Most Intelligent Command Line ("MICL") - pitch for devs

classic Classic list List threaded Threaded
26 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Most Intelligent Command Line ("MICL") - pitch for devs

Stefan Reich
Hi folks.

I'm putting this proposal out in the open now to find some motivated co-conspirators. I know that Lua-L is, historically, not exactly the greatest place to start projects. But then again, this is a list of thousands of Lua lovers... right? And this is a Lua-based project.

Things are looking up btw - I was actually featured in a Hamburg newspaper this week.

(btw, women in Hamburg suck. Sorry... I had to sneak that in there... it needs out =)

So! Back to computers. Here goes our pitch:

***

Most Intelligent Command Line (Shorthand: "MICL" or "ICL")

Abstract >>

MICL is scheduled to be a portable application that is - MATHEMATICALLY PROVABLY - the most intelligent command line possible.

Yes you read that right. The proof is already in my head. Yes, I have studied mathematics =)

MICL will be a bit like using google - you enter what you want your computer to do and the software finds instructions on how to do that. Just that the instructions are then automatically performed for you.

Smart, isn't it? And we know how to do this safely and securely.

Comparison to other command lines >>

OK, there already are command lines, like the Windows command prompt and the Linux shells. How does MICL compare to them?

1. MICL can imitate other shells. It is absolutely no problem to make MICL understand commands that were meant for MS-DOS, Bash or even weird languages like sed and awk. We will do that, and it will be easy.

2. MICL will be WILDLY more tolerant than these legacy command lines. One wrong character and they balk at you. MICL will just figure it out. MICL is something you can use when drunk.

3. MICL is an application that learns. In case it doesn't understand you today yet, wait until tomorrow and it will be able to do it then. Yes this is not a joke.

Community >>

This is another important aspect of the project. We believe that a community will spring into existence around MICL, once a first app exists, because it will be extremely easy to extend MICL - and to share those extensions.

There will be intelligent processes to ensure that good extensions get spread to more users. Write something good and people will use it.

Security >>

Is spreading code around between people like that secure you ask? Yes it will be. We know how to make things secure. Yes we do mean that. We have innovative security concepts ready to be implemented. A secure foundation (Lua) is key for that.

People wanted >>

* Evangelisers
* Users / input givers
* Developers

The one and only requirement if you want to help develop is to learn some Lua. (Lua is easy.) We might need code in other languages too, but just for little things. The bulk is gonna be Lua.

Language >>

We'll develop MICL in English because it's probably going to be an international team. Localizing it afterwards is going to be fun. =)

Time frame >>

Quick. We start assembling the team and developing NOW. Once there are devs, we can make a MVP (minimum viable product) in something like two weeks. Yes, I mean that.

Monetary questions >>

It's an open-source project, so probably no $ payout right at the start. But there's gonna be a lot of public attention, so, one way or another, you'll definitely be glad if you're a part of it.

Let's do it. Join now and become famous. And help making something so good that you'll find yourself using it daily.

Cheers,
Stefan Reich ([hidden email]) / www.pussy-riot-germany.de

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Coda Highland
On Fri, Apr 26, 2013 at 9:16 AM, Stefan Reich
<[hidden email]> wrote:

> Hi folks.
>
> I'm putting this proposal out in the open now to find some motivated
> co-conspirators. I know that Lua-L is, historically, not exactly the
> greatest place to start projects. But then again, this is a list of
> thousands of Lua lovers... right? And this is a Lua-based project.
>
> Things are looking up btw - I was actually featured in a Hamburg newspaper
> this week.
>
> (btw, women in Hamburg suck. Sorry... I had to sneak that in there... it
> needs out =)
>
> So! Back to computers. Here goes our pitch:
>
> ***
>
> Most Intelligent Command Line (Shorthand: "MICL" or "ICL")
>
> Abstract >>
>
> MICL is scheduled to be a portable application that is - MATHEMATICALLY
> PROVABLY - the most intelligent command line possible.
>
> Yes you read that right. The proof is already in my head. Yes, I have
> studied mathematics =)
>
> MICL will be a bit like using google - you enter what you want your computer
> to do and the software finds instructions on how to do that. Just that the
> instructions are then automatically performed for you.
>
> Smart, isn't it? And we know how to do this safely and securely.
>
> Comparison to other command lines >>
>
> OK, there already are command lines, like the Windows command prompt and the
> Linux shells. How does MICL compare to them?
>
> 1. MICL can imitate other shells. It is absolutely no problem to make MICL
> understand commands that were meant for MS-DOS, Bash or even weird languages
> like sed and awk. We will do that, and it will be easy.
>
> 2. MICL will be WILDLY more tolerant than these legacy command lines. One
> wrong character and they balk at you. MICL will just figure it out. MICL is
> something you can use when drunk.
>
> 3. MICL is an application that learns. In case it doesn't understand you
> today yet, wait until tomorrow and it will be able to do it then. Yes this
> is not a joke.
>
> Community >>
>
> This is another important aspect of the project. We believe that a community
> will spring into existence around MICL, once a first app exists, because it
> will be extremely easy to extend MICL - and to share those extensions.
>
> There will be intelligent processes to ensure that good extensions get
> spread to more users. Write something good and people will use it.
>
> Security >>
>
> Is spreading code around between people like that secure you ask? Yes it
> will be. We know how to make things secure. Yes we do mean that. We have
> innovative security concepts ready to be implemented. A secure foundation
> (Lua) is key for that.
>
> People wanted >>
>
> * Evangelisers
> * Users / input givers
> * Developers
>
> The one and only requirement if you want to help develop is to learn some
> Lua. (Lua is easy.) We might need code in other languages too, but just for
> little things. The bulk is gonna be Lua.
>
> Language >>
>
> We'll develop MICL in English because it's probably going to be an
> international team. Localizing it afterwards is going to be fun. =)
>
> Time frame >>
>
> Quick. We start assembling the team and developing NOW. Once there are devs,
> we can make a MVP (minimum viable product) in something like two weeks. Yes,
> I mean that.
>
> Monetary questions >>
>
> It's an open-source project, so probably no $ payout right at the start. But
> there's gonna be a lot of public attention, so, one way or another, you'll
> definitely be glad if you're a part of it.
>
> Let's do it. Join now and become famous. And help making something so good
> that you'll find yourself using it daily.
>
> Cheers,
> Stefan Reich ([hidden email]) / www.pussy-riot-germany.de
>

My primary concern with a "smart" shell is that typos that would
previously have been harmless errors run the risk of getting
reinterpreted as damaging commands. If I put in "rm -rf &" when I
meant to put a path before the &, and the shell says "& is beside *,
you must have meant "rm -rf *", let me blow away the current directory
for you.

I'm all for supporting really broad syntax, but I'd rather have the
"one wrong character" still error out -- or at the very least, prompt
"did you mean ___? y/n".

/s/ Adam

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

steve donovan
On Fri, Apr 26, 2013 at 6:56 PM, Coda Highland <[hidden email]> wrote:
I'm all for supporting really broad syntax, but I'd rather have the
"one wrong character" still error out -- or at the very least, prompt
"did you mean ___? y/n".

I like Stefan's vision, and this interactivity is the only way to resolve ambiguous stuff - conversation is full of questions.  Such a system has to learn from the answers as well.

 
Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Daniel Bolgheroni
In reply to this post by Stefan Reich
On Fri, Apr 26, 2013 at 06:16:03PM +0200, Stefan Reich wrote:
>
> MICL is scheduled to be a portable application that is - MATHEMATICALLY
> PROVABLY - the most intelligent command line possible.

Peter van der Linden, in his book Expert C Programming, has an interesting
point on how flawed is the idea of trying to proof code.

--
Daniel Bolgheroni <[hidden email]>
http://bolgh.eng.br

ASCII ribbon campaign ( )
against HTML e-mail    X
                      / \
(and no top-post also, please)

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Rena
In reply to this post by steve donovan

On 2013-04-26 1:45 PM, "steve donovan" <[hidden email]> wrote:
>
> On Fri, Apr 26, 2013 at 6:56 PM, Coda Highland <[hidden email]> wrote:
>>
>> I'm all for supporting really broad syntax, but I'd rather have the
>> "one wrong character" still error out -- or at the very least, prompt
>> "did you mean ___? y/n".
>
>
> I like Stefan's vision, and this interactivity is the only way to resolve ambiguous stuff - conversation is full of questions.  Such a system has to learn from the answers as well.
>
>  

The problem with "smart" shells is, as someone had pointed out already, they tend to be a little too smart:
$ rm *.o #let's remove any object files that may exist. This command should do nothing if there are no matching files...
> not found *.o, assuming you meant *.c #oh no!
Of course this is much less an issue if it asks what you meant instead of assuming, but it can still be annoying. (and you'd be surprised how many "smart" tools don't ask...)

