Squeak-dev July 1999

squeak-dev@lists.squeakfoundation.org

139 participants
367 discussions

Re: Moving objects... (serialization)
by Marcel Weiher 27 Jul '99

27 Jul '99

Ahh, I was looking for an existing standard, but couldn't find it. I'll have a look. Thanks, Marcel > From: "Eric Ulevik" <eau(a)fast.fujitsu.com.au> > > XMI is a proposal for schema interchange, but I don't really know the > details.

1 0

Namespaces: Pink and Blue
by Augustin Mrazik 27 Jul '99

27 Jul '99

1 0

[Newbie] GraphPane equivalent
by ChrisRath＠aol.com 27 Jul '99

27 Jul '99

I was wanting to futz about with the Pen class. I can instantiate an instance of Pen and draw in the Workspace window. Was wondering how to open a pane just for drawing. The Smalltalk/V code would be: pen := (GraphPane openWindow: 'Simple' extent: 220@220) pen. Is there a simple equivalent to this in Squeak? Thanks.

2 1

Re: Moving objects... (serialization)
by Peter Smet 27 Jul '99

27 Jul '99

Marcel, This is very interesting. So each language would have some implementation of #instantiateFromXMLStream? Is that how it would work? Could you explain a bit about how XML would deal with issues like: Moving behaviour. A class is not an Object in C++ or Java. How does XML serialize classes in C++ and Java? Could these then be understood in a language like Smalltalk? What about class methods, class variables, static variables etc. What happens if my Java XMLreader comes across a Smalltalk class object, can it cope with this? What about my serialized block object, anonymous inner classes etc? These are all language impedance mismatches, where you have a concept in one language, which does not exist in another. Big Endian vs little endian - marshalling stuff. Versioning. A class is changed, and then an old version of the class is read in from an XML stream - the old instance variables don't match the instance variables of the new class definition.. I really like the idea of being able to view and manipulate serialized Objects in a standard way, without them having to be instantiated into their native form first. Peter From: Marcel Weiher <marcel(a)system.de> >I've been looking at cross-platform serialization for a while now >and seriously believe that XML is the way to go here. XML coding is >simple, standardized, self-describing, human readable, etc. Standard >XML mechanisms can be used to describe object-references, even to >objects not within the same file and can even express the difference >between fetching the target of such a reference immediately or via >proxy if referenced (SqueakPages?). > >One of the implications of XML as a serialization format is that you >do not need to have the actual objects in order to understand, >display, modify a serialized representation. As a matter of fact, >purely syntactical tools such as XSL or even grep and vi can be used >to search or transform such formats. The serialization format can >also be close enough to standard XML-based formats that >conversion/encoding is a trivial extension to serialization.

2 1

re: Re: [ENH] Help for Adding Name Space to Squeak
by Alan Lovejoy 27 Jul '99

27 Jul '99

And Namespaces should also each have their own independent Symbol Table, so that objects from different Namespaces can interact without possibility of conflict (Symbols from different Namespaces could then have the same characters, but still not be ==. And it works for message selectors, too!). I agree with the rest of your excellent post (it essentially mirrors what I've thought for years). --Alan > ** Original Sender: Hans-Martin Mosner <hm.mosner(a)cityweb.de> > For Collage, I've pondered the question of Namespaces deeply and came to the > following conclusion: > Classes should have a 'namespace interface' like in VW5i. It just makes a lot > of sense. > Namespaces should subsume Smalltalk, Undeclared and all the shared pools. > Namespaces should be hierarchical, with Smalltalk as the root. > Namespaces should be able to import other Namespaces completely (just as > Classes import shared pools now). > Namespaces should be prominently visible in the Browser, being categorized just > like Classes. > > Questions are: > Should Namespaces implicitly include all variables of their enclosing > Namespaces? > What about variable initializers? I don't really like them; would rather like > to have declarations only. > What about selective variable imports? Perhaps with renaming? > > Hans-Martin > >** --------- End Original Message ----------- ** >

4 3

RE: [ENH] Help for Adding Name Space to Squeak
by Alan Kay 27 Jul '99

27 Jul '99

Stephen -- A good point at the end. The mythical Smalltalk-72 actually could easily make the distinctions you mentioned (although the receiver always had the option to decide whether the sender's interpretation should be evaluated -- e.g. the husband could decide whether to evoke what the wife was thinking when the message was actually sent -- this was similar to but more general than call by name in that the receiver could decide all about how the message would be interpreted). Cheers, Alan ------- At 12:12 PM -0800 7/26/99, Stephen Pair wrote: >Hans-Martin Mosner wrote: >> >> Alan Lovejoy wrote: >> >> > And Namespaces should also each have their own independent Symbol Table, >> > so that objects from different Namespaces can interact without >> possibility of >> > conflict (Symbols from different Namespaces could then have the same >> > characters, but still not be ==. And it works for message >> selectors, too!). >> >> What are the consequences in terms of readability? >> I always thought that in Smalltalk, the meaning of statically >> bound identifiers >> (classes, inst vars, static vars) should be determined in the >> static scope of the >> method in which they appear, whereas the meaning of dynamically >> bound identifiers >> (message sends) should be determined by the receiver object. > >I agree about the readability issue, but message symbols represent an >implied lookup of a "Message" object that is then sent to the reciever >(again, in an implied sense, for performance reasons this stuff is optimized >out). Given that, does it make sense for the receiver to do the look up of >this symbol? The receiver should certainly decide what to do in response to >the object it receives. > >Another point: Should the generice message sending scheme be so tied to >symbols (or actually instances of a Message class)? Or alternatively, >should we be able to send any object as a message to another object? For >instance, we could write: > >'Take out the garbage' sentTo: Husband > >The receiver could then decide what to do (or do nothing). In this case, a >natural language parser might be used to figure out what the intended action >is and depending on the internal state of the object, decide to actually >take out the garbage or not. > >To accomplish this, we would need to make the message sending scheme >slightly more generic, allowing any object to be sent as a message to >another object. > >Getting back to the original thought, if we wrote the above message like: > > Husband takeOutGarbage > >Shouldn't it be the Wife that determines how the #takeOutGarbage symbol is >resolved? Either to a traditional Message, or to the string 'Take out the >garbage.' For example, if the Wife sends this message several times within >a short period of time, she may alter the resolution of #takeOutGarbage such >that it resolves to 'Take out the garbage!' or 'Get off your lazy *^&#@ and >take out the gabage!!!' Variations on the same message, but ones that she >may have a better chance of getting a response. :) > >- Stephen

1 0

3 button Mice
by Harry E. Fassl 26 Jul '99

26 Jul '99

A while back there was a post about 3 button mice. I'm using a Belkin (www.belkin.com) 3 button mouse with Squeak on a Mac. To use all three buttons I'm using the USB Overdrive ($20 shareware...www.usboverdrive.com) which works great. Belkin has a driver but for Windows only. H.E.Fassl

