Hi!

I plan on making a universal compiler called "neos" that can
theoretically compile any scripting/programming language. The compiler
targets custom bytecode used by a custom VM with JIT.

Features:

* Language agnostic: a language schema file describes the syntax and
semantics of the scripting/programming language to use (theoretically
allowing any language to be used).
* Easy to write RJSON (Relaxed JSON) language schema file format.
* Extensible generic (cross-language) semantic concepts (extendable by
providing semantic concept plugins).
* The ability to mix source code from different programming languages
within the same source file.
* Invent your own scripting language to use with neos by writing a new
language schema!

An interesting side effect of the neos architecture that comes almost
for free is the possibility of mixing code from different programming
languages in the same source file. These source files have the .mix
file extension. Here is an example .mix file:

# The following line defines the language selection character (.mix
files are UTF-8 encoded)...
%%%⚛%
⚛⚛c⚛ /* make c the default language and select it */
#include <stdio.h>
⚛c++⚛ /* select the c++ language */
#include <iostream>
⚛ada⚛
with Ada.Text_IO;
use Ada.Text_IO;
⚛⚛ /* select the default language (c in this example) */
void hello_from_c()
{ printf("Hello, world (from C)!\n");
} ⚛ada⚛
procedure Hello is
begin
Put_Line("Hello, world (from Ada)!");
end Hello;
⚛⚛
int main()
{ hello_from_c();
⚛c++⚛
std::cout << "Hello, world (from C++)!" << std::endl;
⚛ada⚛
Hello();
⚛⚛
printf("Hello again from C!\n");
}

Some people reckon what I am doing isn't possible without quite
articulating why. I see it as a fun challenge.

Thoughts?

/Flibble

On 11/18/2022 12:05 PM, Mr Flibble wrote:
> Hi!
>
> I plan on making a universal compiler called "neos" that can
> theoretically compile any scripting/programming language. The compiler
> targets custom bytecode used by a custom VM with JIT.
>
> Features:
>
> * Language agnostic: a language schema file describes the syntax and
> semantics of the scripting/programming language to use (theoretically
> allowing any language to be used).
> * Easy to write RJSON (Relaxed JSON) language schema file format.
> * Extensible generic (cross-language) semantic concepts (extendable by
> providing semantic concept plugins).
> * The ability to mix source code from different programming languages
> within the same source file.
> * Invent your own scripting language to use with neos by writing a new
> language schema!

Sounds ambitious.

Similar ideas go back to the late 1950s:

https://en.wikipedia.org/wiki/UNCOL

https://dl.acm.org/doi/pdf/10.1145/954004.954008

On 11/18/2022 11:05 AM, Mr Flibble wrote:
> Hi!
>
> I plan on making a universal compiler called "neos" that can
> theoretically compile any scripting/programming language. The compiler
> targets custom bytecode used by a custom VM with JIT.
>
> Features:
>
> * Language agnostic: a language schema file describes the syntax and
> semantics of the scripting/programming language to use (theoretically
> allowing any language to be used).
> * Easy to write RJSON (Relaxed JSON) language schema file format.
> * Extensible generic (cross-language) semantic concepts (extendable by
> providing semantic concept plugins).
> * The ability to mix source code from different programming languages
> within the same source file.
> * Invent your own scripting language to use with neos by writing a new
> language schema!
>
> An interesting side effect of the neos architecture that comes almost
> for free is the possibility of mixing code from different programming
> languages in the same source file. These source files have the .mix
> file extension. Here is an example .mix file:
>
> # The following line defines the language selection character (.mix
> files are UTF-8 encoded)...
> %%%⚛%
> ⚛⚛c⚛ /* make c the default language and select it */
> #include <stdio.h>
> ⚛c++⚛ /* select the c++ language */
> #include <iostream>
> ⚛ada⚛
> with Ada.Text_IO;
> use Ada.Text_IO;
> ⚛⚛ /* select the default language (c in this example) */
> void hello_from_c()
> {
> printf("Hello, world (from C)!\n");
> }
> ⚛ada⚛
> procedure Hello is
> begin
> Put_Line("Hello, world (from Ada)!");
> end Hello;
> ⚛⚛
> int main()
> {
> hello_from_c();
> ⚛c++⚛
> std::cout << "Hello, world (from C++)!" << std::endl;
> ⚛ada⚛
> Hello();
> ⚛⚛
> printf("Hello again from C!\n");
> }
>
> Some people reckon what I am doing isn't possible without quite
> articulating why. I see it as a fun challenge.
>
> Thoughts?
>
> /Flibble
>

For conventional procedural programming languages all having the same
underlying low level semantics this should be possible.

Using Lex and YACC to compile source-code into a common abstract syntax
tree would be one way to accomplish this. A single back end could
generate byte code from this AST.

It becomes more difficult when switching programming language paradigms
https://www.coursera.org/articles/types-programming-language

--
Copyright 2022 Pete Olcott "Talent hits a target no one else can hit;
Genius hits a target no one else can see." Arthur Schopenhauer

On Fri, 18 Nov 2022 12:31:55 -0500
DFS <nospam@dfs.com> wrote:

> On 11/18/2022 12:05 PM, Mr Flibble wrote:
> > Hi!
> >
> > I plan on making a universal compiler called "neos" that can
> > theoretically compile any scripting/programming language. The
> > compiler targets custom bytecode used by a custom VM with JIT.
> >
> > Features:
> >
> > * Language agnostic: a language schema file describes the syntax and
> > semantics of the scripting/programming language to use
> > (theoretically allowing any language to be used).
> > * Easy to write RJSON (Relaxed JSON) language schema file format.
> > * Extensible generic (cross-language) semantic concepts (extendable
> > by providing semantic concept plugins).
> > * The ability to mix source code from different programming
> > languages within the same source file.
> > * Invent your own scripting language to use with neos by writing a
> > new language schema!
>
> Sounds ambitious.
>
> Similar ideas go back to the late 1950s:
>
> https://en.wikipedia.org/wiki/UNCOL
>
> https://dl.acm.org/doi/pdf/10.1145/954004.954008

Well I have a common intermediate representation which is JIT'd; the
key difference with neos is to make front ends easy to write.

/Flibble

On 11/18/2022 12:31 PM, Mr Flibble wrote:
> On Fri, 18 Nov 2022 12:31:55 -0500
> DFS <nospam@dfs.com> wrote:
>
>> On 11/18/2022 12:05 PM, Mr Flibble wrote:
>>> Hi!
>>>
>>> I plan on making a universal compiler called "neos" that can
>>> theoretically compile any scripting/programming language. The
>>> compiler targets custom bytecode used by a custom VM with JIT.
>>>
>>> Features:
>>>
>>> * Language agnostic: a language schema file describes the syntax and
>>> semantics of the scripting/programming language to use
>>> (theoretically allowing any language to be used).
>>> * Easy to write RJSON (Relaxed JSON) language schema file format.
>>> * Extensible generic (cross-language) semantic concepts (extendable
>>> by providing semantic concept plugins).
>>> * The ability to mix source code from different programming
>>> languages within the same source file.
>>> * Invent your own scripting language to use with neos by writing a
>>> new language schema!
>>
>> Sounds ambitious.
>>
>> Similar ideas go back to the late 1950s:
>>
>> https://en.wikipedia.org/wiki/UNCOL
>>
>> https://dl.acm.org/doi/pdf/10.1145/954004.954008
>
> Well I have a common intermediate representation which is JIT'd; the
> key difference with neos is to make front ends easy to write.
>
> /Flibble
>

That common intermediate representation would not work for languages in
entirely different paradigms.

https://www.coursera.org/articles/types-programming-language

--
Copyright 2022 Pete Olcott "Talent hits a target no one else can hit;
Genius hits a target no one else can see." Arthur Schopenhauer

On 11/18/2022 10:05 AM, Mr Flibble wrote:
> Hi!
>
> I plan on making a universal compiler called "neos" that can
> theoretically compile any scripting/programming language. The compiler
> targets custom bytecode used by a custom VM with JIT.
>
> Features:
>
> * Language agnostic: a language schema file describes the syntax and
> semantics of the scripting/programming language to use (theoretically
> allowing any language to be used).
> * Easy to write RJSON (Relaxed JSON) language schema file format.
> * Extensible generic (cross-language) semantic concepts (extendable by
> providing semantic concept plugins).
> * The ability to mix source code from different programming languages
> within the same source file.
> * Invent your own scripting language to use with neos by writing a new
> language schema!
>
> An interesting side effect of the neos architecture that comes almost
> for free is the possibility of mixing code from different programming
> languages in the same source file. These source files have the .mix
> file extension. Here is an example .mix file:
>
> # The following line defines the language selection character (.mix
> files are UTF-8 encoded)...
> %%%⚛%
> ⚛⚛c⚛ /* make c the default language and select it */
> #include <stdio.h>
> ⚛c++⚛ /* select the c++ language */
> #include <iostream>
> ⚛ada⚛
> with Ada.Text_IO;
> use Ada.Text_IO;
> ⚛⚛ /* select the default language (c in this example) */
> void hello_from_c()
> {
> printf("Hello, world (from C)!\n");
> }
> ⚛ada⚛
> procedure Hello is
> begin
> Put_Line("Hello, world (from Ada)!");
> end Hello;
> ⚛⚛
> int main()
> {
> hello_from_c();
> ⚛c++⚛
> std::cout << "Hello, world (from C++)!" << std::endl;
> ⚛ada⚛
> Hello();
> ⚛⚛
> printf("Hello again from C!\n");
> }
>
> Some people reckon what I am doing isn't possible without quite
> articulating why. I see it as a fun challenge.
>
> Thoughts?
Just for the record, this sort or proposal isn't new or unique. See
UNCOL (Universal Computer Oriented Language) at Wikipedia. One of the
"prime movers" of that effort was Tom Steel who, with others, made noise
in the late 1950s and early 1960s. Some of the problems were (1) usable
meta compiler technology and meta languages where not available, (2) no
reasonable standards for fixed point, floating point, or character sets,
(3) some languages where idiosyncratic just because that was the whole
point of having them, (4) various languages where prefix, infix, or
postfix and there was no decent agreement on operator naming or
application rules, (5) languages had radically different scoping rules,
(6) various languages provided different guarantees of how various
conditions or exceptions were to be handled as well as order of
evaluation rules, (7) several popular languages allowed machine code to
be mixed with HOL, (8) some languages provided complex capabilities to
expand themselves, (9) miscellaneous issues.
Today, we have a better handle on problems 1, 2 and, 4; problems 3, 5,
6, and 7 are still very tough; and problem 8 depends on whether the
extensions are character sets, syntax, new structures, new control
structures, new operators, etc. UNCOL was talked about for more than a
decade and then faded into the dust. I believe that DARPA added some
money to the pot somewhere along the line.
Some descendants of UNCOL include serious work on meta compiler
technology as well as better, more organized language specifications and
standards works. I had the pleasure of working with Steel for a while
and was impressed by the depth of consideration given the ideas.
--
Jeff Barnett

This seems like a suitable place to dump this generalized way to
disentangle expressions:
>
There is a set things called "symbols". There are two kinds of
symbols - "values" and "operators". There are six kinds of
operators - an operator can be one of "prefix", "infix" or "suffix"
and it can be "strong" or "weak". Each operator has also a
(one or two) precedences - called "left" and "right". I will work
with two stacks - a stack of values called "X" and a stack of
operators called "S".
>
An potential expression is a string of symbols. It is processed
left to right as follows:
>
push a start symbol (here #) onto S
1: read a symbol $
if $ is a prefix operator
.....push $ onto S
,,, go to 1
else is $ is a value
....y = $
2 ...read $
....if $ is a infix or suffix operator
.......t = $
.......3: if left of s (top of S) is greater than right of t
.......... pop s off S
.......... if s is weak
.............y=sy (apply s to y)
.............go to 3
..........else
.............pop x off X
.............y=xsy
.............go to 3
....... else if left of s is less than right of t
.......... if t is weak
............. y=yt
.............. go to 2
.......... else
............ push y onto X
............ push t onto S
.............go to 2
....... else
........... pop s off S
........... if s is weak
.............. y=syt
.............. if s = #
................. return y
...............else
................. go to 2
........... else
.............. pop x off X
.............. y=xsyt
>
Examples of the five different things that happen
y = sy : a = - b
y = xsy : a = b + c
y = yt : a = b++
y = syt : a = ( b )
y = xsyt : a = f ( b )
>
I hope for a minimum of errors.

DFS <nospam@dfs.com> writes:

> On 11/18/2022 12:05 PM, Mr Flibble wrote:
>> Hi!
>> I plan on making a universal compiler called "neos" that can
>> theoretically compile any scripting/programming language. The compiler
>> targets custom bytecode used by a custom VM with JIT.
>> Features:
>> * Language agnostic: a language schema file describes the syntax and
>> semantics of the scripting/programming language to use (theoretically
>> allowing any language to be used).
>> * Easy to write RJSON (Relaxed JSON) language schema file format.
>> * Extensible generic (cross-language) semantic concepts (extendable by
>> providing semantic concept plugins).
>> * The ability to mix source code from different programming languages
>> within the same source file.
>> * Invent your own scripting language to use with neos by writing a new
>> language schema!
>
> Sounds ambitious.

Sounds like it's going backwards. A while ago he talked of it in the
present tense as if it existed. Now it is just a plan!

--
Ben.

On 11/18/2022 1:27 PM, dklei...@gmail.com wrote:
> This seems like a suitable place to dump this generalized way to
> disentangle expressions:
>>
> There is a set things called "symbols". There are two kinds of
> symbols - "values" and "operators". There are six kinds of
> operators - an operator can be one of "prefix", "infix" or "suffix"
> and it can be "strong" or "weak". Each operator has also a
> (one or two) precedences - called "left" and "right". I will work
> with two stacks - a stack of values called "X" and a stack of
> operators called "S".
>>
> An potential expression is a string of symbols. It is processed
> left to right as follows:
>>
> push a start symbol (here #) onto S
> 1: read a symbol $
> if $ is a prefix operator
> ....push $ onto S
> ,,, go to 1
> else is $ is a value
> ...y = $
> 2 ...read $
> ...if $ is a infix or suffix operator
> ......t = $
> ......3: if left of s (top of S) is greater than right of t
> ......... pop s off S
> ......... if s is weak
> ............y=sy (apply s to y)
> ............go to 3
> .........else
> ............pop x off X
> ............y=xsy
> ............go to 3
> ...... else if left of s is less than right of t
> ......... if t is weak
> ............ y=yt
> ............. go to 2
> ......... else
> ........... push y onto X
> ........... push t onto S
> ............go to 2
> ...... else
> .......... pop s off S
> .......... if s is weak
> ............. y=syt
> ............. if s = #
> ................ return y
> ..............else
> ................ go to 2
> .......... else
> ............. pop x off X
> ............. y=xsyt
>>
> Examples of the five different things that happen
> y = sy : a = - b
> y = xsy : a = b + c
> y = yt : a = b++
> y = syt : a = ( b )
> y = xsyt : a = f ( b )
>>
> I hope for a minimum of errors.

I believe everything above was know by the mid 1950s. In fact, your
characterizations are rather incomplete. Trying to complete these things
to a reasonable extent would make a reasonable class exercise. Good luck
and I hope you get a good grade.

For extra credit, gave some formalism for "control structures" and to
"scoping rules".
--
Jeff Barnett

On Sat, 19 Nov 2022 01:27:15 +0000
Ben Bacarisse <ben.usenet@bsb.me.uk> wrote:

> DFS <nospam@dfs.com> writes:
>
> > On 11/18/2022 12:05 PM, Mr Flibble wrote:
> >> Hi!
> >> I plan on making a universal compiler called "neos" that can
> >> theoretically compile any scripting/programming language. The
> >> compiler targets custom bytecode used by a custom VM with JIT.
> >> Features:
> >> * Language agnostic: a language schema file describes the syntax
> >> and semantics of the scripting/programming language to use
> >> (theoretically allowing any language to be used).
> >> * Easy to write RJSON (Relaxed JSON) language schema file format.
> >> * Extensible generic (cross-language) semantic concepts
> >> (extendable by providing semantic concept plugins).
> >> * The ability to mix source code from different programming
> >> languages within the same source file.
> >> * Invent your own scripting language to use with neos by writing a
> >> new language schema!
> >
> > Sounds ambitious.
>
> Sounds like it's going backwards. A while ago he talked of it in the
> present tense as if it existed. Now it is just a plan!
"Plan on making" = release; it hasn't been released yet.

Also, did you knowingly discuss the Halting Problem with a pedophile?
If so then I have deep concerns about you too.

/Flibble

On 11/19/2022 9:13 AM, Mr Flibble wrote:
> On Sat, 19 Nov 2022 01:27:15 +0000
> Ben Bacarisse <ben.usenet@bsb.me.uk> wrote:
>
>> DFS <nospam@dfs.com> writes:
>>
>>> On 11/18/2022 12:05 PM, Mr Flibble wrote:
>>>> Hi!
>>>> I plan on making a universal compiler called "neos" that can
>>>> theoretically compile any scripting/programming language. The
>>>> compiler targets custom bytecode used by a custom VM with JIT.
>>>> Features:
>>>> * Language agnostic: a language schema file describes the syntax
>>>> and semantics of the scripting/programming language to use
>>>> (theoretically allowing any language to be used).
>>>> * Easy to write RJSON (Relaxed JSON) language schema file format.
>>>> * Extensible generic (cross-language) semantic concepts
>>>> (extendable by providing semantic concept plugins).
>>>> * The ability to mix source code from different programming
>>>> languages within the same source file.
>>>> * Invent your own scripting language to use with neos by writing a
>>>> new language schema!
>>>
>>> Sounds ambitious.
>>
>> Sounds like it's going backwards. A while ago he talked of it in the
>> present tense as if it existed. Now it is just a plan!
>
> "Plan on making" = release; it hasn't been released yet.
>
> Also, did you knowingly discuss the Halting Problem with a pedophile?
> If so then I have deep concerns about you too.
>
> /Flibble
>

I was never found guilty of any crime.
What are you hiding behind your fake name?
People that use fake names have something to hide.

--
Copyright 2022 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer

Mr Flibble <flibble@reddwarf.jmc.corp> writes:

> On Sat, 19 Nov 2022 01:27:15 +0000
> Ben Bacarisse <ben.usenet@bsb.me.uk> wrote:
>
>> DFS <nospam@dfs.com> writes:
>>
>> > On 11/18/2022 12:05 PM, Mr Flibble wrote:
>> >> Hi!
>> >> I plan on making a universal compiler called "neos" that can
>> >> theoretically compile any scripting/programming language. The
>> >> compiler targets custom bytecode used by a custom VM with JIT.
>> >> Features:
>> >> * Language agnostic: a language schema file describes the syntax
>> >> and semantics of the scripting/programming language to use
>> >> (theoretically allowing any language to be used).
>> >> * Easy to write RJSON (Relaxed JSON) language schema file format.
>> >> * Extensible generic (cross-language) semantic concepts
>> >> (extendable by providing semantic concept plugins).
>> >> * The ability to mix source code from different programming
>> >> languages within the same source file.
>> >> * Invent your own scripting language to use with neos by writing a
>> >> new language schema!
>> >
>> > Sounds ambitious.
>>
>> Sounds like it's going backwards. A while ago he talked of it in the
>> present tense as if it existed. Now it is just a plan!
>
> "Plan on making" = release; it hasn't been released yet.

So does it exist somewhere other than your head or not?

--
Ben.

On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:

> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>
>> On Sat, 19 Nov 2022 01:27:15 +0000 Ben Bacarisse <ben.usenet@bsb.me.uk>
>> wrote:
>>
>>> DFS <nospam@dfs.com> writes:
>>>
>>> > On 11/18/2022 12:05 PM, Mr Flibble wrote:
>>> >> Hi!
>>> >> I plan on making a universal compiler called "neos" that can
>>> >> theoretically compile any scripting/programming language. The
>>> >> compiler targets custom bytecode used by a custom VM with JIT.
>>> >> Features:
>>> >> * Language agnostic: a language schema file describes the syntax
>>> >> and semantics of the scripting/programming language to use
>>> >> (theoretically allowing any language to be used).
>>> >> * Easy to write RJSON (Relaxed JSON) language schema file format. *
>>> >> Extensible generic (cross-language) semantic concepts (extendable
>>> >> by providing semantic concept plugins).
>>> >> * The ability to mix source code from different programming
>>> >> languages within the same source file.
>>> >> * Invent your own scripting language to use with neos by writing a
>>> >> new language schema!
>>> >
>>> > Sounds ambitious.
>>>
>>> Sounds like it's going backwards. A while ago he talked of it in the
>>> present tense as if it existed. Now it is just a plan!
>>
>> "Plan on making" = release; it hasn't been released yet.
>
> So does it exist somewhere other than your head or not?

It exists both in code and in my head; it will exist entirely in code when
1.0 is released.

/Flibble

Mr Flibble <flibble@reddwarf.jmc.corp> writes:

> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:

>> So does it exist somewhere other than your head or not?
>
> It exists both in code and in my head; it will exist entirely in code when
> 1.0 is released.

Ah. So it does not exist. Nor did it more than a year ago when you
boasted about what it could do. That's not unexpected of course because
the existence of a universal compiler would be headline news, not
something to be fudged in this forgotten corner of the internet.

--
Ben.

On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:

> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>
>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>
>>> So does it exist somewhere other than your head or not?
>>
>> It exists both in code and in my head; it will exist entirely in code
>> when 1.0 is released.
>
> Ah. So it does not exist. Nor did it more than a year ago when you
> boasted about what it could do. That's not unexpected of course because
> the existence of a universal compiler would be headline news, not
> something to be fudged in this forgotten corner of the internet.

It PARTIALLY exists in the form of open source code on GitHub; it will
FULLY exist when 1.0 is released.

/Flibble

Mr Flibble <flibble@reddwarf.jmc.corp> writes:

> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>
>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>
>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>
>>>> So does it exist somewhere other than your head or not?
>>>
>>> It exists both in code and in my head; it will exist entirely in code
>>> when 1.0 is released.
>>
>> Ah. So it does not exist. Nor did it more than a year ago when you
>> boasted about what it could do. That's not unexpected of course because
>> the existence of a universal compiler would be headline news, not
>> something to be fudged in this forgotten corner of the internet.
>
> It PARTIALLY exists in the form of open source code on GitHub; it will
> FULLY exist when 1.0 is released.

So it can't do now, what you claimed it it could do more than a year
ago. I can put some code up on GigHub and claim it will be able to
compile any language, decide halting and compress any file when 1.0 is
released!

--
Ben.

On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:

> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>
>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>
>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>
>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>
>>>>> So does it exist somewhere other than your head or not?
>>>>
>>>> It exists both in code and in my head; it will exist entirely in code
>>>> when 1.0 is released.
>>>
>>> Ah. So it does not exist. Nor did it more than a year ago when you
>>> boasted about what it could do. That's not unexpected of course
>>> because the existence of a universal compiler would be headline news,
>>> not something to be fudged in this forgotten corner of the internet.
>>
>> It PARTIALLY exists in the form of open source code on GitHub; it will
>> FULLY exist when 1.0 is released.
>
> So it can't do now, what you claimed it it could do more than a year
> ago. I can put some code up on GigHub and claim it will be able to
> compile any language, decide halting and compress any file when 1.0 is
> released!

The idea that software isn't released until it is ready is hardly a new or
novel one. Also, why are you talking about deciding halting and
compressing files? I am not working on neos to do either of those things.

/Flibble

On 11/21/22 12:14 PM, Mr Flibble wrote:
> On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:
>
>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>
>>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>>
>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>
>>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>>
>>>>>> So does it exist somewhere other than your head or not?
>>>>>
>>>>> It exists both in code and in my head; it will exist entirely in code
>>>>> when 1.0 is released.
>>>>
>>>> Ah. So it does not exist. Nor did it more than a year ago when you
>>>> boasted about what it could do. That's not unexpected of course
>>>> because the existence of a universal compiler would be headline news,
>>>> not something to be fudged in this forgotten corner of the internet.
>>>
>>> It PARTIALLY exists in the form of open source code on GitHub; it will
>>> FULLY exist when 1.0 is released.
>>
>> So it can't do now, what you claimed it it could do more than a year
>> ago. I can put some code up on GigHub and claim it will be able to
>> compile any language, decide halting and compress any file when 1.0 is
>> released!
>
> The idea that software isn't released until it is ready is hardly a new or
> novel one. Also, why are you talking about deciding halting and
> compressing files? I am not working on neos to do either of those things.
>
> /Flibble

I think the key is that a program "Doesn't Exist" until it is largely
working.

When you say "I have", it implies that you actually HAVE (mostly)
working code that does that.

Code that isn't mostly working should either be referenced as clearly a
work in progress (I am working on a xxx), or if you are confident in
your ability to complete it, in the FUTURE tense, I WILL Have a program
that does xxxx.

Present tense "I have a program that does xxxx" means that the program,
as it currently exists, actually DOES that function. Maybe not quite to
release level of quality, but is functional.

Sounds like you code isn't any where near to that level yet.

On Mon, 21 Nov 2022 12:27:36 -0500, Richard Damon wrote:

> On 11/21/22 12:14 PM, Mr Flibble wrote:
>> On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:
>>
>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>
>>>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>>>
>>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>>
>>>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>>>
>>>>>>> So does it exist somewhere other than your head or not?
>>>>>>
>>>>>> It exists both in code and in my head; it will exist entirely in
>>>>>> code when 1.0 is released.
>>>>>
>>>>> Ah. So it does not exist. Nor did it more than a year ago when you
>>>>> boasted about what it could do. That's not unexpected of course
>>>>> because the existence of a universal compiler would be headline
>>>>> news,
>>>>> not something to be fudged in this forgotten corner of the internet.
>>>>
>>>> It PARTIALLY exists in the form of open source code on GitHub; it
>>>> will FULLY exist when 1.0 is released.
>>>
>>> So it can't do now, what you claimed it it could do more than a year
>>> ago. I can put some code up on GigHub and claim it will be able to
>>> compile any language, decide halting and compress any file when 1.0 is
>>> released!
>>
>> The idea that software isn't released until it is ready is hardly a new
>> or novel one. Also, why are you talking about deciding halting and
>> compressing files? I am not working on neos to do either of those
>> things.
>>
>> /Flibble
>
> I think the key is that a program "Doesn't Exist" until it is largely
> working.
>
> When you say "I have", it implies that you actually HAVE (mostly)
> working code that does that.
>
> Code that isn't mostly working should either be referenced as clearly a
> work in progress (I am working on a xxx), or if you are confident in
> your ability to complete it, in the FUTURE tense, I WILL Have a program
> that does xxxx.
>
> Present tense "I have a program that does xxxx" means that the program,
> as it currently exists, actually DOES that function. Maybe not quite to
> release level of quality, but is functional.
>
> Sounds like you code isn't any where near to that level yet.

I have a program that will do xxxx when it is released.

/Flibble

On 11/21/22 1:30 PM, Mr Flibble wrote:
> On Mon, 21 Nov 2022 12:27:36 -0500, Richard Damon wrote:
>
>> On 11/21/22 12:14 PM, Mr Flibble wrote:
>>> On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:
>>>
>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>
>>>>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>>>>
>>>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>>>
>>>>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>>>>
>>>>>>>> So does it exist somewhere other than your head or not?
>>>>>>>
>>>>>>> It exists both in code and in my head; it will exist entirely in
>>>>>>> code when 1.0 is released.
>>>>>>
>>>>>> Ah. So it does not exist. Nor did it more than a year ago when you
>>>>>> boasted about what it could do. That's not unexpected of course
>>>>>> because the existence of a universal compiler would be headline
>>>>>> news,
>>>>>> not something to be fudged in this forgotten corner of the internet.
>>>>>
>>>>> It PARTIALLY exists in the form of open source code on GitHub; it
>>>>> will FULLY exist when 1.0 is released.
>>>>
>>>> So it can't do now, what you claimed it it could do more than a year
>>>> ago. I can put some code up on GigHub and claim it will be able to
>>>> compile any language, decide halting and compress any file when 1.0 is
>>>> released!
>>>
>>> The idea that software isn't released until it is ready is hardly a new
>>> or novel one. Also, why are you talking about deciding halting and
>>> compressing files? I am not working on neos to do either of those
>>> things.
>>>
>>> /Flibble
>>
>> I think the key is that a program "Doesn't Exist" until it is largely
>> working.
>>
>> When you say "I have", it implies that you actually HAVE (mostly)
>> working code that does that.
>>
>> Code that isn't mostly working should either be referenced as clearly a
>> work in progress (I am working on a xxx), or if you are confident in
>> your ability to complete it, in the FUTURE tense, I WILL Have a program
>> that does xxxx.
>>
>> Present tense "I have a program that does xxxx" means that the program,
>> as it currently exists, actually DOES that function. Maybe not quite to
>> release level of quality, but is functional.
>>
>> Sounds like you code isn't any where near to that level yet.
>
> I have a program that will do xxxx when it is released.
>
> /Flibble

So you don't have a program that does that NOW, so you don't have a
program that DOES that.

As I said, keep your tenses correct. It isn't the future, so don't talk
about now as it if was the future.

Mr Flibble <flibble@reddwarf.jmc.corp> writes:

> On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:
>
>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>
>>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>>
>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>
>>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>>
>>>>>> So does it exist somewhere other than your head or not?
>>>>>
>>>>> It exists both in code and in my head; it will exist entirely in code
>>>>> when 1.0 is released.
>>>>
>>>> Ah. So it does not exist. Nor did it more than a year ago when you
>>>> boasted about what it could do. That's not unexpected of course
>>>> because the existence of a universal compiler would be headline news,
>>>> not something to be fudged in this forgotten corner of the internet.
>>>
>>> It PARTIALLY exists in the form of open source code on GitHub; it will
>>> FULLY exist when 1.0 is released.
>>
>> So it can't do now, what you claimed it it could do more than a year
>> ago. I can put some code up on GigHub and claim it will be able to
>> compile any language, decide halting and compress any file when 1.0 is
>> released!
>
> The idea that software isn't released until it is ready is hardly a new or
> novel one.

As is claiming to have something you don't. Vapourware is very common,
particularly in Usenet groups.

> Also, why are you talking about deciding halting and
> compressing files?

Just an example of what I could claim to "have" (in your sense of the
word) because it's not yet released!

--
Ben.

On Mon, 21 Nov 2022 13:42:28 -0500, Richard Damon wrote:

> On 11/21/22 1:30 PM, Mr Flibble wrote:
>> On Mon, 21 Nov 2022 12:27:36 -0500, Richard Damon wrote:
>>
>>> On 11/21/22 12:14 PM, Mr Flibble wrote:
>>>> On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:
>>>>
>>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>>
>>>>>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>>>>>
>>>>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>>>>
>>>>>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>>>>>
>>>>>>>>> So does it exist somewhere other than your head or not?
>>>>>>>>
>>>>>>>> It exists both in code and in my head; it will exist entirely in
>>>>>>>> code when 1.0 is released.
>>>>>>>
>>>>>>> Ah. So it does not exist. Nor did it more than a year ago when
>>>>>>> you boasted about what it could do. That's not unexpected of
>>>>>>> course because the existence of a universal compiler would be
>>>>>>> headline news,
>>>>>>> not something to be fudged in this forgotten corner of the
>>>>>>> internet.
>>>>>>
>>>>>> It PARTIALLY exists in the form of open source code on GitHub; it
>>>>>> will FULLY exist when 1.0 is released.
>>>>>
>>>>> So it can't do now, what you claimed it it could do more than a year
>>>>> ago. I can put some code up on GigHub and claim it will be able to
>>>>> compile any language, decide halting and compress any file when 1.0
>>>>> is released!
>>>>
>>>> The idea that software isn't released until it is ready is hardly a
>>>> new or novel one. Also, why are you talking about deciding halting
>>>> and compressing files? I am not working on neos to do either of
>>>> those things.
>>>>
>>>> /Flibble
>>>
>>> I think the key is that a program "Doesn't Exist" until it is largely
>>> working.
>>>
>>> When you say "I have", it implies that you actually HAVE (mostly)
>>> working code that does that.
>>>
>>> Code that isn't mostly working should either be referenced as clearly
>>> a work in progress (I am working on a xxx), or if you are confident in
>>> your ability to complete it, in the FUTURE tense, I WILL Have a
>>> program that does xxxx.
>>>
>>> Present tense "I have a program that does xxxx" means that the
>>> program,
>>> as it currently exists, actually DOES that function. Maybe not quite
>>> to release level of quality, but is functional.
>>>
>>> Sounds like you code isn't any where near to that level yet.
>>
>> I have a program that will do xxxx when it is released.
>>
>> /Flibble
>
> So you don't have a program that does that NOW, so you don't have a
> program that DOES that.
>
> As I said, keep your tenses correct. It isn't the future, so don't talk
> about now as it if was the future.

Do you even English? Again: I have a program that will do xxxx when it is
released.

