On 4/28/2010 9:38 AM, Brent Pinkney wrote:
>>> 1. "an object is send a message and responds with an object"
>> It is perfectly consistent with it.
>
> On reading your blog, I think I can grog that: some interrogation message is sent to the class/method and it replies with the
> pragma thing or similar.
>
>>> 2. there is no other "magic" in the system.
>>> Did you make that up? A wise man once said that any sufficiently
>> advanced technology is indistinguishable from magic. I don't recall this
>> ever being mentioned as a "fundamental principle" of Smalltalk.
>
> This was not meant to be confrontational, sorry if it came across as such.

It didn't. It's just not clear what constitutes "magic" and I've never 
seen it mentioned in relation to defining principles of Smalltalk. 
Simplicity, yes, but magic I've never heard. That's why I brought 
Clarke's law into it because while something like "thisContext sender 
selector" is commonplace for many of us, when I first saw it my jaw 
simply dropped. That *was* magic at some point for me.

Cheers,
  - Andreas

> By way of example, I think the general dislike for the
> current compiled method format with its byte code trailer is evidence that we as a community do not like solutions that are
> special and work with unseen code (magic) in the VM.
>
> I am just trying to ensure that what we get is a breakthrough solution that can also motivated from our communities shared
> aesthetics, which I tried to summarise as above.
>
> If so, then cool.
>
> Cheers
>
> Brent
>
>