Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.misc > #74586 > unrolled thread

Re: Favorite Font

Back to article view | Back to comp.os.linux.misc

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: Favorite Font Ethan Carter <ec1828@somewhere.edu> - 2025-09-19 19:16 -0300
    Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-20 00:33 +0000
      Re: Favorite Font Chris Ahlstrom <OFeem1987@teleworm.us> - 2025-09-20 08:58 -0400
        Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-20 16:08 +0100
        Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-20 16:10 +0100
        Re: Favorite Font Rich <rich@example.invalid> - 2025-09-20 16:25 +0000
        Re: Favorite Font Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-20 17:46 +0000
          Re: Favorite Font John Ames <commodorejohn@gmail.com> - 2025-09-22 09:09 -0700
            Re: Favorite Font Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-22 17:22 +0000
              Re: Favorite Font rbowman <bowman@montana.com> - 2025-09-22 19:39 +0000
                Re: Favorite Font Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-23 00:07 +0000
                  Re: Favorite Font "Carlos E.R." <robin_listas@es.invalid> - 2025-09-23 10:00 +0200
                    Re: Favorite Font rbowman <bowman@montana.com> - 2025-09-23 18:45 +0000
                      Re: Favorite Font "Carlos E.R." <robin_listas@es.invalid> - 2025-09-23 22:10 +0200
                Re: Favorite Font Robert Riches <spamtrap42@jacob21819.net> - 2025-09-23 03:36 +0000
                  Re: Favorite Font rbowman <bowman@montana.com> - 2025-09-23 18:42 +0000
            Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-23 12:31 +0100
    Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-20 15:13 +0000
      Re: Favorite Font Farley Flud <ff@linux.rocks> - 2025-09-20 15:51 +0000
        Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-20 18:02 +0000
          Re: Favorite Font Farley Flud <ff@linux.rocks> - 2025-09-20 18:42 +0000
            Re: Favorite Font Jan Panteltje <alien@comet.invalid> - 2025-09-21 08:21 +0000
              Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-21 10:34 +0100
                Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-21 22:53 +0000
                  Re: Favorite Font Jan Panteltje <alien@comet.invalid> - 2025-09-22 07:42 +0000
                    Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-22 22:53 +0000
                      Re: Favorite Font Jan Panteltje <alien@comet.invalid> - 2025-09-23 07:35 +0000
              Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-21 15:26 +0000
                Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-21 18:57 +0100
                  Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-21 22:53 +0000
              Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-21 22:52 +0000
                Re: Favorite Font Jan Panteltje <alien@comet.invalid> - 2025-09-22 07:32 +0000
                  Re: Favorite Font Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-22 17:22 +0000
                    Re: Favorite Font Jan Panteltje <alien@comet.invalid> - 2025-09-22 18:36 +0000
            Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-21 14:59 +0000
              Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-21 16:15 +0100
              Re: Favorite Font Farley Flud <ff@linux.rocks> - 2025-09-21 20:19 +0000
                Re: Favorite Font pothead <pothead@snakebite.com> - 2025-09-23 01:16 +0000
                Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-27 14:44 +0000
                  Re: Favorite Font Farley Flud <fflud@gnu.rocks> - 2025-09-27 15:22 +0000
                    Re: Favorite Font Nuno Silva <nunojsilva@invalid.invalid> - 2025-09-28 11:15 +0100
                      Re: Favorite Font Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2025-09-28 14:03 -0700
                        Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-28 22:51 +0000
                          Re: Favorite Font candycanearter07 <candycanearter07@candycanearter07.nomail.afraid> - 2025-09-30 18:30 +0000
                            Re: Favorite Font Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2025-09-30 11:45 -0700
                              Re: Favorite Font rbowman <bowman@montana.com> - 2025-09-30 20:04 +0000
                            Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-30 22:46 +0000
                        Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-10-04 09:31 +0000
                          Re: Favorite Font Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2025-10-04 10:08 -0700
                            Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-04 21:10 +0000
                      Re: Favorite Font John Ames <commodorejohn@gmail.com> - 2025-09-29 09:11 -0700
                        Re: Favorite Font Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-29 16:45 +0000
                          Date formatting. "Carlos E.R." <robin_listas@es.invalid> - 2025-09-29 19:07 +0200
                            Re: Date formatting. Pancho <Pancho.Jones@protonmail.com> - 2025-09-29 18:30 +0100
                              Re: Date formatting. Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-29 19:40 +0000
                                Re: Date formatting. "Carlos E.R." <robin_listas@es.invalid> - 2025-09-29 21:52 +0200
                              Re: Date formatting. "Carlos E.R." <robin_listas@es.invalid> - 2025-09-29 21:45 +0200
                              Re: Date formatting. rbowman <bowman@montana.com> - 2025-09-30 02:08 +0000
                            Re: Date formatting. c186282 <c186282@nnada.net> - 2025-09-29 23:15 -0400
                          Re: Favorite Font rbowman <bowman@montana.com> - 2025-09-29 18:55 +0000
      Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-20 16:59 +0100
        Re: Favorite Font Richard Kettlewell <invalid@invalid.invalid> - 2025-09-20 18:48 +0100
          Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-20 18:39 +0000
          Re: Favorite Font not@telling.you.invalid (Computer Nerd Kev) - 2025-09-21 09:27 +1000
            Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-21 09:45 +0100
        Re: Favorite Font Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-20 18:18 +0000
          Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-21 09:34 +0100
            Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-21 23:02 +0000
          Re: Favorite Font chrisv <chrisv@nospam.invalid> - 2025-09-21 09:02 -0500
        Re: Favorite Font Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-21 01:39 +0000
        Re: Favorite Font vallor <vallor@cultnix.org> - 2025-09-21 07:20 +0000
          Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-21 10:14 +0100
            Re: Favorite Font vallor <vallor@cultnix.org> - 2025-09-21 22:31 +0000
              Re: Favorite Font The Natural Philosopher <tnp@invalid.invalid> - 2025-09-22 13:34 +0100

