Dark Age Of Computer Programming

It will end on the day when http://www.dictionary.com/search?q=unit%20test yields a result. It certainly will not end before http://wombat.doc.ic.ac.uk/foldoc/foldoc.cgi?query=unit+test&action=Search exists either. And so will begin a NewAgeOfComputerProgramming?.

The latter now exists.

Sorry, I did not mean to be negative. I just wish that everyone in the world knew about unit tests and recognized them as an essential part of any software. So when a software developer delivered a program w/o unit tests the customer would say 'where are the unit tests? This software is not complete.' Few people would purchase a car today without seatbelts or other safety features.

Detroit probably begs to differ; the reason that autos come with many types of safety gear is that the government requires it. Automakers have long refused to provide such equipment otherwise, with a few notable exceptions (like Volvo) who have cultivated reputations based on safety. Most automakers want to make their cars as cheaply as possible. Likewise, many consumers don't necessarily view safety equipment as a value-add.

Many other ComputerScience terminology is not to be found in a general-purpose dictionary....why should UnitTests crack the lexicon of laypeople, and what does that have to do with the sun setting on a hypothetical "dark day"? The world world could know what UnitTests are, and we'd still have buggy software that's over budget and late.

UnitTests are damn useful. But like everything else they are NoSilverBullet.

Take for instance the railroad industry which ignored telegraph for over 30 years before using it to improve safety on trains. Public outcry forced the change after some nasty (and preventable) wrecks. Where is the public outcry for unit tests?

Spread the word!

P.S. I went to UCSD and they do not teach UnitTests. They still teach BigDesignUpFront. Maybe the first thing the community needs to do is get XP into schools.

According to my fiancee, the University of Florida has started (at least since summer 2004) talking about UserStories and UnitTests in at least some of its software engineering courses.

I'm not sure what they teach at college these days, but in my day it was commonly taught that UnitTest (or some form of component test) followed directly on from implementation. In the better courses, they even suggested that you could design unit tests on completion of detailed design. I'll never understand why the software industry has to keep learning lessons again and again.

Of course XPers advocate that UnitTest precedes implementation.

Clarify please why a common English dictionary should contain entries for a two-word technical term? A better link is http://www.google.com/search?q=%22unit+test%22

True: I tried it for http://www.dictionary.com/search?q=reasonable%20argument, and did not get a result. Your point is well made. However, if you try the same and hit the radio button for thesaurus.com, you will get thousands of results by exercising each of the listed sources.

This is another of the pages that seem to be posted lately by a few souls seeking an audience for the negative. This is however a PositiveDialogueCommunity, and most discussions worth reading are in a positive vein. If this is truly the DarkAgeOfComputerProgramming, perhaps its TimeForaParadigmShift, since we are dealing with things in the RiverOfTime.

There are many who have said that our computer industry is "immature" relative to other "science" or "engineering" disciplines, as we've only really been doing computers for about 30 years or so.

They're right.

So why are people so surprised to find that we've found a dramatically more effective way of doing the work? Why are they so surprised that we don't want to continue doing the work "the way we've always done it?" This is exactly what a 'sensible person' should expect in an "immature industry" - change, significant change - often for the better. Yet, it still seems to catch so many people by surprise. ;-> -- JeffGrigg

(Why are you so surprised that people react this way?? ;)

NotWidelyAcceptedPractice does seem to be a rather weak argument, especially in a developing field.

Is NotWidelyAcceptedPractice ever a strong argument? (take the train industry example above as a reference when answering this question).