Re: Squeak evolution 1998 - 2008

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

Re: Squeak evolution 1998 - 2008

Hannes Hirzel
On 10/26/13, Bob Arning <[hidden email]> wrote:
> http://69.251.218.6:9116/ can give you a pretty good picture of the
> evolution in this area.
>
> Cheers,
> Bob


Bob

does it contain _all_ the change sets from 1998 - 2008? It seems that
in 2008 the update mechanism was changed to Monticello files, right?


Did you think it would be possible of running an  from 0001 to 7179,
so to say "replaying the evolution"?

It would be nice to have a graphical display of which areas have been
touched and which ones not. Possibly in the form of a movie.... (e.g.
a rectangle representing a class category sized according to code size
with colors indicating changes, animated gif?)

--Hannes



0001tk_test.doit
"Just a test if the update broadcasting is working"
Transcript show: ' You got an external update'; cr.

0002tk_collapse_RF.cs
'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

.........


7179AdvanceTo3dot10dot2.cs
>From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
2008 at 2:53:19 pm'

Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
Well, it contains all that I have found. There are some gaps in the numbering, so I don't know if I missed some or that gap appeared in the original stream. And, yes, there was a switch to change sets simply loading some Monticello file(s), although there was not a crisp changeover.

I'll give some thought to a graphical version.

Cheers,
Bob


On 10/26/13 8:42 AM, H. Hirzel wrote:
On 10/26/13, Bob Arning [hidden email] wrote:
http://69.251.218.6:9116/ can give you a pretty good picture of the
evolution in this area.

Cheers,
Bob

Bob

does it contain _all_ the change sets from 1998 - 2008? It seems that
in 2008 the update mechanism was changed to Monticello files, right?


Did you think it would be possible of running an  from 0001 to 7179,
so to say "replaying the evolution"?

It would be nice to have a graphical display of which areas have been
touched and which ones not. Possibly in the form of a movie.... (e.g.
a rectangle representing a class category sized according to code size
with colors indicating changes, animated gif?)

--Hannes



0001tk_test.doit
"Just a test if the update broadcasting is working"
Transcript show: ' You got an external update'; cr.

0002tk_collapse_RF.cs
'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

.........


7179AdvanceTo3dot10dot2.cs
>From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
2008 at 2:53:19 pm'





Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Sam Adams-2

This is a great resource, Bob!  Thanks!

I'm wondering how well your infrastructure would work for older change sets, say of the KSC, ParcPlace ST-80 era?

Thar's gold in dem der files!

Regards,
Sam


Sam S. Adams, CTO - Contextual Computing
IBM Distinguished Engineer, IBM Research
Mobile: 919-696-6064, email: [hidden email]
Assistant: Linda R. Morrison. (720) 395-0460 Fax: (845) 491-4318, Tie: 676-0460, [hidden email]
<<Hebrews 11:6, Proverbs 3:5-6, Romans 1:16-17, 1 Corinthians 1:10>>

Inactive hide details for Bob Arning ---10/26/2013 09:54:23 AM---Well, it contains all that I have found. There are some gaps iBob Arning ---10/26/2013 09:54:23 AM---Well, it contains all that I have found. There are some gaps in the  numbering, so I don't know if I

From: Bob Arning <[hidden email]>
To: [hidden email]
Date: 10/26/2013 09:54 AM
Subject: Re: [squeak-dev] Squeak evolution 1998 - 2008
Sent by: [hidden email]





Well, it contains all that I have found. There are some gaps in the numbering, so I don't know if I missed some or that gap appeared in the original stream. And, yes, there was a switch to change sets simply loading some Monticello file(s), although there was not a crisp changeover.

I'll give some thought to a graphical version.

Cheers,
Bob


On 10/26/13 8:42 AM, H. Hirzel wrote:
    On 10/26/13, Bob Arning [hidden email] wrote:


    Bob

    does it contain _all_ the change sets from 1998 - 2008? It seems that
    in 2008 the update mechanism was changed to Monticello files, right?


    Did you think it would be possible of running an  from 0001 to 7179,
    so to say "replaying the evolution"?

