Self-interest September 1992

self-interest@lists.selflanguage.org

18 participants
20 discussions

Inheritance of instance slots, KEVO
by R James Noble 25 Sep '92

25 Sep '92

What I currently do is the "_Define _AddSlots:" cliche Dave Ungar posted. Out of interest, I checked back through my archive of the mailing list - almost _two years ago_, Bill Burdick and Bay-Wei Chang, I think, seemed to work this one out. Urs Hoelzle writes: > 3. Through the programming environment. Something I've seen recently (came over the net) is Kevo, yet another prototype prototype-based OOPL. (this one is a mutant FORTH). Although prototype based, it claims not to provide inheritance; rather, everything is managed by clever copying. So where we inherit, Kevo just copies and does _AddSlots. The catch is that Kevo has another version of _AddSlots with adds the slots to each member of a group of objects. Where we would modify a parent, Kevo modifies each object in the group. Kevo has three levels of "propagation": an individual object; a object and all unmodified copies (clones); an object and all modified copies of the object (descendants). This approach could (in fact) be implemented in Self as it stands, with judicious use of the Mirror references methods. And (I think) were some of the information in the compiler's object maps brought out to the programmer (browse all objects cloned from this object) this would be even easier. I think this would be more within the style of Self than adding complicated constraints to the programming environment - but that's just my opinion. Adding general constraints to the language - say like Brad Myer's KR object system in Lisp - now there's an idea! WRT Kevo, I think the author (Antero Taivalsaari) will be at OOPSLA for those of you lucky enough to be going :-) I don't particularly like Kevo (although I'm happy to believe it may be the best OO FORTH - CUP TEA OF MY FORTH IS NOT). And, as far as I can see, the Kevo scheme (using copying rather than inheritance, with mechanisms for managing copying) is no more powerful than inheritance and has one big drawback: the inheritance hierarchy is no longer manifest in the program's basic structure. kjx (Self bigot flame mode on) Hmm - I guess Self (at over two years in public release) must be considered a "mature" language now. I'm sure that's older than the latest version of C++ for instance! (Self bigot flame mode off) kjx

2 1

Re: Inheritance of instance slots, KEVO
by Michael Grant Wilson 25 Sep '92

25 Sep '92

kjx sais: > (Self bigot flame mode on) > Hmm - I guess Self (at over two years in public release) must be > considered a "mature" language now. I'm sure that's older than the > latest version of C++ for instance! > (Self bigot flame mode off) Since Self has no commercial implementations, no (useful) programming environment, and a microscopic user community, its unlikely to be called "mature" by anyone normal person's standards. Of course, this does not make it any less interesting. .mike

1 0

Re: "is a" relationships
by David.Ungar＠Eng.Sun.COM 24 Sep '92

24 Sep '92

The other way to get inheritance of structure in Self is to _Define your prototypes as copies of the ones you want to extend: ... addSlotsIfAbsent ... traits animal _Define: ( | p* = traits clonable. animal behavior | ) animal _Define: ( | p* = traits animal. animal data slots | ) traits mammal _Define: ( | p* = traits animal. mammal behavior | ) mammal _Define: animal copy _AddSlots: ( | p* = traits mammal. mammal data slots | ) Note the "animal copy _AddSlots:" this does the copy down when you read in the file. Dave

1 0

Re: "is a" relationships
by mcr＠latour.sandelman.ocunix.on.ca 24 Sep '92

24 Sep '92

On Sep 23, 7:43pm, Brad Might wrote: } This is what I have tried, is it the way to go ? } } } prototypes _AddSlots: (| } animal = (| } parent* = traits animal. } "object specific stuff here". } |). } mammal = (| } parent* = traits mammal. } inherit**. "at initialization (copy), this slot will be } filled in with a copy of prototypes animal" You are not the only one to ponder exactly how to do this. Particularily nasty is that changing the animal prototype doesn't affect the inherited slots of mammal at all. Of course, it doesn't affect any existing animals either. I was hacking on `copy-on-write' slots at one point, but this as was with 1.1, and one couldn't use _AddSlots: from within a method... I think that something like that is the solution. -- :!mcr!: | The postmaster never | So much mail, Michael Richardson | resolves twice. | so little time. HOME: mcr(a)sandelman.ocunix.on.ca Bell: (613) 237-5629 SCHOOL: 192228(a)physics.carleton.ca

2 1

"is a" relationships
by jbm＠hal.com 24 Sep '92

24 Sep '92

Trying to determine how "is a" relationships could/should be specified in Self. eg. cat "is a" mammal "is a" animal. c++ does this by subclassing class animal {}; class mammal : public animal {} ; class cat : public mammal {} ; With self we could go several ways (code off the top of my head, please forgive syntax and/or structure): using a c++ like method, we duplicate the Base class in the child, then add our own thing. I don't like this method too much, as changes made to animal may not be reflected into the copy of animal within mammal. Also, mammal must avoid duplicating slot names, or must override them completely, ie. cannot ovverride and then call "super" prototypes _AddSlots: (| animal = ("some stuff") |) prototypes _AddSlots: (| mammal = animal copy |) prototypes mammal _AddSlots: (| "mammal specific stuff here"|) This is what I have tried, is it the way to go ? prototypes _AddSlots: (| animal = (| parent* = traits animal. "object specific stuff here". |). mammal = (| parent* = traits mammal. inherit**. "at initialization (copy), this slot will be filled in with a copy of prototypes animal" "mammal stuff" |). |) inherit is a parent of lesser importance so that methods can be overridden as desired. What bugs me about this one, is the copy required to fill in "inherit" at object initialization (copy) time. Is this how its done ? brad

