A bug reporter surprised by their information going to public mailing list.

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
13 messages Options
Reply | Threaded
Open this post in threaded view
|

A bug reporter surprised by their information going to public mailing list.

David Rowley
In [1] Brian got a bit caught out when he missed the text on the bug
reporting form that says:

"By submitting a bug through this form you agree that all contents of
the bug report, including your personal information as listed, will be
posted to the public pgsql-bugs mailing list, and archived in the
public list archives."

I figured that could have been much worse than it was.

Maybe it's not worth spending too many cycles for people that bypass
the warning text, but I thought I'd mention the idea that this could
be backed up by a check box just above the submit and send email
button which has the text "I understand that this report will be
emailed to a public mailing list".  The submit would be disallowed
until the user checks the checkbox.

Does anyone think it's worth adding?

David

[1] https://www.postgresql.org/message-id/CAAk_qnGumHD6kBiiexd8WQ1jnPpUSXhOu5XiCAvWatZAaKaRfQ@...


Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Jonathan S. Katz-3
On 9/27/20 5:37 PM, David Rowley wrote:

> In [1] Brian got a bit caught out when he missed the text on the bug
> reporting form that says:
>
> "By submitting a bug through this form you agree that all contents of
> the bug report, including your personal information as listed, will be
> posted to the public pgsql-bugs mailing list, and archived in the
> public list archives."
>
> I figured that could have been much worse than it was.
>
> Maybe it's not worth spending too many cycles for people that bypass
> the warning text, but I thought I'd mention the idea that this could
> be backed up by a check box just above the submit and send email
> button which has the text "I understand that this report will be
> emailed to a public mailing list".  The submit would be disallowed
> until the user checks the checkbox.
>
> Does anyone think it's worth adding?
The patch itself is not difficult (attached), though the formatting on
the page itself is not great, I'd need to figure out the CSS to align
the checkbox and label text, as the general form construct has not
required that to date.

I'm not opposed to adding it.

Jonathan

confirm.patch (1K) Download Attachment
signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Tom Lane-2
In reply to this post by David Rowley
David Rowley <[hidden email]> writes:
> Maybe it's not worth spending too many cycles for people that bypass
> the warning text, but I thought I'd mention the idea that this could
> be backed up by a check box just above the submit and send email
> button which has the text "I understand that this report will be
> emailed to a public mailing list".  The submit would be disallowed
> until the user checks the checkbox.

> Does anyone think it's worth adding?

+1.  With that, we could probably simplify the button text back
down to "Submit".  Right now I think it's a bit confusing, because
it makes it sound like there are two different actions there.

                        regards, tom lane


Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Magnus Hagander-2


On Mon, Sep 28, 2020 at 1:39 AM Tom Lane <[hidden email]> wrote:
David Rowley <[hidden email]> writes:
> Maybe it's not worth spending too many cycles for people that bypass
> the warning text, but I thought I'd mention the idea that this could
> be backed up by a check box just above the submit and send email
> button which has the text "I understand that this report will be
> emailed to a public mailing list".  The submit would be disallowed
> until the user checks the checkbox.

> Does anyone think it's worth adding?

I honestly doubt it will help. People have such confirmation-fatigue that if they can't be bothered to read anyway, they're just going to check that box and proceed anyway. To get past that we'd at least have to make it a multi-step process where the confirmation is completely separate, which is a much  bigger patch. (Hmm. Yeah, at some point I had a big patch to change that thing *completely*, which would then also do things like direct pgadmin patches to the pgadmin submission etc, but I never quite finished that one).

It also doesn't massively hurt, so I won't object to adding it --  I just don't think it will work.

That said, if we do add it (like in Jonathans patch), it should *also* be added to the docs form, since they work the same and would have the same basic problem.


+1.  With that, we could probably simplify the button text back
down to "Submit".  Right now I think it's a bit confusing, because
it makes it sound like there are two different actions there.

-1 for changing it to just "Submit". But I can agree it's a bit confusing. Maybe change it to "Submit email" or "Send email" or something like that which again makes it clear what happens. Just "submit" will again direct people away from the fact that they are sending an email. Or maybe "Submit public bug report"?
 
--
Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Jonathan S. Katz-3
(dropping Tom not because I want to but for some reason my mail client
is not resolving his address at the moment).

On 9/28/20 7:56 AM, Magnus Hagander wrote:
>
>
> On Mon, Sep 28, 2020 at 1:39 AM Tom Lane <[hidden email]
> <mailto:[hidden email]>> wrote:
>

>     David Rowley <[hidden email] <mailto:[hidden email]>>
>     writes:
>     > Maybe it's not worth spending too many cycles for people that bypass
>     > the warning text, but I thought I'd mention the idea that this could
>     > be backed up by a check box just above the submit and send email
>     > button which has the text "I understand that this report will be
>     > emailed to a public mailing list".  The submit would be disallowed
>     > until the user checks the checkbox.
>
>     > Does anyone think it's worth adding?
>
>
> I honestly doubt it will help. People have such confirmation-fatigue
> that if they can't be bothered to read anyway, they're just going to
> check that box and proceed anyway. To get past that we'd at least have
> to make it a multi-step process where the confirmation is completely
> separate, which is a much  bigger patch. (Hmm. Yeah, at some point I had
> a big patch to change that thing *completely*, which would then also do
> things like direct pgadmin patches to the pgadmin submission etc, but I
> never quite finished that one).
>
> It also doesn't massively hurt, so I won't object to adding it --  I
> just don't think it will work.
>
> That said, if we do add it (like in Jonathans patch), it should *also*
> be added to the docs form, since they work the same and would have the
> same basic problem.
Agreed with Magnus' analysis.

>     +1.  With that, we could probably simplify the button text back
>     down to "Submit".  Right now I think it's a bit confusing, because
>     it makes it sound like there are two different actions there.
>
>
> -1 for changing it to just "Submit". But I can agree it's a bit
> confusing. Maybe change it to "Submit email" or "Send email" or
> something like that which again makes it clear what happens. Just
> "submit" will again direct people away from the fact that they are
> sending an email. Or maybe "Submit public bug report"?
-1 for changing the language if we take no action.

If we did, coupled with the patch, I would suggest "Submit Bug Report"
as that's exactly what the form is doing, and the new language above it
indicates that it's posting to a mailing list.

Jonathan


signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Tom Lane-2
In reply to this post by Magnus Hagander-2
Magnus Hagander <[hidden email]> writes:
> On Mon, Sep 28, 2020 at 1:39 AM Tom Lane <[hidden email]> wrote:
>> +1.  With that, we could probably simplify the button text back
>> down to "Submit".  Right now I think it's a bit confusing, because
>> it makes it sound like there are two different actions there.

> -1 for changing it to just "Submit". But I can agree it's a bit confusing.
> Maybe change it to "Submit email" or "Send email" or something like that
> which again makes it clear what happens. Just "submit" will again direct
> people away from the fact that they are sending an email. Or maybe "Submit
> public bug report"?

"Submit public bug report" would be OK by me; indeed it seems better
than any of these alternatives, because really the key word here is
"public".

                        regards, tom lane


Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Jonathan S. Katz-3
On 9/28/20 9:46 AM, Tom Lane wrote:

> Magnus Hagander <[hidden email]> writes:
>> On Mon, Sep 28, 2020 at 1:39 AM Tom Lane <[hidden email]> wrote:
>>> +1.  With that, we could probably simplify the button text back
>>> down to "Submit".  Right now I think it's a bit confusing, because
>>> it makes it sound like there are two different actions there.
>
>> -1 for changing it to just "Submit". But I can agree it's a bit confusing.
>> Maybe change it to "Submit email" or "Send email" or something like that
>> which again makes it clear what happens. Just "submit" will again direct
>> people away from the fact that they are sending an email. Or maybe "Submit
>> public bug report"?
>
> "Submit public bug report" would be OK by me; indeed it seems better
> than any of these alternatives, because really the key word here is
> "public".
Alright, after playing around with the checkbox alignment a bit and
reaching the point of diminishing returns, I came up with this: slightly
shorter text around the checkbox, and the button saying "Submit Bug Report."

I'll let this bake a couple of days and after such, if no objections,
I'll commit it.

Thanks,

Jonathan

0001-Add-confirmation-checkbox-for-submitting-bug-report.patch (2K) Download Attachment
signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Tom Lane-2
"Jonathan S. Katz" <[hidden email]> writes:
> Alright, after playing around with the checkbox alignment a bit and
> reaching the point of diminishing returns, I came up with this: slightly
> shorter text around the checkbox, and the button saying "Submit Bug Report."

Seems OK to me.

                        regards, tom lane


Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

David Rowley
In reply to this post by Jonathan S. Katz-3
On Sat, 3 Oct 2020 at 14:35, Jonathan S. Katz <[hidden email]> wrote:
> Alright, after playing around with the checkbox alignment a bit and
> reaching the point of diminishing returns, I came up with this: slightly
> shorter text around the checkbox, and the button saying "Submit Bug Report."

Thank you for working on this.

I'm not familiar enough with the web code to do any meaningful review,
but I'm happy enough with the text you've chosen.

David


Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

Magnus Hagander-2
In reply to this post by Jonathan S. Katz-3
On Sat, Oct 3, 2020 at 3:35 AM Jonathan S. Katz <[hidden email]> wrote:
On 9/28/20 9:46 AM, Tom Lane wrote:
> Magnus Hagander <[hidden email]> writes:
>> On Mon, Sep 28, 2020 at 1:39 AM Tom Lane <[hidden email]> wrote:
>>> +1.  With that, we could probably simplify the button text back
>>> down to "Submit".  Right now I think it's a bit confusing, because
>>> it makes it sound like there are two different actions there.
>
>> -1 for changing it to just "Submit". But I can agree it's a bit confusing.
>> Maybe change it to "Submit email" or "Send email" or something like that
>> which again makes it clear what happens. Just "submit" will again direct
>> people away from the fact that they are sending an email. Or maybe "Submit
>> public bug report"?
>
> "Submit public bug report" would be OK by me; indeed it seems better
> than any of these alternatives, because really the key word here is
> "public".

Alright, after playing around with the checkbox alignment a bit and
reaching the point of diminishing returns, I came up with this: slightly
shorter text around the checkbox, and the button saying "Submit Bug Report."

I still think that is a regression, since fact that it's generating an email is *removed* from the button. See previous comment to that regard, suggesting the addition of "public". But it seems I may be outvoted on that :)

I also wonder if we should just remove the basically identical text from the top of the form, now that it sits with the confirmation checkbox? Or keep them both in the vain hope that people might read at least one of them? :)

I'll let this bake a couple of days and after such, if no objections,
I'll commit it.

The actual objection I have was also mentioned before, which is that the docs comment submission form should be treated the same way.  They're basically the same thing, just to a different mailinglist,  so there is no reason to expect people to read them differently.

--
Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

David Rowley
On Sun, 4 Oct 2020 at 22:42, Magnus Hagander <[hidden email]> wrote:
> I also wonder if we should just remove the basically identical text from the top of the form, now that it sits with the confirmation checkbox? Or keep them both in the vain hope that people might read at least one of them? :)

The original text I proposed was much more simple than what Jonathan's
patch has. I had thought a simple "I understand that this report will
be emailed to a public mailing list" would have been ok. If the user
was surprised by that, then they could go back and read the text we
expect them to have ready already to find out the details.

Perhaps the more verbose wording for the checkbox makes the original
warning partially redundant, but my vote would be to simplify the
checkbox text to something more like what I originally proposed rather
than remove the mention of it from the top.

David


Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

David G Johnston
On Sun, Oct 4, 2020 at 4:05 AM David Rowley <[hidden email]> wrote:
On Sun, 4 Oct 2020 at 22:42, Magnus Hagander <[hidden email]> wrote:
> I also wonder if we should just remove the basically identical text from the top of the form, now that it sits with the confirmation checkbox? Or keep them both in the vain hope that people might read at least one of them? :)

The original text I proposed was much more simple than what Jonathan's
patch has. I had thought a simple "I understand that this report will
be emailed to a public mailing list" would have been ok. If the user
was surprised by that, then they could go back and read the text we
expect them to have ready already to find out the details.

Perhaps the more verbose wording for the checkbox makes the original
warning partially redundant, but my vote would be to simplify the
checkbox text to something more like what I originally proposed rather
than remove the mention of it from the top.

+1

The savebutton probably could just read "Post to Public List"

I'd probably skip the checkbox and add "(public)" after Name, Email, and Details

It puts the word public right next to the three fields that have primary public/private concern for a submitter plus on the final button.  The paragraph next to a checkbox is probably only just a little less eye-glazing than the paragraphs of text they already skimmed over and missed the public detail.  They might just see a checkbox that they need to check and do it to move on.  The field headers are short and so the added text is obvious and succinct.  As David Rowley said, if they have a WTF moment they can read the above verbiage more closely.

David J.

Reply | Threaded
Open this post in threaded view
|

Re: A bug reporter surprised by their information going to public mailing list.

David G Johnston
On Mon, Oct 12, 2020 at 3:59 PM David G. Johnston <[hidden email]> wrote:
The savebutton probably could just read "Post to Public List"

Just experienced a similar situation [1] with someone reading our documentation and sending an email directly to -bugs.