Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

Please read the instructions on that page carefully.  Increment the
participant count.

A good way to identify the mode that you use, is to do

    M-x occur RET mode RET
    C-x b *Occur*

If you are using some minor modes ONLY in some major modes you can
indicate it in the Comments.

http://www.emacswiki.org/emacs/FrequentlyEnabledPackages_Emacs244_Survey

If you have some questions or comments please post it here or in the
Wiki page.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[William G. Gardella]

Jambunathan K <kjambunathan@gmail.com> writes:

> Please read the instructions on that page carefully.  Increment the
> participant count.
>
> A good way to identify the mode that you use, is to do
>
>     M-x occur RET mode RET
>     C-x b *Occur*
>
> If you are using some minor modes ONLY in some major modes you can
> indicate it in the Comments.
>
> http://www.emacswiki.org/emacs/FrequentlyEnabledPackages_Emacs244_Survey
>
> If you have some questions or comments please post it here or in the
> Wiki page.

The implementation of this survey is insane, and as a result, the wiki
page is horribly b0rken. Why not use the automation provided by M-x
report-emacs-bug ?

Many people (particularly those with configs of long standing) are not
necessarily aware of all the modes and libraries they're using.  This
should be automated.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

> The implementation of this survey is insane, and as a result,

Users are discovering (for first time?) features that are on that list
and reporting issues it.  I have also seen bugs when two different minor
modes are mixed and matched together.

I have a strong feeling that some of the bug reports are result of this
survey.

> the wiki page is horribly b0rken. 

I don't think so.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[William G. Gardella]

Jambunathan K <kjambunathan@gmail.com> writes:

>> The implementation of this survey is insane, and as a result,
>
> Users are discovering (for first time?) features that are on that list
> and reporting issues it.  I have also seen bugs when two different minor
> modes are mixed and matched together.
>
> I have a strong feeling that some of the bug reports are result of this
> survey.

On what evidence, exactly?

>> the wiki page is horribly b0rken. 
>
> I don't think so.

Your methodology resulted in a bunch of modes which are very commonly
used being added automatically to the table with erroneous results
(auto-compression-mode and auto-encryption-mode, for example, are
enabled by default, so I doubt vey much that their usage figure is
really 0!).  This wildly distorted evidence is likely to lead to bad
policymaking over at emacs-devel; the best we can hope for is that it's
simply ignored.

If the purpose is to discover which packages are actually frequently
used (for purposes of deciding what should be enabled by defaullt, for
example), it's generating horribly inaccurate and unreadable data, for
no good reason (report-emacs-bug demonstrates how much of this data is
already available in an automated fashion).  You're bringing some
entirely collateral and until-now-unstated purposes of this survey
without any evidence that it's achieving those aims at all.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

"William G. Gardella" <wgg2@member.fsf.org> writes:

> This wildly distorted evidence is likely to lead to bad policymaking
> over at emacs-devel

What will be result of this bad policy making.  Do you have evidence to
suggest that it will lead to apocalypse...

> You're bringing some entirely collateral and until-now-unstated
> purposes of this survey without any evidence that it's achieving those
> aims at all.

I was pointing to you that the survey has (good) side-effects.  You are
talking a language that I don't understand.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

"William G. Gardella" <wgg2@member.fsf.org> writes:

> On what evidence, exactly?

You think you are prosecuting me.  You also sit in the judge's chair.  I
am NOT on the defence benches.

If you could set aside the Hon. Judge's robes and alight the chair you
 you are sitting in, we can have some meaningful discussion.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[William G. Gardella]

Jambunathan K <kjambunathan@gmail.com> writes:

> You think you are prosecuting me.  You also sit in the judge's chair.  I
> am NOT on the defence benches.
>
> If you could set aside the Hon. Judge's robes and alight the chair you
>  you are sitting in, we can have some meaningful discussion.

Ok.  I'll put it more simply.  This survey as presently organized is
stupid drudgery, and the purpose of Emacs (like any good hack) is to
eliminate such.  Resorting to drudgery when automation is available is
doubly stupid.

As I said, if the project wants accurate data about what minor modes are
used, I think the way to do this is with M-x report-emacs-bug, which has
"Minor modes in effect:" as a subset of its data--as opposed to the
drudgery of making users manually enter data on the [unofficial] wiki.

The bad policy-making that could result from an inaccurate survey is
that some modes will be overcounted and some will be undercounted.  That
won't lead to "apocalypse," just to an editor/environment that's worse
than it potentially could be, with less sane defaults.  Bad data is
better than no data, admittedly, but this data is really bad.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

"William G. Gardella" <wgg2@member.fsf.org> writes:

> This survey as presently organized is stupid drudgery

Whose drudgery?  You have no respect for 90+ who has participated in the
survey.

You are angry at my stupidity.  You can do better.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Andreas Röhler]

Am 09.12.2013 10:31, schrieb Jambunathan K:
> "William G. Gardella" <wgg2@member.fsf.org> writes:
>
>> This survey as presently organized is stupid drudgery
>
> Whose drudgery?  You have no respect for 90+ who has participated in the
> survey.


No, Jambunathan, you lurked people into a terrible mess. That "survey" is a shame.
Looks like computing wasn't invented yet.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[William G. Gardella]

Jambunathan K <kjambunathan@gmail.com> writes:

> Whose drudgery?

The drudgery you're subjecting survey respondents to, and the drudgery
you're subjecting emacswiki administrators and editors to as they fix
problems in the document (read the talk page for the survey).

> You have no respect for 90+ who has participated in the survey.

On the contrary, I respect them enough that I want to lower the barrier
for participation, have a more accurate survey, and *use and improve
upon the automation that already exists* for documenting what modes and
features Emacs users are using.  Turnout is pretty low because
participation is onerous.  Low turnout, along with expecting people to
enter the data manually, will unfortunately contribute to an inaccurate
result.

> You are angry at my stupidity.  You can do better.

I already suggested a very simple way to do better.  Use the
self-documenting editor to, you know, document itself, as M-x
report-emacs-bug already does for minor modes, loaded features, and
other information on Emacs's state.  If having an editor wedded to a
Lisp environment is as good as we think it is (and I certainly think
so), then there must be a better way to document our configurations than
updating a website table by hand.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

"William G. Gardella" <wgg2@member.fsf.org> writes:

> I already suggested a very simple way to do better

Pay me $100 and I will implement your suggestion for you.

Jambunathan K.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Emanuel Berg]

We shouldn't be this dismissive to each others
initiatives. If we are, soon no one will dare take
initiatives. Also, we shouldn't get deadlocked over
some detail. A detail can be fixed. Only when all
details are fixed, and the idea still isn't working, it
can be dismissed.

In this particular case, it would seem that the human
factor has indeed produced errors. It is no coincidence
that `auto-compression-mode' and `auto-encryption-mode'
aren't mentioned, because those (and some others) don't
show up in the mode line. I myself was surprised when I
found them - I thought I had disabled everything I
didn't use long ago.

I found that this

(normal-erase-is-backspace-mode 0)
(abbrev-mode            0)
(auto-compression-mode  0)
(auto-encryption-mode   0)
(file-name-shadow-mode  0)
(gpm-mouse-mode         0)
(mouse-wheel-mode       0)
(tool-bar-mode          0)
(tooltip-mode           0)
(menu-bar-mode          0)

sped up my Emacs UX a lot, though that was on a much
slower computer (a laptop that fell from my backpack,
and broke) so I don't know if all that is necessary on
my current system. Anyway, as I don't use it, might
just keep it disabled. But I digress...

If `report-emacs-bug' creates a snapshot of the system,
including the major and minor modes (kind of
essential), that means there is an Elisp subroutine
that does that. We could isolate that and turn it into
a `what-minor-modes' or whatever and that would be it.

-- 
Emanuel Berg, programmer-for-rent. CV, projects, etc at uXu
underground experts united:  http://user.it.uu.se/~embe8573

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Joost Kremers]

William G. Gardella wrote:
> Turnout is pretty low because
> participation is onerous.

Hear hear. I wanted to participate, until I saw how the data are
collected... Then I decided against it.

Now, if I could just do `M-x report-emacs-bug` with "Survey" in the
subject line, that'd be a different matter.


-- 
Joost Kremers                                   joostkremers@fastmail.fm
Selbst in die Unterwelt dringt durch Spalten Licht
EN:SiS(9)

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Emanuel Berg]

Jambunathan K <kjambunathan@gmail.com> writes:

> Please read the instructions on that page carefully.
> Increment the participant count.
>
> A good way to identify the mode that you use, is to do
>
>     M-x occur RET mode RET
>     C-x b *Occur*

I get:

3 matches for "mode" in buffer: *Article*
7:  A good way to identify the mode that you use, is to do
9:  M-x occur RET mode RET
12: If you are using some minor modes ONLY in some major
    modes you can

Or should you do it from a specific buffer?

Is the purpose to find out what minor modes are used in
an Emacs session?

-- 
Emanuel Berg, programmer-for-rent. CV, projects, etc at uXu
underground experts united:  http://user.it.uu.se/~embe8573

Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

http://www.emacswiki.org/emacs/FrequentlyEnabledPackages_Emacs244_Survey

I will make an announcement on user list, once we get a near-stable set
 of packages/modes on that list.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

Jambunathan K <kjambunathan@gmail.com> writes:

> http://www.emacswiki.org/emacs/FrequentlyEnabledPackages_Emacs244_Survey
>
> I will make an announcement on user list, once we get a near-stable set
>  of packages/modes on that list.

Survey trend after 89 users.

I have drawn lines at 50%, 25%, 12.5% levels.  There are duplicate
entries (for eg., ido-mode and ido-everywhere).

Exercise judgement in interpreting the results.

#+CONSTANTS: N=89
| Packages/Modes            | Upvote / User count |  % |
|---------------------------+---------------------+----|
| uniquify                  |                  64 | 71 |
| column-number-mode        |                  62 | 69 |
| show-paren-mode           |                  60 | 67 |
| ido-mode                  |                  55 | 61 |
| transient-mark-mode       |                  45 | 50 |
|---------------------------+---------------------+----|
| ibuffer                   |                  43 | 48 |
| blink-cursor-mode         |                  40 | 44 |
| flyspell-mode             |                  39 | 43 |
| recentf-mode              |                  36 | 40 |
| eldoc-mode                |                  35 | 39 |
| dired-x                   |                  31 | 34 |
| line-number-mode          |                  30 | 33 |
| windmove                  |                  30 | 33 |
| winner-mode               |                  30 | 33 |
| font-lock-mode            |                  29 | 32 |
| auto-fill-mode            |                  27 | 30 |
| global-font-lock-mode     |                  27 | 30 |
| ido-everywhere            |                  27 | 30 |
| abbrev-mode               |                  26 | 29 |
| org-mode                  |                  24 | 26 |
| delete-selection-mode     |                  23 | 25 |
| saveplace                 |                  23 | 25 |
|---------------------------+---------------------+----|
| which-function-mode       |                  20 | 22 |
| desktop-save-mode         |                  19 | 21 |
| hippie-exp                |                  19 | 21 |
| savehist-mode             |                  19 | 21 |
| server-mode               |                  19 | 21 |
| display-time-mode         |                  17 | 19 |
| imenu                     |                  15 | 16 |
| global-subword-mode       |                  13 | 14 |
| size-indication-mode      |                  13 | 14 |
| cua-selection-mode        |                  12 | 13 |
| dired-hide-details-mode   |                  12 | 13 |
| icomplete-mode            |                  12 | 13 |
| org-capture               |                  12 | 13 |
| tooltip-mode              |                  12 | 13 |
|---------------------------+---------------------+----|
| auto-encryption-mode      |                  11 | 12 |
| tramp                     |                  11 | 12 |
| cua-mode                  |                  10 | 11 |
| subword-mode              |                  10 | 11 |
| electric-indent-mode      |                   9 | 10 |
| ffap                      |                   9 | 10 |
| file-name-shadow-mode     |                   9 | 10 |
| global-linum-mode         |                   9 | 10 |
| hs-minor-mode             |                   9 | 10 |
| kill-whole-line           |                   9 | 10 |
| linum-mode                |                   9 | 10 |
| reftex-mode               |                   9 | 10 |
| electric-pair-mode        |                   8 |  8 |
| global-hl-line-mode       |                   8 |  8 |
| auto-composition-mode     |                   7 |  7 |
| auto-compression-mode     |                   7 |  7 |
| compilation-minor-mode    |                   7 |  7 |
| global-auto-revert-mode   |                   7 |  7 |
| ispell-minor-mode         |                   7 |  7 |
| mouse-wheel-mode          |                   7 |  7 |
| edebug-mode               |                   6 |  6 |
| global-whitespace-mode    |                   6 |  6 |
| visible-bell              |                   6 |  6 |
| cl                        |                   5 |  5 |
| global-hi-lock-mode       |                   5 |  5 |
| org-src-mode              |                   5 |  5 |
| bs-show                   |                   4 |  4 |
| iswitchb-mode             |                   4 |  4 |
| shell-dirtrack-mode       |                   4 |  4 |
| visual-line-mode          |                   4 |  4 |
| auto-image-file-mode      |                   3 |  3 |
| delete-by-moving-to-trash |                   3 |  3 |
| epa-file-enable           |                   3 |  3 |
| generic-x                 |                   3 |  3 |
| global-cwarn-mode         |                   3 |  3 |
| hide-ifdef-mode           |                   3 |  3 |
| mouse-avoidance-mode      |                   3 |  3 |
| rectangle-mark-mode       |                   3 |  3 |
| electric-layout-mode      |                   2 |  2 |
| orgstruct++-mode          |                   2 |  2 |
| orgtbl-mode               |                   2 |  2 |
| reveal-mode               |                   2 |  2 |
| semantic-mode             |                   2 |  2 |
| whitespace-mode           |                   2 |  2 |
| auto-completion-mode      |                   1 |  1 |
| autoinsert                |                   1 |  1 |
| descr-text                |                   1 |  1 |
| erc-mode                  |                   1 |  1 |
| outline-minor-mode        |                   1 |  1 |
| pc-selection-mode         |                   1 |  1 |
| printing                  |                   1 |  1 |
| url-handler-mode          |                   1 |  1 |
#+TBLFM: $3= 100.0 * $2/ $N; %d;

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Michael Albinus]

Jambunathan K <kjambunathan@gmail.com> writes:

Hi,

> Exercise judgement in interpreting the results.
>
> | Packages/Modes            | Upvote / User count |  % |
> |---------------------------+---------------------+----|
> | tramp                     |                  11 | 12 |

Tramp is part of the Emacs distribution, and it is enabled by
`file-name-handler-alist'. Does this figure mean, 11 users are not aware
of? Or does this mean, just 11 of 89 users wish to see Tramp?

I don't understand this figure.

Best regards, Michael.

RE: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Drew Adams]

> > Exercise judgement in interpreting the results.

;-)

> > | Packages/Modes            | Upvote / User count |  % |
> > |---------------------------+---------------------+----|
> > | tramp                     |                  11 | 12 |
> 
> Tramp is part of the Emacs distribution, and it is enabled by
> `file-name-handler-alist'. Does this figure mean, 11 users are not aware
> of? Or does this mean, just 11 of 89 users wish to see Tramp?
> 
> I don't understand this figure.

Good question.  The instructions say to increment the count for
a given feature if you use it.  There is no instruction regarding
decrementing.  But the instructions have been ignored or
misinterpreted more than once, so far.

My guess would be that in this case 11 people out of those
participating are (a) aware of Tramp and (b) wanted to mention
that they use it and find it useful.  I would *not* assume that
those who did not increment the Tramp counter are (c) unaware
of it or (d) do not find it useful.