    It would be nice to have a graphical display of which areas have been
    touched and which ones not. Possibly in the form of a movie.... (e.g.
    a rectangle representing a class category sized according to code size
    with colors indicating changes, animated gif?)

    --Hannes



    0001tk_test.doit
    "Just a test if the update broadcasting is working"
    Transcript show: ' You got an external update'; cr.

    0002tk_collapse_RF.cs
    'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

    .........


    7179AdvanceTo3dot10dot2.cs
    >From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
    2008 at 2:53:19 pm'






Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
It would probably work with perhaps a bit of tweaking. The default url points to the change set browser, which simply requires files in chunk format. No concern here about semantics - just search for file(s) containing a string and show them.

The /diffs url might need a bit more work as it tries to understand the meaning of each chunk.

Cheers,
Bob

On 10/26/13 10:05 AM, Sam Adams wrote:

This is a great resource, Bob!  Thanks!

I'm wondering how well your infrastructure would work for older change sets, say of the KSC, ParcPlace ST-80 era?

Thar's gold in dem der files!

Regards,
Sam


Sam S. Adams, CTO - Contextual Computing
IBM Distinguished Engineer, IBM Research
Mobile: 919-696-6064, email: [hidden email]
Assistant: Linda R. Morrison. (720) 395-0460 Fax: (845) 491-4318, Tie: 676-0460, [hidden email]
<<Hebrews 11:6, Proverbs 3:5-6, Romans 1:16-17, 1 Corinthians 1:10>>

Inactive
          hide details for Bob Arning ---10/26/2013 09:54:23 AM---Well,
          it contains all that I have found. There are some gaps iBob Arning ---10/26/2013 09:54:23 AM---Well, it contains all that I have found. There are some gaps in the  numbering, so I don't know if I

From: Bob Arning [hidden email]
To: [hidden email]
Date: 10/26/2013 09:54 AM
Subject: Re: [squeak-dev] Squeak evolution 1998 - 2008
Sent by: [hidden email]





Well, it contains all that I have found. There are some gaps in the numbering, so I don't know if I missed some or that gap appeared in the original stream. And, yes, there was a switch to change sets simply loading some Monticello file(s), although there was not a crisp changeover.

I'll give some thought to a graphical version.

Cheers,
Bob


On 10/26/13 8:42 AM, H. Hirzel wrote:
    On 10/26/13, Bob Arning [hidden email] wrote:


    Bob

    does it contain _all_ the change sets from 1998 - 2008? It seems that
    in 2008 the update mechanism was changed to Monticello files, right?


    Did you think it would be possible of running an  from 0001 to 7179,
    so to say "replaying the evolution"?

    It would be nice to have a graphical display of which areas have been
    touched and which ones not. Possibly in the form of a movie.... (e.g.
    a rectangle representing a class category sized according to code size
    with colors indicating changes, animated gif?)

    --Hannes



    0001tk_test.doit
    "Just a test if the update broadcasting is working"
    Transcript show: ' You got an external update'; cr.

    0002tk_collapse_RF.cs
    'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

    .........


    7179AdvanceTo3dot10dot2.cs
    >From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
    2008 at 2:53:19 pm'







    



Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
In reply to this post by Hannes Hirzel
OK, first iteration of the graphical view is at

http://69.251.218.6:9116/time

each column represents 200 change sets. The numbers in each cell are number of chunks and number of bytes.

I'll clean this up a bit later.

Cheers,
Bob

On 10/26/13 8:42 AM, H. Hirzel wrote:
On 10/26/13, Bob Arning [hidden email] wrote:
http://69.251.218.6:9116/ can give you a pretty good picture of the
evolution in this area.

Cheers,
Bob

Bob

does it contain _all_ the change sets from 1998 - 2008? It seems that
in 2008 the update mechanism was changed to Monticello files, right?


Did you think it would be possible of running an  from 0001 to 7179,
so to say "replaying the evolution"?

It would be nice to have a graphical display of which areas have been
touched and which ones not. Possibly in the form of a movie.... (e.g.
a rectangle representing a class category sized according to code size
with colors indicating changes, animated gif?)

