Re: Tomorrow is PilCon

[Nehal]

Dear List and Pilcon Participants,

I'm adding here the points that I feel will complete the recap given
earlier by Olaf.

Kindly take notice and correct me incase of any error.

1. Database not required for learners. Lists and symbols are sufficient. As
per Alex, symbols are higher level concept than lists in picolisp.
2. Common lisp is more list centric. When Alex was starting out in 1980's
he was advised to understand and work around symbol more than lisp. At the
machine level, symbol is also a cons pair. But apparently it's not needed
to be understood while programming.
3. Example of shopping list. Apples, bananas really useful. This was done
first by setq making cons pairs in definition. And also later defining a
class and object.
4. Great question by another participant (I'm sorry, not knowing name) to
ask why take numbers in car of cons pair (in example given in point 3). car
in cons pair is good for processing. Lot of functions and function
processing can take place.

Geo, my Org Babel works perfectly with Picolisp, I'll still review link
shared by you.

Thanks to Alex for giving us his time and organising the event. It was
interesting and very helpful.

Also thanks to participants for enabling each other communicate easily with
clarity. :)

Best,
Nehal


On Fri, Sep 4, 2020, 6:05 PM George-Phillip Orais <
orais.georgephil...@gmail.com> wrote:

> Hi Olaf,
>
> Thanks for this recap, this is very nice indeed!
>
> I would like to add the hand-made block diagram by Alex for his
> documentation, very cool! Maybe we can call it pilDiagram and make it as
> official diagram for PicoLisp? Lets see :)
>
>
> Hi Alex,
>
> Thank you once again for sharing your time and patiently answering our
> questions, PicoLisp always surprises me with so many nice features, how I
> wish I have a very good Pil knowledge enough to introduce these hidden gems
> to the outside world... but I still believe someone here or you with pil21,
> it will happen someday soon :)
>
>
> Hi Nehal,
>
> I'm not so sure but you mentioned about org-babel, does this help your
> inquiry: https://github.com/tj64/ob-picolisp
> 
>
>
> Thanks again everyone and stay safe always, have a great weekend! Bis blad!
>
>
> BR,
> Geo
>
>
>
>
> On Fri, Sep 4, 2020 at 8:51 PM Alexander Burger 
> wrote:
>
>> Hi Olaf,
>>
>> > I like to recap some topics/items, perhaps others can also enjoy
>> > (disclaimer: there may be misunderstandings or faults in the notes!)
>>
>> Thanks a lot! This is a very good idea!
>>
>> And all correct I think.
>>
>> ☺/ A!ex
>>
>> --
>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>
>>

Re: Pil vs Pil21 performances

[Davide BERTOLOTTO]

Ok, I managed to turn on the optimization (opt -O3) and the results improved

*** recursive
-- picolisp 3.007 sec
-- pil21 4.249 sec

*** iterative
-- picolisp 0.515 sec
-- pil21 0.368 sec

Apparently with the optimization cache does not segfault...

Cheers,
Davide

On Fri, Sep 4, 2020, 22:50 Davide BERTOLOTTO 
wrote:

> Hi all,
>
> I compiled pil21 with llvm on termux and then I ran some simple fibonacci
> benchmark. Here the results
>
> *** recursive
> n 37 fib 24157817
> -- picolisp 3.013 sec
> -- pil21 5.332 sec
> -- python 21.247 sec
>
> *** iterative
> n 8 fib 4589178984+
> -- picolisp 0.518 sec
> -- pil21 0.409 sec
> -- python 0.124 sec
>
> I got also a segmentation fault while trying to use cache to speed up the
> calculation with pil21 (don't know if Alex is aware of that)
>
> From the results it seems that performance of llvm are slightly better for
> iterative solutions and fairly worse for recursive solutions
>
> Question: I used the makefile in the pil21.tgz to compile pil21. Is there
> any optimization in? I didn't dig (yet) into llvm compiler flags :)
>
> Regards,
> Davide
>

Pil vs Pil21 performances

[Davide BERTOLOTTO]

Hi all,

I compiled pil21 with llvm on termux and then I ran some simple fibonacci
benchmark. Here the results

*** recursive
n 37 fib 24157817
-- picolisp 3.013 sec
-- pil21 5.332 sec
-- python 21.247 sec

*** iterative
n 8 fib 4589178984+
-- picolisp 0.518 sec
-- pil21 0.409 sec
-- python 0.124 sec

I got also a segmentation fault while trying to use cache to speed up the
calculation with pil21 (don't know if Alex is aware of that)

>From the results it seems that performance of llvm are slightly better for
iterative solutions and fairly worse for recursive solutions

Question: I used the makefile in the pil21.tgz to compile pil21. Is there
any optimization in? I didn't dig (yet) into llvm compiler flags :)

Regards,
Davide

Re: Tomorrow is PilCon

[George-Phillip Orais]

Hi Olaf,

Thanks for this recap, this is very nice indeed!

I would like to add the hand-made block diagram by Alex for his
documentation, very cool! Maybe we can call it pilDiagram and make it as
official diagram for PicoLisp? Lets see :)


Hi Alex,

Thank you once again for sharing your time and patiently answering our
questions, PicoLisp always surprises me with so many nice features, how I
wish I have a very good Pil knowledge enough to introduce these hidden gems
to the outside world... but I still believe someone here or you with pil21,
it will happen someday soon :)


Hi Nehal,

I'm not so sure but you mentioned about org-babel, does this help your
inquiry: https://github.com/tj64/ob-picolisp


Thanks again everyone and stay safe always, have a great weekend! Bis blad!


BR,
Geo




On Fri, Sep 4, 2020 at 8:51 PM Alexander Burger  wrote:

> Hi Olaf,
>
> > I like to recap some topics/items, perhaps others can also enjoy
> > (disclaimer: there may be misunderstandings or faults in the notes!)
>
> Thanks a lot! This is a very good idea!
>
> And all correct I think.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>
>

Re: Tomorrow is PilCon

[Alexander Burger]

Hi Olaf,

> I like to recap some topics/items, perhaps others can also enjoy 
> (disclaimer: there may be misunderstandings or faults in the notes!)

Thanks a lot! This is a very good idea!

And all correct I think.

☺/ A!ex

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Tomorrow is PilCon

[O.Hamann]

Thank you Alex and all other participants
(and not at least thanks to beneroth (typed correctly?) providing the
space, where we could meet)

again for this nice PilCon Session.

I like to recap some topics/items, perhaps others can also enjoy 
(disclaimer: there may be misunderstandings or faults in the notes!)

- inspiring discussion if literate programming concepts could help
explaining picolisp code (thanks for demonstrating the 'old' Forth way
to enable a similar idea for documentation without mixing doc strings
into the code')

- no assembler language knowledge required for picolisp beginners
-- but knowledge about the cell structure might help understand the core
functionality ... there are docs in the distribution
-- thanks for demonstrating the helpful cells function 'hidden :-)' in
lib/vip/draw.l (hopefully remembered the path correctly)

- use src files and 'load' early to avoid beginners being frustrated by
segfaults
-- segfaults on the way to learn picolisp, for me are a bit like "is the
same as learning the first falls of a toddler while running. after a
short time, they are missing out." (kudos to babelfish)

- easy beginner examples to teach children picolisp are wanted (perhaps
from the physics domain?)

- when starting to program in picolisp start thinking about the data
structure
-- might be lists if you want to train list handling
-- might be classes and objects, if you want an example which can be
easily enhanced or changed later without changing the whole work done so far

- picolisp is different from other LISPs

- using picolisp classes/objects has (very?) little performance constraints

Surely there were more interesting items to remember, but too much for
me :-)


Thank you all,

hopefully we'll meet again in one of the next PilCons :-)

Olaf







 

On 03.09.20 18:13, Alexander Burger wrote:
> Hi all,
>
> as a reminder, tomorrow is PilCon again!
>
> As scheduled, we start at 8:00 UTC on
>
>https://meeting.itship.ch/PilCon
>
> (for those who missed it: PilCon is on the first (8:00 UTC) and on the third
> (16:00 UTC) Friday every month)
>
> ☺/ A!ex
>


--
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe