Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.ai.fuzzy    |    Fuzzy logic... all warm and fuzzy-like    |    1,275 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 571 of 1,275    |
|    Dmitry A. Kazakov to Maxim S. Shatskih    |
|    Re: Fuzzy Logic Operating Systems    |
|    15 Feb 06 14:20:36    |
      XPost: alt.os.development       From: mailbox@dmitry-kazakov.de              On Wed, 15 Feb 2006 14:52:20 +0300, Maxim S. Shatskih wrote:              >> 1. There are technical and theoretical problems with OO languages. Many       >> things (multiple dispatch, multiple inheritance, ad-hoc supertypes,       >> extensible active objects (tasks), extensible protected objects) are quite       >> difficult. Basically nobody knows how to do them right. And this is just on       >> the language level.       >       > Yes, and now note - language is a _tool_, it has no value by itself, its       value       > is only in the tasks it can solve.              That depends, note that the tasks being solved also have no value and so       on...              > So, if you're getting lost if the internals of the _tool_ - this is very bad.       > Maybe use simpler tools? or the simpler features of this complex tool? maybe       > such complex features serve no purpose then self-satisfaction of their       author?              Yes, if that were possible. The things I mentioned are necessary for any       *sound* types system. This practically means, that if you don't have them       in the language, you are doomed to permanently re-invent the wheel, to use       bad patterns, to exercise rigid programming disciplines etc. This costs a       lot of real money.              > For instance, most discussions I've heard from .NET/C# developers are the       > "smart new features" of the language, and IDE settings. Sorry, but how this       > relates to the development itself?       >       > Looks like these people are spending lots of time trying to govern and       > comprehend their tools. Then they get accustomed to extremely complex ways of       > solving the problems, and start to consider this - normal.       >       > The "use the newest and smartest possible new features of the new tool"       > paradigm is evil for me. It increases the number of issues and not decreases       > it.              Well, tools. It think something essential has got lost in the last decades.       For example, a clear understanding that no tool can replace the language.       Languages are bad, and this cannot be fixed by mounting a chain of tools,       one upon another. Tools are parasitizing on language deficiencies.              >> 2. Even if there were a good OO language supporting advanced ADTs, OS       >>would       >> require something more. Present types systems are co-operative. You can       >> call a private method even if you no right to do it. For an OO OS one would       >> need a memory access based protection of dispatching tables and private       >> members. That is another problem to solve.       >       > That's why there is no widely-used OO OSes in the world :-).              No. It is because of the UNIX/Windows disaster, which devastated the market       of operating systems.              > Anyway the       > C++-style private-public division is evil. It mandates that all private stuff       > is declared in the same header as the public part of the class, thus exposing       > the internal details and introducing major build dependencies (add a private       > field to some base class and get the full rebuild).              Egh, what's the point? That there should be no different access rights?       There is Windows 3.11 for that, enjoy it! (:-))              As for exposing implementation details, yes, this is bad, but this is an       unrelated issue. Note that most OO languages simply don't draw that line,       mixing interfaces and implementation in one big mess.              > Also note: OO OS will be 100% tied to the particular OO language.              Now it is 100% tied to ANSI (or even K&R) C, so there is little to lose.       But the question is important. I think that there exists a large common       denominator in all *feasible* types systems. That should allow a language       independent OO interfaces. That's my guess.              --       Regards,       Dmitry A. Kazakov       http://www.dmitry-kazakov.de              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca