Stupid question about Fuel

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

Stupid question about Fuel

Edgar De Cleene
Mariano, Martin, Pavel

> For example, a class can be considered either a global or a regular object. In
> the former case, it will be encoded just its name; in the latter case, the
> class will be encoded in detail, with its method dictionary, etc.

How you do this ?
>the class will be encoded in detail, with its method dictionary, etc.

I export most of 1.4 OneClick classes in http://squeakros.org/Pharo%201dot4/
as cs.gz, like to have .fuel for load classes back in target image with no
compiler.

Thanks in advance

Edgar




Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Marcus Denker-4

On Apr 15, 2012, at 1:20 PM, Edgar J. De Cleene wrote:

> Mariano, Martin, Pavel
>
>> For example, a class can be considered either a global or a regular object. In
>> the former case, it will be encoded just its name; in the latter case, the
>> class will be encoded in detail, with its method dictionary, etc.
>
> How you do this ?
>> the class will be encoded in detail, with its method dictionary, etc.
>
> I export most of 1.4 OneClick classes in http://squeakros.org/Pharo%201dot4/
> as cs.gz, like to have .fuel for load classes back in target image with no
> compiler.

Of course that can not work. .cs is source code, you of course need a compiler
(if Fuel could load cs.gz, it *would be* a compiler).

What you need to do is to serialize the *Objects* that are code (Classes, MethodDictioaries,
CompiledMethods) using Fuel. Then the result can be loaded back using Fuel.

        Marcus

--
Marcus Denker -- http://marcusdenker.de


Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Edgar De Cleene



On 4/15/12 8:24 AM, "Marcus Denker" <[hidden email]> wrote:

> Of course that can not work. .cs is source code, you of course need a compiler
> (if Fuel could load cs.gz, it *would be* a compiler).
 

Yes, I know it.
I have the .cs.gz for normal load, not for Fuel !!!
 
> What you need to do is to serialize the *Objects* that are code (Classes,
> MethodDictioaries,
> CompiledMethods) using Fuel. Then the result can be loaded back using Fuel.

What I ask is how to have the .fuel of classes as regular objects as Mariano
write in his blog.


Then , this Foo.fuel could be loaded into target image.
Some more granular as Pavel made with complete packages some days ago.

And wish made experiments if this Foo.fuel could be loaded into different
fork as Squeak, Cuis having Fuel into.


> Marcus

Thanks !

Edgar



Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Marcus Denker-4
In reply to this post by Edgar De Cleene

On Apr 15, 2012, at 1:44 PM, Edgar J. De Cleene wrote:
>
> And wish made experiments if this Foo.fuel could be loaded into different
> fork as Squeak, Cuis having Fuel into.
>


I would not waste time with that... we will improve Pharo constantly, and it will
be very hard (if not impossible) to keep some kind of compatibility layer on a binary
code level between Pharo, Squeak and Cuis.

And what do you get out of it? Wouldn't it be better to use the time to improve one system
and make sure that system has a future?

        Marcus

--
Marcus Denker -- http://marcusdenker.de


Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Edgar De Cleene



On 4/15/12 8:50 AM, "Marcus Denker" <[hidden email]> wrote:

> I would not waste time with that... we will improve Pharo constantly, and it
> will
> be very hard (if not impossible) to keep some kind of compatibility layer on a
> binary
> code level between Pharo, Squeak and Cuis.
>
> And what do you get out of it? Wouldn't it be better to use the time to
> improve one system
> and make sure that system has a future?
>
> Marcus

And if none of actual forks have all as I wish?
All have some good and all have some problem for me.
It's why try to learn from all here and all in Squeak and from Juan.

Don't' sell me the fish, teach me to fishing.

Edgar



Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Marcus Denker-4
In reply to this post by Edgar De Cleene

On Apr 15, 2012, at 2:11 PM, Edgar J. De Cleene wrote:

>
>
>
> On 4/15/12 8:50 AM, "Marcus Denker" <[hidden email]> wrote:
>
>> I would not waste time with that... we will improve Pharo constantly, and it
>> will
>> be very hard (if not impossible) to keep some kind of compatibility layer on a
>> binary
>> code level between Pharo, Squeak and Cuis.
>>
>> And what do you get out of it? Wouldn't it be better to use the time to
>> improve one system
>> and make sure that system has a future?
>>
>> Marcus
>
> And if none of actual forks have all as I wish?
> All have some good and all have some problem for me.
> It's why try to learn from all here and all in Squeak and from Juan.
>

Learning does not mean to be binary compatible.


--
Marcus Denker -- http://marcusdenker.de


Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Pavel Krivanek-3
In reply to this post by Edgar De Cleene
Hi Edgar,

some weeks ago I played with an idea to have Pharo Kernel as the base
of Squeak. I compared the images to see how much work it would
constitute. I found out two things:
- it would not require too much work. It would be not simple but it is possible
- there is no reason to do that

I can imagine Squeak as an extension of Pharo but I do not see the
purpose and target users of the resultant system.

-- Pavel

On Sun, Apr 15, 2012 at 2:15 PM, Edgar J. De Cleene
<[hidden email]> wrote:

>
>
>
> On 4/15/12 8:50 AM, "Marcus Denker" <[hidden email]> wrote:
>
>> I would not waste time with that... we will improve Pharo constantly, and it
>> will
>> be very hard (if not impossible) to keep some kind of compatibility layer on a
>> binary
>> code level between Pharo, Squeak and Cuis.
>>
>> And what do you get out of it? Wouldn't it be better to use the time to
>> improve one system
>> and make sure that system has a future?
>>
>> Marcus
>
> And if none of actual forks have all as I wish?
> All have some good and all have some problem for me.
> It's why try to learn from all here and all in Squeak and from Juan.
>
> Don't' sell me the fish, teach me to fishing.
>
> Edgar
>
>
>

Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

David T. Lewis
In reply to this post by Marcus Denker-4
On Sun, Apr 15, 2012 at 01:50:06PM +0200, Marcus Denker wrote:

>
> On Apr 15, 2012, at 1:44 PM, Edgar J. De Cleene wrote:
> >
> > And wish made experiments if this Foo.fuel could be loaded into different
> > fork as Squeak, Cuis having Fuel into.
>
> I would not waste time with that... we will improve Pharo constantly, and it will
> be very hard (if not impossible) to keep some kind of compatibility layer on a binary
> code level between Pharo, Squeak and Cuis.
>
> And what do you get out of it? Wouldn't it be better to use the time to improve one system
> and make sure that system has a future?

Fuel seems to be well written and organized. Even though I have only used
Fuel a little (for remote task execution with OSProcess), I was able to
easily understand the code when I loaded it in Squeak.

There is no reason that a well-written package cannot be maintained for
several flavors of the image. That is a sign of competent developers doing
good work :)

Dave


Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Edgar De Cleene
In reply to this post by Pavel Krivanek-3



On 4/15/12 12:02 PM, "Pavel Krivanek" <[hidden email]> wrote:

> I can imagine Squeak as an extension of Pharo but I do not see the
> purpose and target users of the resultant system.

Today we read

> Some is interested in using Smalltalk for prototyping of games.
> http://www.gamedev.net/topic/623380-game-programming-in-smalltalk/
>
> I think we should respond very constructively on gamedev.net to entice
> gamers to try Smalltalk. Programmers are coming to appreciate dynamic
> languages and we can capitalize on that.
>
> Go Smalltalk,
> Aik-Siong Koh

A ideal all purpose system IMHO is Pharo Kernel as base, with all new ideas
Pharo bring us.

Cuis is the best implementation of Morph, but lacks compatibility.
You need re think all and make from zero

Squeak is a mess, no doubt about.
But have necessary backwards compatibility (for me).
I can't do Morphic games without grab Morphs with Halo.
A simple thing like my SmartTrashMorph ported to Cuis or to Pharo takes too
much time to do.

For young people like most here caring about Web 2.0, Pharo is the answer
and Markus is right, we need focus on better Pharo.

But for old people like me, need responses NOW , and know wish more as Web
2.0.

When GsoC ends, several projects give me 95% of info for learn how to build
my ideal system.


As PharoKernel builder and developer, I ask for tutorials for fools.

Cheers

Edgar



Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Mariano Martinez Peck
In reply to this post by David T. Lewis


On Sun, Apr 15, 2012 at 5:52 PM, David T. Lewis <[hidden email]> wrote:
On Sun, Apr 15, 2012 at 01:50:06PM +0200, Marcus Denker wrote:
>
> On Apr 15, 2012, at 1:44 PM, Edgar J. De Cleene wrote:
> >
> > And wish made experiments if this Foo.fuel could be loaded into different
> > fork as Squeak, Cuis having Fuel into.
>
> I would not waste time with that... we will improve Pharo constantly, and it will
> be very hard (if not impossible) to keep some kind of compatibility layer on a binary
> code level between Pharo, Squeak and Cuis.
>
> And what do you get out of it? Wouldn't it be better to use the time to improve one system
> and make sure that system has a future?

Fuel seems to be well written and organized. Even though I have only used
Fuel a little (for remote task execution with OSProcess), I was able to
easily understand the code when I loaded it in Squeak.

There is no reason that a well-written package cannot be maintained for
several flavors of the image. That is a sign of competent developers doing
good work :)


Thanks Dave for the nice words :)
At the beginning (first releases of Fuel) most answers and feedback from the community was "yet another serializer?" "why you have started again from scratch?"
Now, some time later, people is slowly understanding why we have started from scratch ;)

Cheers



--
Mariano
http://marianopeck.wordpress.com

Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Pavel Krivanek-3
In reply to this post by Edgar De Cleene
On Sun, Apr 15, 2012 at 6:05 PM, Edgar J. De Cleene
<[hidden email]> wrote:

>
>
>
> On 4/15/12 12:02 PM, "Pavel Krivanek" <[hidden email]> wrote:
>
>> I can imagine Squeak as an extension of Pharo but I do not see the
>> purpose and target users of the resultant system.
>
> Today we read
>
>> Some is interested in using Smalltalk for prototyping of games.
>> http://www.gamedev.net/topic/623380-game-programming-in-smalltalk/
>>
>> I think we should respond very constructively on gamedev.net to entice
>> gamers to try Smalltalk. Programmers are coming to appreciate dynamic
>> languages and we can capitalize on that.
>>
>> Go Smalltalk,
>> Aik-Siong Koh
>
> A ideal all purpose system IMHO is Pharo Kernel as base, with all new ideas
> Pharo bring us.
>
> Cuis is the best implementation of Morph, but lacks compatibility.
> You need re think all and make from zero
>
> Squeak is a mess, no doubt about.
> But have necessary backwards compatibility (for me).
> I can't do Morphic games without grab Morphs with Halo.
> A simple thing like my SmartTrashMorph ported to Cuis or to Pharo takes too
> much time to do.
>
> For young people like most here caring about Web 2.0, Pharo is the answer
> and Markus is right, we need focus on better Pharo.
>
> But for old people like me, need responses NOW , and know wish more as Web
> 2.0.
>
> When GsoC ends, several projects give me 95% of info for learn how to build
> my ideal system.
>
>
> As PharoKernel builder and developer, I ask for tutorials for fools.
>
> Cheers
>
> Edgar
>

As far as I remember, the last big discussion about Squeak and Pharo
joining was "Meeting Report for 8/18/2010":
http://forum.world.st/Meeting-Report-for-8-18-2010-tt2332389.html#none

We may be happy that this two forks share the same virtual machine. I
didn't see any real Squeak vision and nothing changed from that time.
You may try to change it and convince Squeak community about some.

