ALSACE-LORRAINE-L Archives

Archiver > ALSACE-LORRAINE > 1998-07 > 0900272893


From: Francis BUSSER <>
Subject: [ALSACE-LORRAINE-L] 7 bits GHETTO
Date: Sun, 12 Jul 1998 21:48:13 +0200


"W. David Samuelsen" <> wrote to all list subscribers

==========
" The Rootsweb had installed the filters to keep html formatted messages
out and change any such messages into plain text format. This affect one
item the diatrics (that is e' 'e, etc as found in French and German such
as u"), they get converted into something you might not able to
understand like this message sent by Robert Behra containing the word
de'partment. It got converted into this one, d=E9partement.

Take extra care when typing diatrics. Robert, can you provide
alternative list of diatrics?

W. David Samuelsen, listowner "
=============

Bonjour David,

Until now, I was always very impressed by Rootweb's contribution to
international genealogy, especially through your lists, and I thank you
for your efforts. In the last months, I have tried to also contribute to
Rootsweb's success with several articles about languages, history and
customs in Switzerland, Alsace and South Germany. I have also answered
hundreds of questions about my countries and translated many documents.

But from now on, you have restricted Rootsweb to americano-american
genealogy, fully cut away from your european roots. Your decision to
filter out the messages to 7 bits, throw you back about 30 years in the
the prehistory of computer science and means exclusion for us European.

Allow me please some remarks about your decision.

To take a serious job with French or German names and documents, the
diacritical marks (accents and Umlauts) are absolutely necessary. The
trick with ' or " before or after the marked character is only a
childplay to help by out of age computer systems.

In your note you have given a good example how much this trick is
missleading : at CP/M times (around 1980) it was usage to figure é (e
accent aigu) with 'e and è (e accent grave) with e' (so the word
département had to be written d'epartement and not de'partement). The
accent change the sense of the words. If I wish to write "je décide
d'écaler ma récolte d'amandes" (I decide to hull my crop of almonds) or
"je préfère décaler ma récolte d'amandes" (I find better to gather in my
almonds later), with your trick I must write "je d'ecide d'ecaler ma
r'ecolte d'amandes" and "je pr'efe're d'ecaler ma r'ecolte d'amandes"
(both sentences can mean the same, but in both cases the sense itself is
always ambiguous).

With names, how can I distinguish between D'Evricourt and Dévricourt if
I must write both d'evricourt ? In German the problem is the same. In
Switzerland, Germany and Alsace, I have found documents about many
Buser, Busser, Bußer, Büser, Büsser, Büßer, Bueser, Buesser, Bueßer,
Büeser, Büesser, Büeßer, Bauser, Bausser, Baußer, Bäuser, Bäusser or
Bäußer. To establish they were from several not related lines needed
greatest accuracy in cross-ckecking the documents. Impossible in 7 bits.

Your decision is certainly a major contribution to the international
genealogical chaos and, with the help of the anyway too approximative
registration of names, to the nightmares of many american genealogists
in the future. I think to loose my time if I continue to contribute to
some 7 bits lists and to answer her questions.

By 7 bits, the last bit is set to zero and the number of possible
characters is reduced to 127 as with the old ASCII set (American
Standard Code for Information Interchange) first defined 1966. This set
is limited to the transmission of english textes only. The memory
occupation is the same as with 8 bits, but in the prehistory of
informatics, skilled programmers used the eighth bit to enhance the
display of the characters (as reverse or blinking attributes).

The next step was to extend the character set by changing the character
page with the eighth bit. So with the 8 bits character set ISO-8859-1,
the number of possible characters is 255. This international character
set is derived from ASCII by the International Standard Organisation and
the ANSI, his american member). It is approximatively equivalent with
the international CCITT communication code number 5. The first 127
characters are the same as in the old 7 bits ASCII code.

No technical justification can be found to encode on 7 bits rather than
on 8 bits. Encoding the textes on 7 bits is not shorter and not faster
as on 8 bits. Worse, to go 7 bits, the raw textes are generally
converted from ISO 8 bits to the 'quoted printable' format. So the
transmission of each quoted character needs two bytes more as in ISO and
with your Mc Giver trick, still one character more.

If you are a masochist, you can restore the original characters with
diacritical marks from the quoted-printable format. The marked
characters are displayed as = plus an hexadecimal value. You need only
to convert this value to a decimal base and to type in Alt+0+ the
decimal value to obtain the marked character.

Simple isn't it ?

History and genealogy need greatest accuracy and greatest fidelity to
the textes. If not, we drop in mythology and fancy legends.

One cannot be seriously a 7 bits genealogist. I hope you are aware, your
decision to go 7 bits makes your lists on Rootsweb to an american
ghetto. Was that your intend ?

In fine, a question : have you a definition of the word 'diatrics' you
have used ? I find it in none of my dictionaries but expect it has
something to do with 'diacritical marks' as say the linguists.

Thank you again for all the fine job you and other volunteers make at
Rootsweb. I hope you will install very soon a little less primitive
solution to your filtering problem. Please excuse my poor alsacian
english, but nobody can write French or German on 7 bits...

(uff Elsäßich) Vielmols märssi, unn mache's güet !
(auf Deutsch) Mit freundlichen Grüßen !
(en français) Avec toutes mes amitiés et à bientôt,
(English) With my best regards,

Francis Busser (Alsace, France)

PS : in a week or so, my e-mail will change to :

This thread: