Обсуждение: [Fwd: [Bug 831875] [NEW] "Vacuum Recommended" dialog has no "no" option]

Hi,

AFAICT, the OP's complain is valid. I'm wondering why we have an OK and
a Cancel button. The Cancel button is disabled when force parameter is
set to false. After a quick research, it seems it's always false. Maybe
we should replace the OK/Cancel pair with a Close button.

Any objections to this change?


--
Guillaume
  http://blog.guillaume.lelarge.info
  http://www.dalibo.com
Public bug reported:

I've just had a "Vacuum Recommended" message pop up on my screen, the
only available options are "Vacuum" and "Ok"

Firstly, there are two different UI styles (It should either be "Vacuum
/ Do not vacuum" or "Ok / Cancel"); and secondly, both options seem to
suggest that if I click them then a vacuum will be run. The "Vacuum"
button suggests this slightly stronger, so I would guess that "Ok" means
"Do not vacuum" - but if "Ok" means "do not vacuum" then what does the
"Cancel" button do? It is greyed out, but it is the only button that
seems to be what I want...

ProblemType: Bug
DistroRelease: Ubuntu 11.04
Package: pgadmin3 1.12.2-1
ProcVersionSignature: Ubuntu 3.0.0-4.5-generic 3.0.0-rc6
Uname: Linux 3.0.0-4-generic x86_64
NonfreeKernelModules: nvidia
Architecture: amd64
Date: Tue Aug 23 10:46:34 2011
EcryptfsInUse: Yes
InstallationMedia: Ubuntu 10.10 "Maverick Meerkat" - Release amd64 (20101007)
ProcEnviron:
 LANGUAGE=en_GB:en
 PATH=(custom, user)
 LANG=en_GB.UTF-8
 SHELL=/bin/bash
SourcePackage: pgadmin3
UpgradeStatus: No upgrade log present (probably fresh install)

** Affects: pgadmin3 (Ubuntu)
     Importance: Undecided
         Status: New


** Tags: amd64 apport-bug natty

--
You received this bug notification because you are subscribed to
pgadmin3 in Ubuntu.
https://bugs.launchpad.net/bugs/831875

Title:
  "Vacuum Recommended" dialog has no "no" option

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/pgadmin3/+bug/831875/+subscriptions

On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info> wrote:
> Hi,
>
> AFAICT, the OP's complain is valid. I'm wondering why we have an OK and
> a Cancel button. The Cancel button is disabled when force parameter is
> set to false. After a quick research, it seems it's always false. Maybe
> we should replace the OK/Cancel pair with a Close button.
>
> Any objections to this change?

Yes. We don't have close buttons on dialogs. There's a perfectly good one in the titlebar.

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

On Sat, 2011-09-10 at 18:22 +0100, Dave Page wrote:
> On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info>
> wrote:
> > Hi,
> >
> > AFAICT, the OP's complain is valid. I'm wondering why we have an OK and
> > a Cancel button. The Cancel button is disabled when force parameter is
> > set to false. After a quick research, it seems it's always false. Maybe
> > we should replace the OK/Cancel pair with a Close button.
> >
> > Any objections to this change?
>
> Yes. We don't have close buttons on dialogs. There's a perfectly good one in
> the titlebar.
>

It may make sense to some dialogs, and not to others. On the properties
dialogs, either you have to click OK to accept the changes, or you click
Cancel to close without applying the changes. In the guru hint dialog,
there is no reason (or at least, you didn't give one) to keep the OK and
Cancel buttons. They are disturbing.


