Re: RfD: One-time file loading

In article <2006Jan11.195934@xxxxxxxxxxxxxxxxxxxxxxxxxx>,
Anton Ertl <anton@xxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>Albert van der Horst <albert@xxxxxxxxxxxxxxxxxx> writes:
>>In article <2006Jan7.112050@xxxxxxxxxxxxxxxxxxxxxxxxxx>,
>>Anton Ertl <anton@xxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>>>The proposal is currently for a REQUIRED and REQUIRE that works on
>>>files, not blocks. Words with that functionality have been
>>>implemented widely.
>>
>>My criticism of the proposal is that it should not specify
>>the means to achieve the functionality. That way it will
>>be of wider use.
>
>I don't see this. If the proposal does not specify what the system
>has to do when it executes REQUIRED, then that word will be useless
>for portable programs (an example of this is the word FILE-STATUS).
>And portable programs are what the RfD/CfV process is concerned with.

You don't seem to appreciate the difference between the
what and the how. Standards should shy away from the how.

>
>>Saying that "it just happens not being implemented any other
>>way" is hardly a counter-argument.
>
>If all the current implementors have already implemented a certain
>specification, it usually does not make sense to standardize a weaker
>specification.
>
>>And it is not true, because
>>I implemented the same functionality differently.
>
>"ciforth is insignificant ... In particular I will not complain that
>the word is used in ciforth with a different meaning." (Albert van der
>Horst in <ielhfd.416@xxxxxxxxxxxxxxxxxx>)

An insignificant counter-example is a counter-example and
invalidates the theorem.

>
>>Take the the ``#include xxx'' in C++. The new standard specifically
>>says that the compiler writer is free to supply xxx in whatever
>>way he finds fit, and that it is not allowed to think of xxx.c
>>as a file. Is that the direction you want to take?
>>Changing the definition of INCLUDE of ISO Forth?
>
>No. It seems to me that you want to take this direction. Certainly,
>if this weakening would make sense for REQUIRED (it does not), it
>would also make sense for INCLUDED.

It doesn't make sense to have REQUIRE that is approximately the
same as INCLUDED.

<SNIP>
>
>>>For the internal organization of a library consisting of several
>>>files, the load screen approach of having a single file with INCLUDEs
>>>has fewer disadvantages than for independently-developed libraries,
>>>although REQUIRE may still reduce the maintenance effort.
>>
>>I don't understand what load screens approach that include library
>>files are supposed to do. I hate load screens, and I hate including
>>library files. I do split a project, such that a main file
>>includes a lot of one-issue files.
>
>I don't know what a one-issue file is, but with the "load screen
>approach" I meant the approach of having a single file that does all
>the including. Apparently that's the approach you are taking.

I'm not. I make a distinction between facilities, to be implemented
as libraries, and applications that are subdivided over different
files.
E.g. my manx2 loads a class facility from the library,
and INCLUDEs separate files for separate instruments,
and a separate file for e.g. a the way musical scores is defined.
Now all those need facilities from the library, but in
general there is no problem with the "load file approach"
for the application itself.


>- anton

Groetjes Albert

--
--
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- like all pyramid schemes -- ultimately falters.
albert@xxxxxxxxxxxxxxxxxx http://home.hccnet.nl/a.w.m.van.der.horst
.

Relevant Pages

  • Re: RfD: One-time file loading
    ... >the means to achieve the functionality. ... >>>is a mess that much resembles the dreaded load screens. ... So a directory with a single block file is a mess and contains nested ... >>has fewer disadvantages than for independently-developed libraries, ...
    (comp.lang.forth)
  • Re: client-side Ruby on iPad/iPhone?
    ... Not the standard library, but rather core types. ... before it's a useful Web development language. ... What functionality? ... JQuery and Prototype are huge libraries. ...
    (comp.lang.ruby)
  • Re: Unit Testing an ASP.NET Website using VS2008 Pro
    ... Move your classes from App_Code to their own libraries. ... you will have no problem using a unit test framework to test your functionality. ... Publish and set nUnit to point to those files ... unexpected error in displaying the Create Unit Tests Dialog. ...
    (microsoft.public.dotnet.framework.aspnet)
  • Re: cpu overheating
    ... That's why you have graphics libraries on a headless server. ... Some of these things are required to conform to some standard or other expectation of functionality, but by and large, they're components without which things just won't *work*. ... A bare installation is still somewhere in the neighborhood of six or seven hundred megs (of which, around two hundred is manual pages, documentation, and string translations, vastly out sizing the "optional" dependencies), which is hardly unreasonable. ... Perhaps I mightn't even want to install the documentation. ...
    (Fedora)
  • Re: Question(s) regarding development of proprietary FITS manipulation software. . .
    ... searching through the libraries and all, ... functionality I need shortly. ... A FITS "quickstart" of sorts would be nice also. ... I'm asking for reading recommendations as well as any potential coding ...
    (sci.astro.fits)