--Hannes



0001tk_test.doit
"Just a test if the update broadcasting is working"
Transcript show: ' You got an external update'; cr.

0002tk_collapse_RF.cs
'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

.........


7179AdvanceTo3dot10dot2.cs
>From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
2008 at 2:53:19 pm'





Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Hannes Hirzel
Interesting view, Bob.

Clearly visible is when the idea of testing started to get track, when
Monticello and Etoys arrived.

Morphic seems to be there right from the start. Is that so?

--Hannes

On 10/26/13, Bob Arning <[hidden email]> wrote:

> OK, first iteration of the graphical view is at
>
> http://69.251.218.6:9116/time
>
> each column represents 200 change sets. The numbers in each cell are
> number of chunks and number of bytes.
>
> I'll clean this up a bit later.
>
> Cheers,
> Bob
>
> On 10/26/13 8:42 AM, H. Hirzel wrote:
>> On 10/26/13, Bob Arning <[hidden email]> wrote:
>>> http://69.251.218.6:9116/ can give you a pretty good picture of the
>>> evolution in this area.
>>>
>>> Cheers,
>>> Bob
>>
>> Bob
>>
>> does it contain _all_ the change sets from 1998 - 2008? It seems that
>> in 2008 the update mechanism was changed to Monticello files, right?
>>
>>
>> Did you think it would be possible of running an  from 0001 to 7179,
>> so to say "replaying the evolution"?
>>
>> It would be nice to have a graphical display of which areas have been
>> touched and which ones not. Possibly in the form of a movie.... (e.g.
>> a rectangle representing a class category sized according to code size
>> with colors indicating changes, animated gif?)
>>
>> --Hannes
>>
>>
>>
>> 0001tk_test.doit
>> "Just a test if the update broadcasting is working"
>> Transcript show: ' You got an external update'; cr.
>>
>> 0002tk_collapse_RF.cs
>> 'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'
>>
>> .........
>>
>>
>> 7179AdvanceTo3dot10dot2.cs
>> >From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
>> 2008 at 2:53:19 pm'
>>
>>
>
>

Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
Morphic right from the start? As far as these change set go Morphic is there, but they represent additions to Squeak 1.31. Not sure about earlier versions. So the question is

Was there ever a version of Squeak without Morphic? Jecel, you know?

Cheers,
Bob


On 10/26/13 3:04 PM, H. Hirzel wrote:
Interesting view, Bob.

Clearly visible is when the idea of testing started to get track, when
Monticello and Etoys arrived.

Morphic seems to be there right from the start. Is that so?

--Hannes

On 10/26/13, Bob Arning [hidden email] wrote:
OK, first iteration of the graphical view is at

http://69.251.218.6:9116/time

each column represents 200 change sets. The numbers in each cell are
number of chunks and number of bytes.

I'll clean this up a bit later.

Cheers,
Bob

On 10/26/13 8:42 AM, H. Hirzel wrote:
On 10/26/13, Bob Arning [hidden email] wrote:
http://69.251.218.6:9116/ can give you a pretty good picture of the
evolution in this area.

Cheers,
Bob
Bob

does it contain _all_ the change sets from 1998 - 2008? It seems that
in 2008 the update mechanism was changed to Monticello files, right?


Did you think it would be possible of running an  from 0001 to 7179,
so to say "replaying the evolution"?

It would be nice to have a graphical display of which areas have been
touched and which ones not. Possibly in the form of a movie.... (e.g.
a rectangle representing a class category sized according to code size
with colors indicating changes, animated gif?)

--Hannes



0001tk_test.doit
"Just a test if the update broadcasting is working"
Transcript show: ' You got an external update'; cr.

0002tk_collapse_RF.cs
'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

.........


7179AdvanceTo3dot10dot2.cs
>From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
2008 at 2:53:19 pm'







Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

J. Vuletich (mail lists)

Morphic appeared in Squeak 1.19. You can download Squeak 1.18 from ftp.squeak.org.  It was the last MVC only release of Squeak. It runs well in my Windows netbook, with an appropriate VM. I'm using Andreas' 1.31.1.4 VM from 1998.

