Smalltalk › Pharo › Pharo Smalltalk Users

Data Science with Pharo?

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

20 messages Options

Volkert

Jul 30, 2014; 7:47pm

Data Science with Pharo?

Here a really nice video on interactive data aquiring, analysis und
visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i
understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

Tudor Girba-2

Jul 30, 2014; 7:57pm

Re: Data Science with Pharo?

Moose is not about software analysis, but about data analysis :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

www.tudorgirba.com

"Every thing has its own flow"

stepharo

Jul 30, 2014; 8:12pm

Re: Data Science with Pharo?

On 30/7/14 21:57, Tudor Girba wrote:

Moose is not about software analysis, but about data analysis :)

+1
and you can morph it to your needs and participate :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

Volkert

Jul 30, 2014; 8:38pm

Re: Data Science with Pharo?

Please watch the video.

BW,
Volkert

Am 30.07.2014 um 22:12 schrieb stepharo:

On 30/7/14 21:57, Tudor Girba wrote:

Moose is not about software analysis, but about data analysis :)

+1
and you can morph it to your needs and participate :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

... [show rest of quote]

-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

Volkert

Jul 30, 2014; 8:48pm

Re: Data Science with Pharo?

In reply to this post by Tudor Girba-2

i thought it was only for sw analyse and visualization. At least it was ;-)

Will have a look ;-)

Am 30.07.2014 um 21:57 schrieb Tudor Girba:

Moose is not about software analysis, but about data analysis :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

Tudor Girba-2

Jul 30, 2014; 9:45pm

Re: Data Science with Pharo?

In reply to this post by Volkert

I did watch the movie. The only interesting thing it has are the types. Besides that, from what I can see in the demo, the interactivity and the options of the GTPlayground/GTInspector is, in my opinion, better.

And we just got started :).

Doru

On Wed, Jul 30, 2014 at 10:38 PM, [hidden email] <[hidden email]> wrote:

Please watch the video.

BW,
Volkert

Am 30.07.2014 um 22:12 schrieb stepharo:
On 30/7/14 21:57, Tudor Girba wrote:

Moose is not about software analysis, but about data analysis :)

+1
and you can morph it to your needs and participate :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

... [show rest of quote]
-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

www.tudorgirba.com

"Every thing has its own flow"

Volkert

Jul 31, 2014; 5:55am

Re: Data Science with Pharo?

I like to concept of the Type Providers and the Deedle library.

http://fsharp.github.io/FSharp.Data/

http://msdn.microsoft.com/en-us/library/hh156509.aspx

http://bluemountaincapital.github.io/Deedle/

BW,
Volkert

Am 30.07.2014 um 23:45 schrieb Tudor Girba:

I did watch the movie. The only interesting thing it has are the types. Besides that, from what I can see in the demo, the interactivity and the options of the GTPlayground/GTInspector is, in my opinion, better.

And we just got started :).

Doru
On Wed, Jul 30, 2014 at 10:38 PM, [hidden email] <[hidden email]> wrote:
Please watch the video.

BW,
Volkert

Am 30.07.2014 um 22:12 schrieb stepharo:
On 30/7/14 21:57, Tudor Girba wrote:

Moose is not about software analysis, but about data analysis :)

+1
and you can morph it to your needs and participate :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

... [show rest of quote]
-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
... [show rest of quote]
--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

stepharo

Jul 31, 2014; 6:15am

Re: Data Science with Pharo?

On 31/7/14 07:55, [hidden email] wrote:

I like to concept of the Type Providers and the Deedle library.

http://fsharp.github.io/FSharp.Data/

http://msdn.microsoft.com/en-us/library/hh156509.aspx

http://bluemountaincapital.github.io/Deedle/

Build a library for Pharo and it will be even more exciting. Doing is exciting.
I started to read a book on R and looking at their data manipulation. I think that we can have the same.
Now we can be either contemplative or doing.
I chose my camp. In which one are you?

Stef

BW,
Volkert

Am 30.07.2014 um 23:45 schrieb Tudor Girba:
I did watch the movie. The only interesting thing it has are the types. Besides that, from what I can see in the demo, the interactivity and the options of the GTPlayground/GTInspector is, in my opinion, better.

And we just got started :).

Doru
On Wed, Jul 30, 2014 at 10:38 PM, [hidden email] <[hidden email]> wrote:
Please watch the video.

BW,
Volkert

Am 30.07.2014 um 22:12 schrieb stepharo:
On 30/7/14 21:57, Tudor Girba wrote:

Moose is not about software analysis, but about data analysis :)

+1
and you can morph it to your needs and participate :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

... [show rest of quote]
-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
... [show rest of quote]
--

www.tudorgirba.com

"Every thing has its own flow"
... [show rest of quote]
-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

Volkert

Jul 31, 2014; 7:00am

Re: Data Science with Pharo?

I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)

But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
my problems i move forward ...

BW,
Volkert

Am 31.07.2014 um 08:15 schrieb stepharo:

On 31/7/14 07:55, [hidden email] wrote:

I like to concept of the Type Providers and the Deedle library.

http://fsharp.github.io/FSharp.Data/

http://msdn.microsoft.com/en-us/library/hh156509.aspx

http://bluemountaincapital.github.io/Deedle/

Build a library for Pharo and it will be even more exciting. Doing is exciting.
I started to read a book on R and looking at their data manipulation. I think that we can have the same.
Now we can be either contemplative or doing.
I chose my camp. In which one are you?

Stef
BW,
Volkert

Am 30.07.2014 um 23:45 schrieb Tudor Girba:
I did watch the movie. The only interesting thing it has are the types. Besides that, from what I can see in the demo, the interactivity and the options of the GTPlayground/GTInspector is, in my opinion, better.

And we just got started :).

Doru
On Wed, Jul 30, 2014 at 10:38 PM, [hidden email] <[hidden email]> wrote:
Please watch the video.

BW,
Volkert

Am 30.07.2014 um 22:12 schrieb stepharo:
On 30/7/14 21:57, Tudor Girba wrote:

Moose is not about software analysis, but about data analysis :)

+1
and you can morph it to your needs and participate :)

Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:

http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/

http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/

Doru

On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:

Here a really nice video on interactive data aquiring, analysis und visualization with F#.

http://channel9.msdn.com/posts/Understanding-the-World-with-F

Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
it puts the focus on sw analysis and sw visualization. What about a new
easy to use and interactive data science platform?

BW,
Volkert

--
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

... [show rest of quote]

--

www.tudorgirba.com

"Every thing has its own flow"

... [show rest of quote]

... [show rest of quote]
-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
... [show rest of quote]
--

www.tudorgirba.com

"Every thing has its own flow"
... [show rest of quote]
-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
... [show rest of quote]

... [show rest of quote]

-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

Marcus Denker-4

Jul 31, 2014; 7:07am

Re: Data Science with Pharo?

On 31 Jul 2014, at 09:00, [hidden email] wrote:

> I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)
>
> But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
> my problems i move forward ...
>
Why not investing into what you use now instead of abandoning it? The “outside” solution only is better because someone
build it. If that someone would have just moved on (“I would have done in F# but the other system was better so I moved there”),
it would not exist.

Everything that exists is there because someone did it.

But strangely, most people see the world much more passive: things are there (“ex nihilo”), to be used. Passively.

Marcus

Volkert

Jul 31, 2014; 7:59am

Re: Data Science with Pharo?

Am 31.07.2014 um 09:07 schrieb Marcus Denker:

> On 31 Jul 2014, at 09:00, [hidden email] wrote:
>
>> I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)
>>
>> But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
>> my problems i move forward ...
>>
> Why not investing into what you use now instead of abandoning it? The “outside” solution only is better because someone
> build it. If that someone would have just moved on (“I would have done in F# but the other system was better so I moved there”),
> it would not exist.

I am investing ..... but my time is limited and if i am faster the other
road, i go the road ... stupid, or? Btw, that
is the reason, i am using Pharo and Roassal and i have never been so
fast :-)

>
> Everything that exists is there because someone did it.
Yes ...

>
> But strangely, most people see the world much more passive: things are there (“ex nihilo”), to be used. Passively.
>
> Marcus
My view: The evolution of mankind is about building better tools with/on
top of existing tools ... we
have to choose the tools we use, to create our own tools. If Pharo and
its framework/libraries matches,
to the problems i have, it is perfect ... if not, i can start to add all
the missing pieces or look for better
tools ... that is the way it goes ...

BW,
Volkert

kilon.alios

Jul 31, 2014; 10:38am

Re: Data Science with Pharo?

I dont know anything about data sciences but this looks like the dilema I was into. I love making 3d graphics and I love pharo. But those two dont mix togethere. Available 3d libraries for pharo are nowhere near my needs. As an artist and coder I use Blender and its API. The Blender Api is not even in C so I can wrap it with nativeboost. Its in python. So the whole thing at first seemed too much trouble for little gain.

With the help of great pharo people I found the way to exploit Blender via sockets I can use its python API and I can do this even using pharo syntax. To my surpriseb it was easy, fast and simple.Not that I have an issue coding in python. The project is WIP but so far it has been very succesful and it can only get better.

So I can only agree with the saying "where there is a will there is a way"

Afterall combining programming languages is a very popular choice.

Στις 31 Ιουλ 2014 10:59 π.μ., ο χρήστης "[hidden email]" <[hidden email]> έγραψε:

Am 31.07.2014 um 09:07 schrieb Marcus Denker:

On 31 Jul 2014, at 09:00, [hidden email] wrote:

I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)

But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
my problems i move forward ...

Why not investing into what you use now instead of abandoning it? The “outside” solution only is better because someone
build it. If that someone would have just moved on (“I would have done in F# but the other system was better so I moved there”),
it would not exist.

I am investing ..... but my time is limited and if i am faster the other road, i go the road ... stupid, or? Btw, that
is the reason, i am using Pharo and Roassal and i have never been so fast :-)

Everything that exists is there because someone did it.

Yes ...

But strangely, most people see the world much more passive: things are there (“ex nihilo”), to be used. Passively.

Marcus

My view: The evolution of mankind is about building better tools with/on top of existing tools ... we
have to choose the tools we use, to create our own tools. If Pharo and its framework/libraries matches,
to the problems i have, it is perfect ... if not, i can start to add all the missing pieces or look for better
tools ... that is the way it goes ...

BW,
Volkert

... [show rest of quote]

Volkert

Jul 31, 2014; 11:49am

Re: Data Science with Pharo?

Of course, integration is always an option ... but not a good one in any case, even if there is a way ... ;-)

I have seen your Pharo/Blender video. Nice work ...

BW,
Volkert

Am 31.07.2014 um 12:38 schrieb kilon alios:

I dont know anything about data sciences but this looks like the dilema I was into. I love making 3d graphics and I love pharo. But those two dont mix togethere. Available 3d libraries for pharo are nowhere near my needs. As an artist and coder I use Blender and its API. The Blender Api is not even in C so I can wrap it with nativeboost. Its in python. So the whole thing at first seemed too much trouble for little gain.

With the help of great pharo people I found the way to exploit Blender via sockets I can use its python API and I can do this even using pharo syntax. To my surpriseb it was easy, fast and simple.Not that I have an issue coding in python. The project is WIP but so far it has been very succesful and it can only get better.

So I can only agree with the saying "where there is a will there is a way"

Afterall combining programming languages is a very popular choice.

Στις 31 Ιουλ 2014 10:59 π.μ., ο χρήστης "[hidden email]" <[hidden email]> έγραψε:

Am 31.07.2014 um 09:07 schrieb Marcus Denker:

On 31 Jul 2014, at 09:00, [hidden email] wrote:

I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)

But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
my problems i move forward ...

Why not investing into what you use now instead of abandoning it? The “outside” solution only is better because someone
build it. If that someone would have just moved on (“I would have done in F# but the other system was better so I moved there”),
it would not exist.

I am investing ..... but my time is limited and if i am faster the other road, i go the road ... stupid, or? Btw, that
is the reason, i am using Pharo and Roassal and i have never been so fast :-)

Everything that exists is there because someone did it.

Yes ...

But strangely, most people see the world much more passive: things are there (“ex nihilo”), to be used. Passively.

Marcus

My view: The evolution of mankind is about building better tools with/on top of existing tools ... we
have to choose the tools we use, to create our own tools. If Pharo and its framework/libraries matches,
to the problems i have, it is perfect ... if not, i can start to add all the missing pieces or look for better
tools ... that is the way it goes ...

BW,
Volkert

... [show rest of quote]

... [show rest of quote]

-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

abergel

Jul 31, 2014; 1:29pm

Re: Data Science with Pharo?

In reply to this post by Volkert

> I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)

Glad to ear this :-)

Yes, the vision I see behind Pharo, Moose, Roassal is to provide a solid alternative to R. We are not there yet, but we will get there.

Cheers,
Alexandre

>
> But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
> my problems i move forward ...
>
> BW,
> Volkert
>
> Am 31.07.2014 um 08:15 schrieb stepharo:
>>
>> On 31/7/14 07:55, [hidden email] wrote:
>>> I like to concept of the Type Providers and the Deedle library.
>>>
>>> http://fsharp.github.io/FSharp.Data/
>>>
>>> http://msdn.microsoft.com/en-us/library/hh156509.aspx
>>>
>>> http://bluemountaincapital.github.io/Deedle/
>>
>> Build a library for Pharo and it will be even more exciting. Doing is exciting.
>> I started to read a book on R and looking at their data manipulation. I think that we can have the same.
>> Now we can be either contemplative or doing.
>> I chose my camp. In which one are you?
>>
>> Stef
>>
>>
>>
>>
>>>
>>> BW,
>>> Volkert
>>>
>>>
>>>
>>> Am 30.07.2014 um 23:45 schrieb Tudor Girba:
>>>> I did watch the movie. The only interesting thing it has are the types. Besides that, from what I can see in the demo, the interactivity and the options of the GTPlayground/GTInspector is, in my opinion, better.
>>>>
>>>> And we just got started :).
>>>>
>>>> Doru
>>>>
>>>>
>>>> On Wed, Jul 30, 2014 at 10:38 PM, [hidden email] <[hidden email]> wrote:
>>>> Please watch the video.
>>>>
>>>> BW,
>>>> Volkert
>>>>
>>>> Am 30.07.2014 um 22:12 schrieb stepharo:
>>>>>
>>>>> On 30/7/14 21:57, Tudor Girba wrote:
>>>>>> Moose is not about software analysis, but about data analysis :)
>>>>>
>>>>> +1
>>>>> and you can morph it to your needs and participate :)
>>>>>>
>>>>>> Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:
>>>>>> http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/
>>>>>> http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/
>>>>>>
>>>>>>
>>>>>> Doru
>>>>>>
>>>>>>
>>>>>> On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:
>>>>>> Here a really nice video on interactive data aquiring, analysis und visualization with F#.
>>>>>>
>>>>>> http://channel9.msdn.com/posts/Understanding-the-World-with-F
>>>>>>
>>>>>> Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
>>>>>> it puts the focus on sw analysis and sw visualization. What about a new
>>>>>> easy to use and interactive data science platform?
>>>>>>
>>>>>> BW,
>>>>>> Volkert
>>>>>>
>>>>>> --
>>>>>> www.nivoba.de
>>>>>>
>>>>>> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> www.tudorgirba.com
>>>>>>
>>>>>> "Every thing has its own flow"
>>>>>
>>>>
>>>> --
>>>>
>>>> www.nivoba.de
>>>>
>>>>
>>>> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> www.tudorgirba.com
>>>>
>>>> "Every thing has its own flow"
>>>
>>> --
>>>
>>> www.nivoba.de
>>>
>>>
>>> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>>>
>>
>
> --
>
> www.nivoba.de
>
>
> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>

... [show rest of quote]

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

kilon.alios

Jul 31, 2014; 2:43pm

Re: Data Science with Pharo?

In reply to this post by Volkert

Python works via integration 50% of its code is written in C. Same applies for its libraries. Tons of ported C/C++ libraries.Before that I was using Delphi a Pascal implementation that was using around the same amount of C++ Windows libraries. Two languages with millions of professional developers worldwide.

