Re: [shlomif-beta-reviewers] Re: Message 3 (Re: About Park - points to consider when designing a new programming language)

Hi Omer!

On Saturday 24 June 2006 02:10, Omer Zak wrote:
> On Sat, 2006-06-24 at 01:01 +0300, Shlomi Fish wrote:
> > On Friday 23 June 2006 23:08, Omer Zak wrote:
> > > On Fri, 2006-06-23 at 21:06 +0300, Shlomi Fish wrote:
> > > > (Search for "LISP") he writes about a dozen one-liners a day. So one
> > > > thing I'd like to make Park suitable for is writing short scripts and
> > > > programs.
> > >
> > > To do quick one-liners, I think that a properly designed IDE would be
> > > better than a new language.  I have in mind an IDE which integrates
> > > with a powerful search engine:
> > > You type in concepts as keywords.
> > > The search engine searches through library documentation and cookbooks.
> > > You get, as results, links to relevant libraries (from all programming
> > > languages) and relevant coding cookbook entries.
>
> [... discussion how Shlomif works on his code was snipped ...]
> The issue is that each one has his own style of working.  The optimal
> environment for shlomif is not optimal for me and vice versa.

Right. Someone mentioned here that he was using vim for editing Java code 
after he found that Eclipse was not his cup of tea. I used gvim to edit 
Scheme code for my SICP exercises and other Scheme experimentation.

>
> > I believe most one liners assume the programmer has a ready, instant,
> > knowledge of most of the core functions and concepts he needs to write
> > stuff.
>
> A properly-designed IDE with a properly-designed search engine and
> properly-catalogued Internet-based database of documentation and code
> examples - would extend the effective "ready, instant knowledge" used by
> the programmer.

Possibly. However, many (most ?) programs will just fire-up vim, joe, edit.exe 
or their different idea of a lightweight and convenient command line editor 
to write the script (or edit it directly in the shell prompt), if a one-liner 
is all they want to write.

Only when editing, longer and usually large-scale program does an IDE 
justifies itself.

When I said "ready, instant knowledge" I meant knowledge that you do not have 
to lookup. For example, when programming Perl, I already know the syntax for 
regular expressions, most of the useful functions out of perlfunc, the 
operators that I need, etc. Some people know it for Python or Ruby or PHP or 
a different language. Part of the reason I'm still using Perl for my scripts 
is that looking everything up for a different language would be more time 
consuming, so even if the resultant script would be shorter, it would take me 
longer to write.

>
> > My intention in Park is to have an actually useful Lisp dialect that
> > people could use in place of Perl, Python, etc.
>
> Again, I believe more in designing the whole environment rather than in
> polishing a single component (the programming language).  The Smalltalk
> developers did develop a whole environment.  The FORTH inventor (Chuck
> Moore) developed a whole environment together with FORTH (alebit very
> unconventional one, as he has very unconventional mind).

I've recently started learning Smalltalk. I first tried Squeak Smalltalk which 
is a Smalltalk environment written in C. (by the original Smalltalk 
designers) It looks very nice and all, but its conventions are unconventional 
and therefore annoying. I did not like its own internal editor (or could get 
the hang of it), and tried to find a way to write code in gvim. The people at 
#squeak even told me there was no way to write everything using plaintext. 
(No to mention the code exported by Squeak to a Unix system was \r instead of 
\n terminated.)

Meanwhile, I'm using GNU Smalltalk instead which allows me to edit everything 
as text and uses the native filesystem. Obviously Squeak is probably an 
extreme case of making an environment. Other environments like Eclipse or 
Emacs do play nicely with the other kids on the block.

Guy once posted a message about enhancing the "Unix IDE":

http://groups.yahoo.com/group/hackers-il/message/3681

>
> > I believe PHP would be suitable for writing > n*10M LOCs codebases. But
> > you'll probably need to have very good and discplined programmers for
> > achieving this. (But this goes without saying for any other language
> > you'd like to maintain such codebases in).
>
> For a O(10^7) software development project, you need disciplined
> programmers anyway.
>
> I am now thinking that I have been barking on the wrong tree.
>
> If we want to improve the productivity of software developers, then
> investment of neuron firings will be more profitable, if it is made at
> design and development of a multilingual environment (which I refer to
> as IDE elsewhere), rather than at design of a single new language.
>
> However this is a different problem, which is probably of no interest to
> people, who make an hobby of designing new programming languages.

Right.

>
> > > > One suggestion for handling heisenbugs that has been raised is a
> > > > reversible debugger. See for example:
> > > >
> > > > http://use.perl.org/article.pl?sid=06/06/22/1210201
> > > >
> > > > Do you have any other suggestions?
> > >
> > > Not really.  But can try to use theory
> > > (http://lambda-the-ultimate.org/) to specify a language, in which it is
> > > possible to prove that certain types of heisenbugs cannot occur.  For
> > > example, require explicit declaration of variables, which are shared
> > > among threads, so that the compiler (or interpreter) can check, at
> > > compile time, that accesses to those variables are properly locked.
> >
> > Do you include static run-time checking there? (Like the Stanford
> > checker, etc.)
> >
> > I'd rather not go too much into the direction of a language in which bugs
> > can be proven not to exist for fear of the following quote:
> >
> > <<<
> > "UNIX was not designed to stop its users from doing stupid things, as
> > that would also stop them from doing clever things." – Doug Gwyn
> >
> >
> > So I'm not sure whether such a feature would prevent doing smart things
> > like doing self-writing code, or even generating functions on the fly, or
> > dynamically changing the class inheritance graph.
>
> It is possible to combine both concerns, by language design which
> enforces type checking (including legality of unlocked accesses) by
> default, but allows the software developer to explicitly override it.
>
> An example:  the coercion operator in C/C++.  It allows you to shoot an
> atom bomb at your foot, but you have to explicitly declare that this is
> what you are doing.

I see. Well the best type system I saw was in Haskell and O'Caml which are 
truly strongly typed (but see [1]), but also functional. It is possible a 
similar type system exists in such languages as Module 3 and Ada 95, but I 
don't know them yet.

Only problem is that such a type system does not correspond very well with the 
spirit of Lisp until now, nor with my plans for Park. I personally find such 
types too limiting to think of in day-to-day-programming. Plus, while they 
may help write more bug-free code, they are not a necessity for large scale 
projects. (See my quote about doing smart things and stupid things in the 
previous message).

Python has a type system but it is still dynamically typed where each 
container can contain hetrogeneous values. Perl 6 aims to also have a 
stricter type system than Perl 5, but from what I understood you can do 
Haskell-like type signatures like $foo is a hash whose keys are string and 
his values are hash refs whose "a" element is a string, and "omer" element is 
a float. But I may have mis-concepted the Perl 6 specs.

I don't see such type systems belonging in Park. Park would have an object 
system with a nice Object hierarchy (based on that of another language, but 
possibly somewhat different.) and everything would be an object there. Now, 
containers will generally manage references to the base class (in a similar 
way to Smalltalk or Java) as values. Since Park would be a symbolic language, 
then like Perl 5, you won't need any casting to call a method of any of the 
derived classes the object belongs to.

> > > Can you design Park to support this, even when macros are used?
> > > It may be a good idea to design an IDE, which would allow a developer
> > > to have a view of code, which could be generated by a LISP macro, as
> > > the macro is being developed.
> > >
> > > I think that it is more of IDE issue than language issue, however.
> >
> > I also think Alan Turing has something to say about it too:
> >
> > http://en.wikipedia.org/wiki/Halting_problem
> >
> > If we generalise the Halting problem than we can say a program to
> > conclusively analyse a different computer program would be impossible to
> > write. I don't see how I can get an IDE to analyse a macro and determine
> > how it does what it's supposed to do. The best way would probably be to
> > document the behaviour of the macros in the program a la doxygen or
> > javadoc. But this will require an extra work from the programmer.
>
> One possible (and relatively simple) way to implement it:
> You set the IDE to execute a macro, with arguments which you supply,
> each time you modify its definition.  Then display the resulting code in
> a separate window pane.  Then you can see immediately the effects of
> changes in the macro definition.

Hmmm... wouldn't it still require some work from the programmer? Can you point 
to an implementation of this for a different Lisp dialect?

Regards,

	Shlomi Fish

[1] - http://perl.plover.com/yak/12views/samples/notes.html - search 
for "Strongly Typed".

-- 

---------------------------------------------------------------------
Shlomi Fish      shlomif-ik1l9ssToec+JF/[email protected]
Homepage:        http://www.shlomifish.org/

95% of the programmers consider 95% of the code they did not write, in the
bottom 5%.


------------------------ Yahoo! Groups Sponsor --------------------~--> 
Something is new at Yahoo! Groups.  Check out the enhanced email design.
http://us.click.yahoo.com/SISQkA/gOaOAA/yQLSAA/saFolB/TM
--------------------------------------------------------------------~-> 

 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/hackers-il/

<*> To unsubscribe from this group, send an email to:
    [email protected]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/