What Essays??
Locked
 
|
I am sorry to report that there has been very little activity on the essay front. No one seems willing to make the time to write an essay supporting the Smalltalk language. I find it a little bit sad that people won't fight for the language they love so much. It's as if all the fight has been squeezed out of them. Why? Perhaps they feel dejected by all the previous failed attempts to popularize Smalltalk. It's understandable. But I encourage you to think of Smalltalk Renaissance as a new day. This is a new type of opportunity. There is buzz in the cybersphere once again. People are visiting our websites. You can, and must, help. Take this opportunity to address their concerns about Smalltalk. And reiterate the benefits. Convince them. You may or may not be successful, but at least you will have given your best effort. Please understand: Visitors to our websites who find nothing compelling may never return again. We only get one shot at this. It doesn't require a great deal of time. Most of my posts for the SRP were written within an hour, at least, as a draft. I can polish your essay for you; hence, your contribution is not that demanding. Don't make excuses. Either you love Smalltalk, or you don't. Fight for what you love. You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
|
|
I will fight. I have a blog post I'm working on that will talk about Smalltalk and tenants of OO that everyone else got wrong and still get wrong - outside of Smalltalk. I'm also continuing to develop Redline Smalltalk and contribute a Smalltalk grammar to the Antlr4 library. - James. Redline Smalltalk. You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
|
|
On Fri, Feb 6, 2015 at 8:23 AM, James Ladd <[hidden email]> wrote:
Hi James, Do you have more information what others get wrong about OO? I'd be very interested knowing. I'm a believer but I'd like to be able to express myself better to others. cheers -ben
You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
|
|
Hi Ben, I'm still trying to organise my thoughts and write them down with examples that help. This is hard. In a nutshell I think two things are almost always overlooked when people talk about or do OO - especially those who complain about OO and how it 'failed'. 1. Alan Kay said that message passing is a key concept and I think most people don't understand what he meant. Of course I have not had the opportunity to ask or speak with Mr Kay to clarify (although I have tried to meet him through a mutual friend) so there is the possibility that I'm wrong. I have seen the concept of message passing boiled down to mean calling methods and while this is a result somewhere in the process of message passing I don't believe method calls are the implementation of this key concept rather it is a requirement of an OO system to enable all things via message passing. For example, flow control, looping and other constructs are implemented through message passing. Languages that have the compiler/interpreter handle these constructs (C++, Java etc) miss the point and therefore the benefits. Thinking in terms of message passing at this level changes your thought process and the design of the resulting system. Lisp and Smalltalk are the only languages I know to get this right. 2. Most developers who practice OO are simply bundling a structure (variables/fields) with some procedures that operate on them, failing to actually model behaviour. The fields of an object will fall out of modelling behaviour as they should only be there to support that behaviour yet defining the 'data' of an object is a starting point for most and the wrong place to start. Almost always the 'data' of an object does not need to be made available to the outside world yet writing a getter and a setter is the starting place for so many. This is a quick note and I hope you don't mind that it is just a quick scribble. - James. On Fri, Feb 6, 2015 at 3:35 PM, Ben Coman <[hidden email]> wrote:
You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
|
|
thanks james. On Sat, Feb 7, 2015 at 10:46 AM, James Ladd <[hidden email]> wrote:
You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
|
|
In reply to this post by James Ladd
> > > Thinking in terms of message passing at this level changes your thought process and the design of the resulting > system. Lisp and Smalltalk are the only languages I know to get this right. > The designers of NewSpeak took this idea to an extreme and created an OO system that had message passing as the main paradigm (they removed assignment from the set of available language primitives). When they did this they found that several other software construction concepts can easily be carried by a simple framework of nested classes. So they mapped binding of global names to classes and namespaces to nested classes. They created a concept of binding environment (yep, that is a class too) that parameterises code loaded ‘onto’ such an environment. They could have gone further, for example by implementing constant binding as message sends too but decided not to go that far… http://www.newspeaklanguage.org/ > 2. Most developers who practice OO are simply bundling a structure (variables/fields) with some procedures that > operate on them, failing to actually model behaviour. The fields of an object will fall out of modelling behaviour > as they should only be there to support that behaviour yet defining the 'data' of an object is a starting point for > most and the wrong place to start. Almost always the 'data' of an object does not need to be made available to > the outside world yet writing a getter and a setter is the starting place for so many. ‘Mister MVC’ Trygve Reenskaug has introduced a new OO paradigm called DCI where data resides in exactly such dumb objects and their system behaviour is separated out into role objects (ephemeral Traits) that are available only in specific contexts (use cases). It is interesting to see that a ‘dumb objects are wrong’ meme has been turned on its head by him… http://en.wikipedia.org/wiki/Data,_context_and_interaction Hope you find this inspiring, Reinout ——————— -- You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
|
|
In reply to this post by Ben Coman-3
Any feedback/comments? Sent from the holodeck. -- You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
|
|
In reply to this post by Reinout Heeck
Thank you for your feedback.
I'll look into newspeak. I have done a lot of work DCI and it is an interesting pattern. I find it useful on the edges of a system but less so where the behaviour is captured well from those inputs around the edges. I don't agree that DCI shows dumb objects are ok. Sent from the holodeck. > On 7 Feb 2015, at 11:02 pm, Reinout Heeck <[hidden email]> wrote: > > >> >> >> Thinking in terms of message passing at this level changes your thought process and the design of the resulting >> system. Lisp and Smalltalk are the only languages I know to get this right. > > The designers of NewSpeak took this idea to an extreme and created an OO system that had message passing as the main paradigm (they removed assignment from the set of available language primitives). > > When they did this they found that several other software construction concepts can easily be carried by a simple framework of nested classes. So they mapped binding of global names to classes and namespaces to nested classes. They created a concept of binding environment (yep, that is a class too) that parameterises code loaded ‘onto’ such an environment. They could have gone further, for example by implementing constant binding as message sends too but decided not to go that far… > > > http://www.newspeaklanguage.org/ > > > > >> 2. Most developers who practice OO are simply bundling a structure (variables/fields) with some procedures that >> operate on them, failing to actually model behaviour. The fields of an object will fall out of modelling behaviour >> as they should only be there to support that behaviour yet defining the 'data' of an object is a starting point for >> most and the wrong place to start. Almost always the 'data' of an object does not need to be made available to >> the outside world yet writing a getter and a setter is the starting place for so many. > > > ‘Mister MVC’ Trygve Reenskaug has introduced a new OO paradigm called DCI where data resides in exactly such dumb objects and their system behaviour is separated out into role objects (ephemeral Traits) that are available only in specific contexts (use cases). > > It is interesting to see that a ‘dumb objects are wrong’ meme has been turned on its head by him… > > > http://en.wikipedia.org/wiki/Data,_context_and_interaction > > > > > > > > Hope you find this inspiring, > > Reinout > ——————— > > -- > You received this message because you are subscribed to a topic in the Google Groups "Smalltalk Research" group. > To unsubscribe from this topic, visit https://groups.google.com/d/topic/smalltalk-research/HudsRj0qDjo/unsubscribe. > To unsubscribe from this group and all its topics, send an email to [hidden email]. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Smalltalk Research" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. For more options, visit https://groups.google.com/d/optout. |
«
Return to English
|
1 view|%1 views
| Free forum by Nabble | Edit this page |