Guido van Rossum: Before Python | Hacker News Reader

Guido van Rossum: Before Python

neopythonic.blogspot.com

171 pointstathagatadg19 comments

> "But punch cards are the reason that some software still limits you (or just defaults) to 80 characters per line."

Neat. I had always wondered about this.

Visually: http://en.wikipedia.org/wiki/Characters_per_line

Actually, that's debatable. First, most 80 columns cards had the last 8 columns ignored (or used for metadata). Second, it is much more probable that 80 columns limit comes from 80x25 text resolution of EGA/VGA displays.

There were multiple punch card sizes and formats; some were as small as 45 and some as large as 96 columns. Eighty does seem to be the one everyone remembers.

Guido punctuates with more emoticons than a 14 year old girl. :)

I almost laughed out loud at the part "Have you ever invented a programming language? :-)"

Off topic: It annoys me when people go "Oh you don't know this person? But they're so famous!" because not everyone knows all the famous people in every field.

And his use of parentheses is different than I expected:

(blabla :-) vs (blabla :-))

Relevant XKCD: http://xkcd.com/541/

I usually do the second way, but put a space between the emoticon and parenthesis (like so :) )

(cons "lisp joke" here)

To me, he always seemed to be a very jolly person. After seeing and hearing him talk, I'm not very surprised by a lot of emoticons.

I find these old computing stories fascinating. All programming luminaries should consider putting their history into books for us normal nerds to read ;)

It's amazing how many programmers think that programming is going to look roughly the same 50 years from now. They imagine the transition from punch cards to text files was a one time thing... That from here on out it's just going to be more and more refined text languages.

I think history repeating itself, and programming becoming unrecognizably different, is much more likely.

But I've also come to terms that those who lack the imagination to start picturing such things aren't going to believe it until after it happens.

I'm actually looking at taking programming beyond text by moving the control logic to graphs. The idea is based on the concept of Flow-Based Programming:

http://en.wikipedia.org/wiki/Flow-based_programming

It might be a misstep, as graphical programming has been tried before without much success. But I like the concept, and maybe new tools like multi-touch tablets finally make it viable.

Flow-based programming is definitely something I'm (and some people I know) looking at. The coordination language idea is worthwhile for language-independence. And functional languages with lazy sequences are very similar.

Then again, aircraft have looked pretty much the same for the last 50 years. Maybe programming isn't going to change as drastically as it has in the past, when more experimenting was taking place. (I'm saying this semi-tongue in cheek, who can predict these things, really).

"Pascal really had only one new feature compared to Algol-60, pointers." -- Well, records were pretty significant too.

> I realized that a pointer was just an address. Then I finally understood them.

This is quite interesting. It seems to be a universal pattern. Has anybody understood pointers without first understanding memory addressing?

Reply on news.ycombinator.com