Is Smalltalk a weird cause to fight for?
medium.com
medium.com
Sure, there might be explanations elsewhere, just like there are evangelists for Lisp ... or Perl 6. And with similar arguments.
Take http://flownet.com/gat/jpl-lisp.html from 2002 as an example:
> I did try to introduce Lisp to Google. Having had some experience selling Lisp at JPL I got all my ducks in a row, had a cool demo going, showed it to all the other members of the ads team, and had them all convinced that this was a good idea. The only thing left was to get approval from the VP of engineering. The conversation went something like this:
Me: I'd like to talk to you about something...
Him: Let me guess - you want to use Smalltalk.
Me: Er, no...
Him: Lisp?
Me: Right.
Him: No way.
I ended up putting this essay into the pile of other advocacy-for-language-X pieces I've read, with no interest in reading other blog postings.
But I did anyway. https://medium.com/@richardeng/how-does-python-make-thinking... ("Pythonistas will also tell you that Python’s syntax reads like English"). 1) It's a strawman argument, as I'm a Pythonista and I don't say that, and 2) AppleScript is more English-like than Smalltalk, so if "English-like" were really a goal, that's more like what we should be using.
https://levelup.gitconnected.com/python-vs-pharo-df47599a8ed... ("Python vs Pharo") asserts that Pharo is easier to learn than Python. However, it provides no empirical evidence that the chosen metrics are appropriate. The ABC language, which was a strong influence on Python, show how empirical tests can be done.
As an example from Python, a high school teacher presented at a Python conference almost 20 years ago. The AP test at that time used C++ or Java - I can't remember which. He found it was easier to teach Python first, then C++ or Java, than to do everything in C++ or Java. While this is anecdotal evidence, it's still more substantial than an argument based on being able to fit the syntax on a post card. (If that were important, use Lisp, Forth, or APL, all of which have simpler syntax.)
And so on.
Please note that Perl 6 has been renamed to Raku (https://raku.org using the #rakulang tag on social media).
</evangelism>
As a point of pedantry, should Perl 6 (the programming language of 4 months ago) be considered a Raku dialect?
Or vice versa?
The C-like languages, plus Pascal, Ada, and others, are part of the Algol superfamily of languages.
"Family" and "superfamily" aren't well defined, in the way that it is for biological taxonomy, but I think it's intuitively clear enough for this discussion.
The Python family includes the various CPython language implementations, plus JPython/Jython, Iron Python, and PyPy, as well as wider-away dialects like Cython. At some point it's part of the Python superfamily.
What's the name for the family of languages which includes Perl 6 and Raku?
I think "Perl-like" includes Perl 4, 5, and 6, Raku, as well as (further away), Ruby.
I think Perl 6 and Raku are parts of the Perl 6 family, following the usual tradition of naming language families after the first instance in that family. (It isn't always that, see https://wiki.c2.com/?WirthLanguages ).
What would you suggest as the family name for Raku and Perl 6 which excludes the more distant members of the family, like Perl 5?
Delphi is an Object Pascal, for example.
Lisp has seen direct lineage from the 60s until today with lots of backwards compatibility.
"Lisp" refers to a family of languages, not a specific implementation.
Similarly, the term "Lisp evangelist" probably does not mean someone who only promotes Scheme, or only Clojure, or some other specific Lisp dialect.
Among other things, they don't have syntax made of mutable cons cells, terminated by the symbol nil, which is also self-evaluating and serves as (the only) Boolean false.
There are basically almost no languages with single implementations. C has many. Python has many. Ruby has several, Scheme has dozens, Logo has probably a hundred, Clojure has several, ...
We have a clear view of what C means: the language even has standards. C then has a lot of derived languages: Objective C, C++, and a bunch of others. Still I would think that there are many implementations/languages which are very near to C, but don't implement the full ISO C standard or can differ from it in various ways (-> Dialects).
Similar for Lisp. There are basically two standards for Lisp: ANSI Common Lisp (with >30 implementations - full or subsets) and ISO Lisp (called ISLisp). There are also a few (often) historical variants of Lisp: Emacs Lisp, AutoLisp / Visual Lisp, Standard Lisp, Interlisp, LeLisp, ZetaLisp, ... These languages have a core, which is more or less compatible to / derived from Lisp I.
See for example the Lisp dialects mentioned in 'History of Lisp': http://www.softwarepreservation.org/projects/LISP/ These are the historic core of Lisp.
There is a tendency to speak of a wider 'Family of Lisp' - but if you look closely, then languages like JavaScript/ECMAScript tend to be a part of that 'Family' - which makes it more or less meaningless. I tend to avoid that, because it does not set the expectations for practical usage right: JavaScript is practically not compatible, though it shares some features with Lisp: functions, anonymous functions, symbols, garbage collection, data syntax, evaluation, interactive use, ...
Similar: Scheme is now its own language family with its own standards: RNRS Scheme, IEEE Scheme, ... and many more or less conforming implementations.
> As a point of pedantry, should Perl 6 (the programming language of 4 months ago) be considered a Raku dialect? Or vice versa?
My point was that "Lisp", the original implementation line, died in the 1960s. We have a bunch of implementations now in the greater Lisp family. And we call them "Lisp"s.
So, what is the greater family name which includes Perl 6 and Raku? Is it "a Perl 6"?
The "Family of Lisp" definition in the broad sense you refer to has also been used to include Python, so I know what you mean. However, I think that's excessively broad.
In https://news.ycombinator.com/item?id=21873228 I wrote '"Family" and "superfamily" aren't well defined'. I cannot offer a better definition. However, given my description that the "Perl" family may extend to include Ruby, I think my further request for a "family name for Raku and Perl 6 which excludes the more distant members of the family, like Perl 5" is understandable.
You said 'language'. It did not die at all. The original Lisp and the code written in it was moved through the decades. There is even Lisp code from the 60s which was moved through all the implementations / Lisp evolutions over time:
Lisp I -> a bunch of Lisp 1.5 implementations -> Lisp 1.6 / PDP-6 Lisp -> Maclisp -> Zetalisp, ...
Example: Macsyma was based on code coming from Lisp 1.5 and moved through several generations of Lisp.
I thought it was clear that I was referring to the specific implementation which died with Lisp 1.5.
Because "Raku" refers to a specific implementation name in the Raku-like languages, of which Perl 6 is one. And the Raku evangelist seems to argue that it was incorrect to describe a Raku evangelist as "a Perl 6" evangelist.
I was drawing a parallel. If a modern Lisp system can be referred to as a Lisp, then why can't Raku be referred to as a Perl 6?
Now, I'm perfectly fine with saying this is a sociological question, and that the members of the group get to decide their name. I also think there are limits to that framework. For one, it leads to gatekeeping. If a Scheme or Clojure users says they are using a language in the Lisp family .... I think you see the issue.
The "implementation democracy" model has benefits, but the cost seems to be high, hurting availability of libraries. The network effect is dampened.
Usually the compatibility is not great and thus the lead implementation is mostly used.
> Java, Haskell
Both have standards. Java has extensive standardization.
On the other hand, it's ok to post your own stuff as part of a diverse mix of stories and comments that are about nothing in particular but just gratify intellectual curiosity. Then you'll be participating in HN as a community member and the community will see your account that way. It's particularly good if you submit stories on weird or out-of-the-way topics that rarely or never get attention. Those are the best!
Sorry if I seem to be just doing PR. I'm doing a lot more than that. I'm one of the core developers of Raku (née Perl 6). And I publish the Rakudo Weekly blog (https://rakudoweekly.blog )
FWIW, I do think I'm adding material to HN that is interesting to at least some visitors of HN, especially the ones interested in programming, with an emphasis on Raku. The reason I'm triggered by "Perl 6" is that I think it is important that visitors of HN know that it is no longer called that. Once everybody is using Raku instead of Perl 6, there won't be any of these comments.
Whenever there's a new issue of the Rakudo Weekly, I post this on HN. Are you saying that that is not of interest of HN visitors?
I'd therefore like to know what point of the guidelines I'm exactly in violation of.
> Don't solicit upvotes, comments, or submissions. Users should do those things when they run across something they personally find interesting, not because someone has content to promote.
But there was no submission solicitation here. If that's what you're referring to, perhaps this could be reworded to clarify?