FreeLists / arachne / [arachne] Re: code deletions :(((

["Ray Andrews" <randrews@xxxxxxxxx>]

Arachne at FreeLists---The Arachne Fan Club!

On Tue, 08 Jan 2008 22:46:15 -0500, jasorn <jasorn@xxxxxxxxx> wrote:

> Arachne at FreeLists---The Arachne Fan Club!

Well, this has proven to be a fruitfull discussion:

> Kinda makes me wish I had the time and inclination to compare the two
> sources.  I'm kinda curious now :)

I hope you do.    
    
    > What has to be done in that regard is for the others to adopt my
    > source code, since only it has the efficiency  and free memory
    > to make those upgrades possible in a 16 bit environment.

    But is there a reason to restrict development to a 16 bit environment?
    That, I think is the real question.

No, not 'restrict', but clean 16 bit code might have become the
starting point for an efficient, fast and bug free port to many other
platforms and environments.    
    
    Michael Polak had much more in mind than a rewrite of the existing
    code to make it neater and more legible. Even I understand that and I'm
    not a programmer. He envisioned a completely different and efficient
    modular design, that he likened to a "flowerpot" concept, with the same
    pot holding both DOS and UNIX versions. What that might have meant in
    Michael's hands nobody knows.

Correct, and I've started down that road a fair distance. Modularity
is one of my main goals.    
    
    Without SSL and Javascript support the state of the code just doesn't
    matter. At least Michael got things done. His hyperactivity wasn't
    neat but it was creative and productive in major ways. All we've gotten
    from the team so far is minor bug fixes and various significant
    enhancements--but no major upgrades that would make Arachne really
    functional and usable in today's web environment. Sure, with a great
    deal of ingenuity requiring editing the HTML of visited sites and other
    desperate measures, workarounds can be found can be found for many
    problems. But on the real front of functionality nothing is any better
    now than with Michael's last official release many years ago.

Quite right, and part of the reason for that is that the old code
became so unworkable that MP gave up on it. I've tried to fix that,
but alas, few get the point. You can't build on a rotten foundation,
so I tried to rebuild that foundation strong and square.    
    
    > Without SSL and Javascript support the state of the code just doesn't
    > matter.
    >   
    That's the point IMO.

With the state of the old code, SSL and JS will never happen. 
Fixing the one makes the others possible.    
    
Glenn's philosophy:    
    
    >> I have been working with this code for almost 3 years.
    >> Q) How much of the code that I started with did I delete ?
    >> A) None of it.
    >> That's right... absolutely no deletions were ever done.
    
... and he's entitled to it, but few programmers agree. Most of us
see working code as working code. Naturaly older versions are kept
for reference.    
    
    >> If we do not learn from history... we are destined to repeat the same 
    >> By seeing how a particular section of code was before and now is...
    >> This helps a programmer to better understand why the change was made.
    
All true, but that's what archives are for.    

    All of those _and_ the one that Joe is currently working on are compiled
    using the 'old dirty' sources.

    Let me ask you Ray....

    How many public releases have been made which are the result of
    your 'cleaned sources' which are in your words......

    ".... I move forward 10x faster than the old fork,
    and the reson is simple -- my code is clean, readable, optimized,
    organized, and efficient, it's 10x easier to do some work in my sources!"

I'm a one man show and it took me 3 years just to get the sources
clean, never mind releases, the 'count' is irrelevant. What is for sure
is that had the cleaned sources been in use all this time, we'd have
moved much further forward. As you know very well, we're just now
getting the opt pages in order. However, if you'd given me a hand
(besides testing), I'd have had a release years ago. I can't do this
all on my own. 

You can lead a horse to water but you can't make it drink.




Regards,


Ray Andrews
-- Arachne:A1b:x9

                  Arachne at FreeLists                  
-- Arachne, The Premier GPL Web Browser/Suite for DOS --