Cheers,
-- Pavel

Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

philippeback
In reply to this post by Pavel Krivanek-3
I think that pharo serves a need. The need for something solid and free from restrictions related to software.

Got interested in Smallatalk and Squeak quite a long time ago. But never moved past the toy-stage with them. Went homoiconic with Tcl and "OO" with the whole Java chain.

With Pharo, there is a sense of progress, cleanliness, and will to push the boundary by making some hard choices.

And this made me interested in how things worked inside, and invest time and energy in making it a key target platform for my ideas-materialization activities in my business. Because business thrives on something solid, and innovation doesn't like arbitrary restrictions.

Owning and mastering the whole chain from VM up to Image detail is huge. In Squeak, it is accretion. I think Pharo leans more towards the "crystallization" side of things. (Yep, I drank the kool-aid).

Philippe

2012/4/15 Pavel Krivanek <[hidden email]>
Hi Edgar,

some weeks ago I played with an idea to have Pharo Kernel as the base
of Squeak. I compared the images to see how much work it would
constitute. I found out two things:
- it would not require too much work. It would be not simple but it is possible
- there is no reason to do that

I can imagine Squeak as an extension of Pharo but I do not see the
purpose and target users of the resultant system.

-- Pavel

On Sun, Apr 15, 2012 at 2:15 PM, Edgar J. De Cleene
<[hidden email]> wrote:
>
>
>
> On 4/15/12 8:50 AM, "Marcus Denker" <[hidden email]> wrote:
>
>> I would not waste time with that... we will improve Pharo constantly, and it
>> will
>> be very hard (if not impossible) to keep some kind of compatibility layer on a
>> binary
>> code level between Pharo, Squeak and Cuis.
>>
>> And what do you get out of it? Wouldn't it be better to use the time to
>> improve one system
>> and make sure that system has a future?
>>
>> Marcus
>
> And if none of actual forks have all as I wish?
> All have some good and all have some problem for me.
> It's why try to learn from all here and all in Squeak and from Juan.
>
> Don't' sell me the fish, teach me to fishing.
>
> Edgar
>
>
>




--
Philippe Back
"Helping you hit the top 3 outcomes you really want to achieve"

Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 Mail: [hidden email] | Web: http://philippeback.eu | Blog:

High Octane SPRL
rue cour Boisacq 101
1301 Bierges

Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Stéphane Ducasse
In reply to this post by Edgar De Cleene
> I can't do Morphic games without grab Morphs with Halo.

there are halos in Pharo.

> A simple thing like my SmartTrashMorph ported to Cuis or to Pharo takes too
> much time to do.


Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Stéphane Ducasse
In reply to this post by philippeback

On Apr 15, 2012, at 7:59 PM, [hidden email] wrote:

> I think that pharo serves a need. The need for something solid and free from restrictions related to software.
>
> Got interested in Smallatalk and Squeak quite a long time ago. But never moved past the toy-stage with them. Went homoiconic with Tcl and "OO" with the whole Java chain.
>
> With Pharo, there is a sense of progress, cleanliness, and will to push the boundary by making some hard choices.

Thanks and I can tell you that we have more plans :)

> And this made me interested in how things worked inside, and invest time and energy in making it a key target platform for my ideas-materialization activities in my business. Because business thrives on something solid, and innovation doesn't like arbitrary restrictions.
>
> Owning and mastering the whole chain from VM up to Image detail is huge. In Squeak, it is accretion. I think Pharo leans more towards the "crystallization" side of things. (Yep, I drank the kool-aid).
>

I'm not sure. But we want a system to invent the next one.THis is why we work on the bootstrap and soon on new class builder and first instance variable work of camillo and toon.
Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Marcus Denker-4
In reply to this post by Marcus Denker-4

On Apr 15, 2012, at 5:52 PM, David T. Lewis wrote:
>
>
> There is no reason that a well-written package cannot be maintained for
> several flavors of the image. That is a sign of competent developers doing
> good work :)

What I want to avoid is the following: Imagine we have X. But sadly, X is not
easy to do in Squeak. Will we not do X *just because* it will be hard to port code using
X to Squeak?

        Marcus
       

--
Marcus Denker -- http://marcusdenker.de


Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Edgar De Cleene
In reply to this post by Pavel Krivanek-3



On 4/15/12 2:53 PM, "Pavel Krivanek" <[hidden email]> wrote:

> We may be happy that this two forks share the same virtual machine. I
> didn't see any real Squeak vision and nothing changed from that time.
> You may try to change it and convince Squeak community about some.
>
> Cheers,
> -- Pavel

I running for the Board this year and if I become elected, try to convince
SOB members to wake up of comatose status of Squeak.

You was right, Pharo have a vision and SOB members complain about Andreas
not leading the community.

Also I wish Andreas come back, but is time to grow up.

Cheers

Edgar

P.S. Apple should survive Steve pass away .....



Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Edgar De Cleene
In reply to this post by philippeback
Re: [Pharo-project] Stupid question about Fuel


On 4/15/12 2:59 PM, "[hidden email]" <[hidden email]> wrote:

I think that pharo serves a need. The need for something solid and free from restrictions related to software.

Got interested in Smallatalk and Squeak quite a long time ago. But never moved past the toy-stage with them. Went homoiconic with Tcl and "OO" with the whole Java chain.

With Pharo, there is a sense of progress, cleanliness, and will to push the boundary by making some hard choices.

And this made me interested in how things worked inside, and invest time and energy in making it a key target platform for my ideas-materialization activities in my business. Because business thrives on something solid, and innovation doesn't like arbitrary restrictions.

Owning and mastering the whole chain from VM up to Image detail is huge. In Squeak, it is accretion. I think Pharo leans more towards the "crystallization" side of things. (Yep, I drank the kool-aid).

Philippe


Agree with all

Edgar
Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Edgar De Cleene
In reply to this post by Stéphane Ducasse



On 4/15/12 6:14 PM, "Stéphane Ducasse" <[hidden email]> wrote:

> there are halos in Pharo.

Can you as stupid example grab a Browser window , duplicate and rotate it in
no more as two clicks?

Maybe I do not know Pharo well...

Give me a Pharo image which behave as Squeak for Morph in the surface and
have all deeper as clear as Juan Cuis and I sign.

And I atart to like Pharo , only need some PharoForGames in Morphic NOW.

Cheers

Edgar




Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Stéphane Ducasse
You can get the halos but not all the time in your face.
Click + shift + option on mac bring the halos.
We do value direct manipulation but only when we want it, not all the time
Stef

On Apr 16, 2012, at 10:52 AM, Edgar J. De Cleene wrote:

>
>
>
> On 4/15/12 6:14 PM, "Stéphane Ducasse" <[hidden email]> wrote:
>
>> there are halos in Pharo.
>
> Can you as stupid example grab a Browser window , duplicate and rotate it in
> no more as two clicks?
>
> Maybe I do not know Pharo well...
>
> Give me a Pharo image which behave as Squeak for Morph in the surface and
> have all deeper as clear as Juan Cuis and I sign.
>
> And I atart to like Pharo , only need some PharoForGames in Morphic NOW.
>
> Cheers
>
> Edgar
>
>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: Stupid question about Fuel

Edgar De Cleene



On 4/16/12 7:45 AM, "Stéphane Ducasse" <[hidden email]> wrote:

> You can get the halos but not all the time in your face.
> Click + shift + option on mac bring the halos.
> We do value direct manipulation but only when we want it, not all the time
> Stef


Mmmm.

Yes, you select some submorph, not the root one.
And a preference could do the trick...

In others forks you click middle button morph and manipulate Debug, Browser,
Workspace.
One click and duplicate and you have two Debug windows and could go several
steps back in one and see where you start in the other.

Show this to game developers !!!


Very thanks !!

Edgar



12