How did you find newLISP?

[Fanda]

I am being curious, how members of this forum found newLISP (except Lutz ;-)



The first time I found it, was here:

http://shootout.alioth.debian.org/">http://shootout.alioth.debian.org/



Website doesn't show newLISP benchmarks anymore. It used to say something like a "LISP implementation with Tk built-in".

At that time, I was looking for a powerful language with Tk interface to do explorative programming. I also wanted to learn LISP (everybody was saying, how cool it is), so it was a good match.



I feel that there are people looking for a language like newLISP and many are pretty happy when they find it. Can we make newLISP more visible?



Fanda

[Jeff]

I was learning LISP using clisp and sbcl and getting incredibly sick of the amount of work it takes to do a simple regex in most ansi lisps (since ansi lisp doesn't have any regex and you depend on the implementation or 3rd party libs).  Lisp traditionally looks down on regex.



I found newLISP when looking for a lisp with a more modern interface and set of built-in functionality.

[cormullion]

I am being curious, how members of this forum found newLISP (except Lutz ;-)

if you really want to know, here's my take http://unbalanced-parentheses.nfshost.com/index.cgi?view-post-id=20060411043500">//http://unbalanced-parentheses.nfshost.com/index.cgi?view-post-id=20060411043500

I feel that there are people looking for a language like newLISP and many are pretty happy when they find it. Can we make newLISP more visible?

I'm doing what I can, and so are you :-)! I think a boost could result if it became a scripting language for some major application. But this requires coding skills I don't have... :(

[newdep]

I was seeking at that time (as always) for an

alternative programming language that both gramaticaly-eyed nice

and included networking..  (Im stuck since then to newlisp ;-)

[rickyboy]

I think I saw it for the first time http://wiki.alu.org/Implementation">here, but I can't be sure.

[m i c h a e l]

It seems so long ago now. Let's see if I can blow the cobwebs from my mind and remember.



I was attempting to use OCaml again after a year of using Ruby exclusively. I was once again getting bogged down with OCaml's complicated object system and decided to give up and begin looking for another language.



I decided to go with a language from the Lisp family, and after test-driving a couple of the other command-line-oriented scripting Lisps, I found out about newLISP from ______. Blank. This means I can't remember exactly where I first saw a reference to newLISP (sorry). I printed up a copy of the manual and began to study it in earnest. My initial superficial understanding of newLISP led to much misunderstanding and confusion. I was close to giving up, when someone made a ridiculous comment about the quality of the writing in the manual. This compelled me to rush to newLISP's defense, so I decided to stick around and try to help out. Getting involved has probably been the single biggest reason for my staying with newLISP. I'm still struggling without a true object system, but Lutz's recent macro gives me hope that one day, newLISP can be used to do OOP as easily as it currently does FP.

I feel that there are people looking for a language like newLISP and many are pretty happy when they find it.

I think so, too.

Can we make newLISP more visible?

This is something I have thought about a lot. Far too many interesting and worthy languages get thrown onto the junk heap of history. In many cases, it was simply because not enough people were made aware of them. This, of course, is called "marketing."



Every post by cormullion on (newLISPer) or Jeff on Artful Code helps with this awareness. Each new GS program written by Norman or newLISP-wiki-based website like Fanda's helps make newLISP more visible. Hans-Peter's NeoBook work. Dmi's Russian introduction. Peter's code snippets. Ryon's fan club. Everyone creating something from their passion for newLISP, contributing to the overall work of making this melodious language heard above the din.



m i c h a e l

[pjot]

I also was looking for a language which included everything. Norman pointed me to newLisp, since then I am using it for most my programming ;-)



Peter

[HPW]

As an longtime autolisp programmer I was always searching for lisp-alternatives running outside autocad but fairly compatibel with autolisp.

At that time (around 2001/2002) I evaluated every lisp I get my fingers on. (xlisp,corman-lisp,Lispworks etc.)



One day I ran across newLISP and the old forum.

I start learning and asking for features and Lutz was so kind to add much of the stuff (from setq to infix.lsp etc.)

The most important thing for me was the release of newLISP.dll which allows the easy embedding in other enviroments.

Building around newLISP.dll I worked on the integration into the windows authoring tool neobook

which grow to a powerfull framework for windows development.

The results out of this framework are very similar to newLISP itself: small, fast, powerfull, single file an self contained.

newLISP is the perfect scripting language for such a GUI tool.



And naturally it fits best to my motto: Use the right tool to get the job done.

;-)

[Cyril]

I think I saw it for the first time http://wiki.alu.org/Implementation">here, but I can't be sure.

This is the my case too.

[Cyril]

The first time I found it, was here:

http://shootout.alioth.debian.org/">http://shootout.alioth.debian.org/



Website doesn't show newLISP benchmarks anymore.

I wasn't aware that they have newLISP some time ago. Funny that I am now trying to re-implement some benchmarks from that site (just for fun and for learning the newLISP way, no intent to post them to the site). Maybe it is a good idea to open a local contest  for these benchmarks implementations here on forum? Writing such a code snippets may be a good way to learn some new tricks.

[cormullion]

A set of benchmarks would be good. Not to compare with other languages - I think that's another thing altogether - but to compare performance of newLISP on different machines, different release versions, etc...



The trouble with benchmarking is simply that the code starts to get obsessed with raw speed, and things like elegance, maintainability, and readability all suffer. Eventually you think "why not use assembler if it's got  to be the fastest.."

[Lutz]

see also here:



http://newlisp.org/benchmarks/">http://newlisp.org/benchmarks/



Note that this was not a subjective selection of tests, but rather offered (almost) all, which could be implemented in newLISP at that time. I remember there were two missing, for using p-threads, which newLISP doesn't have, because it uses UNIX fork().

...but to compare performance of newLISP on different machines...

whatever computer I can get my hands on, I compile newLISP and benchmark it. I am obsessed with speed ;-). I have seen the strangest things happening. Different OS's or a different generation of Intel/AMD type of i386 CPU can turn results on it's head. One has to do a lot of benchmarking, to really get to comparitive results. Because of CPU caches, just running a program a second or third time will skew results. I typically run a program three times and then take the last reading.



Lutz

[notchent]

I recently discovered NewLisp in a link on the Rebol mailing list.  It seems to share many of the features that've made Rebol useful to me (small file size, built-in GUI functionality, cross platform binaries, "batteries included" philosophy, etc.).  I love that it enables and embodies a more mainstream approach to licensing, community involvement, and connection with other common development tools (dlls, sqlite, etc.).  Moving from Rebol to Newlisp seem pretty intuitive so far - I'm looking forward to learning more :)

[newBert]

I was (and I'm still) a fan of Logo and I wanted to learn Lisp, the "forerunner" of Logo, but didn't find a Lisp that satisfied me (neither CLisp, nor XLisp, etc...). So I tried Scheme, a little too complex for me and too "heavy" in its implementation. I tried Python, simpler but still so heavy.



And by dint of searching (Google, etc.) I found NewLISP that I tested out. And then Rebol, very close to Logo, with its limitations which make me come back to NewLISP.



Now I'm using http://www.physik.uni-augsburg.de/~micheler/">aUCBLogo, http://www.elica.net/site/index.html">Elica Logo, http://www.fbsl.net/">FBSL (Freestyle Basic Script Language) and http://newlisp.org/">NewLISP. I abandoned http://www.rebol.com/">Rebol for a couple of reasons and above all to avoid dissipation. I'm just a hobbyist, not a student or a pro (professional programer).

;)

[m35]

Now I'm using http://www.physik.uni-augsburg.de/~micheler/">aUCBLogo, http://www.elica.net/site/index.html">Elica Logo, http://www.fbsl.net/">FBSL (Freestyle Basic Script Language) and http://newlisp.org/">NewLISP.

Thanks newBert! I like checking out these lesser-known languages.

And WOW, Nothing has been clearer than the 5 minutes I spent reading http://www.elica.net/download/papers/TheEasyGuide.pdf">this elica paper. This might finally get me into Logo like newLISP got me into Lisp :)