July 2009


Jul 15, 2009: 23. Squeak, Pharo, more forks and quality time


As my health is bad, I write very little in this section, but there is some news about Squeak, about which I wrote earlier in 13.  About Squeak, 15. More about Squeak, 20. Documented Squeak.

I suppose in what follows that you have read these or else are somewhat familiar with Squeak or Smalltalk.

First, what is a fork? A new implementation of a project, here Squeak, that is itself a fork of Smalltalk. Other forks of Squeak are Tweak, Cuis, OpenCroquet, Sophie, Pier, Newspeak and quite a few more.

Google will reveal all if you search the above with an added "Smalltalk". Indeed, if you are interested in Smalltalk at all, it makes sense to look up these and other projects: Some are somewhat succesful or useful, and some of those that are not (yet) may be theoretically interesting. An example of a somewhat succesful forked project based on Squeak is Sophie, that attempts to help one built e-books; an example of a theoretically interesting fork is Newspeak, that is an attempt to improve Smalltalk.

The Pharo in the title is another recent fork of Squeak by a group of its developers who were not happy with the drift and development of Squeak.

About this much could be said, and indeed has been said on Squeak's developers' mailing list, quite often none too friendly, and sometimes also not honestly, I think.

Indeed, as in most socalled "communities", the majority of "We The People" when mailing to the developers' list nearly always pretends to speak for "We" when they mean "Me and my own current interests", and often uses that pretended We-position to scrub anybody who disagrees as an opponent of "Us".

Personally, I can quite well handle this linguistically, logically and satirically in appropriate language - which would explode a mailing list like Squeak if I honestly write what I think, in the language I consider appropriate, fitting and fair, so I do not do this, and only read the list, with a psychologically trained logician's mind, that does not make me happier about people on average.

So I will only make four points here:

First, Squeak has been very slowly developed over the last years (regardless of its direction(s)), in part because it is quite difficult to contribute new source code, while it may last years after such source code as one has succeeded in contributing to a code-repository gets evaluated and included or not in a new Squeak distribution as part of the image.

Second, Squeak has been very slowly developed the last years because there are far more wishes, plans, e-mails and claims than there are people with the time, knowledge and will to implement even a small part of these.

Third, I will soon remove the Squeak Help material on my site, because I have reached the conviction that those naive to Smalltalk are much better of with VisualWorks (also freely available, and hundreds of times better working, better looking and better documented than Squeak, if one is interested in Smalltalk) or with Pharo if one is a software-developer with some prior knowledge of Smalltalk.

Fourth, there is currently a protracted debate going on Squeak's developers' list that pisses me off thoroughly because of its dishonesty and impoliteness, for it is a debate that seems to be based on the following kind of stance

"my balls are bigger and better than yours because I contributed more to Squeak than you did, so We The Squeakpeople think that you-who-criticize-us should shut up and fork off!".

The current tone is as follows - and I cite from today's developers list, a Briton writing to a Briton in the tradition of "Britons Never Will Be Slaves" I suppose:

Remember, there are CCTV cameras everywhere in UK. 
We watch you Keith. :P

Your attitude stinks. Really.

and also niceties like

The community is focusing on you rather than meaningful discussions. For example, (X) spent quality time replying to your message rather than, let's say, commenting on my idea about a migration tool. I am spending the time to write this letter. A real waste of time. (*)

So... I don't think that I can waste more of my own quality time, and it seems to me that I wasted a great lot of my quality time and quality brain - also ill since 32 years, always in pain, always very poor, always without any help any invalid should and would get in any real human civilization - to write documentation for Squeak, which, it seems to me, is still not written because the people who do make money by way of Squeak or Smalltalk and who should be perfectly well capable of writing a properly documented Smalltalk or Squeak in at most a month of work (8 hour day) have simply effectively refused to do so, and refuse to do so since 30 years or so, apparently because they feel too good to write decent documentation of code for others. (**)

This was so in 2002 when Ted Kaehler tried to yoke me to do the documentation work he should have done in 1980 at the latest, and it is still the same with Andreas Raab who is currently lording Squeak in the name of We The People. (***)

It was all an intentional waste of my quality time; my good will; my capacities; and my wish to do applied mathematics and logic, so it is better that people like me fork off, as the Squeaky developers phrase this, to something better, more rewarding, better documented, and with more polite would be developers.

The people interested in spending time and effort on Smalltalk (that still is theoretically a very interesting system) should spend it on VisualWorks, because that is well-documented, works well, and it also is freely available for non-commercial use.

It is presently the only Smalltalk I know that is decently documented and has a tolerable interface, and that one fairly and honestly could propose to people who cannot program or who can program and do not know Smalltalk, to learn Smalltalk in.

P.S. If Eliott Miranda - who used to be at VisualWorks - ever gets his Cog working, which is a new VM for Squeak and forks, possibly Squeak might become more interesting, provided it gets forked, and if the more maoistic e-mailers on Squeak developers' list stick with developing their Vision To Teach All Chidren From 8 To 80 To Program In Squeak (Like A Real Pro Does!).

(*) Although I am a 59-year old psychologist (among other degrees) this keeps amazing me on a list of would be intelligent would be freely collaborating open source developers.

This also is the reason I hardly ever wrote to that list: I have no quality time available for e-mailing about pretentious would be honest bullshit, si vous excusez mon Francais, sans cedille.

(**) It may be the case though - and I am again speaking as a psychologist - that the vast majority of those capable of writing source are simply not really capable of writing clear explanations of their own or others' code, rather like the majority of poytechnically educated people (engineers of various kinds) tend to be quite unable to explain the mathematics they can work with. (Ask an engineer to explain the Schröder-Bernstein theorem, or infinitary reasoning used in the integral calculus, and you'll be surprised! Or at least: I was, and quite a few times, though indeed the vast majority of humans learn to speak without ever knowing grammar consciously, and engineers usually are not really stupid.)

Anyway.... in a comparable if not equivalent way, only the best mathematical people, such as Poincaré or Russell or Feynman, really could write.

Likewise, very few are truly gifted both in languages and in mathematics, even if also on a much lower level than Poincaré and Russell.

(***) Thousands of people must have wasted thousands of hours of their own unpaid quality time trying to puzzle out the usually undocumented or else atrociously "documented" massively intertwined source code in a Squeak image.

Personally, I do not think it is moral to promise people (and small children!) that they can learn to program in a system most professional programmers gave up if they tried it at all as an obscure mess of pretentious code that's almost wholly undocumented.

To me, that simply is either lying directly or else speaking the thing that is not, rather like honest if cognitively challenged Salvation Army Soldiers also honestly do.

But it is indeed true that unfortunately you and I live in postmodern times, where even the people who received a university education were defrauded of science, truth and morality, and - outside real quality unuversities like MIT or Harvard - nearly everybody was fed, believed and probably still believes that, ultimately

"There is one thing a professor can be absolutely certain of: almost every student entering the university believes, or says he believes that truth is relative."

And for me Smalltalk and Squeak have been a waste of the quality time of my own life, because it took me a year to understand a system that I could have understood and learned in a few weeks at most if it had been properly documented - after which I learned that most developers of Squeak in practice like and maintain this situation:

That naive people get lured to Squeak because of its promises and pretensions, because its developers feel themselves to be too good to even try to document the system for those who, unlike them, were not so happy to be taught Smalltalk in their courses at university, and who don't earn money with it in their own little firms.

Maarten Maartensz


        home - index - top - mail