Older Americans (like me) remember when US and Canadian dollars
were equal in value and not only could you spend bills relatively
freely back and forth (at least when you were close to the
border), you could spend coins the same way.
I must remember to take my leftover European money to Montreal for
the Worldcon, in case I can spend it in the Dealers Room with
European vendors. In Edinburgh in 1995, I bought an alternate
history fanzine from a German fan. It was DM8, and since I had
some left-over DM from our stop-over on the way to India, I gave
him DM10, getting 90p change. Thus is international finance
carried out at conventions.
(Kurlansky refers to the language as Euskera, but that is the
Basque word for it, and seems to me a bit strange, the same as
referring to Francais, Deutsch, or Ellinika, instead of French,
German, or Greek would be.)
The first 90% of the code accounts for the first
90% of the development time.
The remaining 10% of the code accounts for the other
90% of the development time.
--Tom Cargill
On Sun, 6 Sep 2026 09:18:42 -0400, Evelyn C. Leeper wrote:
(Kurlansky refers to the language as Euskera, but that is the
Basque word for it, and seems to me a bit strange, the same as
referring to Francais, Deutsch, or Ellinika, instead of French,
German, or Greek would be.)
rCLFran|oaisrCY, surely ...
On a related note, the Turks have now decreed that the official name
of their country shall be rCLT|+rkiyerCYrCa. I guess theyrCOre fed up of jokes
about the birds what go gobble-gobble ...
On 9/6/26 6:15 PM, Lawrence DrCOOliveiro wrote:
On Sun, 6 Sep 2026 09:18:42 -0400, Evelyn C. Leeper wrote:
On a related note, the Turks have now decreed that the official name
of their country shall be rCLT|+rkiyerCYrCa. I guess theyrCOre fed up of jokes
about the birds what go gobble-gobble ...
The only trouble is that Usenet readers don't always handle non-ASCII >characters well. They look fine for me (Thunderbird on a Mac), but I
suspect some other people will see garbage characters.
The only trouble is that Usenet readers don't always handle
non-ASCII characters well.
Identical-looking characters in different languages aren't always
the same Unicode character.
Lawrence D\377Oliveiro wrote:
\342\200\234Fran\347ais\342\200\235, surely ...
On a related note, the Turks have now decreed
that the official name of their country shall be
\342\200\234T\374rkiye\342\200\235\342\200\240. I guess
they\342\200\231re fed up of jokes about the birds what go
gobble-gobble ...
The only trouble is that Usenet readers don't always handle
non-ASCII characters well. They look fine for me (Thunderbird on a
Mac), but I suspect some other people will see garbage characters.
But I still don't know what the non-ASCII character in Lawrence's
name is supposed to be; it never seems to be the same character
twice. I'm not sure he's not just trolling us.
Keith F. Lynch wrote:
But I still don't know what the non-ASCII character in Lawrence's
name is supposed to be; it never seems to be the same character
twice. I'm not sure he's not just trolling us.
So your reader is not only buggy, it is not even consistently buggy. That\342\200\231s your fault, not mine.
On Sun, 6 Sep 2026 19:52:04 -0400, Gary McGath wrote:
The only trouble is that Usenet readers don't always handle
non-ASCII characters well.
That must be very old software, is all I can say. Unicode has been
around for about a third of a century now, and is universally
supported across all major computer platforms.
For me, the weird character shows up as a lower case y with an
umlaut. Somehow I doubt that's what you want me to see.
Several people here have helpfully warned you that your text looks
like gibberish. But of course it's all our fault for not using the
exact same version of the exact same software as you, as all
right-thinking people do.
I'm not using any "reader." I'm viewing the text as plain ASCII.
Someone Else wrote:
For me, the weird character shows up as a lower case y with an
umlaut. Somehow I doubt that's what you want me to see.
My posting contains this header line:
Content-Type: text/plain; charset=UTF-8
How much more explicit do I need to be?
