Text 6879, 174 rader
Skriven 2013-02-24 15:17:55 av Michiel van der Vlist (2:280/5555)
Kommentar till en text av mark lewis (1:3634/12.42)
Ärende: LOL!
============
Hello mark,
On Wednesday February 20 2013 14:13, you wrote to me:
ml>> then how come björn's name appears as an o umlate sometimes and
ml>> as a division symbol at other times
MvdV>> Because some use CP437/CP850 encoding and some use Latin-1
MvdV>> encoding and the CP437 only crowd is unable to do the
MvdV>> conversion.
ml> my software does perform conversions...
The problem lies with the CP437 only systems that do NOT convert. They leave
the byte representing the ö as is, but when quoting it with their editor, they
send the message witj a CP437 CHRS kludge no matter what. So if the original
message was in Latin-1, it will show up as a division sign in the next step.
ml> that doesn't explain why the incorrect characters /arrive/ in the
ml> PKTs... no tosser software should be performing any transliterations
ml> on any messages passing through it...
It is not the tosser that does any translation. It is the reader that does the
translation when transfering the bytes from the message base to the display
buffer. It translates from the ecoding defined vy the CHTS kludge to the native
code page of the local system.
When reply commenting, the conten of the buffer is edited and when writing back
to the message base, it is translated to the encoding that the user has
selected as the encoding for sending messages. Ever so often that is the native
code [age of his system, so that no further translation takes place
ml> that means that the characters leave the originating system as
ml> the incorrect characters to start with...
No.
ml>> and even worse as two incomprehensible graphical characters
MvdV>> Because some use UTF-8 encoding and not all can properly deal
MvdV>> with that.
ml> provide or point to the transliteration tables to be used going from
ml> UTF-8 to CP437, CP850, Latin-1, etc and i'm sure they will be used...
I have only one for UTF-8 to CP850
ml> that should put a halt to all the bickering about character sets, too
ml> ;)
I doubt it.
MvdV>> Because some that convert, convert the wrong things.
ml> then that means that their tables are wrong yet they are the same
ml> tables used by everyone else AFAIK...
Or they have not configured their system to do any translation at all. They
have not isntalled any transation tables.
ml> my understanding is and has been that everyone software is using the
ml> same tables and table format used back when FSC-0054 (and its
ml> predecessor(s)) were developed...
Not evryone has installed conversion tables for all conceivable encodings.
ml> i know that GoldEd uses them
But not everyone has them installed, or has it installed incorrectly. You e.eg
do not seem to have a table installed for Latin-9.
ml> and i'm pretty sure that i've used GoldEd's in TimED and TimED's in
ml> GoldEd in the past...
I never got Timed to do it properly, so i dumped it.
ml>> that's not fouling up the existing works??
MvdV>> As I wrote: hardly. For the ASCII/CP437 only crowd there is no
MvdV>> problem as long as they stay on their own turf. For the
MvdV>> multilangual crowd it is no problem as they have adapted years
MvdV>> ago. If it is in Russian, you can't read it anyway.
ml> so you are actually promoting even more segregation and separation
On the contrary. I promote that we move to ONE encoding scheme for all and dump
this mess with dozens of code pages and dozens to the square conversion tables
to convert to and from.
ml> by your use of the phrase "stay on their own turf"?? it sure reads
ml> that way...
When moving forward, those that won't follow and stay behind, stay on their own
turf. It is not something I promote, but it is what happens if they so
choose...
MvdV>> It is only in these so called 'international' echos where the
MvdV>> ASCII/CP437 only crowd engages with the rest of the world that
MvdV>> we see an occasional mangled character. Not nice, but nothing
MvdV>> really breaks and the problem is manageable.
ml> true but it is ugly to say the least and that's what folks are
ml> complaining about...
It can't be helped.
ml> message text... those using speech readers being among the first in
ml> line when dealing with these character set issues... one can only
ml> imagine whan björn's name sounds like when a speech reader gets it
ml> with a division symbol instead of an ö or even worse when it arrives
ml> with two unpronouncible characters in UTF-8 encoding...
Is it any worse than when just looking at the screen? Anyway it can't be
heleped. There is no way going back to CP437 only. And if people do not
prepoerly install the stuff to deal with other encoding schemes, they will see
these problems. speech readers or not.
ml> think about it... "you" (inclusive) sighted folks don't stop to think
ml> about those with vision problems nor to "you" stop to think about
ml> those with other handicaps that lead to problems with their tools when
ml> someone decides to change things up just because...
Things are not changed 'just because'. They have been changed because what we
had did not work for those who need more that what ASCII only or CP437 has to
offer.
You acuse me of not stopping to think about the problems of others, but it is
actually you that does not look further than the needs of the Anglo American
language crowd. How many times do I have to say it to get it across: What you
want does not work for them. And so they have gone their own way. A long time
ago already, way over a decode ago. There is no turnig back.
MvdV>> Anyway, the foul up is no more than it ever was. It only fouls
MvdV>> up the stone age stuff and that is nothing new. It has been
MvdV>> doing that for over 15 years too.
ml> yes and it goes completely against the grain of the network which
ml> supported all kinds of machines... PDP-11, C=64, various mainframes,
ml> etc, etc and so on... some of those machines are being brought back
ml> into the network, ya know?
What does that have to do with it?
ml>>> how can you read UTF-8 on my CP437 BBS system if you telnet to
ml>>> it??
MvdV>> I have never telnetted inti your BBS system, and I have no
MvdV>> intention of ever doing it. But if I wanted I could configure
MvdV>> Putty for CP437. Or any other encoding scheme.
ml> it sounds like an interesting experiment...
Nothing interesting about it. I just fire up Putty, click on the "translation"
button on the left of the screen, and select the encoding scheme. Three mouse
clicks.
MvdV>> Face reality: outside Northern America, Fidonet has become a
MvdV>> sysops and points only network. Has been for almost 20 years.
MvdV>> No reason to shut it down for that reason.
ml> with no users, there is no need for the network is there? if an ISP
ml> has no users, they shutdown... their managers and admins don't keep
ml> everything running for their own usage, do they?
If Fido were a commercial enterprise, it would have gone belly up 15 years ago.
But it isn't. I wrote it before: Fido is like a shipping company without
passengers or cargo, but the crew keeps the ships sailing because they like to
play with ships.
Cheers, Michiel
--- GoldED+/W32-MINGW 1.1.5-b20110320
* Origin: http://www.vlist.org (2:280/5555)
|