I assume F# is a .NET language with direct access to .NET libraries. And of course we should not forget several popular java languages with Clojure being the latest example.

Not only integration is a popular very successful choice but it wont be an exaggeration to state that it has been the foundation of the modern software industry.

So no I don't agree even though my own implementation has a long way to go before its ideal.

Of course, integration is always an option ... but not a good one in any case, even if there is a way ... ;-)

I have seen your Pharo/Blender video. Nice work ...

BW,
Volkert

Am 31.07.2014 um 12:38 schrieb kilon alios:

I dont know anything about data sciences but this looks like the dilema I was into. I love making 3d graphics and I love pharo. But those two dont mix togethere. Available 3d libraries for pharo are nowhere near my needs. As an artist and coder I use Blender and its API. The Blender Api is not even in C so I can wrap it with nativeboost. Its in python. So the whole thing at first seemed too much trouble for little gain.

With the help of great pharo people I found the way to exploit Blender via sockets I can use its python API and I can do this even using pharo syntax. To my surpriseb it was easy, fast and simple.Not that I have an issue coding in python. The project is WIP but so far it has been very succesful and it can only get better.

So I can only agree with the saying "where there is a will there is a way"

Afterall combining programming languages is a very popular choice.

Στις 31 Ιουλ 2014 10:59 π.μ., ο χρήστης "[hidden email]" <[hidden email]> έγραψε:

Am 31.07.2014 um 09:07 schrieb Marcus Denker:

On 31 Jul 2014, at 09:00, [hidden email] wrote:

I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)

But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
my problems i move forward ...

Why not investing into what you use now instead of abandoning it? The “outside” solution only is better because someone
build it. If that someone would have just moved on (“I would have done in F# but the other system was better so I moved there”),
it would not exist.

I am investing ..... but my time is limited and if i am faster the other road, i go the road ... stupid, or? Btw, that
is the reason, i am using Pharo and Roassal and i have never been so fast :-)

Everything that exists is there because someone did it.

Yes ...

But strangely, most people see the world much more passive: things are there (“ex nihilo”), to be used. Passively.

Marcus

My view: The evolution of mankind is about building better tools with/on top of existing tools ... we
have to choose the tools we use, to create our own tools. If Pharo and its framework/libraries matches,
to the problems i have, it is perfect ... if not, i can start to add all the missing pieces or look for better
tools ... that is the way it goes ...

BW,
Volkert

... [show rest of quote]

... [show rest of quote]

-- 
www.nivoba.de

"The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW

Offray

Jul 31, 2014; 5:01pm

Data Narratives with Pharo (Re: Data Science with Pharo?)

In reply to this post by abergel

Hi,

Is really nice to see this talk happening. I want to use Pharo for data
narratives (it can be open/garage/citizen science or data Journalism)
and I have recently talked about this on this list and the ones of Moose
and SciSmalltalk, but seems that the thread initialized by Volkert got
more traction, so I will continue here, which seems like a better place.

I would like to advocate that the conditions for a Notebook for visual
data narratives and open/citizen/garage science made in Smalltalk are
there now that we have Pharo 3, Roassal and Moose Playgrounds. I think
that this notebook could be made essentially by putting these
playgrounds ordered in a tree/outliner. I have made a similar argument
for a tree like interface for IPython notebooks on [1] and I'm using
this tree-like writing experience for my own writing of my PhD Thesis
[2] (Spanish) and I have started to explore some visual data narratives
in Roassal/Moose[3] (Spanish).

[1] http://mutabit.com/mutabit/default/wiki/ipython-deepness
[2]
http://mutabit.com/offray/static/blog/output/posts/la-forma-en-que-escribo-para-el-doctorado.html
[3]
http://mutabit.com/offray/static/blog/output/posts/borrachos-bochinche-futbol.html

I'm just making a mockup of this using Moose browsers while I learn
Pharo with this problem oriented approach. I will ask in another thread
my specific questions about the issues I found while working on this
project. I think that the Tudor's video about PostgreSQL and
visualization is a glimpse of the possibilities and Kilion work
interfacing with blender/python shows that this outline could be even
interface with IPython notebooks. Outliners can be a proper metaphor for
exploratory computing and writing and as org-mode has shown, this simple
but powerful metaphor can be a game changer. This are good times for
data narratives in Pharo. I would try to help with this.

Cheers,

Offray

On 07/31/2014 08:29 AM, Alexandre Bergel wrote:

>> I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)
>
> Glad to ear this :-)
>
> Yes, the vision I see behind Pharo, Moose, Roassal is to provide a solid alternative to R. We are not there yet, but we will get there.
>
> Cheers,
> Alexandre
>
>
>
>>
>> But it is always good the see, what is happening outside the "camp" ... and yes, if i find outside better solutions to
>> my problems i move forward ...
>>
>> BW,
>> Volkert
>>
>> Am 31.07.2014 um 08:15 schrieb stepharo:
>>>
>>> On 31/7/14 07:55, [hidden email] wrote:
>>>> I like to concept of the Type Providers and the Deedle library.
>>>>
>>>> http://fsharp.github.io/FSharp.Data/
>>>>
>>>> http://msdn.microsoft.com/en-us/library/hh156509.aspx
>>>>
>>>> http://bluemountaincapital.github.io/Deedle/
>>>
>>> Build a library for Pharo and it will be even more exciting. Doing is exciting.
>>> I started to read a book on R and looking at their data manipulation. I think that we can have the same.
>>> Now we can be either contemplative or doing.
>>> I chose my camp. In which one are you?
>>>
>>> Stef
>>>
>>>
>>>
>>>
>>>>
>>>> BW,
>>>> Volkert
>>>>
>>>>
>>>>
>>>> Am 30.07.2014 um 23:45 schrieb Tudor Girba:
>>>>> I did watch the movie. The only interesting thing it has are the types. Besides that, from what I can see in the demo, the interactivity and the options of the GTPlayground/GTInspector is, in my opinion, better.
>>>>>
>>>>> And we just got started :).
>>>>>
>>>>> Doru
>>>>>
>>>>>
>>>>> On Wed, Jul 30, 2014 at 10:38 PM, [hidden email] <[hidden email]> wrote:
>>>>> Please watch the video.
>>>>>
>>>>> BW,
>>>>> Volkert
>>>>>
>>>>> Am 30.07.2014 um 22:12 schrieb stepharo:
>>>>>>
>>>>>> On 30/7/14 21:57, Tudor Girba wrote:
>>>>>>> Moose is not about software analysis, but about data analysis :)
>>>>>>
>>>>>> +1
>>>>>> and you can morph it to your needs and participate :)
>>>>>>>
>>>>>>> Look at the Roassal demos, or at the work around the GTPlayground or GTInspector:
>>>>>>> http://www.humane-assessment.com/blog/browsing-files-with-gtinspector-video/
>>>>>>> http://www.humane-assessment.com/blog/dynamic-exploration-of-a-postgres-db-with-the-gtinspector/
>>>>>>>
>>>>>>>
>>>>>>> Doru
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Jul 30, 2014 at 9:47 PM, [hidden email] <[hidden email]> wrote:
>>>>>>> Here a really nice video on interactive data aquiring, analysis und visualization with F#.
>>>>>>>
>>>>>>> http://channel9.msdn.com/posts/Understanding-the-World-with-F
>>>>>>>
>>>>>>> Is this the way Pharo + Roassal + ... goes? Ok, we have moose, but a i understand
>>>>>>> it puts the focus on sw analysis and sw visualization. What about a new
>>>>>>> easy to use and interactive data science platform?
>>>>>>>
>>>>>>> BW,
>>>>>>> Volkert
>>>>>>>
>>>>>>> --
>>>>>>> www.nivoba.de
>>>>>>>
>>>>>>> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> www.tudorgirba.com
>>>>>>>
>>>>>>> "Every thing has its own flow"
>>>>>>
>>>>>
>>>>> --
>>>>>
>>>>> www.nivoba.de
>>>>>
>>>>>
>>>>> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> www.tudorgirba.com
>>>>>
>>>>> "Every thing has its own flow"
>>>>
>>>> --
>>>>
>>>> www.nivoba.de
>>>>
>>>>
>>>> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>>>>
>>>
>>
>> --
>>
>> www.nivoba.de
>>
>>
>> "The more complex an object, the larger the investment in learning to use it, and the greater the resistance to abandon it.", NW
>>
>

... [show rest of quote]

Sean P. DeNigris

Jul 31, 2014; 9:08pm

Re: Data Narratives with Pharo (Re: Data Science with Pharo?)

Administrator

Offray wrote

Is really nice to see this talk happening

+1. And please keep posting screencasts/shots on work with data. I'm frequently at non-Smalltalk tech meetings in NYC and there is demand for this. The more videos I have to explain the better chance of bringing in new users :)

Cheers,
Sean

Thierry Goubier

Aug 01, 2014; 3:38pm

Re: Data Science with Pharo?

In reply to this post by abergel

Le 31/07/2014 15:29, Alexandre Bergel a écrit :
>> I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)
> Glad to ear this :-)
>
> Yes, the vision I see behind Pharo, Moose, Roassal is to provide a solid alternative to R. We are not there yet, but we will get there.
Cool. I'm interested by that objective :)

Thierry

Volkert

Aug 01, 2014; 10:20pm

Re: Data Science with Pharo?

Am 01.08.2014 um 17:38 schrieb Thierry Goubier:

> Le 31/07/2014 15:29, Alexandre Bergel a écrit :
>>> I am successful using Pharo + Roassal/Mondrian in analyzing and
>>> visualizing IT architecture models. And it is real fun. :-)
>> Glad to ear this :-)
>>
>> Yes, the vision I see behind Pharo, Moose, Roassal is to provide a
>> solid alternative to R. We are not there yet, but we will get there.
> Cool. I'm interested by that objective :)
>
> Thierry
>

Me too

BW,
Volkert

Tudor Girba-2

Aug 03, 2014; 1:56pm

Re: Data Science with Pharo?

Talking about the vision of Moose: The stated goal of Moose is to enable an engineer to craft a data analysis in under 15 minutes, whatever the analysis: metric, query, clustering, visualization, browsing etc.

The goal of Moose is to enable an engineer to craft a data analysis in under 15 minutes ... whatever the analysis: metric, query, clustering, visualization, browsing etc.

On the one hand, this will change software development. For example, you should not manually search for code or objects. Instead, you should build tools that do it for you. The core idea around the Glamorous Toolkit is to offer a moldable development environment that allows you to do just that. This is not a regular toolkit, but it one that embodies a new philosophy for software development.

On the other hand, we want to build pieces that are generic enough so that they can be instantiated for any domain while still being accessible. For example, we think that one should not require a PhD to write a visualization or to build an interactive browser, and we showed how this is possible :). This is why it is so easy and inexpensive to build a completely novel kind of inspector that capture the imagination quite rapidly. And at this point, people did not even pay much of an attention to the moldable debugger :).

It's true that we are not where R is in terms of number crunching analyses, but we have other things around interaction and analysis composition that are quite unique. We would indeed benefit from more R-like capabilities, but at this point Moose does no longer play a catch-up game. It is forging its own separate place that others start to look at. And this is possible precisely because Moose is almost indistinguishable from Pharo.

I do believe we are sitting on a goldmine, and that it is up to us to exploit it.

Doru

On Sat, Aug 2, 2014 at 12:20 AM, [hidden email] <[hidden email]> wrote:

Am 01.08.2014 um 17:38 schrieb Thierry Goubier:

Le 31/07/2014 15:29, Alexandre Bergel a écrit :

I am successful using Pharo + Roassal/Mondrian in analyzing and visualizing IT architecture models. And it is real fun. :-)

Glad to ear this :-)

Yes, the vision I see behind Pharo, Moose, Roassal is to provide a solid alternative to R. We are not there yet, but we will get there.

Cool. I'm interested by that objective :)

Thierry

Me too

BW,
Volkert

... [show rest of quote]

www.tudorgirba.com

"Every thing has its own flow"