[PATCH] Clarify "Press any key to continue..." message

[PATCH] Clarify "Press any key to continue..." message

[Colin Watson]

When grub_wait_after_message says "Press any key to continue...", it
really means that it will continue anyway after a short delay, but that
you can press a key to skip the delay.  Unfortunately, the delay is just
long enough that in practice a number of users have time to see it,
press a key, and then report a bug saying that their system won't boot
without manual intervention, even though it would have booted just fine
if they'd left it alone.  Rephrase this message to make it clearer
what's really happening.

* grub-core/normal/menu.c (grub_wait_after_message): Rephrase
message to make it clear that GRUB will continue even if the user
does not press a key.
---
 ChangeLog               | 6 ++++++
 grub-core/normal/menu.c | 4 +++-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/ChangeLog b/ChangeLog
index 6eca73e..53c2aff 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,9 @@
+2014-04-03  Colin Watson  <cjwatson@ubuntu.com>
+
+	* grub-core/normal/menu.c (grub_wait_after_message): Rephrase
+	message to make it clear that GRUB will continue even if the user
+	does not press a key.
+
 2014-03-31  Thomas Falcon <tlfalcon@linux.vnet.ibm.com>
 
 	btrfs: fix get_root key comparison failures due to endianness
diff --git a/grub-core/normal/menu.c b/grub-core/normal/menu.c
index b47991a..a6fa93c 100644
--- a/grub-core/normal/menu.c
+++ b/grub-core/normal/menu.c
@@ -63,7 +63,9 @@ grub_wait_after_message (void)
 {
   grub_uint64_t endtime;
   grub_xputs ("\n");
-  grub_printf_ (N_("Press any key to continue..."));
+  grub_printf_ (N_(
+    "Waiting a moment so that you can read previous messages.\n"
+    "Press any key to skip this delay..."));
   grub_refresh ();
 
   endtime = grub_get_time_ms () + 10000;
-- 
1.9.1

Re: [PATCH] Clarify "Press any key to continue..." message

[Prarit Bhargava]

On 04/03/2014 10:23 AM, Colin Watson wrote:
> When grub_wait_after_message says "Press any key to continue...", it
> really means that it will continue anyway after a short delay, but that
> you can press a key to skip the delay.  Unfortunately, the delay is just
> long enough that in practice a number of users have time to see it,
> press a key, and then report a bug saying that their system won't boot
> without manual intervention, even though it would have booted just fine
> if they'd left it alone.  Rephrase this message to make it clearer
> what's really happening.

I have mixed feelings about changing this message since it is such a well known
message.  OTOH, I've never really liked it myself.

Maybe a better idea is to do a timeout message

Press any key for menu (boot will continue in 5 seconds) ...
Press any key for menu (boot will continue in 4 seconds) ...
...
etc.

?

> 
> * grub-core/normal/menu.c (grub_wait_after_message): Rephrase
> message to make it clear that GRUB will continue even if the user
> does not press a key.
> ---
>  ChangeLog               | 6 ++++++
>  grub-core/normal/menu.c | 4 +++-
>  2 files changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/ChangeLog b/ChangeLog
> index 6eca73e..53c2aff 100644
> --- a/ChangeLog
> +++ b/ChangeLog
> @@ -1,3 +1,9 @@
> +2014-04-03  Colin Watson  <cjwatson@ubuntu.com>
> +
> +	* grub-core/normal/menu.c (grub_wait_after_message): Rephrase
> +	message to make it clear that GRUB will continue even if the user
> +	does not press a key.
> +
>  2014-03-31  Thomas Falcon <tlfalcon@linux.vnet.ibm.com>
>  
>  	btrfs: fix get_root key comparison failures due to endianness
> diff --git a/grub-core/normal/menu.c b/grub-core/normal/menu.c
> index b47991a..a6fa93c 100644
> --- a/grub-core/normal/menu.c
> +++ b/grub-core/normal/menu.c
> @@ -63,7 +63,9 @@ grub_wait_after_message (void)
>  {
>    grub_uint64_t endtime;
>    grub_xputs ("\n");
> -  grub_printf_ (N_("Press any key to continue..."));
> +  grub_printf_ (N_(
> +    "Waiting a moment so that you can read previous messages.\n"
> +    "Press any key to skip this delay..."));
>    grub_refresh ();
>  
>    endtime = grub_get_time_ms () + 10000;
> 

Re: [PATCH] Clarify "Press any key to continue..." message

[Colin Watson]

On Thu, Apr 03, 2014 at 12:49:50PM -0400, Prarit Bhargava wrote:
> On 04/03/2014 10:23 AM, Colin Watson wrote:
> > When grub_wait_after_message says "Press any key to continue...", it
> > really means that it will continue anyway after a short delay, but that
> > you can press a key to skip the delay.  Unfortunately, the delay is just
> > long enough that in practice a number of users have time to see it,
> > press a key, and then report a bug saying that their system won't boot
> > without manual intervention, even though it would have booted just fine
> > if they'd left it alone.  Rephrase this message to make it clearer
> > what's really happening.
> 
> I have mixed feelings about changing this message since it is such a
> well known message.  OTOH, I've never really liked it myself.

Surely nobody can really be depending on it; I don't see a problem with
changing that kind of text, in and of itself.

> Maybe a better idea is to do a timeout message
> 
> Press any key for menu (boot will continue in 5 seconds) ...
> Press any key for menu (boot will continue in 4 seconds) ...
> ...
> etc.
> 
> ?

It's not necessarily "for menu" - you get this for errors after a menu
item has been selected too, and then it's just a delay before booting.

Seems like a lot of effort.  The actual delay is 2.5 seconds so a
countdown would look jumpy at the end ...

-- 
Colin Watson                                       [cjwatson@ubuntu.com]

Re: [PATCH] Clarify "Press any key to continue..." message

[Prarit Bhargava]

On 04/03/2014 12:52 PM, Colin Watson wrote:
> On Thu, Apr 03, 2014 at 12:49:50PM -0400, Prarit Bhargava wrote:
>> On 04/03/2014 10:23 AM, Colin Watson wrote:
>>> When grub_wait_after_message says "Press any key to continue...", it
>>> really means that it will continue anyway after a short delay, but that
>>> you can press a key to skip the delay.  Unfortunately, the delay is just
>>> long enough that in practice a number of users have time to see it,
>>> press a key, and then report a bug saying that their system won't boot
>>> without manual intervention, even though it would have booted just fine
>>> if they'd left it alone.  Rephrase this message to make it clearer
>>> what's really happening.
>>
>> I have mixed feelings about changing this message since it is such a
>> well known message.  OTOH, I've never really liked it myself.
> 
> Surely nobody can really be depending on it; I don't see a problem with
> changing that kind of text, in and of itself.

Yeah ... but it is one of those things like the NMI dazed and confused message.
 Everyone knows about it, everyone agrees it could be better, but no one wants
to change it because so much open documentation references it.

Try googling linux + "Press any key to continue" and you'll see what I mean.

But again, I've never really cared for the message so I'm favour of changing it ;)

> 
>> Maybe a better idea is to do a timeout message
>>
>> Press any key for menu (boot will continue in 5 seconds) ...
>> Press any key for menu (boot will continue in 4 seconds) ...
>> ...
>> etc.
>>
>> ?
> 
> It's not necessarily "for menu" - you get this for errors after a menu
> item has been selected too, and then it's just a delay before booting.

Hmm ... really?  After a menu item has been selected?  I've never done that before.
> 
> Seems like a lot of effort.  The actual delay is 2.5 seconds so a
> countdown would look jumpy at the end ...
> 

Just do a countdown and change the delay to an even 3 seconds.  Who is really
going to care about a 1/2 second?

P.

Re: [PATCH] Clarify "Press any key to continue..." message

[Colin Watson]

On Thu, Apr 03, 2014 at 12:57:22PM -0400, Prarit Bhargava wrote:
> On 04/03/2014 12:52 PM, Colin Watson wrote:
> > On Thu, Apr 03, 2014 at 12:49:50PM -0400, Prarit Bhargava wrote:
> >> I have mixed feelings about changing this message since it is such a
> >> well known message.  OTOH, I've never really liked it myself.
> > 
> > Surely nobody can really be depending on it; I don't see a problem with
> > changing that kind of text, in and of itself.
> 
> Yeah ... but it is one of those things like the NMI dazed and confused message.
>  Everyone knows about it, everyone agrees it could be better, but no one wants
> to change it because so much open documentation references it.
> 
> Try googling linux + "Press any key to continue" and you'll see what I mean.

Of the first two pages of hits, only two relate to GRUB 2; and given
that there's still another instance of "Press any key to continue" in
the code, even those might not be relevant.

Anyway, I really don't think this is at all important given that most of
those hits are not exactly very informative anyway, and we should not
shy away from improving messages just because people have made terrible
attempts to explain what's going on with a previous version of the
message with an open-ended set of causes.

> > It's not necessarily "for menu" - you get this for errors after a menu
> > item has been selected too, and then it's just a delay before booting.
> 
> Hmm ... really?  After a menu item has been selected?  I've never done that before.

Yes; for instance save_env in a menu item when /boot isn't writeable by
GRUB will do it.

> > Seems like a lot of effort.  The actual delay is 2.5 seconds so a
> > countdown would look jumpy at the end ...
> 
> Just do a countdown and change the delay to an even 3 seconds.  Who is really
> going to care about a 1/2 second?

I don't think a countdown is necessary here, so I'm not going to spend
time writing code for it.  But of course you're free to do so ...

-- 
Colin Watson                                       [cjwatson@ubuntu.com]

Re: [PATCH] Clarify "Press any key to continue..." message

[Andrey Borzenkov]

В Thu, 3 Apr 2014 17:52:57 +0100
Colin Watson <cjwatson@ubuntu.com> пишет:

> 
> Seems like a lot of effort.  The actual delay is 2.5 seconds

May be I misunderstand it, but how 10000ms becomes 2.5 seconds?

Re: [PATCH] Clarify "Press any key to continue..." message

[Colin Watson]

On Thu, Apr 03, 2014 at 09:15:45PM +0400, Andrey Borzenkov wrote:
> В Thu, 3 Apr 2014 17:52:57 +0100
> Colin Watson <cjwatson@ubuntu.com> пишет:
> > Seems like a lot of effort.  The actual delay is 2.5 seconds
> 
> May be I misunderstand it, but how 10000ms becomes 2.5 seconds?

Sorry, I was looking at DEFAULT_ENTRY_ERROR_DELAY_MS at the top of that
file and didn't notice that grub_wait_after_message actually uses a
different delay.  (A countdown here still seems to me like code we don't
need, though.)

-- 
Colin Watson                                       [cjwatson@ubuntu.com]

Re: [PATCH] Clarify "Press any key to continue..." message

[Vladimir 'φ-coder/phcoder' Serbinenko]

[-- Attachment #1: Type: text/plain, Size: 1941 bytes --]


Go ahead
On 03.04.2014 16:23, Colin Watson wrote:
> When grub_wait_after_message says "Press any key to continue...", it
> really means that it will continue anyway after a short delay, but that
> you can press a key to skip the delay.  Unfortunately, the delay is just
> long enough that in practice a number of users have time to see it,
> press a key, and then report a bug saying that their system won't boot
> without manual intervention, even though it would have booted just fine
> if they'd left it alone.  Rephrase this message to make it clearer
> what's really happening.
> 
> * grub-core/normal/menu.c (grub_wait_after_message): Rephrase
> message to make it clear that GRUB will continue even if the user
> does not press a key.
> ---
>  ChangeLog               | 6 ++++++
>  grub-core/normal/menu.c | 4 +++-
>  2 files changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/ChangeLog b/ChangeLog
> index 6eca73e..53c2aff 100644
> --- a/ChangeLog
> +++ b/ChangeLog
> @@ -1,3 +1,9 @@
> +2014-04-03  Colin Watson  <cjwatson@ubuntu.com>
> +
> +	* grub-core/normal/menu.c (grub_wait_after_message): Rephrase
> +	message to make it clear that GRUB will continue even if the user
> +	does not press a key.
> +
>  2014-03-31  Thomas Falcon <tlfalcon@linux.vnet.ibm.com>
>  
>  	btrfs: fix get_root key comparison failures due to endianness
> diff --git a/grub-core/normal/menu.c b/grub-core/normal/menu.c
> index b47991a..a6fa93c 100644
> --- a/grub-core/normal/menu.c
> +++ b/grub-core/normal/menu.c
> @@ -63,7 +63,9 @@ grub_wait_after_message (void)
>  {
>    grub_uint64_t endtime;
>    grub_xputs ("\n");
> -  grub_printf_ (N_("Press any key to continue..."));
> +  grub_printf_ (N_(
> +    "Waiting a moment so that you can read previous messages.\n"
> +    "Press any key to skip this delay..."));
>    grub_refresh ();
>  
>    endtime = grub_get_time_ms () + 10000;
> 



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 213 bytes --]