1 0

Re: Moving objects... (serialization)
by Marcel Weiher 26 Jul '99

26 Jul '99

Hi, the answer is XML. What was your question again? :-) I've been looking at cross-platform serialization for a while now and seriously believe that XML is the way to go here. XML coding is simple, standardized, self-describing, human readable, etc. Standard XML mechanisms can be used to describe object-references, even to objects not within the same file and can even express the difference between fetching the target of such a reference immediately or via proxy if referenced (SqueakPages?). One of the implications of XML as a serialization format is that you do not need to have the actual objects in order to understand, display, modify a serialized representation. As a matter of fact, purely syntactical tools such as XSL or even grep and vi can be used to search or transform such formats. The serialization format can also be close enough to standard XML-based formats that conversion/encoding is a trivial extension to serialization. Of course, I also see a couple of drawbacks, one is that the format is pretty verbose, and the other is that it can't handle raw binary data without quoting. WAP, a compact binary format for XML might actually address both of these issues, and for plain XML, you can either go with external entities (move the binary data into a separate file) or bite the bullet and quote the data. I've also been thinking about an object/value representation that works without byte-encoding everything (useful for inter-language calling within the same address space). > From: "Peter Smet" <peter.smet(a)flinders.edu.au> > The achilles heel of all this stuff seems to lie with the serialization. For > example, I can send a serialized object or class across the net with BOSS, > but it will collapse in Squeak, Dolphin, and even more so in Java. There > are JVM's spread all across the net, but we cannot talk to the objects in > them, since the serialized objects are not compatible with Smalltalk. Does > ANSI say anything about serialization across Smalltalks? Could serialization > be improved by MetaObject Protocol extensions? I am thinking that all > serialized objects could contain some serialized meta-information that could > be instantiated into a MetaObject (across all languages). The MetaObject > would then aid in reconstructing the state and behaviour of the 'dead' > serialized object (sorry, just rambling). Here's some sample output from an XML-serializer I just wrote (in Objective-C) last weekend. I manually cut out some data, so it probably doesn't nest properly any longer. <object id=1d9f30 class=MPWAdTemplate> <epsTemplate> <!-- instance variable named 'epsTemplate' --!> <object id=31e478 class=MPWEPSFile> <!-- first reference is coded inline --!> <dsc> <!-- instance var of MPWEPSFile object in 'epsTemplate' var --!> <object id=31ada8 class=NSMutableArray> <count valuetype='i'>41</count> <arrayelement> <object id=31ae30 class=MPWDSCComment> <name>%!PS-Adobe-</name> <value>2.0 EPSF-1.2</value> </object> </arrayelement> <arrayelement> <object id=31aef0 class=MPWDSCComment> <name>%%Creator:</name> <value>QuarkXPress(R) 4.03</value> </object> </arrayelement> <!-- array elements omitted --!> <arrayelement> <object id=31d090 class=MPWSubData> <myData> <object id=31cfa0 class=MPWSubData> <myData> <object id=31b2c8 class=NSData> <data length=392143/></object> <!-- actual data omitted for now --!> </myData> <offset valuetype='i'>0</offset> <myLength valuetype='I'>392143</myLength> </object> </myData> <offset valuetype='i'>168</offset> <myLength valuetype='I'>33</myLength> </object> </arrayelement> <arrayelement> <object id=31a2f8 class=MPWSubData> <myData idref=31cfa0/> <!-- reference to previous MPWSubData --!> <offset valuetype='i'>364</offset> <myLength valuetype='I'>58</myLength> </object> </arrayelement> <arrayelement> <object id=31a388 class=MPWSubData> <myData idref=31cfa0/> <offset valuetype='i'>162124</offset> <myLength valuetype='I'>1</myLength> </object> </arrayelement> <arrayelement> <object id=31dd88 class=MPWDSCEmbeddedFile> <name>%%BeginDocument:</name> <value>kurt:Desktop Folder:-> W 10:AudiBW_S.eps</value> <unnamed_1> <object id=31abf0 class=MPWEPSFile> <dsc> <object id=31d968 class=NSMutableArray> <count valuetype='i'>30</count> <arrayelement> <object id=31ddf8 class=MPWDSCComment> <name>%!PS-Adobe-</name> <value>2.0 EPSF-1.2</value> </object> </arrayelement> <!-- lots of stuff omitted --!> </object> </arrayelement> <arrayelement> <object id=31e758 class=MPWDSCComment> <name>%%EndDocument</name> <value></value> </object> </arrayelement> <arrayelement> <object id=31d720 class=MPWSubData> <myData idref=31cfa0/> <offset valuetype='i'>296029</offset> <myLength valuetype='I'>183</myLength> </object> </arrayelement> <!-- again, lots of stuff omitted --!> </dsc> <subObjects> <object id=31cb30 class=NSMutableArray> <count valuetype='i'>4</count> <arrayelement idref=31abf0/> <arrayelement idref=323388/> <arrayelement idref=3236d0/> <arrayelement idref=323a18/> </object> </subObjects> </object> </epsTemplate> <slots> <object id=323de0 class=NSMutableArray> <count valuetype='i'>4</count> <arrayelement> <object id=323e98 class=MPWRectSlot> <original idref=31abf0/> <rect valuetype='{?={?=ff}{?=ff}}'>{{1764, 1984}, {284, 176}}</rect> <name>AudiBW_S</name> <active valuetype='c'>1</active> </object> </arrayelement> <!-- similar array elements omitted --!> </object> </slots> <replacements> <object id=306a8 class=NSMutableArray> <count valuetype='i'>2</count> <arrayelement> <!-- array elements omitted --!> </arrayelement> </object> </replacements> </object>

1 0

hi, i'm working on a project
by larry a price 26 Jul '99

26 Jul '99

I'm new to squeak, but not to object oriented programming, i'd like to be able to do a couple of things with smalltalk namely, get down to the hardware and a. talk to the modem -general case any modem b. tt modem -particular case this modem c. build, in cooperation with others a suite of parsers and display objects that let me view the web the way I want. - thanks all for the great work so far So, any advice, experience, suggestions on strategies for device drivers etc. would be much appreciated as would advice on how to make things go faster, (i'd like to dispense with the OS...and replace it with an Environment myself). thanx Larry A. Price laprice(a)efn.org (541)-988-5869 "only a system in a state far from equilibrium can move to higher orders of complexity in its organization."-Ilya Prigogine

1 0

Re: Moving objects...
by Peter Smet 26 Jul '99

26 Jul '99

>Peter Smet wrote: > >> I have been reading a book on mobile objects... The book is called 'Programming Mobile Objects with Java' by Jeff Nelson, Wiley, ISBN 0-471-25406-1. I like it because it is not padded out with too much speculation or raw code (like Orfali's CORBA and Java, which spends far too much time implementing simple benchmarking programs). I have only read about half of it, but it has good descriptions of RMI, DCOM, CORBA etc. There are probably lots of applications that could benefit from mobile code, and I would like the book to talk about such applications and why these are best implemented with mobile code much more (Things like testing across remote images... :-). Certainly, having a mobile object bring it's own UI across the net should at the very least simplify thin clients compared to installing a UI interface onto 100,000 client machines.... Mobile objects seem to me the next logical step to moving the state of objects. (The funny thing is that computer viruses have been doing a very similar thing all along). My impression from the book is that having a VM as a place to receive objects and 'breathe life' into them is a HUGE advantage, and languages without a VM are going to have a much harder time adapting to code moving across the net. The achilles heel of all this stuff seems to lie with the serialization. For example, I can send a serialized object or class across the net with BOSS, but it will collapse in Squeak, Dolphin, and even more so in Java. There are JVM's spread all across the net, but we cannot talk to the objects in them, since the serialized objects are not compatible with Smalltalk. Does ANSI say anything about serialization across Smalltalks? Could serialization be improved by MetaObject Protocol extensions? I am thinking that all serialized objects could contain some serialized meta-information that could be instantiated into a MetaObject (across all languages). The MetaObject would then aid in reconstructing the state and behaviour of the 'dead' serialized object (sorry, just rambling). >Peter, > >Thanks for this post. I have been Smalltalking for a number of years now, I've >used BOSS a bit, and I've done a decent amount of multi processing in ST, but >the notion of mixing the two had never occured to me until today. Since, I have >VW here at work up and running, I ran an experiment there, which was precisely >as you asked. Travis, thanks for checking this out! (Had to make a start on my tax return this weekend, so no coding). While searching the web I noticed that IBM are working on 'aglets' (agents/applets) which is the most disgusting name. They might as well call it 'Roaches', since cockroaches have invaded more niches than almost any creature... Anyway, they are setting some guidelines and protocols, and doing it all in Java by the looks of it. All of it would work just as well in Smalltalk. >How totally cool. I'll give this a whirl with Squeak tonite (if somebody else >hasn't already by then). I imagine it ought to work. The ramifications of what >one could do with this are pretty cool. I could use sockets to stream active >processes out across the net to other machines. To those who do agents and stuff >like that, maybe this is old hat, but it was an "aha" for me... just had to >share. I think Alan Kay or someone else once mentioned the term 'network ecology', which I really like. Like you, I think the ramifications of this stuff are incredible. Objects will become more and more autonomous, interacting with other remote objects on the net in ways their authors may never have imagined... </speculative ramble> Peter

2 1

← Newer
1
...
6
7
8
9
10
11
12
...
37
Older →

Jump to page: