Path: utzoo!censor!geac!torsqnt!lethe!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!ccut!wnoc-tyo-news!sranha!srava!erik
From: erik@srava.sra.co.jp (Erik M. van der Poel)
Newsgroups: comp.std.internat
Subject: Re: Reneging on promises (Internationalization)
Message-ID: <4939@srava.sra.co.jp>
Date: 22 Jan 91 08:26:55 GMT
References: <2435@enea.se> <4934@srava.sra.co.jp> <2445@enea.se>
Organization: Software Research Associates, Inc., Japan
Lines: 47

In article <2445@enea.se> sommar@enea.se (Erland Sommarskog) writes:
> Also sprach Erik M. van der Poel (erik@srava.sra.co.jp):
> > Yes, but Stavros is talking about *existing* programs, so how can it
> > be anything other than a question of updating? (Not that I am for
> > updating...)
> 
> Hm, the starting point for our discussion was a programming language
> (Eiffel) who use []{}\ as special character, and whereof it would
> be very easy to provide alternatives for the first four.

I assume that what you want is that new languages are designed such
that the ISO 646 substitutable characters are not used. Personally, I
think that new languages should go ahead and use any ASCII characters
that they wish, so that there will be obstacles for ISO 646 users,
hopefully eventually leading to the decline of ISO 646.


> > I wonder which of the following 2 alternatives will be less expensive
> > to the company in the long run:
> > 
> >  (a) replacing or adjusting the 7-bit hardware, or
> >  (b) countless frustrating man-hours battling software incompatibility
> 
> There is no software incompatibility to battle. It is just that you 
> have to chose between reading Swedish with brackets and braces or 
> reading programming code where letters appears as special characters.

I guess I should have made myself more clear. Assuming that you do not
like the current situation of having brackets in Swedish words and
Swedish characters in programs, and assuming that you want to do
something about it, which of the above alternatives would be less
costly? I.e. (a) is where you get the hardware to deal with 8-bit
codes, and (b) is where you get the software to deal with ISO 646 by
avoiding the use of ISO 646 substitutable characters in programming
constructs, etc. With (b) you create incompatibility between e.g. 
Scandinavia and USA, which I would think is a very big disadvantage.


> There probably is a cost in man-hours, but none that you just get
> some figures and then walk into management and say: "hey, let's
> throw out this stone-age equipment."

Yes, it may be difficult to convince the management.
-
-- 
Erik M. van der Poel                                      erik@sra.co.jp
Software Research Associates, Inc., Tokyo, Japan     TEL +81-3-3234-2692