User talk:Galileo

Ulam_spiral_(for_primes)

Seeing your {{trans|Phix}}, I reviewed my code and removed the spurious definition of s, in both the Phix and Ulam_spiral_(for_primes)#Yabasic entries --Pete Lomax (talk) 14:13, 30 December 2018 (UTC)

Hello, again

I just stumbled on phixmonti and wanted to say: well done, makes me feel like a proud grandparent.
I haven't done anything with it yet, but let me know (right here on your talk page, or mine) if you ever give it a proper GUI or get anywhere writing a transpiler to javascript (I might well be interested in helping out) or otherwise get it running in a browser.

You have probably already seen it, but in case you want more speed, investigate the toy compiler at Compiler/code_generator#Phix - as mentioned on Compiler/virtual_machine_interpreter#Phix there is potentially a 500-fold performance improvement right there. --Pete Lomax (talk) 21:57, 6 January 2020 (UTC)

Hello, again

Hi. I'm glad you liked Phixmonti.

In answer to your question, I created Phixmonti because I found Monti interesting, but when I contacted its creator, he informed me that he had abandoned it. So I made my own version of the language, with the modifications I wanted to make. I chose Phix because it was fast and it suited the project very well.

As for the compiler, thanks for the warning, because I didn't know it existed. I'll take a look at it.

I haven't thought about converting Phixmonti to Javascript (although it would be nice if it could be ported to Phix code).

As far as providing a graphical interface, what I'd like to do is provide it with basic graphical capabilities so that it can draw with simple primitives (lines, circles, polygons, ... with or without fill, color commands, etc.). But this is not a priority.

I really created Phixmonti firstly for the fun of creating a programming language, and secondly to use it for programming in a different way than I used to do in Yabasic. That is, to program using new ideas and solutions, abandoning the old ways.

This just popped up in my inbox (because of the addition below), so 16 months later I thought I would quickly let you know that I am in the final stages of writing a transpiler for Phix to JavaScript so things can be run in the browser/online, and a simple canvas as you described seems to work a charm. Do you still do anything with Phixmonti or Klingphix? --Pete Lomax (talk) 12:48, 23 May 2021 (UTC)

Hello, Petelomax.

From time to time I solve some Rosetta Code problems, but at the moment I don't continue developing either language. As I told you, I did it for fun, with no major pretensions. In fact, with Klingphix, I focused more on realizing my ideas than on optimizing its performance (it is very slow).

Improve Klingphix Sieve of Eratosthenes

In your recent addition of a Klingphix implementation of the SoE (http://rosettacode.org/wiki/Sieve_of_Eratosthenes#Klingphix), it isn't exactly SoE as you use multiplicative indexing as ( 2 $limit 1 - $i / int ) [ $i * false swap set ] for; better would be to use additive indexing, which is more true to the algorithm, as ( $i dup * $limit $i ) [ false swap set ] for. This may actually be faster and it also uses the accepted optimization of starting the composite number culling at the square of the base prime, since lesser composites have already been culled by lower base primes. Regards - GordonBGood (talk) 21:10, 22 May 2021 (UTC)

Thank you for your comment. I was surprised that anyone was interested in looking at the examples in Klingphix. I would like to know your honest opinion on the language concept, if you would be so kind as to give it to me.