Cheers,

Juan Vuletich

Quoting Bob Arning <[hidden email]>:

Morphic right from the start? As far as these change set go Morphic is there, but they represent additions to Squeak 1.31. Not sure about earlier versions. So the question is

Was there ever a version of Squeak without Morphic? Jecel, you know?

Cheers,
Bob


On 10/26/13 3:04 PM, H. Hirzel wrote:
Interesting view, Bob.

Clearly visible is when the idea of testing started to get track, when
Monticello and Etoys arrived.

Morphic seems to be there right from the start. Is that so?

--Hannes

On 10/26/13, Bob Arning [hidden email] wrote:
OK, first iteration of the graphical view is at

http://69.251.218.6:9116/time

each column represents 200 change sets. The numbers in each cell are
number of chunks and number of bytes.

I'll clean this up a bit later.

Cheers,
Bob

On 10/26/13 8:42 AM, H. Hirzel wrote:
On 10/26/13, Bob Arning [hidden email] wrote:
http://69.251.218.6:9116/ can give you a pretty good picture of the
evolution in this area.

Cheers,
Bob
Bob

does it contain _all_ the change sets from 1998 - 2008? It seems that
in 2008 the update mechanism was changed to Monticello files, right?


Did you think it would be possible of running an  from 0001 to 7179,
so to say "replaying the evolution"?

It would be nice to have a graphical display of which areas have been
touched and which ones not. Possibly in the form of a movie.... (e.g.
a rectangle representing a class category sized according to code size
with colors indicating changes, animated gif?)

--Hannes



0001tk_test.doit
"Just a test if the update broadcasting is working"
Transcript show: ' You got an external update'; cr.

0002tk_collapse_RF.cs
'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

.........


7179AdvanceTo3dot10dot2.cs
>From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
2008 at 2:53:19 pm'



Cheers,
Juan Vuletich



Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Hannes Hirzel
It would be nice to add this information in a ReadMe.txt file in
http://ftp.squeak.org/1.18/

On 10/27/13, J. Vuletich (mail lists) <[hidden email]> wrote:

>
>
> Morphic appeared in Squeak 1.19. You can download Squeak 1.18 from
> ftp.squeak.org.[1]  It was the last MVC only release of Squeak. It runs well
> in my Windows netbook, with an appropriate VM. I'm using Andreas' 1.31.1.4
> VM from 1998.
>
> Cheers,
>
> Juan Vuletich
>
> Quoting Bob Arning <[hidden email]>:
>
>> Morphic right from the start? As far as these change set go Morphic is
>> there, but they represent additions to Squeak 1.31. Not sure about earlier
>> versions. So the question is
>>
>> Was there ever a version of Squeak without Morphic? Jecel, you know?
>>
>> Cheers,
>> Bob
>>
>> On 10/26/13 3:04 PM, H. Hirzel wrote:
>>
>>
>> > Interesting view, Bob.Clearly visible is when the idea of testing
>> > started to get track, when Monticello and Etoys arrived.Morphic seems to
>> > be there right from the start. Is that so?--HannesOn 10/26/13, Bob
>> > Arning <[hidden email]> wrote:
>> >
>> > > OK, first iteration of the graphical view is
>> > > athttp://69.251.218.6:9116/timeeach column represents 200 change sets.
>> > > The numbers in each cell are number of chunks and number of bytes.I'll
>> > > clean this up a bit later.Cheers, BobOn 10/26/13 8:42 AM, H. Hirzel
>> > > wrote:
>> > >
>> > > > On 10/26/13, Bob Arning <[hidden email]> wrote:
>> > > >
>> > > > > http://69.251.218.6:9116/ can give you a pretty good picture of
>> > > > > the evolution in this area.Cheers, Bob
>> > > >
>> > > > Bobdoes it contain _all_ the change sets from 1998 - 2008? It seems
>> > > > that in 2008 the update mechanism was changed to Monticello files,
>> > > > right?Did you think it would be possible of running an  from 0001 to
>> > > > 7179, so to say "replaying the evolution"?It would be nice to have a
>> > > > graphical display of which areas have been touched and which ones
>> > > > not. Possibly in the form of a movie.... (e.g. a rectangle
>> > > > representing a class category sized according to code size with
>> > > > colors indicating changes, animated gif?)--Hannes0001tk_test.doit
>> > > > "Just a test if the update broadcasting is working" Transcript show:
>> > > > ' You got an external update'; cr.0002tk_collapse_RF.cs 'From Squeak
>> > > > 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32
>> > > > pm'.........7179AdvanceTo3dot10dot2.cs >From Squeak3.10.2beta of 5
>> > > > June 2008 [latest update: #7175] on 5 June 2008 at 2:53:19 pm'
>
> Cheers,
> Juan Vuletich
>
> Links:
> ------
> [1] http://ftp.squeak.org
>

Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Jecel Assumpcao Jr
In reply to this post by Bob Arning-2
Bob Arning wrote:

> Morphic right from the start? As far as these change set go Morphic is
> there, but they represent additions to Squeak 1.31. Not sure about
> earlier versions. So the question is
>
> Was there ever a version of Squeak without Morphic? Jecel, you know?

Juan has already replied that Morphic was added to 1.19 (the versions I
remember before that were 1.13, 1.16, 1.17 and 1.18). Note that even
1.31 started up in a MVC project and you had to open a new Morphic one
to play around with that. This was inverted in 2.0, I think.

One issue is that 1998 is not "the start" at all. You might think that
not much is missing when you lack under two years, but it is important
to remember that the rate of development of the Squeak project was
amazing in the beginning compared to what we are now used to. That is
actually common for a low of projects. So one version doesn't have
networking at all, and then next you have a web browser that tries to
play Flash pages.

In terms of lines of code, on the other hand, the rate of development
got faster as time went on rather than slower. This isn't so obvious if
you do some statistics on the various images because there were several
times when a lot of code was removed from the official images. So if you
look at one image and it has 40% more classes/methods/lines of code than
an older one but the older one can sing "Silent Night" and the new one
can't, then the number is smaller than it should be.

Sam Adam asked about extending this back in time (I guess it would have
to be through Apple Smalltalk-80 and then back through Xerox Smalltalks)
and I mentioned a two year gap. It might be a neat idea to convert the
earlier changes into the same format (a lot of the needed information
can be found in the archives for this mailing list) and number them -1,
-2, etc...

-- Jecel


Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
In reply to this post by Bob Arning-2
Munging around with old mailing lists sounds like work. ;-)

What I have done is to take each of the old changes files, subtract the previous changes file and treat the result as if it were a single update. These files will be identified as '**v1' '*113' '*115' etc so that they will sort better.

This is now available on the /diffs page:

http://69.251.218.6:9116/diffs

I'll see about adding this data to the /browse and /time pages.

Cheers,
Bob

sample:



rootForGrabOf:


*119
rootForGrabOf: aMorph
    "Like root, but can be overridden by a morph that wants to allow its submorphs to be extracted just by picking them up."

    (owner = nil or: [owner isWorldOrHandMorph]) ifTrue: [↑ self].
    owner allowSubmorphExtraction
        ifTrue: [↑ self]
        ifFalse: [↑ owner rootForGrabOf: aMorph].

*131
rootForGrabOf: aMorph
    "Like root, but can be overridden by a morph that wants to allow its submorphs to be extracted just by picking them up."

    (owner = nil or: [owner isWorldOrHandMorph]) ifTrue: [↑ self].
    owner allowSubmorphExtraction
        ifTrue: [↑ self]
        ifFalse: [↑ owner rootForGrabOf: aMorph].
    "Be sticky if sticky property is set. Normally this would be recompiled in the unique subclass of each morph, so this is a backstop only, and for development."
    ↑ self isSticky
        ifTrue:
            [nil]
        ifFalse:
            [(owner = nil or: [owner isWorldOrHandMorph])
                ifTrue:
                    [self]
                ifFalse:
                    [owner allowSubmorphExtraction
                        ifTrue: [self]
                        ifFalse: [owner rootForGrabOf: aMorph]]]


0149Flood-sw.cs
rootForGrabOf: aMorph
    "Be sticky if sticky property is set. Normally this would be recompiled in the unique subclass of each morph, so this is a backstop only, and for development."
    ↑ self isSticky
    ↑ (self isSticky and: [self isPartsDonor not])
        ifTrue:
            [nil]
        ifFalse:
            [(owner = nil or: [owner isWorldOrHandMorph])
                ifTrue:
                    [self]
                ifFalse:
                    [owner allowSubmorphExtraction
                        ifTrue: [self]
                        ifFalse: [owner rootForGrabOf: aMorph]]]

2776TruthAndBeauty.cs
rootForGrabOf: aMorph
    self flag: #arNote. "TODO: Remove the entire rootForGrabOf: logic"
    ↑ (self isSticky and: [self isPartsDonor not])
        ifTrue:
            [nil]
        ifFalse:
            [(owner = nil or: [owner isWorldOrHandMorph])
                ifTrue:
                    [self]
                ifFalse:
                    [owner allowSubmorphExtraction
                        ifTrue: [self]
                        ifFalse: [owner rootForGrabOf: aMorph]]]

2783DnD-Day.cs
rootForGrabOf: aMorph
    self flag: #arNote. "TODO: Remove the entire rootForGrabOf: logic"
    ↑ (self isSticky and: [self isPartsDonor not])
        ifTrue:
            [nil]
        ifFalse:
            [(owner = nil or: [owner isWorldOrHandMorph])
                ifTrue:
                    [self]
                ifFalse:
                    [owner allowSubmorphExtraction
                        ifTrue: [self]
                        ifFalse: [owner rootForGrabOf: aMorph]]]
** explicitly removed **

On 10/27/13 2:53 PM, Jecel Assumpcao Jr. wrote:
Bob Arning wrote:

Morphic right from the start? As far as these change set go Morphic is
there, but they represent additions to Squeak 1.31. Not sure about
earlier versions. So the question is

Was there ever a version of Squeak without Morphic? Jecel, you know?
Juan has already replied that Morphic was added to 1.19 (the versions I
remember before that were 1.13, 1.16, 1.17 and 1.18). Note that even
1.31 started up in a MVC project and you had to open a new Morphic one
to play around with that. This was inverted in 2.0, I think.

One issue is that 1998 is not "the start" at all. You might think that
not much is missing when you lack under two years, but it is important
to remember that the rate of development of the Squeak project was
amazing in the beginning compared to what we are now used to. That is
actually common for a low of projects. So one version doesn't have
networking at all, and then next you have a web browser that tries to
play Flash pages.

In terms of lines of code, on the other hand, the rate of development
got faster as time went on rather than slower. This isn't so obvious if
you do some statistics on the various images because there were several
times when a lot of code was removed from the official images. So if you
look at one image and it has 40% more classes/methods/lines of code than
an older one but the older one can sing "Silent Night" and the new one
can't, then the number is smaller than it should be.

Sam Adam asked about extending this back in time (I guess it would have
to be through Apple Smalltalk-80 and then back through Xerox Smalltalks)
and I mentioned a two year gap. It might be a neat idea to convert the
earlier changes into the same format (a lot of the needed information
can be found in the archives for this mailing list) and number them -1,
-2, etc...

-- Jecel






Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
OK, the

http://69.251.218.6:9116/time

has been updated with the pre-update stream stuff.

I'm having a little confusion here - I dont' seem to have where SameGame
entered the image. Anybody know?

Cheers,
Bob

Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
http://69.251.218.6:9116/time2

has been added to sort categories in order of earliest modification

Cheers,
Bob

On 10/27/13 6:01 PM, Bob Arning wrote:
OK, the

http://69.251.218.6:9116/time

has been updated with the pre-update stream stuff.

I'm having a little confusion here - I dont' seem to have where SameGame entered the image. Anybody know?

Cheers,
Bob





Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Bob Arning-2
In reply to this post by Bob Arning-2
Mystery mostly resolved. It *is* in the 2.1 changes file, but not in any separate change set from that era. This is from back when it was more common to download a new version image rather than try to update through.