Keith F. Lynch wrote:
I'm not using any "reader." I'm viewing the text as plain ASCII.
I wonder why?
My posting contains this header line:
Content-Type: text/plain; charset=UTF-8
How much more explicit do I need to be?
On Mon, 07 Sep 2026 00:16:36 -0400, Someone Else wrote:
For me, the weird character shows up as a lower case y with an
umlaut. Somehow I doubt that's what you want me to see.
My posting contains this header line:
Content-Type: text/plain; charset=UTF-8
How much more explicit do I need to be? ThatrCOs a standard MIME-type
header line, invented back in the 20th century and commonly adopted
across email, HTTP ... everything, really. USENET is supposed to use
the same sorts of header lines.
On 9/6/26 9:18 AM, Evelyn C. Leeper wrote:
Older Americans (like me) remember when US and Canadian dollars
were equal in value and not only could you spend bills relatively
freely back and forth (at least when you were close to the
border), you could spend coins the same way.
I can remember having to explain to a cashier why I didn't want a
Canadian dime in my change. Canadian coins once turned up fairly often
in New Hampshire, though they're much less common now.
Lawrence D\377Oliveiro <ldo@nz.invalid> wrote:
Keith F. Lynch wrote:
I'm not using any "reader." I'm viewing the text as plain ASCII.
I wonder why?
I just explained why: Treating newsgroup posts as code would let any
rogue Usenet poster trash my computer.
On Mon, 7 Sep 2026 03:07:40 -0000 (UTC), Keith F. Lynch wrote:
Several people here have helpfully warned you that your text looks
like gibberish. But of course it's all our fault for not using the
exact same version of the exact same software as you, as all
right-thinking people do.
Lots of other people use entirely different software from me, and have
no trouble.
Lawrence D\377Oliveiro <ldo@nz.invalid> wrote:
Keith F. Lynch wrote:
I'm not using any "reader." I'm viewing the text as plain ASCII.
I wonder why?
I just explained why: Treating newsgroup posts as code would let any
rogue Usenet poster trash my computer. As I said, the worst case I
can think of is that it would be turned into a child porn server,
and that I would then be given a life sentence.
Unlikely? Yes. But the downside is so huge, and the upside is so
tiny, it's no contest. Similarly, I would never try to race a train
across a closed railroad gate, since I'd much rather be delayed by
five minutes than to risk being torn limb from limb by a train.
Indeed, I think anyone who makes the opposite choice simply doesn't
know as much about the risks of computers as I do.
But I've only been a computer professional for 46 years, so what do I
know? It must be that you're right, and that the several people who
have reported seeing gibberish from you are all doing it wrong. You
should tell all of us exactly what computer hardware and software you
are using, so that we can all throw ours in the trash and replace it
with a setup identical to yours.
On Sun, 6 Sep 2026 19:52:04 -0400, Gary McGath wrote:
The only trouble is that Usenet readers don't always handle
non-ASCII characters well.
That must be very old software, is all I can say. Unicode has been
around for about a third of a century now, and is universally
supported across all major computer platforms. Fallback fonts that
give legible renditions for every character (e.g. GooglerCOs rCLNotorCY >series) are freely available. So IrCOm not really sure what the excuse
is, any more.
Everyone, feel free to argue this to your heart's content.snip.
(I am not totally against change. I just recently started using only one >space rather than two between sentences.)
On Sun, 6 Sep 2026 19:52:04 -0400, Gary McGath wrote:
The only trouble is that Usenet readers don't always handle
non-ASCII characters well.
That must be very old software, is all I can say. Unicode has been
around for about a third of a century now, and is universally
supported across all major computer platforms. Fallback fonts that
give legible renditions for every character (e.g. GooglerCOs rCLNotorCY series) are freely available. So IrCOm not really sure what the excuse
is, any more.
Identical-looking characters in different languages aren't always
the same Unicode character.
That is a known pitfall, with no easy answer.
For what it's worth, I was also a computer professional for my whole
career, and I can't conceive of any way that rendering text as Unicode
could cause code to be executed on my computer.
I don't think anyone is objecting to anything that you've posted.
It's just thread drift about Lawrence D<some character>Oliveiro's
diatribes against those of us using software he doesn't like.
Someone Else <SueDominus@example.com.invalid> wrote:
I don't think anyone is objecting to anything that you've posted.
It's just thread drift about Lawrence D<some character>Oliveiro's
diatribes against those of us using software he doesn't like.
To be fair, I should also mention the diatribes against Lawrence for
using characters we can't all see correctly.
On 9/7/26 7:12 AM, Gary McGath wrote:
For what it's worth, I was also a computer professional for my
whole career, and I can't conceive of any way that rendering text
as Unicode could cause code to be executed on my computer.
Other than the code to render the characters (because I'm sure
someone will nitpick my words).
And one reason I do this is that I use vi to produce the plain text,
and a lot of these things show up as weird blue codes rather than
letters.
On Mon, 7 Sep 2026 03:08:56 -0400, Evelyn C. Leeper wrote:
And one reason I do this is that I use vi to produce the plain text,
and a lot of these things show up as weird blue codes rather than
letters.
Surely even vi can cope with Unicode nowadays ...
On 9/7/26 17:48, Lawrence DrCOOliveiro wrote:
On Mon, 7 Sep 2026 03:08:56 -0400, Evelyn C. Leeper wrote:
And one reason I do this is that I use vi to produce the plain
text, and a lot of these things show up as weird blue codes rather
than letters.
Surely even vi can cope with Unicode nowadays ...
I don't know. All I know is sometimes when someone sends me
something for inclusion, weird blue codes show up in vi.
The basic fact is that Usenet is ancient, and so is a lot of the
software used to access it. When lots of people report the issue,
telling them all to use different software is never a successful
strategy.
On Mon, 7 Sep 2026 18:59:52 -0400, Evelyn C. Leeper wrote:
On 9/7/26 17:48, Lawrence DrCOOliveiro wrote:
On Mon, 7 Sep 2026 03:08:56 -0400, Evelyn C. Leeper wrote:
And one reason I do this is that I use vi to produce the plain
text, and a lot of these things show up as weird blue codes rather
than letters.
Surely even vi can cope with Unicode nowadays ...
I don't know. All I know is sometimes when someone sends me
something for inclusion, weird blue codes show up in vi.
Do you have the rCLunicoderCY tool available at your command line? E.g.
ldo@theon:~> unicode u+00b1
U+00B1 PLUS-MINUS SIGN
UTF-8: c2 b1 UTF-16BE: 00b1 Decimal: ± Octal: \0261
-#
Category: Sm (Symbol, Math); East Asian width: A (ambiguous)
Unicode block: 0080..00FF; Latin-1 Supplement
Bidi: ET (European Number Terminator)
Age: Assigned as of Unicode 1.1.0 (June, 1993)
Would the same command show the character legibly on your system?
On 9/7/26 17:48, Lawrence DrCOOliveiro wrote:
On Mon, 7 Sep 2026 03:08:56 -0400, Evelyn C. Leeper wrote:
And one reason I do this is that I use vi to produce the plain text,
and a lot of these things show up as weird blue codes rather than
letters.
Surely even vi can cope with Unicode nowadays ...
I don't know. All I know is sometimes when someone sends me something
for inclusion, weird blue codes show up in vi.
I've been editing/publishing the MT VOID for 48 years this way, through
2448 issues, and I am not going to twist myself into knots for non- English-alphabet letters.-a The only time *that I know of* when it
changes meaning is in Spanish ("año" vs. "ano"), and I will just avoid that.
Similarly, the esszett, which looks a little like a
lower-case Greek beta, can be replaced with "ss."
In running ConReg for Dundracon, I get files of comp'd members
generated by Windows programs. Once transferred to a Linux
system, I use vi to--among other things--remove the trailing ^M
(carriage return) from each line.
When I generate badge images to print, my program inserts the
needed PCL coding, and those codes invariably begin with an
escape character. If you look at the generated files in vi, they
are full of ^[ characters.
Don't get me started on people who send HTML formatted files to a
*nix mailbox expecting it to be formatted when opened by "mail".
On the other hand, the inability to format HTML or display
graphics--let alone execute anything--is an effective way to
avoid malware infections.
Don't get me started on people who send HTML formatted files to a
*nix mailbox expecting it to be formatted when opened by "mail". On
the other hand, the inability to format HTML or display
graphics--let alone execute anything--is an effective way to avoid
malware infections.
I've been editing/publishing the MT VOID for 48 years this way,
through 2448 issues, and I am not going to twist myself into knots
for non-English-alphabet letters.
In Message-ID:<117nn3u$36dk1$4@dont-email.me>,
Gary McGath <garym@mcgath.com> wrote:
Similarly, the esszett, which looks a little like a lower-case
Greek beta, can be replaced with "ss."
This reminds me of something I saw on the Internet long ago. (I'll
use "b" as that German letter.)
Someone wanted to have an awesome "signature", so he used that
symbol in his name and ended up with "b"illy "b"ad ass. Someone
pointed out what the pronunciation of that would be.
We can all also be annoyed at the Greekish signs (usually in a
chiseled font) that use capital sigma for capital E.
Dorothy J Heydt <djheydt@kithrup.com> wrote:
In running ConReg for Dundracon, I get files of comp'd members
generated by Windows programs. Once transferred to a Linux
system, I use vi to--among other things--remove the trailing ^M
(carriage return) from each line.
MUCH faster to do it from the command line with
tr '\015' ' '
Don't get me started on people who send HTML formatted files to a
*nix mailbox expecting it to be formatted when opened by "mail".
On the other hand, the inability to format HTML or display
graphics--let alone execute anything--is an effective way to
avoid malware infections.
Legacy mail doesn't know about mime, but you can save it, munpack it,
and run it into lynx if you wish.
On Mon, 7 Sep 2026 15:01:14 -0400, Gary McGath wrote:
On 9/7/26 7:12 AM, Gary McGath wrote:
For what it's worth, I was also a computer professional for my
whole career, and I can't conceive of any way that rendering text
as Unicode could cause code to be executed on my computer.
Other than the code to render the characters (because I'm sure
someone will nitpick my words).
rCL... could cause code controlled by some external untrusted entity to
be executed on my computerrCY, perhaps. ;)
#NitpickProofing
On Mon, 07 Sep 2026 21:28:31 -0400, Someone Else wrote:
We can all also be annoyed at the Greekish signs (usually in a
chiseled font) that use capital sigma for capital E.
Remember rCLMY BIG FAT GRSSK WSDDINGrCY ...
Similarly, the esszett, which looks a little like a
lower-case Greek beta, can be replaced with "ss."
On 9/7/2026 5:46 PM, Lawrence DrCOOliveiro wrote:
On Mon, 7 Sep 2026 15:01:14 -0400, Gary McGath wrote:
On 9/7/26 7:12 AM, Gary McGath wrote:
For what it's worth, I was also a computer professional for my
whole career, and I can't conceive of any way that rendering text
as Unicode could cause code to be executed on my computer.
Other than the code to render the characters (because I'm sure
someone will nitpick my words).
rCL... could cause code controlled by some external untrusted entity to
be executed on my computerrCY, perhaps. ;)
#NitpickProofing
At the limit, any time you use a program to process data provided by
someone else, there is a risk that it will exploit a bug in your
program to make the program do something undesired and unexpected,
up to and including running code sent as part of the data.
Back when I was a wee hacker, buffer overflows were the gateway -
send much longer text than expected, or number staggeringly huge,
and you could compromise many programs.
Keith may think by eschewing Unicode and HTML, he's escaped the
problem, but trn still has bugs.
Possibly on the more general topics under discussion... Dorothy's
novel _The Interior Life_ was written using vi and the submitted
manuscript was formatted using nroff.
On Tue, 8 Sep 2026 18:37:46 GMT, Dorothy J Heydt wrote:
Possibly on the more general topics under discussion... Dorothy's
novel _The Interior Life_ was written using vi and the submitted
manuscript was formatted using nroff.
nroff is Turing-complete, I believe. So if you were to accept
arbitrary nroff/troff markup from untrusted sources, you could indeed
be tricked into executing dodgy code on your system ...
On Mon, 7 Sep 2026 07:07:28 -0400, Gary McGath wrote:
The basic fact is that Usenet is ancient, and so is a lot of the
software used to access it. When lots of people report the issue,
telling them all to use different software is never a successful
strategy.
How far back do we need to go, I wonder? I mentioned that Unicode is
older now than ASCII was when Unicode was invented. But even then,
ASCII was still younger than Baudot code was when ASCII was invented.
Is that an argument for going back to Baudot code?
Even folks using Windows -- unless that Windows version is older than
XP, it is built on the Windows NT kernel. Which was one of the first
to embrace Unicode right into its deepest levels, way back to its
first release in 1993.
It would be ironic, indeed, if peoplerCOs Windows systems are having
trouble with Unicode, would it not?
You could, if you chose to do so, use an ASCII apostrophe U+0027
rather than U+2019. I believe that would be semantically more
appropriate, since the character does not introduce a quotation,
whereas an apostrophe is exactly what fits in that place: "Lawrence D'Oliveiro". It would be displayed correctly even by ASCII-only
client software.
Of course it's your name, and you can spell it how you like. If you
want your name to contain an unmatched right quotation mark, you
can do so. If you chose to use an apostrophe instead, more people
would be able to read your name.
There are also visual issues. Historically, the ASCII apostrophe
has done double duty as an opening and closing single quote, so
many typefaces represent it as a vertical stroke for the sake of
neutrality. The right quotation mark often looks more like what
we expect an apostrophe in an attractive font to look like.
Gary McGath <garym@mcgath.com> wrote:
There are also visual issues. Historically, the ASCII apostrophe
has done double duty as an opening and closing single quote, so
many typefaces represent it as a vertical stroke for the sake of
neutrality. The right quotation mark often looks more like what
we expect an apostrophe in an attractive font to look like.
Again, the perfect is the enemy of the good. It might look slightly
better to some people, but only at the expense of looking much worse
to most others, e.g. like "\342\200\231."
On Mon, 7 Sep 2026 04:48:42 -0000 (UTC), Keith F. Lynch wrote:
Lawrence D\377Oliveiro <ldo@nz.invalid> wrote:
Keith F. Lynch wrote:
I'm not using any "reader." I'm viewing the text as plain ASCII.
I wonder why?
I just explained why: Treating newsgroup posts as code would let any
rogue Usenet poster trash my computer.
It???s text, not code.
Imagine it???s 1994, and you were refusing to use this newfangled ASCII >thing, you wanted to stick to EBCDIC or something. You???d get laughed
out of the room.
Unicode is older now than ASCII was then.
In <117lm25$30p2c$1@dont-email.me> Lawrence <ldo@nz.invalid> writes:
You may be thinking of BCD, which was a 6 bit code (EBCDIC was 8 bit, >designed around Bytes, which earlier machines like the 7090, which didn't >use bytes, but words, or earlier machines like the 1620, which used
decimal digits).
The 026 keypunch was built for BCD. The 029 came out with System/360.
Lawrence <ldo@nz.invalid> writes:
Unicode is older now than ASCII was then.
| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 76 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 56:00:03 |
| Calls: | 1,265 |
| Calls today: | 1 |
| Files: | 1,362 |
| D/L today: |
3 files (2,684K bytes) |
| Messages: | 285,718 |