"consider M-x recover-file" too important to just flash

"consider M-x recover-file" too important to just flash

[Dan Jacobson]

You can't just flash
  index.html has auto save data; consider M-x recover-file
in the minibuffer for one second and expect people to see it.

No, two seconds, three seconds, etc. no good either.  We might have
been answering the telephone.

Therefore you need to hit us over the head with a message that needs a
yes or no.  Like "... Do you understand, Holmes? (yes or no)".

In firefox this is run of the mill: the "OK?" box that doesn't go away
unless you click it.  Odd, I don't recall encountering those "got
that, Holmes?" one-choice interactive messages in emacs.

Hmmm, no wonder. There are always some choices to give. Here we would
probably want to give choices like 'yes' to recover-file, 'no' to get
the file without recovering, and 'quit' to not do anything at all.

emacs-version "21.3.1"

Re: "consider M-x recover-file" too important to just flash

[Richard Stallman]

    In firefox this is run of the mill: the "OK?" box that doesn't go away
    unless you click it.  Odd, I don't recall encountering those "got
    that, Holmes?" one-choice interactive messages in emacs.

That behavior is terribly annoying--we certainly shouldn't make Emacs
act that way.

Re: "consider M-x recover-file" too important to just flash

[Dan Jacobson]

DJ>     In firefox this is run of the mill: the "OK?" box that doesn't go away
DJ>     unless you click it.

rms> That behavior is terribly annoying--we certainly shouldn't make Emacs
rms> act that way.

Well, how shall we make sure we saw that important message? What if we
reached down to scratch our ankle during the one second or whatever we
were given to have noticed that message before it went away or was
covered up by some other message?

No big deal if #backup_file# exists for you perhaps, but I want emacs
to make sure I am aware of it!

You could ask the user if he wants to recover("yes") or blunder on
anyway("no"), "or hit C-g" to not open the file at all.

Re: "consider M-x recover-file" too important to just flash

[Richard Stallman]

    rms> That behavior is terribly annoying--we certainly shouldn't make Emacs
    rms> act that way.

    Well, how shall we make sure we saw that important message?

It's a secondary issue.  I've told you my decision on this question.

Re: "consider M-x recover-file" too important to just flash

[Dan Jacobson]

"Didn't I fix paragraph 503 a few months ago?  Now I have doubts about
the integrity of the whole file!"

The above statement would likely be made by a user who's head was
turned during the vital "consider M-x recover-file" message in the
minibuffer, many edits ago that fateful day.  The message was probably
replaced by a flyspell message or whatever by the time he turned his
head back.  M-x mark-my-words: one day it will bite you too.

If you are not going to use a stronger method of warning than fleeting
messages in the minibuffer, then well, at least you admit that you are
not going to impede emacs -batch jobs whereas I would!

Re: "consider M-x recover-file" too important to just flash

[Mathias Dahl]

Dan Jacobson <jidanni@jidanni.org> writes:

> rms> That behavior is terribly annoying--we certainly shouldn't make Emacs
> rms> act that way.

I agree.

> Well, how shall we make sure we saw that important message? What if we
> reached down to scratch our ankle during the one second or whatever we
> were given to have noticed that message before it went away or was
> covered up by some other message?
>
> No big deal if #backup_file# exists for you perhaps, but I want emacs
> to make sure I am aware of it!
>
> You could ask the user if he wants to recover("yes") or blunder on
> anyway("no"), "or hit C-g" to not open the file at all.

Seems like emacs lacks some sort of more permanent status/message/echo
area that is both visible (for some time maybe) and both not in the
way. Hard problem to crack though...

/Mathias

Re: "consider M-x recover-file" too important to just flash

[Ted Zlatanov]

On Fri, 01 Apr 2005, brakjoller.rem0veth1s@gmail.com wrote:

> Seems like emacs lacks some sort of more permanent status/message/echo
> area that is both visible (for some time maybe) and both not in the
> way. Hard problem to crack though...

Emacs has the modeline, which is appropriate because it belongs to the
buffer in question, so it's not a global alert.  A flashing (or
otherwise noticeable) message in the modeline might help.

Ted

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Re: "consider M-x recover-file" too important to just flash

[Dan Jacobson]

>> Hard problem to crack though...
No.
Ted> Emacs has the modeline, which is appropriate because it belongs to the
Ted> buffer in question, so it's not a global alert.  A flashing (or
Ted> otherwise noticeable) message in the modeline might help.
No. No. No. You may be looking at the modeline, but I don't. Plus it
may be obscured.  What is obvious to you may not be to others, and
colors, blinks, etc. might not be "seen" by all users all the time.

The answer is simple and occurs 1000 times in emacs already at much
less important times: the user should be asked in the minibuffer how
to proceed. And an accidental RET or SPC etc. should not be considered
an answer.

Only a (setq recover-file-warnings 'ho-hum) should avoid the question.

RMS> It's a secondary issue.
Yes, No need for "OK?" screens. But do make sure the user gives his
informed consent to ignore the newer #backup#, which currently happens
these days as there are too many ways for a user to not see certain
message amongst so many others these days. Etc.