Re: [dev] less(1) replacement?

From: Stéphane Aulery <lkppo_AT_free.fr>
Date: Mon, 28 Aug 2017 23:31:38 +0200

Le 28/08/2017 à 21:04, sylvain.bertrand_AT_gmail.com a écrit :
> On Mon, Aug 28, 2017 at 07:22:58PM +0200, Stéphane Aulery wrote:
>> Le 28/08/2017 à 11:44, sylvain.bertrand_AT_gmail.com a écrit :
>>> On Mon, Aug 28, 2017 at 02:41:42AM +0200, Stéphane Aulery wrote:
>>>> Le 27/08/2017 à 19:29, sylvain.bertrand_AT_gmail.com a écrit :
>>>>> On Sun, Aug 27, 2017 at 05:27:24PM +0200, Stéphane Aulery wrote:
>>>>>> My idea is how to reconcile the implementation of programs and a kernel
>>>>>> that is a multiplexer like plan9 with a language and a sound compilation
>>>>>> environment like that of Oberon.
>>>>>
>>>>> Once you have a nice working kernel with a vulkan stack and GPUs drivers, we'll
>>>>> talk about it again. In the meantime, good luck and fair winds.
>>>>>
>>>>> But really, you should try ada and rust, they "solved" probably all what you
>>>>> are talking about, already.
>>>>> Even rust has a less worse syntax than go, as far as I can recall. Don't forget
>>>>> to have a look at ada, a very strongly backed language... oh! and I was told
>>>>> something about mathematical proof integration with the D language.
>>>>> You also have the ML family, the "beauty" of "functional" languages:
>>>>> formal proof software is usually written using them, and lucky you, one of them
>>>>> has object orientation straight in the syntax, ocaml. I expect the addition of
>>>>> "aspect programming" straight in the syntax too!
>>>>> A good start would be to write your own compiler (not optimizing first). I
>>>>> suggest a c++17 compiler could be a good warm up, what do you think of that?
>>>>
>>>> That one feels the anger through irony. Thanks anyway.
>>>
>>> Irony, probably, but there is probably more, like a bit of truth and reality?
>>>
>>> It may be time to come down from the silver tower.
>>
>> Then explains.
>
> You are on suckless. _You_ have to explain how you end up here missing some core
> points of suckless that much.

I had read extensively *.cat-v.org, plan9 materials, used WMII for a
long time, Debian since 2007, but I have been mostly limited to web
development since the beginning of my career. It does not stop me from
being interested in fundamentals.

I have never had the choice to use a scripting language or not, but I
have always had a way to approach programming that is in sync with the
suckless philosophy. At least I think so : clarity, conciseness,
simplicity, throw out all the features that are not essential, reduce
dependencies to zero if possible, not depend on the temple merchants for
something as essential as its information system.

As I mentioned in my first post, I will not want to write a piece of
lower level or intermediate software with a scripting language, but one
can still write a user software whose only dependency is the
interpreter. The principles can be transposed.

For Oberon I can not remember how I discovered it (maybe via Rob Pike's
Plan9 Acme inspiration). Niklaus Wirth literally spent his life
simplifying Algol to retrieve only the essential in Oberon, a language
like C but with less caltrop. Its aesthetic is not the same but the
spirit is the same. It was not for nothing that Uriel was pushing Go
(https://www.reddit.com/r/programming/comments/11h7ab/9fans_uriel_has_passed_away/)
which has some relationship with it.

-- 
Stéphane Aulery
Received on Mon Aug 28 2017 - 23:31:38 CEST

This archive was generated by hypermail 2.3.0 : Mon Aug 28 2017 - 23:36:17 CEST