What I'd find more useful than a smart shell is smart tab completion. E.g. I might start typing "SELECT " and hit tab, and have that turned into something like [[psql "SELECT "]], with the cursor keeping its position, and then further tab presses listing available databases and columns much like existing shells' tab complete lists directories.

Of course there's a good reason this doesn't exist (that I know of): it's difficult to do, and even more difficult to do well.

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Elias Barrionovo
In reply to this post by Daniel Bolgheroni
I think Wolfram Alpha [1] is a step toward this "smart" CLI. It has a
somewhat loose syntax and can compute some peculiar tasks. Too bad it
isn't free software.

On Fri, Apr 26, 2013 at 3:17 PM, Daniel Bolgheroni <[hidden email]> wrote:
> Peter van der Linden, in his book Expert C Programming, has an interesting
> point on how flawed is the idea of trying to proof code.

Sad Dijkstra is sad (though he was more into deriving the program from
the proof instead trying to prove an already developed program).

[1] http://www.wolframalpha.com

--
NI!

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Rena

On 2013-04-26 3:10 PM, "Elias Barrionovo" <[hidden email]> wrote:
>
> I think Wolfram Alpha [1] is a step toward this "smart" CLI. It has a
> somewhat loose syntax and can compute some peculiar tasks. Too bad it
> isn't free software.
>
> On Fri, Apr 26, 2013 at 3:17 PM, Daniel Bolgheroni <[hidden email]> wrote:
> > Peter van der Linden, in his book Expert C Programming, has an interesting
> > point on how flawed is the idea of trying to proof code.
>
> Sad Dijkstra is sad (though he was more into deriving the program from
> the proof instead trying to prove an already developed program).
>
> [1] http://www.wolframalpha.com
>
> --
> NI!
>

Yes, Wolfram does a pretty good job, but even then sometimes I can't get it to understand a seemingly simple calculation at all. Google does similar with their search.

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Tim Hill
In reply to this post by Stefan Reich
If it's proven to be the MOST intelligent shell POSSIBLE, then I should be able to just type "do it" and have it figure out what I mean right? In fact, couldn't it write itself from nothing? Or was that the infinite improbability drive? :)

--Tim


On Apr 26, 2013, at 9:16 AM, Stefan Reich <[hidden email]> wrote:

MICL is scheduled to be a portable application that is - MATHEMATICALLY PROVABLY - the most intelligent command line possible.

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Coda Highland
Unless of course that requires an impossible level of intelligence, at
which point the most intelligent shell couldn't be expected to satisfy
it. ;)

/s/ Adam

On Fri, Apr 26, 2013 at 10:59 PM, Tim Hill <[hidden email]> wrote:

> If it's proven to be the MOST intelligent shell POSSIBLE, then I should be
> able to just type "do it" and have it figure out what I mean right? In fact,
> couldn't it write itself from nothing? Or was that the infinite
> improbability drive? :)
>
> --Tim
>
>
> On Apr 26, 2013, at 9:16 AM, Stefan Reich
> <[hidden email]> wrote:
>
> MICL is scheduled to be a portable application that is - MATHEMATICALLY
> PROVABLY - the most intelligent command line possible.
>
>

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

steve donovan
In reply to this post by Rena
On Fri, Apr 26, 2013 at 9:12 PM, Rena <[hidden email]> wrote:
>Yes, Wolfram does a pretty good job, but even then sometimes I can't get it to understand a seemingly >simple calculation at all. Google does similar with their search.

Ah, but imagine it was interactive - then any point of ambiguity would result in a question, and so on.

I once caught Stefan's vision and wanted to do a natural language shell, but it turns out to be a distinctly non-trivial problem to work out what 'this' means, as Tim Hill notes ;)  For one thing, it would have to have deep hooks into the desktop environment so that it had full context;  for traditional shells the context is mostly the current working directory.



Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Benjamin Heath
In reply to this post by Stefan Reich

I think I'd rather the CLI be good, thin, and stupid.

On Apr 26, 2013 9:17 AM, "Stefan Reich" <[hidden email]> wrote:
Hi folks.

I'm putting this proposal out in the open now to find some motivated co-conspirators. I know that Lua-L is, historically, not exactly the greatest place to start projects. But then again, this is a list of thousands of Lua lovers... right? And this is a Lua-based project.

Things are looking up btw - I was actually featured in a Hamburg newspaper this week.

