Ancient use of generators

Guido van Rossum reminisces a bit about early discussions of generators in the Python community (read the other messages in the thread as well). I think we talked about the articles he mentions way back when. Earlier still, and beyond the discussion by Guido here, was Icon, a clever little language that I have a soft spot for. i don't think we ever fully assessed its influence on Python and other languages.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Icon/UnIcon - a language still in use today

I don't have a soft spot for the language as I use it for practical purposes today and have used Icon since V6. A language that you can pick up in a matter of days and still be learning the intricacies 25 - 30 years later.

It has made solving many problems over the years easy, particularly in the light of rapidly changing requirements.

It does have its quirks and there are a number of features that I would like to see added to it. If my reimplementation of the virtual machine is ever completed, then they will be.

Have you ever tried to

Have you ever tried to install Icon on an IBM Mainframe running MVS? Well, I did. So there.

Installing on an IBM mainframe?

Hmm? No. But after installing it, did you put it to good purpose? What sort of applications did you end up writing in it?

I found that it was/is excellent for data analysis and data manipulation, better than other systems that I have tried.

I suppose what I found delightful was the ability to completely rewrite a data analysis package that had taken me some 6-8 weeks to create in Turbo-Pascal. 3 days from getting the icon system and and to completing the rewrite. The string analysis functions, the tables, generators, success/fail as a control methodology. The end result was so much simpler and much easier to modify.

I have used it in many of the various positions that I have worked in since. It required so little to install and then start writing my tools. I am no longer employed in the industry (now a full-time carer) but I still do most of my programming in Icon/Unicon.

We never managed to install

We never managed to install it from the tape the Icon project sent us. And we spent a good bit of time trying to get it to load. The goal was to process log files. Many years before that someone actually wrote SNOBOL4 code, but nobody was using it anymore (or SNOBOL in general). I did manage to get the SNOBOL4 to run, which was fun. I was especially when I discovered that it used the DD definitions of the FORTRAN I/O routines. Icon on the mainframe, alas, remained a nice dream.

Honeywell DPS8 front-end configuration file control

at one location I worked at was written in SNOBOL4. I had used it at uni and was surprised that it was used to generate the DSA configuration files. Very effective too. I still have a copy of the Vanilla SNOBOL4 interpreter on my current system.

I am surprised that you didn't manage to get the icon interpreter up and running though. I found that Ralph Griswold and his team to be quite helpful, even though I was only communicating via letter from Australia.

Certainly, both Clinton and Jafar help out in relation to the Unicon version today when questions arise and problems are encountered.

It's been several years

It's been several years since I had access to IBM Mainframes. At the time, we tried everything including contacting the Icon team, IIRC, but eventually gave up.

Question in retrospect

What kind of response did you get back from the Icon team? Did you try the newsgroup as well? I don't recall anything appearing there though. I know it doesn't get much traffic but there is a variety of help there.