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

#75342

On 2025-09-27, Farley Flud wrote:

> On 27 Sep 2025 14:44:29 GMT, Stéphane CARPENTIER wrote:
>
>> [En-tête "Followup-To:" positionné à comp.os.linux.advocacy.]
>> Le 21-09-2025, 
>>
>
> What the fuck is this "21-09-2025?"
>
> Dates are supposed to be specified as year-month-day, which
> in this case would be "2025-09-25."
>
> <https://en.wikipedia.org/wiki/ISO_8601>
>
> Your software is non-conformant with modern standards as is
> your brain.

Some countries use or have used different standards for date display,
and some software will use localization settings to decide what to
output.

DD-MM-YYYY and D.M.YYYY are sometimes seen and used, for example.

Honestly, I'd not call DD-MM-YYYY "what the fuck", it at least is in
order and this one is not even ambiguous, you can tell which is the
month and which is the year. 01-09-2025 could perhaps raise that
question, although I wonder if it's common for USA date order to use
dashes.

-- 
Nuno Silva
21-09-25, now *that* would be a headache.
And 01-09-05 would be even more painful.

[toc] | [prev] | [next] | [standalone]

#75371

On 9/28/25 03:15, Nuno Silva wrote:
> On 2025-09-27, Farley Flud wrote:
> 
>> On 27 Sep 2025 14:44:29 GMT, Stéphane CARPENTIER wrote:
>>
>>> [En-tête "Followup-To:" positionné à comp.os.linux.advocacy.]
>>> Le 21-09-2025,
>>>
>>
>> What the fuck is this "21-09-2025?"
>>
>> Dates are supposed to be specified as year-month-day, which
>> in this case would be "2025-09-25."
>>
>> <https://en.wikipedia.org/wiki/ISO_8601>
>>
>> Your software is non-conformant with modern standards as is
>> your brain.
> 
> Some countries use or have used different standards for date display,
> and some software will use localization settings to decide what to
> output.
> 
> DD-MM-YYYY and D.M.YYYY are sometimes seen and used, for example.
> 
> Honestly, I'd not call DD-MM-YYYY "what the fuck", it at least is in
> order and this one is not even ambiguous, you can tell which is the
> month and which is the year. 01-09-2025 could perhaps raise that
> question, although I wonder if it's common for USA date order to use
> dashes.
> 

	I think it is backward as I use 2025-09-28 for today and on my files
to sort them I use a simllar if not identical system of numbering for
example when I pull the last week from my daily journal I will file it
as 2025-09.20-09.26 to indicate the period covered.
	It is common to use "/" as a separator but I save that for
directories.

	bliss

[toc] | [prev] | [next] | [standalone]

#75372

On Sun, 28 Sep 2025 14:03:00 -0700, Bobbie Sellers wrote:

> It is common to use "/" as a separator but I save that for
> directories.

You could use “∕” instead.

[toc] | [prev] | [next] | [standalone]

#75479

Lawrence D’Oliveiro <ldo@nz.invalid> wrote at 22:51 this Sunday (GMT):
> On Sun, 28 Sep 2025 14:03:00 -0700, Bobbie Sellers wrote:
>
>> It is common to use "/" as a separator but I save that for
>> directories.
>
> You could use “∕” instead.


I feel like that would cause a lot of confusion, especially in a
terminal env...
-- 
user <candycane> is generated from /dev/urandom

[toc] | [prev] | [next] | [standalone]

#75481

On 9/30/25 11:30, candycanearter07 wrote:
> Lawrence D’Oliveiro <ldo@nz.invalid> wrote at 22:51 this Sunday (GMT):
>> On Sun, 28 Sep 2025 14:03:00 -0700, Bobbie Sellers wrote:
>>
>>> It is common to use "/" as a separator but I save that for
>>> directories.
>>
>> You could use “∕” instead.
> 
> 
> I feel like that would cause a lot of confusion, especially in a
> terminal env...

	I can  only agree since the idea is stupid since it would increase the
number of strokes to add "" to every separator.
	The US position vis dates and the proper way to write them is
bass-ackwards in my opinion as is the 12 hour clock in caparison
to the 24 hour clock.

	Yes I know very well I am out of step with the greater(?) society
and on many other topics as well but my computer so I use whatever
I want to use and translate back to 12 hours when I involve non-computer
people.

	bliss - time and dating deviant

[toc] | [prev] | [next] | [standalone]

#75483

On Tue, 30 Sep 2025 11:45:06 -0700, Bobbie Sellers wrote:

> 	Yes I know very well I am out of step with the greater(?) society
> and on many other topics as well but my computer so I use whatever I
> want to use and translate back to 12 hours when I involve non-computer
> people.

At one time I had a digital watch rather than an analog or at least an 
electronic watch with a analog display. If someone asked me the time I 
would glance at the watch and say '17:42'. After getting the confused look 
I would translate. 

At the moment the computer says it's Sep 30 14:01. Work was always a 24 
hour clock so it seems normal.

[toc] | [prev] | [next] | [standalone]

#75502

On Tue, 30 Sep 2025 18:30:03 -0000 (UTC), candycanearter07 wrote:

> Lawrence D’Oliveiro <ldo@nz.invalid> wrote at 22:51 this Sunday (GMT):
>>
>> On Sun, 28 Sep 2025 14:03:00 -0700, Bobbie Sellers wrote:
>>
>>> It is common to use "/" as a separator but I save that for
>>> directories.
>>
>> You could use “∕” instead.
> 
> I feel like that would cause a lot of confusion, especially in a
> terminal env...

Confusion for whom? The computer would not be confused. And humans are
good at interpreting things in context, in situations which could
cause confusion for computers.

[toc] | [prev] | [next] | [standalone]

#75688

Le 28-09-2025, Bobbie Sellers <bliss-sf4ever@dslextreme.com> a écrit :
> 	I think it is backward as I use 2025-09-28 for today and on my files
> to sort them I use a simllar if not identical system of numbering for
> example when I pull the last week from my daily journal I will file it
> as 2025-09.20-09.26 to indicate the period covered.
> 	It is common to use "/" as a separator but I save that for
> directories.

Printing a date for people to read it is not the same thing as printing
a date for naming files which is not the same as printing dates for
exchanges between systems.

-- 
Si vous avez du temps à perdre :
https://scarpet42.gitlab.io

[toc] | [prev] | [next] | [standalone]

#75703

On 10/4/25 02:31, Stéphane CARPENTIER wrote:
> Le 28-09-2025, Bobbie Sellers <bliss-sf4ever@dslextreme.com> a écrit :
>> 	I think it is backward as I use 2025-09-28 for today and on my files
>> to sort them I use a simllar if not identical system of numbering for
>> example when I pull the last week from my daily journal I will file it
>> as 2025-09.20-09.26 to indicate the period covered.
>> 	It is common to use "/" as a separator but I save that for
>> directories.
> 
> Printing a date for people to read it is not the same thing as printing
> a date for naming files which is not the same as printing dates for
> exchanges between systems.
> 
	Indeed is is not but people can learn to use the computer version
as easlly as the computer can be set to render such dates.  But the
computer date version can easily be used to fnd documents of a
specific date much more easily than sorting by named months, which
names differ between cultures, as do named days of the week..
	The numbering of the years of the Common Era and the months
and days of those years does not differ unless you move to another
calendar than the CE Solar Year.
	
	bliss

[toc] | [prev] | [next] | [standalone]

#75715

On Sat, 4 Oct 2025 10:08:56 -0700, Bobbie Sellers wrote:

> On 10/4/25 02:31, Stéphane CARPENTIER wrote:
>>
>> Printing a date for people to read it is not the same thing as printing
>> a date for naming files which is not the same as printing dates for
>> exchanges between systems.
>> 
> Indeed is is not but people can learn to use the computer version
> as easlly as the computer can be set to render such dates.

Different regions use different formats, and in an international world, 
that can, and does, lead to confusion. That’s why we have an international 
standard, ISO 8601, that we can agree on for everybody to use.

[toc] | [prev] | [next] | [standalone]

#75406

On Sun, 28 Sep 2025 11:15:32 +0100
Nuno Silva <nunojsilva@invalid.invalid> wrote:

> although I wonder if it's common for USA date order to use
> dashes.

US date formats normally use a forward slash and MM/DD/YYYY is the more
common ordering (which reflects spoken US English - it's more normal to
say "June 5th" than "the fifth of June.") Dashes get used mostly in a
computing context where software may be prone to barf on encountering
the slash (*nix treating it as a directory separator, Win/DOS as a flag
indicator.) Computing also does see YYYY/MM/DD used instead, since that
ASCII-sorts automagically, but it's not common in any other context.

[toc] | [prev] | [next] | [standalone]

#75408

On 2025-09-29, John Ames <commodorejohn@gmail.com> wrote:

> On Sun, 28 Sep 2025 11:15:32 +0100
> Nuno Silva <nunojsilva@invalid.invalid> wrote:
>
>> although I wonder if it's common for USA date order to use
>> dashes.
>
> US date formats normally use a forward slash and MM/DD/YYYY is the more
> common ordering (which reflects spoken US English - it's more normal to
> say "June 5th" than "the fifth of June.") Dashes get used mostly in a
> computing context where software may be prone to barf on encountering
> the slash (*nix treating it as a directory separator, Win/DOS as a flag
> indicator.) Computing also does see YYYY/MM/DD used instead, since that
> ASCII-sorts automagically, but it's not common in any other context.

Not for lack of trying.  ISO 8601 is being gently but persistently pushed
in a lot of places, and I'm proud to say that I'm doing my bit; all report
programs that I write format dates as yyyy-mm-dd, and so far no user has
found it sufficiently objectionable to demand a change.

I can still remember the day in 1967 when I realized that year-month-day
is the only reasonable date format.  That's probably because I was already
seeing too much ambiguity with things like 4/5/1967 - is it April 5 or
May 4?  The answer depends on whether you're in the US or the UK, and
if you're in Canada all bets are off.  Better to toss the whole thing
out and replace it with an unambiguous standard, i.e. ISO 8601.

BTW year-month-day also EBCDIC-sorts, as well as ASCII-sorts. :-)

-- 
/~\  Charlie Gibbs                  |  Growth for the sake of
\ /  <cgibbs@kltpzyxm.invalid>      |  growth is the ideology
 X   I'm really at ac.dekanfrus     |  of the cancer cell.
/ \  if you read it the right way.  |    -- Edward Abbey

[toc] | [prev] | [next] | [standalone]

#75410 — Date formatting.

On 2025-09-29 18:45, Charlie Gibbs wrote:
> On 2025-09-29, John Ames <commodorejohn@gmail.com> wrote:
> 
>> On Sun, 28 Sep 2025 11:15:32 +0100
>> Nuno Silva <nunojsilva@invalid.invalid> wrote:
>>
>>> although I wonder if it's common for USA date order to use
>>> dashes.
>>
>> US date formats normally use a forward slash and MM/DD/YYYY is the more
>> common ordering (which reflects spoken US English - it's more normal to
>> say "June 5th" than "the fifth of June.") Dashes get used mostly in a
>> computing context where software may be prone to barf on encountering
>> the slash (*nix treating it as a directory separator, Win/DOS as a flag
>> indicator.) Computing also does see YYYY/MM/DD used instead, since that
>> ASCII-sorts automagically, but it's not common in any other context.
> 
> Not for lack of trying.  ISO 8601 is being gently but persistently pushed
> in a lot of places, and I'm proud to say that I'm doing my bit; all report
> programs that I write format dates as yyyy-mm-dd, and so far no user has
> found it sufficiently objectionable to demand a change.
> 
> I can still remember the day in 1967 when I realized that year-month-day
> is the only reasonable date format.  That's probably because I was already
> seeing too much ambiguity with things like 4/5/1967 - is it April 5 or
> May 4?  The answer depends on whether you're in the US or the UK, and
> if you're in Canada all bets are off.  Better to toss the whole thing
> out and replace it with an unambiguous standard, i.e. ISO 8601.
> 
> BTW year-month-day also EBCDIC-sorts, as well as ASCII-sorts. :-)

As soon as I learned of this iso standard I switched to it. But it is 
certainly not mandatory as someone claimed angrily in this thread. No 
manner of mad jumping will make banks switch from the standard they use, 
which is whatever national standard is in the nation they are, which 
here is DD/MM/YYYY.

cer@Telcontar:~> date
2025-09-29T19:04:05 CEST
cer@Telcontar:~>

I don't like the date-time separator, though.



-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]

#75412 — Re: Date formatting.

On 9/29/25 18:07, Carlos E.R. wrote:
> On 2025-09-29 18:45, Charlie Gibbs wrote:
>> On 2025-09-29, John Ames <commodorejohn@gmail.com> wrote:
>>
>>> On Sun, 28 Sep 2025 11:15:32 +0100
>>> Nuno Silva <nunojsilva@invalid.invalid> wrote:
>>>
>>>> although I wonder if it's common for USA date order to use
>>>> dashes.
>>>
>>> US date formats normally use a forward slash and MM/DD/YYYY is the more
>>> common ordering (which reflects spoken US English - it's more normal to
>>> say "June 5th" than "the fifth of June.") Dashes get used mostly in a
>>> computing context where software may be prone to barf on encountering
>>> the slash (*nix treating it as a directory separator, Win/DOS as a flag
>>> indicator.) Computing also does see YYYY/MM/DD used instead, since that
>>> ASCII-sorts automagically, but it's not common in any other context.
>>
>> Not for lack of trying.  ISO 8601 is being gently but persistently pushed
>> in a lot of places, and I'm proud to say that I'm doing my bit; all 
>> report
>> programs that I write format dates as yyyy-mm-dd, and so far no user has
>> found it sufficiently objectionable to demand a change.
>>
>> I can still remember the day in 1967 when I realized that year-month-day
>> is the only reasonable date format.  That's probably because I was 
>> already
>> seeing too much ambiguity with things like 4/5/1967 - is it April 5 or
>> May 4?  The answer depends on whether you're in the US or the UK, and
>> if you're in Canada all bets are off.  Better to toss the whole thing
>> out and replace it with an unambiguous standard, i.e. ISO 8601.
>>
>> BTW year-month-day also EBCDIC-sorts, as well as ASCII-sorts. :-)
> 
> As soon as I learned of this iso standard I switched to it. But it is 
> certainly not mandatory as someone claimed angrily in this thread. No 
> manner of mad jumping will make banks switch from the standard they use, 
> which is whatever national standard is in the nation they are, which 
> here is DD/MM/YYYY.
> 
> cer@Telcontar:~> date
> 2025-09-29T19:04:05 CEST
> cer@Telcontar:~>
> 
> I don't like the date-time separator, though.
> 
> 

I like YYYYMMDD too, but the real epiphany moment for me was when I 
realised that all dates should be handled internally UTC, Unix time, 
and that output format, display, is just a user preference.

It is so obvious, and yet I worked on so many systems that messed up 
date conversions/different time zones.

[toc] | [prev] | [next] | [standalone]

#75425 — Re: Date formatting.

On Mon, 29 Sep 2025 18:30:06 +0100, Pancho wrote:

> I like YYYYMMDD too, but the real epiphany moment for me was when I
> realised that all dates should be handled internally UTC, Unix time,
> and that output format, display, is just a user preference.

As far as I can tell, it was Unix that pioneered the idea of maintaining 
system time as UTC (actually GMT in the early days), and converting back 
and forth as necessary for human interpretation. Other systems continued 
to resist the idea for decades afterwards. Windows still doesn’t do it 
today. macOS --- who can tell?

> It is so obvious, and yet I worked on so many systems that messed up
> date conversions/different time zones.

All too common, on systems like Microsoft Windows that maintain system 
time as local time, to forget that they have done the daylight-saving 
switch once, and do it a second time. That kind of bug can’t happen when 
system time is maintained as UTC.

There is an article on Bitsavers which mentions that IBM mainframes needed 
to be rebooted to switch daylight saving on and off.

[toc] | [prev] | [next] | [standalone]

#75430 — Re: Date formatting.

On 2025-09-29 21:40, Lawrence D’Oliveiro wrote:
> On Mon, 29 Sep 2025 18:30:06 +0100, Pancho wrote:
> 
>> I like YYYYMMDD too, but the real epiphany moment for me was when I
>> realised that all dates should be handled internally UTC, Unix time,
>> and that output format, display, is just a user preference.
> 
> As far as I can tell, it was Unix that pioneered the idea of maintaining
> system time as UTC (actually GMT in the early days), and converting back
> and forth as necessary for human interpretation. Other systems continued
> to resist the idea for decades afterwards. Windows still doesn’t do it
> today. macOS --- who can tell?

Windows has a registry boolean option to store the CMOS time as UTC. I 
have no idea if this also means internally.

> 
>> It is so obvious, and yet I worked on so many systems that messed up
>> date conversions/different time zones.
> 
> All too common, on systems like Microsoft Windows that maintain system
> time as local time, to forget that they have done the daylight-saving
> switch once, and do it a second time. That kind of bug can’t happen when
> system time is maintained as UTC.
> 
> There is an article on Bitsavers which mentions that IBM mainframes needed
> to be rebooted to switch daylight saving on and off.


-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]

#75427 — Re: Date formatting.

On 2025-09-29 19:30, Pancho wrote:
> On 9/29/25 18:07, Carlos E.R. wrote:
>> On 2025-09-29 18:45, Charlie Gibbs wrote:
>>> On 2025-09-29, John Ames <commodorejohn@gmail.com> wrote:
>>>
>>>> On Sun, 28 Sep 2025 11:15:32 +0100
>>>> Nuno Silva <nunojsilva@invalid.invalid> wrote:
>>>>
>>>>> although I wonder if it's common for USA date order to use
>>>>> dashes.
>>>>
>>>> US date formats normally use a forward slash and MM/DD/YYYY is the more
>>>> common ordering (which reflects spoken US English - it's more normal to
>>>> say "June 5th" than "the fifth of June.") Dashes get used mostly in a
>>>> computing context where software may be prone to barf on encountering
>>>> the slash (*nix treating it as a directory separator, Win/DOS as a flag
>>>> indicator.) Computing also does see YYYY/MM/DD used instead, since that
>>>> ASCII-sorts automagically, but it's not common in any other context.
>>>
>>> Not for lack of trying.  ISO 8601 is being gently but persistently 
>>> pushed
>>> in a lot of places, and I'm proud to say that I'm doing my bit; all 
>>> report
>>> programs that I write format dates as yyyy-mm-dd, and so far no user has
>>> found it sufficiently objectionable to demand a change.
>>>
>>> I can still remember the day in 1967 when I realized that year-month-day
>>> is the only reasonable date format.  That's probably because I was 
>>> already
>>> seeing too much ambiguity with things like 4/5/1967 - is it April 5 or
>>> May 4?  The answer depends on whether you're in the US or the UK, and
>>> if you're in Canada all bets are off.  Better to toss the whole thing
>>> out and replace it with an unambiguous standard, i.e. ISO 8601.
>>>
>>> BTW year-month-day also EBCDIC-sorts, as well as ASCII-sorts. :-)
>>
>> As soon as I learned of this iso standard I switched to it. But it is 
>> certainly not mandatory as someone claimed angrily in this thread. No 
>> manner of mad jumping will make banks switch from the standard they 
>> use, which is whatever national standard is in the nation they are, 
>> which here is DD/MM/YYYY.
>>
>> cer@Telcontar:~> date
>> 2025-09-29T19:04:05 CEST
>> cer@Telcontar:~>
>>
>> I don't like the date-time separator, though.
>>
>>
> 
> I like YYYYMMDD too, but the real epiphany moment for me was when I 
> realised that all dates should be handled internally UTC, Unix time, and 
> that output format, display, is just a user preference.
> 
> It is so obvious, and yet I worked on so many systems that messed up 
> date conversions/different time zones.

