On Jul 31, 2014, at 6:05 AM, Helge Nowak hknowak@yahoo.de wrote:

Well, the question may also be: what IS Smalltalk? The ANSI Standard? ST-80? Whose implementation IS Smalltalk in that case? I think many current implementations may safely say that they are Smalltalk-inpired.

I think you are overgeneralizing. Doru is saying that Pharo is not sticking to (let’s say) the historical concept of Smalltalk. Sure, you may imply from that that some Smalltalks ARE sticking to that concept. But IMO it is not valid to conclude that he is saying that ALL of them are like that.

I don’t follow your reasoning. Divorce is a very strong and aggressive statement. I don’t see that from what Doru is saying. Could you provide more detail for your train of thought?

Just my 0.02EUR ...

---> Save our in-boxes! http://emailcharter.org <---

Johan Fabry  -  http://pleiad.cl/~jfabry
PLEIAD lab  -  Computer Science Department (DCC)  -  University of Chile

2014-07-31 12:05 GMT+02:00 Helge Nowak hknowak@yahoo.de:

Well, I tried to open a smalltalk-80 image recently and it didn't work on
the Pharo VM.
In addition, it seems that recent Pharo images cannot run on the Xerox-D
microcoded machines.

So I think clearly Pharo is not Smalltalk for obvious backward
compatibility issues.

I love mails about this topic it really makes my day each time :-).

Clement

On 31 Jul 2014, at 12:05, Helge Nowak hknowak@yahoo.de wrote:

The whole thing started by the fact that there are lots of great things to do that Smalltalk did not yet do. (That are 100% in line with the idea of Smalltalk, and that
are very natural and clearly the next step).

But if we say that we are Smalltalk, then people will immediately request 100% compatibility forever.

We actually had exactly that case. Someone told us that for every change we need to first sync with all Smalltalk Vendors and all other open source
Smalltalks and only of they are ok (and implement the same at the same time) we are allowed to e.g. add a new API in Pharo core classes.
Reason: “Pharo is Smalltalk, it says right so on the website”.

So Pharo is Smalltalk in the sense of being a reflective, dynamic, always improving environment. It is not Smalltalk in the sense of a finished, non-changing artefact.

Marcus

Dear Johan,

I think there is a common understanding of what Smalltalk is and what is not: Ruby, Java and many others ARE Smalltalk inspired. Up to now I considered all Smalltalks that carry that term and also Squeak and the Pharo fork as "true" Smalltalks.
By Doru's words he considers Pharo more in the "inspired, non-Smalltalk" corner. That is ok, yet has consequences. One of those is that in this light Pharo is leaving the Smalltalk community. Whether you reckon the word "divorce" for that departure aggressive is your personal feeling. Divorce is a usual term describing the splitting of a former whole. I don't find it aggressive.

Smalltalk as a technology, philosophy and community has always evolved - to the future (to what else?). Claiming that one leaves a community in emphasizing that one wants to move to the future
silently implies that that community didn't move to the future, i. e.
stays as is. I don't think that this is a correct observation with regards to Smalltalk. If it was intended to say that Pharo wants to move away from some Smalltalk
concept (technical or non-technical) it should be said so, and whith regards to which concept and
into what differing direction. There are many ways to the future. No problem with that. Smalltalk will walk its path.

Cheers
Helge

„Wege entstehen dadurch, dass man sie geht.“, Franz Kafka

Von: Johan Fabry jfabry@dcc.uchile.cl
An: Helge Nowak hknowak@yahoo.de
CC: "pharo-business@lists.pharo.org" pharo-business@lists.pharo.org; "pharo-consortium@lists.gforge.inria.fr" pharo-consortium@lists.gforge.inria.fr; ESUG esug-list@lists.esug.org
Gesendet: 12:24 Donnerstag, 31.Juli 2014
Betreff: Re: [Esug-list] "Pharo is Smalltalk inspired"

On Jul 31, 2014, at 6:05 AM, Helge Nowak hknowak@yahoo.de wrote:

Well, the question may also be: what IS Smalltalk? The ANSI Standard? ST-80? Whose implementation IS Smalltalk in that case? I think many current implementations may safely say that they are Smalltalk-inpired.

I think you are overgeneralizing. Doru is saying that Pharo is not sticking to (let’s say) the historical concept of Smalltalk. Sure, you may imply from that that some Smalltalks ARE sticking to that concept. But IMO it is not valid to conclude that he is saying that ALL of them are like that.

I don’t follow your reasoning. Divorce is a very strong and aggressive statement. I don’t see that from what Doru is saying. Could you provide more detail for your train of thought?

Just my 0.02EUR ...

---> Save our in-boxes! http://emailcharter.org <---

Johan Fabry   -  http://pleiad.cl/~jfabry
PLEIAD lab  -  Computer Science Department (DCC)  -  University of Chile

Dear Clément,

thank you for your valuable contribution. So, what you are saying is that none of the Smalltalks around (VA Smalltalk, GemStone, Smalltalk/X, Dolphin, VisualWorks, ObjectStudio  etc, pp.) is Smalltalk.

Kent Beck once used the term "balkanization". The point of my email was to open eyes to not run into that behaviour! It won't help the Smalltalk community - to which I still count Pharo ... and hope you do too!
Cheers
Helge

Von: Clément Bera bera.clement@gmail.com
An: Helge Nowak hknowak@yahoo.de
CC: "pharo-business@lists.pharo.org" pharo-business@lists.pharo.org; "pharo-consortium@lists.gforge.inria.fr" pharo-consortium@lists.gforge.inria.fr; ESUG esug-list@lists.esug.org
Gesendet: 13:00 Donnerstag, 31.Juli 2014
Betreff: Re: [Esug-list] "Pharo is Smalltalk inspired"

2014-07-31 12:05 GMT+02:00 Helge Nowak hknowak@yahoo.de:

Dear Pharoers,

upon Doru’s (BTW excellent, as usual) presentation on Live Objects at NDC 2014.
In there he states “Pharo is Smalltalk inspired. … we want to point ourselves
that we are Smalltalk inspired because we want to move towards the future”. This
implies three things:

the future

divorced from the community that gave them birth

wondering whether this is indeed the official position of the Pharo community?
And how the Smalltalkers think about it.

Well, I tried to open a smalltalk-80 image recently and it didn't work on the Pharo VM. 
In addition, it seems that recent Pharo images cannot run on the Xerox-D microcoded machines.

So I think clearly Pharo is not Smalltalk for obvious backward compatibility issues.

I love mails about this topic it really makes my day each time :-).

Clement
 

On 7/31/14 4:09 , Helge Nowak wrote:

So, how do you define "Smalltalk community"?

Andres.

On 31 Jul 2014, at 13:09, Helge Nowak hknowak@yahoo.de wrote:

Squeak definitly never had that goal. When I discovered it it was explicitly stated that is was “Work in progress based on Smalltalk 80 with which it is still reasonably compatible”.

Yes, the freedom to do things that are not in all the other Smalltalks already. Not because these are “not Smalltalk”, only because they where not there in the past, even though maybe they
should have.

Marcus

Dear Helge, Johan et al,
to me, it seems obvious that Pharo is Smalltalk, in a sense in which
Ruby et al are not Smalltalk although they are Smalltalk-inspired.  Of
course, all Smalltalks are also inspired by earlier Smalltalks and by
other current Smalltalks.

All Smalltalk dialects share a good deal of base class library
compatibility but of course even as regards non-UI, what runs on one
Smalltalk may DNU on another, unless written with a cross-dialect goal
in view or using a compatibility layer or with some code differences in
dialect versions.  When you get into the area of UI, compatibility is
more the exception than the rule.

It may be that the talk's phrasing was accidentally a bit misleading -
or was intentionally exaggerated to avoid a "Smalltalk - isn't that old"
reaction from that particular audience.

I don't mind exactly what is said in any particular talk oriented to a
particular audience, but it seems clear to me that Pharo is a dialect of
Smalltalk;  it is not just Smalltalk-inspired.

      Just my 0.02p FWIW
            Niall Ross

Helge Nowak wrote:

The problem with the Smalltalk community is that it holds itself back.
For example there are no Smalltalks with a decent namespace
implementation (although there was Dave Simmons' S# for a while). The
single survivor seems to be NewSpeak which clearly chose to 'divorce'
itself from Smalltalk 'proper'.

I recall Pharo was created because Squeak did not cater to the
professional market and Pharo would conquer the FLOSS portion of that
niche. Seeing that Pharo did not get host window support, nor decent
namespaces it seems fair to say that the community did not get its act
together -- and if we stretch it we might say that the community held
Pharo back.

What I see in the Smalltalk community is a giant circle-jerk (direct
object manipulationz! refactoringz! TDDz! xUnitz! we are greatz!) with
people wallowing in past greatness.

The reality (a 'correct' observation as per the above?) is that people
are experimenting with new(ish) software development methodologies in
other environments nowadays.
The example-du-jour is of course Bret Victor who proposes an IDE where
we can flatten abstractions (like time) into 2d so our brains can
readily grasp and predict consequences of code alterations. Seeing that
Apple xCode IDE adopts this paradigm (with Swift) before the Smalltalk
IDE did is telling.

Seeing that Pharo and Squeak are still producing browser framework after
browser framework and -oh yeah- let's reify packages as objects and we
still need a JIT, I observe a lot of ant-like activity at ground level
and very little 'giants' activity at the 'how to encode my abstractions'
level. (Another example: there still is no accepted paradigm in the
Smalltalk community that instructs us how to document implementation
requirements and decisions -- go figure).
So the programmer's discourse with the Smalltalk machine has not changed
many times in the past, perhaps just once during the introduction of
refactorings and TDD at roughly the same time.

What the Smalltalk community seems to miss is that 'Smalltalk 2.0 is
dead, long live Smalltalk 3.0' feeling.

Perhaps it behooves ESUG to create a session where the community
finally buries Smalltalk 1.0 and perhaps also pick the date where we
sunset Smalltalk 2.0.

Go Doru,
Pharo desperately wants to escape Smalltalk 1.0 but the community
won't let you.

Reinout