Another thing to consider is that the table was empty to start
with, and different entries appeared at different points, when
someone thought to add them.

So, e.g., if the first person happened to add feature `blah'
to the list, then subsequent "voters" saw `blah' and considered
it.  If `floop' was not added until much later (for whatever
reason), subsequent voters did not see `floop' until then.

It is what it is.

RE: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Drew Adams]

> But the instructions have been ignored or misinterpreted more
> than once, so far.

Now there's a superfluous "so far"!  It's not likely that with
additional voters this will drop below 2. ;-)

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

Drew Adams <drew.adams@oracle.com> writes:

> Another thing to consider is that the table was empty to start
> with, and different entries appeared at different points, when
> someone thought to add them.

Actually, it was seeded with my .emacs + Tom's initial list + Andrew's
list - 1 vote for each of the features.

> So, e.g., if the first person happened to add feature `blah'
> to the list, then subsequent "voters" saw `blah' and considered
> it.  If `floop' was not added until much later (for whatever
> reason), subsequent voters did not see `floop' until then.

It was for this reason that at around the time when the participant
count stood at 17, I added a list of all major and minor modes (by
rgrepping for define-derived-mode and define-minor-mode in the src
tree).  It was generally felt by others that such a big and
comprehensive list would actually drive people away.  (This reasoning is
perfeclty understandable.)

I have dumped the list of major and minor modes in to a separate page of
it's own.  There are around 244 minor modes and 293 major/derived modes.
See

     http://www.emacswiki.org/emacs/List_Of_Major_And_Minor_Modes

I couldn't think of any good reason why such a list would be useful.  I
generated the list and I felt why not save it for posterity?

> It is what it is.

Sure it is.  It is Emacs users telling how they relate to their Emacs.

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Matthias Meulien]

 
> (...) There are around 244 minor modes and 293 major/derived 
> modes. 

Note that `visual-line-mode' appears twice!
-- 
Matthias

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Rüdiger Sonderfeld]

On Sunday 08 December 2013 14:56:55 Michael Albinus wrote:
> Jambunathan K <kjambunathan@gmail.com> writes:
> > Exercise judgement in interpreting the results.
> > 
> > | Packages/Modes            | Upvote / User count |  % |
> > |
> > |---------------------------+---------------------+----|
> > |
> > | tramp                     |                  11 | 12 |
> 
> Tramp is part of the Emacs distribution, and it is enabled by
> `file-name-handler-alist'. Does this figure mean, 11 users are not aware
> of? Or does this mean, just 11 of 89 users wish to see Tramp?

The problem is that everybody started to mark and add features at different 
points.  I wouldn't really worry whether this represents actual usage for a 
default enabled feature.  I didn't increase tramp (I think) and it probably 
wasn't even on the list when I voted.  But I'm a regular tramp user.

The format of the survey is not really optimal.  I guess a better way would 
have been to use a piece of elisp to collect the data.  Similar to what the 
org-mode folks did in their survey.  But anyway the survey is probably good 
enough to consider the first 5 to 10 data points.  Which is probably all that 
matters for the actual question at hand.

I'd say the survey suggests that `column-number-mode' and `show-paren-mode' 
(maybe `recentf-mode') should be enabled as default.  `ido-mode' is popular 
but changes the behaviour to drastically to be a default imo.

Regards,
Rüdiger

Re: Frequently Enabled Packages - Emacs-24.4 survey (Please participate)

[Jambunathan K]

Jambunathan K <kjambunathan@gmail.com> writes:

>> http://www.emacswiki.org/emacs/FrequentlyEnabledPackages_Emacs244_Survey
>>
>> I will make an announcement on user list, once we get a near-stable set
>>  of packages/modes on that list.
>
> Survey trend after 89 users.

Creating a xref to the other thread.

For results after 120 users, see

    http://permalink.gmane.org/gmane.emacs.devel/166539

Permalink:

   http://www.emacswiki.org/emacs/FrequentlyEnabledPackages_Emacs244_Survey_Results