (The consequence was that their system failed to start.) Apart from malware, nowadays, hidden and system attribute are nothing short of trouble: A Windows user creates a folder, stores files in it for a while and now wants to delete it. A cryptic message shows up, asking whether he wants to delete a hidden system file called “desktop.ini”; the message warns the end of the world comes if system files are deleted. (Of course, deleting this particular file safe.)
Starting with Windows 7, Microsoft has started storing boot files on System Reserved partitions, which are never mounted in Windows. (Not hidden, just not in the way.) Linux derivatives have been doing this for a long time. So, these attributes have already lost their initial purpose. The time has come for the operating systems to discard these attributes and implement a policy of transparency and tamper-protection.
Scroll lock key
I have been using computers for 20 years and as far as I can remember, Scroll Lock was a redundant key. Occasionally, computer games or exotic system utilities use it but no one misses this key. Architectures younger than this key have come and gone; the fact that it is neither used nor removed shows no one really cares about it.
Plain text files (.txt)
There was a time when information interchange between computers was so limited that it did not matter if two computers in the same room were totally incompatible. It was in those dark times that files containing text were created. But how were they created? At storage level, computers see things as “bytes”. A byte is an atomic unit of storage than can have 256 different states. So, computer programmers thought: It is like Morse code, only instead of the dot and line states, we have 256 states. We can assign a character (a letter, digit, symbol or glyph) to each state and so files can contain text.
Except not all the developers in the world checked with each other. One assigned state 1 to letter “a”; another assigned 1 to “A”; yet another thought “one should be one” and assigned state 1 to digit “1”. As a result, when text files went from one computer to another, they became junk and remained junk until they returned to the computer that created them. Eventually, national standards were created which ensured that standard-compliant computers assigns the same character to the same state and can exchange text files. Ultimately, Internet, which is by no means bound to a single nation, brought about Unicode, a standard that assigns a storage state to all existing characters of all languages of the world. But since the Internet is not governed by any nation, national standards survived along Unicode.
However, a huge mistake has gone completely unnoticed: Plain text files contain text that adheres to some standard, but there is no telling which standard! It is like you trying to read a text without knowing in which language it is written. Nowadays, computers resort to heuristics but still the process of parsing a plain text file is hit and miss. Older unpatched versions of Windows before Vista used to have a logic problem in parsing text, known as “Bush hid the facts” bug: Opening a plain text containing “Bush hid the facts” (without newline character or quotation marks) in ASCII encoding in Notepad results in Notepad displaying the nonsensical “畂桳栠摩琠敨映捡獴“. (Of course, if you are a hardcore fan of Bush, you would be interested to know that any sequence of characters with similar space distribution triggers the same bug; e.g. “this app can break”, or “
aaaa aaa aaa aaaaa“.)
Fortunately, text files written in Unicode are exempt from this problem but Windows, as of yet, does not have a Unicode-friendly command-line interface. (Even PowerShell, in spite of being derived from .NET Framework, is very Unicode-unfriendly.)
E-mail is nowadays an indispensable part of the modern life.
So is spam.
The problem is that e-mail could have been invented in a way that spam was impossible, only if the major sponsors of e-mail had not forgotten about postage stamp’s history.
Before the invention of postage stamp, recipients had to pay to receive their mail. So, the recipient could easily deny payment for a letter if he or she was not interested, incurring irredeemable expense on the postal system. (But I am speaking hypothetically; the reality back then was far more complex.) E-mail is just like that. The e-mail service at the recipient side must maintain a server computer or server farm that accepts and stores incoming mail. The burden of being ready to receive mail is on the recipient’s service provider. The sender could be anyone with a cheap computer; he or she could send millions of e-mail to any address. Or at least, this was the case at the dawn of e-mail era. Gradually, anti-spam software relying on heuristics were employed and sender identity infrastructures were invented. But still, no one thought that the burden of hosting contents of the message must be on the sender.
Posted on 20 May 2014, in Computers and Internet and tagged dim, e-mail, hidden, human inventions, malware, mojibake, power saving, screen saver, Scroll Lock, spam, system, text files. Bookmark the permalink. 2 Comments.