Re: emacs mode line suggestions

[Richard Riley]

Ian Eure <ian@digg.com> writes:

> On Nov 16, 2008, at 10:20 AM, Richard Riley wrote:
>
>> Ian Eure <ian@digg.com> writes:
>>>
>>> My objection is to the idea that you don't want star buffers in the
>>> list. These are also used for interaction with external processes:
>>> *ssh: host*, *SQL: foo*, *Twit-recent*, *compilation*, *shell*,
>>> *Python*. It seems ill advised to exclude those from the list.
>>
>> The kind of user that might want to see them is clued in enough to
>> use a
>> prefix or customise their setup accordingly.
>>
> That's not the point. A behavior is being suggested which has a
> serious usability negative. Your argument cuts both ways; anyone who
> might use the existing mechanism surely is clued in enough to know how
> to use it, so why bother changing it, either?
>
> Consider:
>
> 1. Edit stuff.
> 2. Create a *shell* buffer, do work in the shell.
> 3. Switch back to editing.
> 4. Try to go back to the shell - but wait, it's gone! What happened?
>
> It's equally confusing. If you want to improve things, great, but
> replacing one crappy behavior with a different crappy behavior is a
> waste.

Not really. The buffer list does get large. I use IDO and that works
well enough. But you are arguing how it would be if you keep the shell
as it is. Another way might be simply to have a regexp to show which
buffers are included - in fact I would be surprised if it wasnt already
there:-;

I was merely idly agreeing with Xah that the default list of ALL buffers
might not be right one. It should be a more advanced one.

>
>
>> I must say I agree with Xah and the "well thats the way its always
>> been"
>> kind of reply is not constructive in the slightest.
>>
> Did you actually read my reply? Because I criticized the proposal on
> it's merits, and never said anything remotely like that.

I was not referring to your reply. I was referring to the thread. Your
reply was fair and considered.