home .. forth .. colorforth mail list archive ..

Re: [colorforth] TCP State Engine


On Fri, 16 Apr 2004, John Drake wrote:
>
> Mark, I'm afraid you are the one that's lost sight
> of the context here.  While I don't agree with
> everything Sam says he's on the money this time.
> At least if I understand him right.  Jonah was
> the one saying (if I understood him right) that
> somehow TCP would need to be more complicated
> if you wanted to support a non-trivial browser.

He was speculating.

> Sam's counterpoint is that the complexity of
> TCP really has NOTHING to do with the complexity
> of a web browser.  The same basic TCP protocol
> that supported HTML 1.0 now supports HTML 4.0,
> Flash, JavaScript, DHTML ect.

This is obvious. Did Jonas think that TCP and a web browser are the same
thing? If so he is more clouded than I thought.

>
> Now you asked Sam where he got his "5 block"
> estimate for wget.  As you well know Chuck
> has estimated 3 blocks for TCP.  I'm not sure
> what Chuck is basing this on, but I'm guessing
> it's on his experiences at iTV.  Obviously they
> had a working TCP/IP stack that was written
> in machineForth and/or ANS Forth.  (I'm not
> sure who did what at iTV, I just know from
> Jeff's stories that there were MF and AF
> programmers).  So (correct me if I'm wrong
> Sam) 5 blocks come from 3 block estimate for
> TCP + 2 blocks estimate for the rest of wget.

Chuck says on his web site:

TCP/IP: 3 blocks?

That is where I get the number 3 for both TCP and IP. I have know idea how
many blocks we'll need. 3 would be great, and I think it is something to
shoot for.

>
> Also the following statement seems to have
> gotten you off kilter.
>
> ===========================================
> I couldn't agree more with this.  Fortunately for us,
> most problems are quite trivial, as long as you let go
> of certain basic assumptions and presuppositions
> ===========================================
>
> You seem to be stuck on the idea that the "you"
> in the above quote is "you" as in "Mark Slicker".
> I think (again correct me if I'm wrong Sam) the
> "you" is a generic for "anyone".  After all you
> (Mark Slicker) are clearly not the one that was
> talking about "refrigerator browsers".  The above
> could be written as follows.

No one has talked about "refrigerator browsers". And yes I really do
question why Sam is going into a lecture about letting go of assumptions.
Maybe he is the one with all these assumtions and he is convincing himself
a browser doesn't need to be as complex as say Mozilla.

I still reject the idea that all problems are trivial. I don't think the
problems Chuck Moore solves are trivial. They may have a simple expression
in code, but to achieve that expression is not trivial. I don't think TCP
is trivial, and I don't a Browser is trivial. To get these things right
is a very real challenge. If anyone disagrees, we should be seeing code
from you very soon.

Further, I reject the idea that is commonly spread in comp.lang.forth that
Forth is only appropriate for "embedded systems", or systems with minimal
resources.

>
> ===============================================
> Fortunately for us most problems are quite
> trivial as long as the programmer can let go
> of certain basic assumptions and presuppositions.
> ===============================================
>
> Sam mentioned a few.  Here are a few more.

Please don't, unless these are personal hangups, and you wish to say how
you found these assumptions are not well founded.

Mark

---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com