in reply to Re^11: Perl 6 Module manager
in thread Perl 6 Module manager

I really don't understand how you can argue against this?

Actually, I think we are agreeing :)

My point was that I don't think the complexity issue is the source of the problem with interoperability. It is the lack of documentation for that complexity, and the lack of the will to document it. It's just another form of the universal ill of protectionism.

However, I don't think that protectionism through proprietary formats is the preserve of monopolies--though it is most effective and damaging there. For example, there is no monopoly in the RDBMS field, but the big four achieve a similar form of protectionism--usually termed 'lock-in'--through their proprietary extensions to the open standards.

In part, this is the fault of the conservatism of the standards bodies that are generally reluctant to incorporate new features and extensions to the standards before some (nearly impossible) level of universal consensus is reached. This means that the standards usually lag the requirements and possibilities of the leading edge by 5 or 10 years. In doing so, it creates the possibility for each producer to provide for the shortfall in proprietary extensions that effectively lock-in all their users that need the features and make use of the extensions.

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
Lingua non convalesco, consenesco et abolesco. -- Rule 1 has a caveat! -- Who broke the cabal?
"Science is about questioning the status quo. Questioning authority".
In the absence of evidence, opinion is indistinguishable from prejudice.

Replies are listed 'Best First'.
Re^13: Perl 6 Module manager
by BerntB (Deacon) on Apr 22, 2006 at 04:16 UTC
    Your point about the standard process was interesting. I haven't thought about it like that, but I think I buy your argument.

    It fits well with e.g. extensions to C/C++ compilers. Few would use vendor specific extensions (outside Windows) today for those languages, but (more than) a decade ago -- there might have been a case. For sql, vendor specific extensions might be a case today. And, as you write, the standardisation process have ground to cover there.

[reply]
      It fits well with e.g. extensions to C/C++ compilers. Few would use vendor specific extensions (outside Windows) today for those languages, ...

      gcc supports quite a few extensions to the latest C99, and they do get used. I've encountered the ternary operator used as an lvalue in a couple of perl modules, and computed-goto is used by one flavour of Parrot runloop. The C++ practice of intermingling declarations and statements, and declaring for loop variables inline was very wide spread long before they were finally adopted by the C99 revision.

      Indeed, without you applying --ansi --pedantic, gcc accepts these and a bunch of other extensions without warning. I have several versions of MS VC and VC++ compiler, and none of them accept extensions without explicitly enabling them as far as I am aware.

      I guess my point is that it's too easy and very common to embue MS with "evil intent", as opposed to "over-exploitation of a dominant market position". It's also easy for the younger generations to loose sight of the fact that history is just repeating itself. Google for AT&T "baby bells" 1984 and further back IBM antitrust 1972".

      Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
      Lingua non convalesco, consenesco et abolesco. -- Rule 1 has a caveat! -- Who broke the cabal?
      "Science is about questioning the status quo. Questioning authority".
      In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]
[d/l]
[select]
        Oh, so VC++ follows standards, these days? In the .NET C++, too?? Wonder will never cease... :-)

        Some people still use GCC extensions in new code? Ah well...

        I do remember other monopolists abusing power. (Don't ask me about food prices and Swedish oligopolies.. it is too long.)

        A counter example (to file systems protocols and file formats, here) proves that Microsoft isn't doing the "embrace and extend" thing everywhere. I didn't argue that, of course.

[reply]

In Section Meditations