--
Guillaume
  http://blog.guillaume.lelarge.info
  http://www.dalibo.com

On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info> wrote:
> On Sat, 2011-09-10 at 18:22 +0100, Dave Page wrote:
>> On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info>
>> wrote:
>> > Hi,
>> >
>> > AFAICT, the OP's complain is valid. I'm wondering why we have an OK and
>> > a Cancel button. The Cancel button is disabled when force parameter is
>> > set to false. After a quick research, it seems it's always false. Maybe
>> > we should replace the OK/Cancel pair with a Close button.
>> >
>> > Any objections to this change?
>>
>> Yes. We don't have close buttons on dialogs. There's a perfectly good one in
>> the titlebar.
>>
>
> It may make sense to some dialogs, and not to others. On the properties
> dialogs, either you have to click OK to accept the changes, or you click
> Cancel to close without applying the changes. In the guru hint dialog,
> there is no reason (or at least, you didn't give one) to keep the OK and
> Cancel buttons. They are disturbing.

Cancel on that dialog means "cancel the operation that triggered the hint so I can correct the problem and try again". It's no different fthan "do you want to delete? OK/Cancel"

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

On Sat, 2011-09-10 at 18:43 +0100, Dave Page wrote:
> On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info>
> wrote:
> > On Sat, 2011-09-10 at 18:22 +0100, Dave Page wrote:
> >> On Saturday, September 10, 2011, Guillaume Lelarge <
> guillaume@lelarge.info>
> >> wrote:
> >> > Hi,
> >> >
> >> > AFAICT, the OP's complain is valid. I'm wondering why we have an OK and
> >> > a Cancel button. The Cancel button is disabled when force parameter is
> >> > set to false. After a quick research, it seems it's always false. Maybe
> >> > we should replace the OK/Cancel pair with a Close button.
> >> >
> >> > Any objections to this change?
> >>
> >> Yes. We don't have close buttons on dialogs. There's a perfectly good one
> in
> >> the titlebar.
> >>
> >
> > It may make sense to some dialogs, and not to others. On the properties
> > dialogs, either you have to click OK to accept the changes, or you click
> > Cancel to close without applying the changes. In the guru hint dialog,
> > there is no reason (or at least, you didn't give one) to keep the OK and
> > Cancel buttons. They are disturbing.
>
> Cancel on that dialog means "cancel the operation that triggered the hint so
> I can correct the problem and try again". It's no different fthan "do you
> want to delete? OK/Cancel"
>

It's really different because on the latter, you have a real question.
On the former, there is no question which makes it difficult to
understand what the OK button will do. What will the OK button do? close
the dialog? VACUUM the table (on the VACUUM hint)? one cannot guess that
without trying. Actually, I expect a user to hit more easily the cancel
button than the OK button. Unfortunately, the Cancel button is always
disabled, so it has no meaning at all.


--
Guillaume
  http://blog.guillaume.lelarge.info
  http://www.dalibo.com

On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info> wrote:
> On Sat, 2011-09-10 at 18:43 +0100, Dave Page wrote:
>> On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info>
>> wrote:
>> > On Sat, 2011-09-10 at 18:22 +0100, Dave Page wrote:
>> >> On Saturday, September 10, 2011, Guillaume Lelarge <
>> guillaume@lelarge.info>
>> >> wrote:
>> >> > Hi,
>> >> >
>> >> > AFAICT, the OP's complain is valid. I'm wondering why we have an OK and
>> >> > a Cancel button. The Cancel button is disabled when force parameter is
>> >> > set to false. After a quick research, it seems it's always false. Maybe
>> >> > we should replace the OK/Cancel pair with a Close button.
>> >> >
>> >> > Any objections to this change?
>> >>
>> >> Yes. We don't have close buttons on dialogs. There's a perfectly good one
>> in
>> >> the titlebar.
>> >>
>> >
>> > It may make sense to some dialogs, and not to others. On the properties
>> > dialogs, either you have to click OK to accept the changes, or you click
>> > Cancel to close without applying the changes. In the guru hint dialog,
>> > there is no reason (or at least, you didn't give one) to keep the OK and
>> > Cancel buttons. They are disturbing.
>>
>> Cancel on that dialog means "cancel the operation that triggered the hint so
>> I can correct the problem and try again". It's no different fthan "do you
>> want to delete? OK/Cancel"
>>
>
> It's really different because on the latter, you have a real question.
> On the former, there is no question which makes it difficult to
> understand what the OK button will do. What will the OK button do? close
> the dialog? VACUUM the table (on the VACUUM hint)? one cannot guess that
> without trying. Actually, I expect a user to hit more easily the cancel
> button than the OK button. Unfortunately, the Cancel button is always
> disabled, so it has no meaning at all.

I'm sure you must be mistaken there. I'm certain we don't force people to obey the hints in all cases (eg. The one about editing functions in the query tool). Can't check ATM though.

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

On Sat, 2011-09-10 at 20:07 +0100, Dave Page wrote:
> On Saturday, September 10, 2011, Guillaume Lelarge <guillaume@lelarge.info>
> wrote:
> > On Sat, 2011-09-10 at 18:43 +0100, Dave Page wrote:
> >> On Saturday, September 10, 2011, Guillaume Lelarge <
> guillaume@lelarge.info>
> >> wrote:
> >> > On Sat, 2011-09-10 at 18:22 +0100, Dave Page wrote:
> >> >> On Saturday, September 10, 2011, Guillaume Lelarge <
> >> guillaume@lelarge.info>
> >> >> wrote:
> >> >> > Hi,
> >> >> >
> >> >> > AFAICT, the OP's complain is valid. I'm wondering why we have an OK
> and
> >> >> > a Cancel button. The Cancel button is disabled when force parameter
> is
> >> >> > set to false. After a quick research, it seems it's always false.
> Maybe
> >> >> > we should replace the OK/Cancel pair with a Close button.
> >> >> >
> >> >> > Any objections to this change?
> >> >>
> >> >> Yes. We don't have close buttons on dialogs. There's a perfectly good
> one
> >> in
> >> >> the titlebar.
> >> >>
> >> >
> >> > It may make sense to some dialogs, and not to others. On the properties
> >> > dialogs, either you have to click OK to accept the changes, or you
> click
> >> > Cancel to close without applying the changes. In the guru hint dialog,
> >> > there is no reason (or at least, you didn't give one) to keep the OK
> and
> >> > Cancel buttons. They are disturbing.
> >>
> >> Cancel on that dialog means "cancel the operation that triggered the hint
> so
> >> I can correct the problem and try again". It's no different fthan "do you
> >> want to delete? OK/Cancel"
> >>
> >
> > It's really different because on the latter, you have a real question.
> > On the former, there is no question which makes it difficult to
> > understand what the OK button will do. What will the OK button do? close
> > the dialog? VACUUM the table (on the VACUUM hint)? one cannot guess that
> > without trying. Actually, I expect a user to hit more easily the cancel
> > button than the OK button. Unfortunately, the Cancel button is always
> > disabled, so it has no meaning at all.
>
> I'm sure you must be mistaken there. I'm certain we don't force people to
> obey the hints in all cases (eg. The one about editing functions in the
> query tool). Can't check ATM though.
>

You're right, the "edit function" hint doesn't force people to do the
VACUUM. I still think we should get a Close button, instead of the
OK/Cancel pair. On the "edit function", we would have a "Edit" button
(and the Close button). IOW, I still think we have a bad UI for hints.


--
Guillaume
  http://blog.guillaume.lelarge.info
  http://www.dalibo.com

On Mon, Oct 3, 2011 at 9:42 PM, Guillaume Lelarge
<guillaume@lelarge.info> wrote:
>
> You're right, the "edit function" hint doesn't force people to do the
> VACUUM.

Of course it doesn't - that would be ridiculous. What I was saying was
that the "edit function" hint doesn't force you to use the query tool.
It suggests you use the query tool, and if you click Cancel it cancels
the original operation to open the properties dialogue, and if you
click OK it continues.

> I still think we should get a Close button, instead of the
> OK/Cancel pair. On the "edit function", we would have a "Edit" button
> (and the Close button). IOW, I still think we have a bad UI for hints.

What would make more sense is to have a "Cancel" button to cancel the
original operation and an "Ignore" button meaning "Ignore the hint". I
don't know if that would be appropriate on all hints though.

--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company