(Ugh! I forgot to post this last week. My apologies)
Programming, as I’ve probably said before is a strange occupation. You start with a blank sheet, steal bits and pieces from where ever you can find them and glue them together modify them, add some bits of original (to you) code and try to think of all the possible ways your program can go wrong.
Then you try and break your code (and usually succeed at first). Programming is still very much an art form. Of course things have changed a lot over the years, and we are able to use the work of others to help us in our endeavours, but my first paragraph is still true.
In the beginning there was “Hello World”. This is probably the simplest program that does something visible. It doesn’t take any information in and its output, the words “Hello” and “World” are not very useful in themselves. Actually, I’d say that there is an even simpler program that takes no input, produces no output, and in the process changes nothing. A “null” program if you like.
A programmer writing a new program may well jump in and start coding by grabbing some other code that he or she has access to, but that stolen code was developed, ultimately, from “Hello World” or the null program.
A good programmer is one who steals code from elsewhere and modifies it to do what he or she wants. There is no stigma of plagiarism attached to this process, and it is in fact strongly encouraged that programmers share code. A spoof news item that I came across stated that all programming courses would be replaced with a course on how to find code on “Stack Overflow“. I’ve been unable to find the link again, but I believe that the item was on “The Onion“, a well known satirical website.
Of course, such a process may propagate errors or bugs across many programs, but it is such an effective strategy that it is used more often than not. If code exists to solve a problem then it would be silly to pass it by and write it ones self, maybe introducing bugs to the code. The advantage of “borrowing” code is that while errors and bugs may be in the borrowed code, many eyes will have looked at the code and there is nothing more that programmers like than pointing out bugs in the code of other programmers.
Stack Overflow allows anyone to post code and comment (up to a point), so code posted may not be top quality, but other programmers are quick to jump if they see bugs or inefficiencies in code. Contributors will also point out code which doesn’t follow standards or conventions in the programming language being used. This is considered useful, as the code, if modified, can be accessed and understood more easily, and may often be safer and free of more bugs than unconventional code.
When a program is written it starts out as literally a few lines of code or even an empty file. Any programmer knows that a program grows swiftly and in ways that can’t be foreseen until it may be of enormous size. It won’t be all written in one sitting but is usually written in stages. I personally like to write my programs in very small chunks, building on what has gone before. I think that many programmers use this process, though there may be others who write a sizeable chunk of code before testing it.
Ah, testing! Testing is the less enthralling parts of writing programs. Any program must be tested, to ensure that it does all that is required and nothing else. Generally the program being written doesn’t do all that is required and does things that shouldn’t happen, and initially it is likely to crash or produce cryptic error messages under some conditions.
Testing is supposed to reduce the number of such unwanted happenings, and the programmer may do some rudimentary testing and may handle at least some errors. However the programmer will realise that users who are unfamiliar with how the program is written may well do something that he has not expected.
So clever people have developed ways of automatically testing programs. To do this they have had to write the programs that are used to test programs. And of course those testing programs may have bugs. You can see where that leads to!
When a programmer knows a programming language really well, he is able to literally think in that language. The word “literally” has been devalued in recent time, but I am using it in the true sense of the word. This is hard for some people to understand as they think of language as something like French or Tagalog, and they can’t understand how one can think in a programming language, which is qualitatively different from a spoken language.
An interesting thing happens when a true programmer is programming something. His thought processes become so involved in the process of programming and in thinking in the programming language that he loses track of the outside world. That’s why programmers are whimsically thought to subsist on fizzy energy drinks and dialled in pizza. It is because those things are easily acquired and the programmer can keep programming.
A programmer “in the zone” is so embedded in the world of the program that he or she may often be reluctant to leave that world and respond to irritations like bodily needs and colleagues. I doubt that there is a real programmer who has not surfaced from a deep dive into the depths of a programming problem and realised that all his colleagues have left and it is late at night or very early in the morning. That’s the reason programmers stay after all other people have left – they know that they can slay the current bug with just a few more changes and a few more runs of the program.
The zone has similarities to the state of meditation. While meditation is passive though, programming is an active state. In both cases the person basically disconnects from the world, so far as he or she can, and the concentration is directed internally. Now that I think about it, any deep thought, be it meditation, programming, or philosophising, even playing a sport at a very high level, needs such concentration that much of the world is disregarded and the exponent enters the zone.