1 0

Re: Argument blocks, value: and -infinity.
by David.Ungar＠Eng.Sun.COM 23 Sep '92

23 Sep '92

Glad you are using it! I think the following should work: blahMethod: withArg SpecialBlock: myBlock = ( .... myBlock value: 10 With: 20. ). I want to call it with: blahMethod: myArg SpecialBlock: [ | :firstArg. :secondArg | " do neat things with the arguments " ]. Note value:With: instead of value:Value: and also note the period after ":firstArg". I am doing this from memory, so I hope I have got right. good luck! Dave Ungar

1 0

Argument blocks, value: and -infinity.
by richards＠turing.scs.carleton.ca 23 Sep '92

23 Sep '92

I am having difficulty finding my copy of the Self manual, and Danny Epstein isn't to be found, so I can't borrow his. I thought I was beyond having to look up syntax :-) How do I declare a multi-argument block? I.e. given blahMethod: withArg SpecialBlock: myBlock = ( .... myBlock value: 10 Value: 20. ). I want to call it with: blahMethod: myArg SpecialBlock: [ | :firstArg :secondArg | " do neat things with the arguments " ]. Actually, I haven't written the body of the blahMethod: yet. I am getting syntax errors on the call... It is possible that a redesign would get rid of this problem, but I doubt it, since I want to have a generic function that does something intelligent to a property of a sub-object. Specifically, it expands or contracts one of the dimensions, the dimension being changed should be specified by the *enclosing* object, and *changes* for one case (the y dimension has an up from baseline and down from baseline component). Maybe I'm just too much of a LISP hacker :-) Also, I created an object that I call a `boundedNum' that I use for the dimensions. It is an integer that can grow (shrink) to a certain maximum (minimum) size, but no greater (smaller). Some boundedNum's have only one bound, or zero bounds --- i.e. one of the bounds is infinity. I was pleased to type `infinity' and have it work. I tried -infinity, but that got Self 1.3 upset. Okay, negInfinity: 0 - infinity, and it is happy. (it even prints as "-Inf") However, once I've create negInfinity, any subsequent code loaded fails. Does this happen with 2.0/2.1, or is this particular to my image... Thanks. p.s: I have done too much work under 1.3 to switch to 2.0 at this junction. Probably later. R James Noble's X stuff works quite well. I understand that the 2.0 X stuff uses bitmaps?

1 0

concurrency
by Paulo Henrique Cavalcante Lisboa 18 Sep '92

18 Sep '92

Could anyone answer me some simple questions: 1) Is there any concurrency facility in Self (I didnt find it in the manuals). 2) Is there any time limit for Self to become a available comercial language. 3) In my point of view, reflective facilities in Self make it a good language fo r OODB. Is there any work in this area (OODB) for self. 4) Is any known software Enterprise (Sun, Microsoft, IBM, etc.) trying to release a comercial version of self. Thank you in advance, Paulo Lisboa ==> phcl(a)di.ufpe.br

2 1

Self 2.0.1 available by anonymous ftp in UK
by Mario Wolczko 18 Sep '92

18 Sep '92

A copy of the Self 2.0.1 distribution is now available from mushroom.cs.man.ac.uk (130.88.13.70), in /pub/self/2.0.1. There are also copies of some of the Self Group's papers, the sources to the Richards benchmark, and an archived copy of this mailing list, in Emacs Rmail format. UK users will probably find it faster to ftp from this site. It's not clear (to me, at least) whether it will be faster for the rest of Europe -- the wire connecting the UK to the Continent is reputedly rather thin. Note that the distribution is broken out into the constituent files and directories. If you want the whole thing, cd to /pub/self and ask for 2.0.1.tar or 2.0.1.tar.Z -- the ftp daemon will generate the appropriate kind of file. Mario Wolczko ______ Dept. of Computer Science Internet: mario(a)cs.man.ac.uk /~ ~\ The University uucp: mcsun!uknet!man.cs!mario ( __ ) Manchester M13 9PL JANET: mario(a)uk.ac.man.cs `-': :`-' U.K. Tel: +44-61-275 6146 (FAX: 6236) ____; ;_____________the mushroom project___________________________________

1 0

Compiling the Self Source Code
by jrd＠cadre.com 17 Sep '92

17 Sep '92

i have a couple of questions about Self: 1. when i try to compile the VM source code, i get an error because the file "std.h" can't be found. this file also doesn't exist as part of the Gnu compiler. is this file missing or is my Gnu compiler installation incomplete? 2. will Self ever be compatible with the Sun C++ compiler? 3. when will there be a native X version of Self that doesn't use Pixrects? thanks! john r. durand Cadre Technologies Inc. O_o 222 Richmond Street =( )= Ack! Providence, Rhode Island U 0 2 9 0 3 email: jrd(a)cadre.com Phone: (401) 351-5950 FAX: (401) 351-7380

2 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

Self-interest September 1992