Why did we bother with line numbers at all? [close

Question

When you write something in BASIC, you are required to use line numbers. Like:

10 PRINT "HOME"
20 PRINT "SWEET"
30 GOTO 10

But I wonder: who came up with the idea to use line numbers at all? It is such a nuisance, and left quite an "echo" in the developing (pun intended) world!

Answer 1

In the early days, most programs were entered with punch cards. The punch cards were usually entered in sequence, usually one instruction per card, with labels (JMP/JSR targets) being a separate instruction card.

To edit your program, you replaced the card.

Later implementations added an optional sequence number on the right end of the line, so that when/if they got out of order, they could be resequenced by an automated reader.

Fortran used both numeric target labels on the left (col 1-5) and left a reserved block on the right (73-80) for sequence or comment.

When BASIC was initially written, it was decided to move the sequence numbers to the left, into FORTRAN's label field, and to allow overwriting prior cards' memory footprint... as an editing mode. This was intended for the interactive dev environment, but worked just as well with cards. And cards were used in some early implementations for a variety of reasons.

Keep in mind: Many computers were card-reader and printer interface right through the late 1970's. Even tho' interactive mode basics were available, card punched basic programs were frequently used. Since many simply were feeding into the IDE, they worked exactly the same way. Including needing a "Run" card at the end. In such cases, one could simply tack a correction card and another Run card to rerun with a variation on some variable; likewise, in complex programs, simply adding a corrected line of a card before the run was adequate to edit out problems without spending precious time finding the errant card itself.

Answer 2

I find them very helpful when pairing. I don't have to point at a line when my pair has the keyboard, I can just say, "on line 74, shouldn't that really be getMoreBeer()?"

Answer 3

Back in the fifties, when high programming languages were in their early beginnings, there were no terminals, no editors, no monitors (yes, no monitors), just card punchers and readers (for writing and reading the contents of cards into memory of a computer) and printers (for printing results, naturally). Later, tape was introduced, but that's another story.

Each punch card had its own number. There were several reasons for that; from purely keeping them in order, to determining the sequence of execution. Each card was one line of code (in today's terms). Since, at that time, there were no constructs like if..then..else, or whatever variant of the like, the sequence of execution had to be determined somehow. So GOTO statements were introduced. They were the basis of loops. The term "spaghetti code" comes from that time period also, since badly written code was relatively hard to follow, like spaghetti on a plate :)

Answer 4

They originated in FORTRAN, from which BASIC was derived. However, in FORTRAN only lines referenced by other lines (like GOTO targets) needed numbers. In BASIC they had a secondary use, which was to allow editing of specific lines.

Answer 5

It has a loong-loong history.

Line numbering actually comes from Dartmouth BASIC, which was the original version of the BASIC programming language and was the integral part of a so called Dartmouth Time Sharing System. That DTSS had a rudimentary IDE, which was nothing more than an interactive command line.

So every line typed inside this "IDE", and beginning with a line number, was added to the program, replacing any previously stored line with the same number; anything else was assumed to be a DTSS command and immediately executed.

Answer 6

I like the robot church on Futurama, on the walls were written stuff like

10 SIN
20 GOTO HELL

On the Speccy you couldn't edit a line without the line number.