Re: Why There are no Asm Apps

From: The Wannabee (faq_at_.@.@.@.@SZMyggenPV.Com)
Date: 06/16/04

  • Next message: f0dder: "Re: Why There are no Asm Apps" 
    Date: Tue, 15 Jun 2004 15:07:40 -0700

    På Tue, 15 Jun 2004 12:20:03 GMT, skrev Randall Hyde
    <randyhyde@earthlink.net>:

    >
    > "The Wannabee" <faq@.@.@.@.@SZMyggenPV.Com> wrote in message
    > news:opr9nj1fnsv9uht2@news.fasthost.no...
    >> På Mon, 14 Jun 2004 18:58:07 GMT, skrev Robert Redelmeier
    >> <redelm@ev1.net.invalid>:
    >>
    >> > Ray M. Ransom wrote:
    >> >> Come on now. Office????? If you are going to talk about writing
    >> >> Windoze application in assembly, at least start with a realistic
    >> task.
    >> >
    >> > I'm not sure it's unrealistic.
    >>
    >> Its not unrealistic - to do it. But why do it. There are Open Sources
    >> Open
    >> Office? If it works, let it. Writing Office is a very involving task. I
    >> think you would need years to do it.
    >
    > Well, if the "Open Source" code you refer to is Sun's "Star Office"
    > (or whatever they are calling it these days), you're talking about a
    > *Java* application. Of course, it's *easy* to demonstrate the
    > superiority of an assembly app over a Java app in terms of speed.

    Surely.

    > Alas, people won't compare the performance of an assembly
    > office system against a Java app, they'll compare it against MS Office.

    Sure.

    > As for it not being unrealistic - MS Office represents several *million*
    > lines of code. Though I've probably crossed the threshold and joined
    > the "million-line club" at some point or another, this is not a trivial
    > undertaking.

    No its not. And imo unneeded. But it fully doable in asm as well as HLL,
    thats my point, the main point you seem to have missed.

    > I've been relatively productive (in terms of cranking out
    > lines of code) over the past 30 years, and I've just barely cracked that
    > number. Now, before you start insulting me about my inability to
    > crank out millions of lines of code, I'd point out that your "idol" Rene
    > has spent six years (by his estimates) cranking out a 70,000 line
    > application. Even considering the lines of code that he has thrown
    > away, you're talking *60 years* at that rate to produce a million
    > lines of code.

    Yes, oki. I agree basically. It will be very involving. But Excel was
    created by 5 programmers. But you are right. M$ Office is a superduper
    bloat app, that allmost have every possible functionality known to man, so
    it would really be a large large work. But if someone offer me money for
    each working hour, I am sure I could do it within at least 10-15 years, in
    asm, alone. It wouldnt be much fun for me, it would be hell on earth, and
    I would require a large sum of money. I acknowledge, that even to document
    it features, before starting the rewrite, would probably take one whole
    year, if not two. Office is a very involving beast. But it IS a beast, and
    it _could_ be done much simpler, of that I am quite sure. Those x million
    lines could most likely be recreated with maybe x*100.000 codelines, and
    then it would be fully possible. But a nightmare, for sure. You must
    remember that M$ programmers are on the avarage 100 times worse than most
    mediocre asm programmers, so you could cut the time consirably. Also of
    course, M$ keeps updating the *** all the time, and its a great
    timesaver, to know what to implemment from the start, in terms of
    strategy, when you rewrite. You could work nonstop on large sections. With
    coffie in one hand and a smoke in the other, as you allready knew what to
    do (and then type with you something else :-)) :-)) )

    Yeah. It would be extremly hard. But not impossible. But really very
    fruitless. And imo pointless. And as fun as commiting daily suiside.

    >
    > Keeping in mind that MS Office is *several* million lines of code,
    > I'm afraid that it *is* unreleastic to expect an individual to do this
    > task. You'd need more than *years*, you'd need *lifetimes*.
    > The only realistic way to attack such a project is to have a *large*
    > team of programmers work on it. And as I pointed out in my
    > original post, you can't get a bunch of maverick assembly programmers
    > to work together on such a project.

    No. But my point is, it is AS possible in assembly, no MORE possible in
    asm. You just need the right tools, the right mindset and de(termination).
    To say that it could be done by one man is unfair, my point was originally
    that asm would be a better choise of tools. And with RosAsm = completly
    possible, and even more likely to happen then with many other assemblers.

    > So writing a faster GUID implementation convinces you that you can
    > write an Office app? Go for it!

    You missed the point. The point was that it could just as easily be done
    in asm.

    > Cheers,
    > Randy Hyde
    >
    > 

    -- 
Sender med M2, Operas revolusjonerende e-postprogram: http://www.opera.com/
  • Next message: f0dder: "Re: Why There are no Asm Apps"
    • Quantcast