/Flibble

On Mon, 21 Nov 2022 19:26:28 +0000, Ben Bacarisse wrote:

> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>
>> On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:
>>
>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>
>>>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>>>
>>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>>
>>>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>>>
>>>>>>> So does it exist somewhere other than your head or not?
>>>>>>
>>>>>> It exists both in code and in my head; it will exist entirely in
>>>>>> code when 1.0 is released.
>>>>>
>>>>> Ah. So it does not exist. Nor did it more than a year ago when you
>>>>> boasted about what it could do. That's not unexpected of course
>>>>> because the existence of a universal compiler would be headline
>>>>> news,
>>>>> not something to be fudged in this forgotten corner of the internet.
>>>>
>>>> It PARTIALLY exists in the form of open source code on GitHub; it
>>>> will FULLY exist when 1.0 is released.
>>>
>>> So it can't do now, what you claimed it it could do more than a year
>>> ago. I can put some code up on GigHub and claim it will be able to
>>> compile any language, decide halting and compress any file when 1.0 is
>>> released!
>>
>> The idea that software isn't released until it is ready is hardly a new
>> or novel one.
>
> As is claiming to have something you don't. Vapourware is very common,
> particularly in Usenet groups.

All software is vapourware until it is released.

/Flibble

Mr Flibble <flibble@reddwarf.jmc.corp> writes:

> Again: I have a program that will do xxxx when it is released.

I have a made a start on my own universal compiler:

int main(void) { }

It's early days, but it will compile any language when it is released.
But I would /not/ do what you did in 2021 and describe it as

"my universal compiler that /can/ compile any programming language"
(emphasis mine)

Will and can are very different.

--
Ben.

On 11/21/22 4:16 PM, Mr Flibble wrote:
> On Mon, 21 Nov 2022 13:42:28 -0500, Richard Damon wrote:
>
>> On 11/21/22 1:30 PM, Mr Flibble wrote:
>>> On Mon, 21 Nov 2022 12:27:36 -0500, Richard Damon wrote:
>>>
>>>> On 11/21/22 12:14 PM, Mr Flibble wrote:
>>>>> On Mon, 21 Nov 2022 00:47:20 +0000, Ben Bacarisse wrote:
>>>>>
>>>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>>>
>>>>>>> On Sun, 20 Nov 2022 20:38:29 +0000, Ben Bacarisse wrote:
>>>>>>>
>>>>>>>> Mr Flibble <flibble@reddwarf.jmc.corp> writes:
>>>>>>>>
>>>>>>>>> On Sat, 19 Nov 2022 23:25:17 +0000, Ben Bacarisse wrote:
>>>>>>>>
>>>>>>>>>> So does it exist somewhere other than your head or not?
>>>>>>>>>
>>>>>>>>> It exists both in code and in my head; it will exist entirely in
>>>>>>>>> code when 1.0 is released.
>>>>>>>>
>>>>>>>> Ah. So it does not exist. Nor did it more than a year ago when
>>>>>>>> you boasted about what it could do. That's not unexpected of
>>>>>>>> course because the existence of a universal compiler would be
>>>>>>>> headline news,
>>>>>>>> not something to be fudged in this forgotten corner of the
>>>>>>>> internet.
>>>>>>>
>>>>>>> It PARTIALLY exists in the form of open source code on GitHub; it
>>>>>>> will FULLY exist when 1.0 is released.
>>>>>>
>>>>>> So it can't do now, what you claimed it it could do more than a year
>>>>>> ago. I can put some code up on GigHub and claim it will be able to
>>>>>> compile any language, decide halting and compress any file when 1.0
>>>>>> is released!
>>>>>
>>>>> The idea that software isn't released until it is ready is hardly a
>>>>> new or novel one. Also, why are you talking about deciding halting
>>>>> and compressing files? I am not working on neos to do either of
>>>>> those things.
>>>>>
>>>>> /Flibble
>>>>
>>>> I think the key is that a program "Doesn't Exist" until it is largely
>>>> working.
>>>>
>>>> When you say "I have", it implies that you actually HAVE (mostly)
>>>> working code that does that.
>>>>
>>>> Code that isn't mostly working should either be referenced as clearly
>>>> a work in progress (I am working on a xxx), or if you are confident in
>>>> your ability to complete it, in the FUTURE tense, I WILL Have a
>>>> program that does xxxx.
>>>>
>>>> Present tense "I have a program that does xxxx" means that the
>>>> program,
>>>> as it currently exists, actually DOES that function. Maybe not quite
>>>> to release level of quality, but is functional.
>>>>
>>>> Sounds like you code isn't any where near to that level yet.
>>>
>>> I have a program that will do xxxx when it is released.
>>>
>>> /Flibble
>>
>> So you don't have a program that does that NOW, so you don't have a
>> program that DOES that.
>>
>> As I said, keep your tenses correct. It isn't the future, so don't talk
>> about now as it if was the future.
>
> Do you even English? Again: I have a program that will do xxxx when it is
> released.
>
> /Flibble

So you don't have a program that does x NOW, so you shouldn't say you
have a program that does X, as that, because the statement is present
tense, implies NOW. The Qualifier is important.

Ben was commenting that in the past, you made the statement without the
future qualifier, and thus adding the qualifier shows backwards progress
or inaccurate claims in the past.

IF before, you HAD the program that DID x, and now you have a program
that WILL do x, you have lost progress.