Cheers,
Bob

On 10/27/13 6:01 PM, Bob Arning wrote:
I'm having a little confusion here - I dont' seem to have where SameGame entered the image. Anybody know?



Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Jecel Assumpcao Jr
In reply to this post by Bob Arning-2
Bob Arning wrote:

> Munging around with old mailing lists sounds like work. ;-)

I should have mentioned that what we can get from the mailing list is
the actual author for every piece of code. People would send stuff to
the list and Dan Ingalls would copy/paste the most interesting code to
his image, which would later become the official image. For the
relicensing effort this was important information to have, but for
casual browsing we don't really need it.

> What I have done is to take each of the old changes files, subtract the
> previous changes file and treat the result as if it were a single update.
> These files will be identified as '**v1' '*113' '*115' etc so that they will
> sort better.

Great job! Congratulations on the results.

-- Jecel


Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Tim Olson-4
In reply to this post by Bob Arning-2

On Oct 28, 2013, at 7:13 AM, Bob Arning <[hidden email]> wrote:

Mystery mostly resolved. It *is* in the 2.1 changes file, but not in any separate change set from that era. This is from back when it was more common to download a new version image rather than try to update through.

Wow, a blast from the past!  I originally wrote the Squeak version of SameGame to better learn some of the ins and outs of the then new Morphic environment which was released in Squeak 2.0.  I supplied it as a change set downloadable from my personal home page, and Dan Ingalls included it in the 2.1 version, along with my JPEGReadWriter code.

Digging through the mailing list archives, it looks like I announced it to the mailing list on 23 May 1998.

— tim



Reply | Threaded
Open this post in threaded view
|

Re: Squeak evolution 1998 - 2008

Eliot Miranda-2
In reply to this post by Bob Arning-2



On Sat, Oct 26, 2013 at 12:46 PM, Nicolas Cellier <[hidden email]> wrote:

"The image is somewhat larger (around 1.9Mb), since it includes both MVC and a fairly complete Morphic system with several running examples. However it is still right around a megabyte if you pare back the interpreter and Morphic, and you can get under 650k by executing Smalltalk>>majorShrink."

wow :-)
 



2013/10/26 Bob Arning <[hidden email]>
Morphic right from the start? As far as these change set go Morphic is there, but they represent additions to Squeak 1.31. Not sure about earlier versions. So the question is

Was there ever a version of Squeak without Morphic? Jecel, you know?

Cheers,
Bob


On 10/26/13 3:04 PM, H. Hirzel wrote:
Interesting view, Bob.

Clearly visible is when the idea of testing started to get track, when
Monticello and Etoys arrived.

Morphic seems to be there right from the start. Is that so?

--Hannes

On 10/26/13, Bob Arning [hidden email] wrote:
OK, first iteration of the graphical view is at

http://69.251.218.6:9116/time

each column represents 200 change sets. The numbers in each cell are
number of chunks and number of bytes.

I'll clean this up a bit later.

Cheers,
Bob

On 10/26/13 8:42 AM, H. Hirzel wrote:
On 10/26/13, Bob Arning [hidden email] wrote:
http://69.251.218.6:9116/ can give you a pretty good picture of the
evolution in this area.

Cheers,
Bob
Bob

does it contain _all_ the change sets from 1998 - 2008? It seems that
in 2008 the update mechanism was changed to Monticello files, right?


Did you think it would be possible of running an  from 0001 to 7179,
so to say "replaying the evolution"?

It would be nice to have a graphical display of which areas have been
touched and which ones not. Possibly in the form of a movie.... (e.g.
a rectangle representing a class category sized according to code size
with colors indicating changes, animated gif?)

--Hannes



0001tk_test.doit
"Just a test if the update broadcasting is working"
Transcript show: ' You got an external update'; cr.

0002tk_collapse_RF.cs
'From Squeak 1.31 of Feb 4, 1998 on 8 May 1998 at 4:31:32 pm'

.........


7179AdvanceTo3dot10dot2.cs
>From Squeak3.10.2beta of 5 June 2008 [latest update: #7175] on 5 June
2008 at 2:53:19 pm'



      

    











--
best,
Eliot