TidBITS TidBITS TidBITS Talk 
Comments about 10.4.3 Chris Page (apparently) - 01:33pm Nov 1, 2005 PSTvia emailTidBITS: “…upper ASCII characters in account passwords no longer potentially prevent login or crash the Mac OS X startup sequence.”Apple: “Addresses an issue in which high ASCII characters in a password could lead to a blue screen at startup, or prevent log in.”
This is one of my mantras: There is no such thing as “high ASCII.” ASCII is a character set with 128 values from 0 to 127. Any numeric value 128 or greater is by definition not ASCII in any way, shape or form. Rather, any such non-ASCII value necessarily belongs to some other character set/encoding. (On a Mac, MacRoman, ISO Latin-1 and Unicode are probably the most commonly encountered.)
It seems TidBITS was merely repeating Apple’s error, but Apple should know better than to use a vague and sometimes misleading non-term like “high ASCII” when meaningful, specific descriptions are readily available, e.g., “Unicode values 128 and higher”, “character values…” or even “non-ASCII characters”—whatever best describes the actual problem.
Pedantically yours,
Chris
P.S. The Apple article at
says “High ASCII” characters are listed in table II of
Note that although table I does indeed contain the ASCII character set, that page incorrectly identifies table II as ASCII as well. In fact, table II contains what is probably a Windows/DOS character set (again, not ASCII in any sense). To their credit, Apple does list some specific characters that cause problems, and I find it interesting to note that those characters, as far as I recall, can all be found in MacRoman and may be typed using the standard Mac U.S. keyboard layout. In other words, Apple is telling you to use only ASCII characters and the character set or encoding of the problematic characters is actually irrelevant.
-- Chris Page - Computer Professional
The most likely way for the world to be destroyed, most experts agree, is by accident. That's where we come in; we're computer professionals. We cause accidents. -- Nathaniel Borenstein
Mark as Read
|
|
Re: 10.4.3 - Interesting Behaviour On Cmd-S Boot-up
Just installed 10.4.3 Combo on my Rev B iMac G5 (Repair Permissions,
install 10.4.3 Combo, reboot, R'Perm again.)
Then Shut Down, powered up and ddid Command-S to do an <fsck -fy>. As
soon as the screen started to fill with the usual text, the iMac's
internal fan(s) fired up and ran at full blast all through the fsck
and did not shut up until the machine had just about finished booting
up into OS X.
So far,
<Cross fingers>
all seems OK.
</Cross fingers>
We wait and see what happens ...
Mike Millard.
|
|
 |  |
|
|
via email - Dunedin, New Zealand |
|
|
Re: Comments about 10.4.3
On 2/11/2005 9:39 AM, "mike.millard" <mike.millard  shaw.ca> spake thus:
> As soon as the screen started to fill with the usual text, the iMac's
> internal fan(s) fired up and ran at full blast all through the fsck
> and did not shut up until the machine had just about finished booting
> up into OS X.
My G5 has done that ever since I upgraded to Tiger (it may have even done it
before that, I can't remember). My guess is that it's something along the
lines of "hey, the system appears to be busy, but I haven't loaded the
environmental management software yet, so I'll just turn all the fans on to
just be safe".
--
Nigel Stanger, Dunedin, NEW ZEALAND.
http://public.xdi.org/=nigel.stanger
|
|
 |  |
tebogue
-
Nov 1, 2005 3:41 pm
(#3 Total: 15)
|
 |
|
|
After 10.4.3 updt Lacie Firewire no longer visible G4 500Mhz
My Lacie Firewire 160gb hd no longer shows up after the 10.4.3 update. Works fine on my pb 12". The computer is a early G4 500Mhz tower, 640mb memory, one 40gb and a second 80gb drive. I did all the right stuff, repair, permissions, & PRAM.
I expect it's a bug that will show up fixed some time in the future.
tks all.
Tim
|
|
 |  |
|
|
Re: Comments about 10.4.3
On 2/11/2005, at 9:39 AM, mike.millard wrote:
> Just installed 10.4.3 Combo on my Rev B iMac G5 (Repair Permissions,
> install 10.4.3 Combo, reboot, R'Perm again.)