Oh, yes.

MsDOS did not have initially a way to cope with timezones or locales. 
They had localtime only, and you had to type it at boot. And they had 
different charsets to adapt to several... er... european? languages.

With that inheritance, trying to adapt to multizone and multilanguage, 
it is terrible situation.


-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]

#75440 — Re: Date formatting.

On Mon, 29 Sep 2025 18:30:06 +0100, Pancho wrote:

> I like YYYYMMDD too, but the real epiphany moment for me was when I
> realised that all dates should be handled internally UTC, Unix time,
> and that output format, display, is just a user preference.

That would sometimes throw our support people. When you harvest logs from 
a site the times match their wall clock time. If you grab raw data where 
the time was stored as a time_t value, reporting tools will use 
localtime() to when formatting the data, leading to 'Why are the times off 
by two hours?'

[toc] | [prev] | [next] | [standalone]

#75445 — Re: Date formatting.

On 9/29/25 13:07, Carlos E.R. wrote:
> On 2025-09-29 18:45, Charlie Gibbs wrote:
>> On 2025-09-29, John Ames <commodorejohn@gmail.com> wrote:
>>
>>> On Sun, 28 Sep 2025 11:15:32 +0100
>>> Nuno Silva <nunojsilva@invalid.invalid> wrote:
>>>
>>>> although I wonder if it's common for USA date order to use
>>>> dashes.
>>>
>>> US date formats normally use a forward slash and MM/DD/YYYY is the more
>>> common ordering (which reflects spoken US English - it's more normal to
>>> say "June 5th" than "the fifth of June.") Dashes get used mostly in a
>>> computing context where software may be prone to barf on encountering
>>> the slash (*nix treating it as a directory separator, Win/DOS as a flag
>>> indicator.) Computing also does see YYYY/MM/DD used instead, since that
>>> ASCII-sorts automagically, but it's not common in any other context.
>>
>> Not for lack of trying.  ISO 8601 is being gently but persistently pushed
>> in a lot of places, and I'm proud to say that I'm doing my bit; all 
>> report
>> programs that I write format dates as yyyy-mm-dd, and so far no user has
>> found it sufficiently objectionable to demand a change.
>>
>> I can still remember the day in 1967 when I realized that year-month-day
>> is the only reasonable date format.  That's probably because I was 
>> already
>> seeing too much ambiguity with things like 4/5/1967 - is it April 5 or
>> May 4?  The answer depends on whether you're in the US or the UK, and
>> if you're in Canada all bets are off.  Better to toss the whole thing
>> out and replace it with an unambiguous standard, i.e. ISO 8601.
>>
>> BTW year-month-day also EBCDIC-sorts, as well as ASCII-sorts. :-)
> 
> As soon as I learned of this iso standard I switched to it. But it is 
> certainly not mandatory as someone claimed angrily in this thread. No 
> manner of mad jumping will make banks switch from the standard they use, 
> which is whatever national standard is in the nation they are, which 
> here is DD/MM/YYYY.


   Nonsense ... MM/DD/YYYY is the CORRECT way !

[toc] | [prev] | [next] | [standalone]

#75421

On Mon, 29 Sep 2025 16:45:25 GMT, Charlie Gibbs wrote:

>  Better to toss the whole thing
> out and replace it with an unambiguous standard, i.e. ISO 8601.

I've found it annoying that DB2 uses '1965-07-27-00.00.00'rather than   
'1965-07-27 00:00:00'.  Another DB2 quirk is it will accept 24:00:00. We 
used '9999-12-31-24.00.00' to indicate an unknown time in DB2 and had to 
throw in a bunch of #ifdef and conditionals.

[toc] | [prev] | [next] | [standalone]

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

Back to top | Article view | comp.os.linux.misc

csiph-web