FC7 plan comments

[Jeremy Katz]

On Wed, 2006-12-20 at 20:13 -0500, Dave Jones wrote:
> On Wed, Dec 20, 2006 at 02:35:26PM -0500, Jeff Garzik wrote:
>  > On Wed, Dec 20, 2006 at 08:25:30PM +0100, Tomas Mraz wrote:
>  > > For example the egate cards are USB devices so the pcscd could
>  > > theoretically be started by udev when the card is plugged in. But the
>  > > library currently doesn't work right when the daemon is (re)started when
>  > > it was already initialized. And even if this was fixed the reaction time
>  > > for an app to actually discover that a smartcard was plugged in is
>  > > already pretty long. So making it even longer due to having to start the
>  > > pcscd is not too good idea.
>  > 
>  > That's just a one-time cost.  Once its running, you know the user uses
>  > smart cards, so no need add idle-out, daemon-stop complexity.
>  > 
>  > I think adding a miniscule pause the first time a user uses a SmartCard
>  > after booting the system is just fine, compared to what it gains us.
> 
> Right. Theres the option of doing it when its actually useful inconveniencing
> that user once each time they plug the device in, which takes maybe a few
> seconds.

And if the reader is plugged in at system boot, then it actually doesn't
end up being a net difference as the daemon will _still_ be getting
started during boot but the advantage is it will only be happening for
the users who want it.

I should really start filing this set of bugs as I think it's a pretty
straight-forward set of fixes that should have a noticeable impact on
boot time.

Jeremy