>
> Then Shut Down, powered up and ddid Command-S to do an <fsck -fy>. As
> soon as the screen started to fill with the usual text, the iMac's
> internal fan(s) fired up and ran at full blast all through the fsck
> and did not shut up until the machine had just about finished booting
> up into OS X.
I understand this is normal behavior, kind of like a fail safe.
It's the operating system that controls the fan speed and I'm
assuming that if the control signals don't appear after a certain
time. presumably early in the boot process, the fans get switched
into full "on" mode.
We have a new G5 X-serve that did the same thing when booted into
target disk mode, it's loud enough to make you want to leave the room
after a few minutes
-------------------------------------
Jason Campbell
Technician
Psychology Department
University of Otago
Ph (03) 479 7668
|
|
 |  |
|
|
Re: Comments about 10.4.3
At 12:39 PM 11/01/2005 -0800, Chris Page wrote:
>This is one of my mantras: There is no such thing as "high ASCII." ASCII
>is a character set with 128 values from 0 to 127. Any numeric value 128 or
>greater is by definition not ASCII in any way, shape or form.
To expand on this, the reason for the 7-bit code is that until recently,
the eighth bit was used for parity. This was going on at least as far back
as the days of teletypes and paper tape -- that's as far back as I go.
Paper tape had eight channels, so a 7-bit code allowed for a parity bit.
Most pre-Internet dial-up services also used 7-bit transmission codes, as
I'm sure many here remember. IP, however, uses 8-bit frames, and so has
mostly pushed the 7-bit codes out of use. The single-byte code was most
useful when checking an entire message was infeasible -- certainly true on
a teletype and partly true on systems through the 1970s. Modern
communications protocols use more sophisticated error checking codes which
obviate the need for the single-byte code.
I don't remember whether the parity code was part of the ASCII spec, or if
the bit was simply left unused due to conventional use.
Edward
Art Works by Melynda Reid: http://paleo.org
|
|
 |  |
-
Nov 3, 2005 10:55 am
(#6 Total: 15)
|
  |
|
|
Digest from TidBITS Talk
I installed the update and when i went back to work in InDesign CS2, my HP5550 printer would not print, in spite of reinstalling the printer driver, rebooting etc. I could only print when I choose another partition which has 10.4.2. There seems to be no way to tell Apple via email that there are serious bugs in this up date.
Jack Hodges, Hendersonville, NC
|
|
 |  |
|
|
Re: Comments about 10.4.3
--On November 1, 2005 12:39:02 PM -0800 "mike.millard"
<mike.millard  shaw.ca> wrote:
> Just installed 10.4.3 Combo on my Rev B iMac G5 (Repair Permissions,
> install 10.4.3 Combo, reboot, R'Perm again.)
Out of curiosity, are people really finding all these steps necessary? I
haven't found the need for a combo install (vs. the delta install) since
early in the 10.3 cycle. And I haven't repaired permissions in at least as
long (when I installed 10.4 I did an archive and install, i didn't bring
anyting back except for data files -- no settings or apps)
The most I do is make sure i've updated the backup of my home folder and
let 'er rip. Am I just lucky, or is proactively taking these steps a little
overkill?
|
|
 |  |
|
|
Comments about 10.4.3
On or about 11/2/05 7:39 AM, thus spake "Edward Reid" <edward  paleo.org>: At 12:39 PM 11/01/2005 -0800, Chris Page wrote: > This is one of my mantras: There is no such thing as "high ASCII." ASCII > is a character set with 128 values from 0 to 127. Any numeric value 128 or > greater is by definition not ASCII in any way, shape or form. To expand on this, the reason for the 7-bit code is that until recently, the eighth bit was used for parity. This was going on at least as far back as the days of teletypes and paper tape -- that's as far back as I go. Paper tape had eight channels, so a 7-bit code allowed for a parity bit. Nevertheless, the insistence that one can't say "high ASCII" at all is just
a knee-jerk canard. During the 80s we said it all the time. And we knew (as
does Chris Page) exactly what it meant: it meant a character byte with a
value between 128 and 255. And we knew exactly what high ASCII did on our
particular machine; for example, in an Apple II, a high ASCII character sent
to the screen appeared inverted (highlighted) from its low ASCII
counterpart. Even when different uses of the character bytes 128-255 had appeared (e.g.
MacRoman vs. WinLatin), there was *still* no doubt what "high ASCII" meant:
it still meant those character byte values (though as glyphs they were
rendered differently on different machines, which is why to this day you see
web sites with "1" and "2" instead of curly quotes, and so forth). It is also certainly true that, now that characters are represented by
multiple bytes and a glyph's numeric value can be much higher than 255,the
phrase "high ASCII" used out of context can be a bit confusing. But as soon
as you have some context, it stops being confusing. For example, AppleScript
strings are MacRoman. Once you know that, you know exactly what I mean when
I describe an AppleScript character as high-ASCII, and pretending you don't
is just that - pretending. m. --
matt neuburg, phd = matt  tidbits.com, http://www.tidbits.com/matt/
pantes anthropoi tou eidenai oregontai phusei
AppleScript: the Definitive Guide -
http://www.amazon.com/exec/obidos/ASIN/0596005571/somethingsbymatt
Take Control of Word 2004, Tiger, and more -
http://www.takecontrolbooks.com/tiger-customizing.html
Subscribe to TidBITS! It's free and smart. http://www.tidbits.com/
|
|
 |  |
|
|
Re: Comments about 10.4.3
On Nov 3, 2005, at 10:55 AM, <tidbits-talk  tidbits.com> <tidbits-
talk  tidbits.com> wrote:
> I installed the update and when i went back to work in InDesign
> CS2, my HP5550 printer would not print, in spite of reinstalling
> the printer driver, rebooting etc. I could only print when I
> choose another partition which has 10.4.2. There seems to be no
> way to tell Apple via email that there are serious bugs in this up
> date.
Have you tried to go into the Printer Setup Utility, and then choose
Reset Printing System under the Printer Setup Utility menu? This
does some cleanup and then readd the printer. You should see if there
is another print driver download on the HP site, too. Let us know if
it works.
|
|
 |  |
|
|
Re: Comments about 10.4.3
At 9:58 AM -0800 11/3/05, Kevin van Haaren wrote:
>--On November 1, 2005 12:39:02 PM -0800 "mike.millard"
><mike.millard  shaw.ca> wrote:
>
>>Just installed 10.4.3 Combo on my Rev B iMac G5 (Repair Permissions,
>>install 10.4.3 Combo, reboot, R'Perm again.)
>
>Out of curiosity, are people really finding all these steps necessary? I
>haven't found the need for a combo install (vs. the delta install) since
>early in the 10.3 cycle. And I haven't repaired permissions in at least as
>long (when I installed 10.4 I did an archive and install, i didn't bring
>anyting back except for data files -- no settings or apps)
>
>The most I do is make sure i've updated the backup of my home folder and
>let 'er rip. Am I just lucky, or is proactively taking these steps a little
>overkill?
As usual, John Gruber nailed it...
< http://daringfireball.net/2004/12/software_update>
< http://daringfireball.net/2005/01/misc_updates>
Summary: no!
Best,
Ken P.
|
|
 |  |
|
|
Re: Comments about 10.4.3
On Nov 3, 2005, at 12:58 PM, Kevin van Haaren wrote:
> --On November 1, 2005 12:39:02 PM -0800 "mike.millard"
> <mike.millard  shaw.ca> wrote:
>> Just installed 10.4.3 Combo on my Rev B iMac G5 (Repair Permissions,
>> install 10.4.3 Combo, reboot, R'Perm again.)
>
> Out of curiosity, are people really finding all these steps necessary? I
> haven't found the need for a combo install (vs. the delta install)
> since early in the 10.3 cycle. And I haven't repaired permissions in at
> least as long (when I installed 10.4 I did an archive and install, i didn't
> bring anyting back except for data files -- no settings or apps)
>
> The most I do is make sure i've updated the backup of my home
> folder and let 'er rip. Am I just lucky, or is proactively taking these steps
> a little overkill?
Knock wood, I haven't yet had problems following the same update
strategy for years. I've been using this computer for 2 years, and
only repaired permissions once, when something got funky.
Archive and install, which first I read about in the excellent "Take
Control Of Upgrading To Panther," has worked just fine.
Marilyn
|
|
 |  |
|
|
Re: Comments about 10.4.3
On Nov 3, 2005, at 11:55 AM, <tidbits-talk  tidbits.com> wrote:
> I installed the update and when i went back to work in InDesign
> CS2, my HP5550 printer would not print, in spite of reinstalling
> the printer driver, rebooting etc. I could only print when I
> choose another partition which has 10.4.2. There seems to be no
> way to tell Apple via email that there are serious bugs in this up
> date.
The work around is first insert the 10.4 installer DVD.
Then, go to Library, Printers, delete the HP printer folder, then in
the same Library go to Receipts and remove any HP reference. Once
done that, in the 10.4 installer DVD go to the Optional
Installs.mpkg and reinstall just the HP drivers. Restart the
computer, being sure that the printer is turned on so it can be
recognized.
Hector I. Macedo
|
|
 |  |
|
|
Re: Comments about 10.4.3
[This is the last post on this particular branch... -Adam]
On Nov 3, 2005, at 9:58 AM, Matt Neuburg wrote:
> Nevertheless, the insistence that one can't say "high ASCII" at all
> is just
> a knee-jerk canard. During the 80s we said it all the time. And we
> knew (as
> does Chris Page) exactly what it meant: it meant a character byte
> with a
> value between 128 and 255.
But it was nonetheless wrong then (and annoyed me then just as much
as it does now).
Perhaps in my case that partly comes from sharing an office--in
itself that tells you it was long ago--with NCR's representative on
the committee which standardized ASCII in the 8 bit environment.
[ASCII in the 7-bit environment was already standard.]
This was one of the few standards battles IBM lost (until more recent
times): they wanted the high bit (conventionally "bit 8" in those
days) to be a copy of the second bit over ("bit 6"). Everyone else
wanted bit 8 to be zero (at least after some discussion).
IBM took their ball and went home and the world had to suffer with
EBCDIC for a long time. Notice what happens to various assumptions
given IBM's rule: The lower case letters are no longer 32+<value of
the upper case letter>; neither alphabet is a simple contiguous run
of values; the controls are not all together at the "bottom" (except
for DEL). [DEL is 127 because you can't go back and unpunch a line
(character) of paper tape--you can go back and punch out all the
holes, though.
But all this has remarkably little to do with 10.4.3.
--John (who doesn't care that Apple's spell checker doesn't like
unpunch[sic])
|
|
 |  |
|
|
Re: Comments about 10.4.3
On Nov 4, 2005, at 7:53 AM, Marilyn Matty wrote:
> Knock wood, I haven't yet had problems following the same update
> strategy for years. I've been using this computer for 2 years, and
> only repaired permissions once, when something got funky.
My backup strategy is a combination of version control and disk
cloning, and before I do an update I make sure that I've committed
all outstanding changes and that I have cloned the disk recently.
I've never had a problem, and I've only ever run "Repair Permissions"
to see what it does.
Charlton
--
Charlton Wilbur
cwilbur  chromatico.net
|
|
 |  |
|
|
10.4.3 and isync
Has anyone else found that isync has suddenly popped up saying that
there are 1200 odd changes to the address book from dotmac, and when
you press ok (which was a big mistake and I should have known
better), each line in your address book is given its own address
label (i.e. instead of having 1 address connect to an entry, you've
got 5 addresses, one for each field, which means you can't do mail
merge from the address book any more amongst other things).
its going to take a while to sort that one out.
Paul
|
|
|
TidBITS TidBITS TidBITS Talk Comments about 10.4.3
|
|