(btw, women in Hamburg suck. Sorry... I had to sneak that in there... it needs out =)

So! Back to computers. Here goes our pitch:

***

Most Intelligent Command Line (Shorthand: "MICL" or "ICL")

Abstract >>

MICL is scheduled to be a portable application that is - MATHEMATICALLY PROVABLY - the most intelligent command line possible.

Yes you read that right. The proof is already in my head. Yes, I have studied mathematics =)

MICL will be a bit like using google - you enter what you want your computer to do and the software finds instructions on how to do that. Just that the instructions are then automatically performed for you.

Smart, isn't it? And we know how to do this safely and securely.

Comparison to other command lines >>

OK, there already are command lines, like the Windows command prompt and the Linux shells. How does MICL compare to them?

1. MICL can imitate other shells. It is absolutely no problem to make MICL understand commands that were meant for MS-DOS, Bash or even weird languages like sed and awk. We will do that, and it will be easy.

2. MICL will be WILDLY more tolerant than these legacy command lines. One wrong character and they balk at you. MICL will just figure it out. MICL is something you can use when drunk.

3. MICL is an application that learns. In case it doesn't understand you today yet, wait until tomorrow and it will be able to do it then. Yes this is not a joke.

Community >>

This is another important aspect of the project. We believe that a community will spring into existence around MICL, once a first app exists, because it will be extremely easy to extend MICL - and to share those extensions.

There will be intelligent processes to ensure that good extensions get spread to more users. Write something good and people will use it.

Security >>

Is spreading code around between people like that secure you ask? Yes it will be. We know how to make things secure. Yes we do mean that. We have innovative security concepts ready to be implemented. A secure foundation (Lua) is key for that.

People wanted >>

* Evangelisers
* Users / input givers
* Developers

The one and only requirement if you want to help develop is to learn some Lua. (Lua is easy.) We might need code in other languages too, but just for little things. The bulk is gonna be Lua.

Language >>

We'll develop MICL in English because it's probably going to be an international team. Localizing it afterwards is going to be fun. =)

Time frame >>

Quick. We start assembling the team and developing NOW. Once there are devs, we can make a MVP (minimum viable product) in something like two weeks. Yes, I mean that.

Monetary questions >>

It's an open-source project, so probably no $ payout right at the start. But there's gonna be a lot of public attention, so, one way or another, you'll definitely be glad if you're a part of it.

Let's do it. Join now and become famous. And help making something so good that you'll find yourself using it daily.

Cheers,
Stefan Reich ([hidden email]) / www.pussy-riot-germany.de

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Stefan Reich
In reply to this post by Stefan Reich
OK, I'll try to address the various answers in a later post. I do find them a bit more positive than what I got in earlier years :o)

However, there is one thing that bewilders me. I posted the proposal for something that is both NotThatMuchWork and ReallyGood, and it will be a huge service for Lua (as it exploits Lua's unique properties to do something unprecedented). And this is the central list for Lua.

But still - so far, exactly no developers came forward.

None.

Out of 2,000 people, NOBODY is open to join a new project? Like, really, nobody?

What is it? Is the claim too outrageous? The concept really is solid, I can tell you that.

Or is it that everybody works for money and has no time?

Or is this list cursed in some way?

Really, I want to find out. I am going crazy if nobody joins my projects forever. It can't be like this, it makes no sense. They're too good for that :)

What's going on?
Stefan
Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Peter Hickman-3
Well your reputation can hardly help and neither did this line in your post:

(btw, women in Hamburg suck. Sorry... I had to sneak that in there... it needs out =)

Actually it didn't need to be said. At all. Ever.

As to the concept being solid, that is doubtful. The mathematical proof of computer programs is up there alongside hard AI. And quite frankly I expect hard AI to be cracked, P = NP proven and flying pigs before we can prove programs mathematically. In all probability you will need P = NP if you are going to have a mathematical proof of a computer program anyhow.

If you think it is NotThatMuchWork then post your solution to the halting problem and prove to us you have the chops to prove a computer program mathematically.


Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Stefan Reich
>> (btw, women in Hamburg suck. Sorry... I had to sneak that in there... it needs out =)
> Actually it didn't need to be said. At all. Ever.

What, that is now a crime? Saying what is on your heart? Letting it out? How insane are you?

> The mathematical proof of computer programs is up there alongside hard AI.

No. Also you don't know the concept yet, and neither the proof. You just don't understand anything yet. Talk about something you don't understand? Is that your thing?

> prove to us you have the chops to prove a computer program mathematically.

I've done that, and quite extensively. It's a research topic I have engaged in.

But what do I say, you're just one more negative person in this world...

Stefan



On Sat, Apr 27, 2013 at 10:33 PM, Peter Hickman <[hidden email]> wrote:
Well your reputation can hardly help and neither did this line in your post:

(btw, women in Hamburg suck. Sorry... I had to sneak that in there... it needs out =)

Actually it didn't need to be said. At all. Ever.

As to the concept being solid, that is doubtful. The mathematical proof of computer programs is up there alongside hard AI. And quite frankly I expect hard AI to be cracked, P = NP proven and flying pigs before we can prove programs mathematically. In all probability you will need P = NP if you are going to have a mathematical proof of a computer program anyhow.

If you think it is NotThatMuchWork then post your solution to the halting problem and prove to us you have the chops to prove a computer program mathematically.



Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Corey Richardson
In reply to this post by Stefan Reich
On Sat, Apr 27, 2013 at 4:09 PM, Stefan Reich
<[hidden email]> wrote:

> OK, I'll try to address the various answers in a later post. I do find them
> a bit more positive than what I got in earlier years :o)
>
> However, there is one thing that bewilders me. I posted the proposal for
> something that is both NotThatMuchWork and ReallyGood, and it will be a huge
> service for Lua (as it exploits Lua's unique properties to do something
> unprecedented). And this is the central list for Lua.
>
> But still - so far, exactly no developers came forward.
>

We're busy people with jobs and school and family. Generally how it
works in open source is you do work on your thing, make an early
release, and people attracted to the idea will try and work on it.

Also, from my perspective, the idea sounds extremely difficult, if not
impossible, definitely NotThatMuchWork. I, and I doubt many others,
don't want to commit time to a huge unknown like that.

So, work on a proof of concept, and go from there. If it actually
works as claimed, I'll be one of the first to jump on board.

--
Corey

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Corey Richardson
On Sat, Apr 27, 2013 at 4:58 PM, Corey Richardson <[hidden email]> wrote:

> On Sat, Apr 27, 2013 at 4:09 PM, Stefan Reich
> <[hidden email]> wrote:
>> OK, I'll try to address the various answers in a later post. I do find them
>> a bit more positive than what I got in earlier years :o)
>>
>> However, there is one thing that bewilders me. I posted the proposal for
>> something that is both NotThatMuchWork and ReallyGood, and it will be a huge
>> service for Lua (as it exploits Lua's unique properties to do something
>> unprecedented). And this is the central list for Lua.
>>
>> But still - so far, exactly no developers came forward.
>>
>
> We're busy people with jobs and school and family. Generally how it
> works in open source is you do work on your thing, make an early
> release, and people attracted to the idea will try and work on it.
>
> Also, from my perspective, the idea sounds extremely difficult, if not
> impossible, definitely NotThatMuchWork. I, and I doubt many others,
> don't want to commit time to a huge unknown like that.
>
> So, work on a proof of concept, and go from there. If it actually
> works as claimed, I'll be one of the first to jump on board.
>

A related quote, a favorite of mine:

An extraordinary claim requires extraordinary proof.

- Marcello Truzzi

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Sean Conner
In reply to this post by Stefan Reich
It was thus said that the Great Stefan Reich once stated:

> OK, I'll try to address the various answers in a later post. I do find
> them a bit more positive than what I got in earlier years :o)
>
> However, there is one thing that bewilders me. I posted the proposal for
> something that is both NotThatMuchWork and ReallyGood, and it will be a
> huge service for Lua (as it exploits Lua's unique properties to do
> something unprecedented). And this is the central list for Lua.
>
> But still - so far, exactly no developers came forward.
>
> None.
>
> Out of 2,000 people, NOBODY is open to join a new project? Like, really,
> nobody?
>
> What is it? Is the claim too outrageous? The concept really is solid, I can
> tell you that.

  The claims are out there and there's nothing to see.  Even Linus Torvalds
had *something* to show people when we asked for help.  It's one thing to
have a grandiose idea; it's another to execute on said idea.

  Furthre more, you said:

> MICL is scheduled to be a portable application that is - MATHEMATICALLY
> PROVABLY - the most intelligent command line possible.
>
> Yes you read that right. The proof is already in my head. Yes, I have
> studied mathematics =)

but the proof does me no good because I can't read it (nor your mind).
Would it be so hard to write the proof down for the rest of us?  That might
convince a few more people that you are on to something.

> Or is it that everybody works for money and has no time?
>
> Or is this list cursed in some way?

  I can only speak for me, but why would I join an empty project?  I have
plenty of empty grandiose projects of my own to work on, and one or two
not-so-empty grandiose projects.  

  You keep saying that MICL is this wonderful project, but even if I wanted
to help, what now?  Do you have any design plans that are more specific than
"MICL can imitate other shells"?  

  A past project I worked on, Gnugol [1], started with *something* that
could be worked with, feedback could be given, requirements hashed out,
designed argued about.  It wasn't just David Taht going "Hey everybody!
Let's make a command line search client!"

> Really, I want to find out. I am going crazy if nobody joins my projects
> forever. It can't be like this, it makes no sense. They're too good for
> that :)

  Write the damn code!  Don't expect people to write your project for you
(that's called "work" and people get "paid" to do this).  Get something that
people can look at.  And if the "minimum viable product" can be done in two
weeks, DO IT!  You!  You need to do the "minimum viable product" and show
it.  

  You need to give us something more than just "it'll be great!"

  -spc

[1] https://github.com/dtaht/Gnugol/tree/master/src



Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Marco Salamone
You need to provide some additional information-- your pitch includes no technical details as to how you would accomplish your design goals.


On Sat, Apr 27, 2013 at 4:44 PM, Sean Conner <[hidden email]> wrote:
It was thus said that the Great Stefan Reich once stated:
> OK, I'll try to address the various answers in a later post. I do find
> them a bit more positive than what I got in earlier years :o)
>
> However, there is one thing that bewilders me. I posted the proposal for
> something that is both NotThatMuchWork and ReallyGood, and it will be a
> huge service for Lua (as it exploits Lua's unique properties to do
> something unprecedented). And this is the central list for Lua.
>
> But still - so far, exactly no developers came forward.
>
> None.
>
> Out of 2,000 people, NOBODY is open to join a new project? Like, really,
> nobody?
>
> What is it? Is the claim too outrageous? The concept really is solid, I can
> tell you that.

  The claims are out there and there's nothing to see.  Even Linus Torvalds
had *something* to show people when we asked for help.  It's one thing to
have a grandiose idea; it's another to execute on said idea.

  Furthre more, you said:

> MICL is scheduled to be a portable application that is - MATHEMATICALLY
> PROVABLY - the most intelligent command line possible.
>
> Yes you read that right. The proof is already in my head. Yes, I have
> studied mathematics =)

but the proof does me no good because I can't read it (nor your mind).
Would it be so hard to write the proof down for the rest of us?  That might
convince a few more people that you are on to something.

> Or is it that everybody works for money and has no time?
>
> Or is this list cursed in some way?

  I can only speak for me, but why would I join an empty project?  I have
plenty of empty grandiose projects of my own to work on, and one or two
not-so-empty grandiose projects.

  You keep saying that MICL is this wonderful project, but even if I wanted
to help, what now?  Do you have any design plans that are more specific than
"MICL can imitate other shells"?

  A past project I worked on, Gnugol [1], started with *something* that
could be worked with, feedback could be given, requirements hashed out,
designed argued about.  It wasn't just David Taht going "Hey everybody!
Let's make a command line search client!"

> Really, I want to find out. I am going crazy if nobody joins my projects
> forever. It can't be like this, it makes no sense. They're too good for
> that :)

  Write the damn code!  Don't expect people to write your project for you
(that's called "work" and people get "paid" to do this).  Get something that
people can look at.  And if the "minimum viable product" can be done in two
weeks, DO IT!  You!  You need to do the "minimum viable product" and show
it.

  You need to give us something more than just "it'll be great!"

  -spc

[1]     https://github.com/dtaht/Gnugol/tree/master/src




Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Craig Barnes
Wow, syntax bike-shedding I can tolerate, but
these Stefan Reich threads need to be killed
with fire.

Sorry, had to be said.

Reply | Threaded
Open this post in threaded view
|

Re: Most Intelligent Command Line ("MICL") - pitch for devs

Greg Lewin
isn't this all a troll?


On 27 April 2013 23:57, Craig Barnes <[hidden email]> wrote:
Wow, syntax bike-shedding I can tolerate, but
these Stefan Reich threads need to be killed
with fire.

Sorry, had to be said.


12