On 25 November 2015 at 22:01, Adam Williamson
On Wed, 2015-11-25 at 15:40 -0500, Przemek Klosowski wrote:
> On 11/25/2015 03:25 PM, drago01 wrote:
> > On Wed, Nov 25, 2015 at 9:17 PM, Adam Williamson
> > <adamwill(a)fedoraproject.org> wrote:
> > > On Fri, 2015-11-20 at 14:34 +0000, Ian Malone wrote:
> > > >  Apparently middle mouse buttons are rare. I'm in an office
> > > This seems an odd assertion [...]
> > Its not odd ... its plain wrong.
> I think Ian meant to say that the mice WITHOUT middle button are rare.
> The quote above continues on like this:
> I'm in an office surrounded by them and them only computers I've used
without one for roughly the past decade are my old laptop
> Am I right, Ian?
Well, it was what I said. I was going to add that it would be possible
to check in the archives, but I can't actually find my original post
there. Not sure why this thread has woken up again.
The wiki page explaining the GNOME-on-Wayland approach to
paste - https://wiki.gnome.org/Initiatives/Wayland/PrimarySelection
makes this claim as one of the reasons why it wasn't initially planned
for inclusion in Wayland:
"Additionally, reasons against keeping it:
the middle click is a hard-to-discover easter egg
there are few middle mouse buttons in the world"
Ian was, I think with reason, questioning the second of those. I was
pointing out that those are only a couple of *supplementary* reasons,
so it's not really worth spending much time disputing that assertion,
even though it does seem like an odd one. The primary reasons why
Wayland wasn't initially going to implement a PRIMARY selection
mechanism are given earlier in the page:
I was questioning those and a tone that's become very common:
1. This is obscure, only crusty old dinosaurs know about it and use
it. (I'm fairly sure I'm neither crusty or old.)
2. Not relevant any more because Y.
Throw these in to support any breakage you would like to introduce and
suddenly you're a breath of fresh air blowing all the old cruft away
and anyone who disagrees is stuck in the dark ages. Except in this
case they're trivially demonstrated as untrue (well documented, unlike
documenting all new features in blog posts and leaving them to rot,
hardware support widespread). So this feature is being re-introduced
because "many longterm X users have become reliant on this easter egg"
(still an easter egg) and it is, "to ease the transition for long-term
X users." (Who just can't cope with the modern world, so let's throw
them a bone.) They look like very little consideration was given.
"Among the arguments for eschewing the PRIMARY selection were:
It makes it easy to unintentionally paste passwords, snippets of
private conversations and other non-public information,
into online communication.
security concerns with unexpected data stealing if the mere act of
selecting a text fragment makes it available to all running
That is a genuine concern, but compare it to having a clipboard of the
past N selections which most users are probably not aware of either.
It looks like they've come up with a scheme to tie it down.
and it goes on to propose that the primary selection should in fact
And turned off by default of course.
Actually, my main comment (snipped) was that the utility of having
separate buffers is not even discussed, and it is not actually clear
that this wont simply be reintroduced as an option to turn on a form
of auto-copy on select. Which the discussion of the signalling
mechanism involved and the "not just text" aspect seem to suggest it