Re: Should Inline be private in the private part of a package spec?

[Georg Bauhaus <rm.dash-bauhaus@futureapps.de>]

On 15.08.12 09:51, Dmitry A. Kazakov wrote:
> On Tue, 14 Aug 2012 17:04:09 -0500, Randy Brukardt wrote:
> 
>> "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote in message 
>> news:1mchat48i3fos.1ksbz02nuzf5f$.dlg@40tude.net...
>>> On Tue, 14 Aug 2012 07:44:42 -0700 (PDT), Shark8 wrote:
>>>
>>>> "XML combines the efficiency of text files with the readability of binary 
>>>> files" -- unknown
>>>
>>> That one was great! (:-))
>>
>> And accurate! Text files are quite a bit slower to access than binary files, 
>> which is OK if they are intended to be processed (read/modified) by humans. 
>> But XML is lousy at that (humans are not good at counting bracketing 
>> constructs).
> 
> It has a vital psychological effect. Something simple to read and
> understand will never be considered worth of money and efforts.
> 
> On the contrary, a highly complicated mess requiring a heap of tools
> mounted on each other, is a cool thing. Clearly, spending man-centuries on
> designing all the garbage must indicate its great value in all fields of
> software engineering!

Like the situation that created Ada, the situation before agreed-upon
means of data exchange between totally heterogeneous systems
had created a highly complicated mess. To get at some piece of
data it was required not only to have tools mounted each other,
not only to know what section of bits may mean what,
but integrating the mess also required a second order set of tools
to make the first order set of tools work together.

Fortunately, for non-internal data formats, getting at pieces of
information in "foreign" data is now possible in a validating
form by having the heterogeneous systems interchange data in
an agreed-upon simple, structured, descriptive format.

I would say that a network connecting the electrical devices
of a car does not counts as heterogeneous.