Re: Chuck's SVFIG talk

[Luis Commins <Luis@xxxxxxxxxx>]

Thanks for the reply!! Comments below.

>>> "Lloyd R. Prentice" <pai@tiac.net> 13/07/99 17:34:02 >>>
>Hi, Luis,

>Many thanks for your thoughtful responses.

>Luis Commins wrote:

> 1. Complexity. Very subjective term. Looking at nature and the world of >physics from a 'distant' perspective it all seems very complex but, digging a >little further and narrowing down the focus, you can see simple building blocks. >Apply that to an 'application' and you can see that they seem complex but all >they boil down to is a finite set of assembly language/machine code >instructions. Complexity as much as simplicity have a beauty of their own in >that structure, whether regular or irregular has an appeal.
>> The refuge of the complex is in the simple and vice versa.
>> I don't know what a more appropriate term would be without having to make >clear what you are actually referring to when calling something 'complex' but >'clean' is a word I would use, as in 'clean code'.

>I agree that this really is a slippery issue. My instincts tell me that I should be >able to build software that is simple and clean. But every time it somehow ends >up confusing and messy. I ask myself whether this is a result of the tools I'm >using or my own stupidity. I seem to require many iterations to come up with >optimal factoring. And all too often my deadlines won't allow it.

Time can be a great healer. I suppose it's different for everybody, that 'roll your own' comment from Vic really hit home for some. You do the best you can with what you have at that particular moment. If you feel that you can do it differently then explore other approaches, some people even have an imaginary critic that looks at their code over their shoulders, some plan plan plan 'till there's little time to decide on which plan and end up with the one that feels right at that moment.
Confusing and messy is a personal perspective, if it's validated by others it just proves they have a similar point of view. It's only when you feel you have got it 'right' that you feel comfy, even if you change it later, at least those changes are steps forward. Even a step back can be evolutionarily beneficial when you turn around.

>> 5. This reminds me of Genetix and Rebol.

>Is Genetix for real? What's Rebol?

Yep, go to www.genetixsw.com and www.rebol.com . Rebol seems to be going from strength to strength.

>> Much as a manifesto is a good thing I think the main issue/problem is >gumption.

>Actually, I launched the manifesto a bit tongue-in-cheek. I thought I could stimulate >some more enlightening discussion, but it doesn't seem to have helped the list much.

At our stage of development anything's worth a shot!!

>> It's nice to know what the group is aiming at but without the 'push' there is >little we can shove the way of the general industry.
> If we can agree on some structure, and take charge of certain issues, we >could go a long way. Lets get a website, lets think of applications in both >hardware and software and flesh out the bones, MuP21, FPGA whatever, lets >set up a central repository of information.
>> Compare what we have with what we want and see how to get what we >need.
>
> MISC is a concept/philosophy and Chuck isn't the only one to have ventured >down that path. He has become the 'masthead' of the idea and we can learn >a lot from what he is doing only if we know exactly what he is doing, even if it's >'way over our heads' or whatever, without the grain there is no field.
>
> Poo, not another long email...anyway, all the .com, .org and .net variants of >www.misc have been taken. www.minimum.org is NOT taken, if you all agree I >can buy this name and set up a simple site to start with. I'd welcome other >suggestions!!  For example: www.minima.org is free too.
>> Please let me know as soon as possible, net names are very volatile.

>I fully agree about the gumption part. Problem is that everybody has to make a >living so things will undoubtably move in fits and starts.

Chity Chitty Bang Bang could fly.

>Some months ago I offered to organize the website. I've just wrapped up two >major 80-hour/week projects, one in Barcelona, Spain, and the other in >Scottsdale, Arizona, so now I can start thinking about it again. I've got to do >some of my own marketing first, so it looks like mid-August before I'll be able to >anything for MISC. Also, I don't want to do anything that steps on Jeffs's toes. >Do you have any idea about how he'd feel about another MISC site?

Cor!! Travelin' man!! I've been trying to get myself out of my current rutt for a few years now...
Hmmm...what has it got to do with Jeff? (Directly I mean). It would be nice to have his 'blessing' so to speak and contributions, but I wasn't aiming to push anyone in particular, rather the concept. RISC doesn't belong to anyone, it's a concept, which to a degree MISC is an extreme of. The incorporation of Forth into the system is not the essence, it's the Forth philosophy in the system that counts for me. I mean, you can change the mnemonics in your assembler to virtually anything you want, they are just labels for zeroes and ones (the only grief is translating it to the manufacturers 'preferred' mnemonics in the manual...) so you could have a Forth 'like' assembler or change the mnemonics to place/road names.
Hmmm.. rambled on again.
Basically, lets get something out there.

Regards,

Luis.


>Best wishes,

>Lloyd