Re: Some developement questions

[hw]

Radon Rosborough <address@hidden> writes:

>> How is it relevant whether it takes 50ms more or less to start
>> Emacs?
>
> Because I can notice a difference of 50ms in a delay that happens very
> frequently. And yes, I restart Emacs very frequently. Easily a dozen
> times a day if I am doing active development on my Emacs
> configuration, which happens often.

Are you sure you notice a difference of 0.05 seconds?

Even if, this difference adds up to 0.6 seconds over a day, i. e. over
16 hours, when you start Emacs a dozen times.  That means within a
period of 57600 seconds, you notice a difference of 0.6 seconds, which
is 0.001%.  Seriously?

>> Is that even within the error margin of the measurements?
>
> Sure, there's some variation (although usually less than 150ms). But
> the mean value going down by 50ms is equally valuable regardless of
> the variance.

Are you sure you notice a difference of 0.05 seconds when the variance
is around 0.15 seconds anyway?

>> I just keep Emacs running on my server in a tmux session.
>
> I know that works for you (and many others), but it doesn't work for
> me. The reason it doesn't work for me is that I develop very actively
> on my Emacs configuration. This means that Emacs needs to be restarted
> frequently to avoid outdated configuration from causing problems
> (which, yes, can be worked around easily by evaluating code in the
> current session -- but it's a lot easier to just restart and wait half
> a second, with no thought required).

Ok, keeping it running won't work for you.

> The Emacs server also has some big usability problems for me (not
> bugs, but just things that make me personally prefer not to use it).
> Firstly, the mode line highlighting is shared between all sessions,
> and doesn't behave in the way one might expect if one is used to
> launching Emacs sessions directly.

Oh, I didn't mean the Emacs server; I can just use Emacs in a tmux
session and a terminal, connecting with ssh to the server.

> Secondly, it's not easy to share an Emacs server between graphical and
> terminal sessions -- so I'd need to run two servers, and the problem
> of having Emacs servers with outdated configuration gets twice as bad.

ssh -Y server
emacsclient -nc
emacsclient -nw

How is this difficult?

>> you can't keep up with me anyway ;P
>
> Fair enough. My configuration loads and quits a tty frame in 500ms,
> whereas emacsclient does the same in only 100ms. For me, the usability
> problems of the server are worth 400ms to avoid, since I don't pop in
> and out of Emacs to edit different files, only to reload my
> configuration.

no fair comparison when you can't let it run

How do you even measure such tiny amounts of time?