path: root/2023/captions/emacsconf-2023-devel--emacs-development-updates--john-wiegley--main.vtt

WEBVTT


00:00:02.419 --> 00:00:08.480
2 seconds. All right. I think we are live.

00:00:08.480 --> 00:00:10.280
Yes. So, hi again, everyone.

00:00:10.580 --> 00:00:13.620
I have the pleasure to welcome John Wiegley

00:00:13.620 --> 00:00:15.440
in person to this EmacsConf.

00:00:15.700 --> 00:00:17.260
Hi, John. Hello there.

00:00:17.680 --> 00:00:18.820
How are you doing, Leo?

00:00:18.820 --> 00:00:21.100
I am doing fantastic, and even more now that

00:00:21.100 --> 00:00:23.540
I am in a room with you because we've been,

00:00:24.279 --> 00:00:25.640
we were reminiscing with Sacha.

00:00:25.640 --> 00:00:30.860
So you had been there in person in 2013 And

00:00:30.860 --> 00:00:32.680
since we started doing those online,

00:00:32.680 --> 00:00:35.020
Juan, since 2019, I think you've always been

00:00:35.020 --> 00:00:38.460
online, right? Usually it's a pre-recorded

00:00:38.520 --> 00:00:40.440
video. I think this will be the first 1 I do

00:00:40.440 --> 00:00:42.540
live in a long time. You're right.

00:00:42.540 --> 00:00:44.239
I'm saying we are online right now,

00:00:44.239 --> 00:00:45.920
but I just meant pre-recorded video.

00:00:45.920 --> 00:00:48.400
So it's good to have you almost in person or

00:00:48.400 --> 00:00:50.900
at least live and we are excited to hear

00:00:50.900 --> 00:00:52.360
about some of the Emacs news.

00:00:52.360 --> 00:00:54.780
So the floor is yours.

00:00:55.080 --> 00:00:56.900
All right, well welcome everybody.

00:00:57.260 --> 00:01:00.239
This is the yearly state of the Emacs union,

00:01:00.239 --> 00:01:02.860
I guess, about how Emacs development is

00:01:02.860 --> 00:01:05.379
going. Just to note, I am not currently a

00:01:05.379 --> 00:01:07.760
maintainer of Emacs. So what I do to get

00:01:07.760 --> 00:01:09.920
these notes is I call up my friend,

00:01:09.920 --> 00:01:12.040
Eli Zaretsky, 1 of the current Emacs

00:01:12.040 --> 00:01:13.820
maintainers, and he and I sit down for an

00:01:13.820 --> 00:01:17.160
hour, and he just gives me his dump of what's

00:01:17.160 --> 00:01:19.400
been going on. So I'm sort of just the

00:01:19.400 --> 00:01:22.200
messenger here. But thanks to Eli for these

00:01:22.200 --> 00:01:24.400
notes and all of the efforts that he

00:01:24.400 --> 00:01:27.400
contributes. So what he's been telling me is

00:01:27.400 --> 00:01:30.060
that this Emacs 29 release that we had

00:01:30.060 --> 00:01:31.720
recently looks to have been very,

00:01:31.720 --> 00:01:33.580
very successful, which is some good news,

00:01:33.580 --> 00:01:35.420
because there were a lot of new features,

00:01:35.660 --> 00:01:37.360
and some of those features were actually

00:01:37.360 --> 00:01:39.520
quite radical. So far,

00:01:39.520 --> 00:01:40.680
it's been quite a success,

00:01:40.680 --> 00:01:42.440
no serious problems with it,

00:01:42.440 --> 00:01:44.080
and we have Emacs 29.2

00:01:45.140 --> 00:01:46.740
will be released very soon.

00:01:47.260 --> 00:01:50.020
They are thinking now about starting the

00:01:50.020 --> 00:01:53.080
Emacs 30 release cycle soon after 29.2

00:01:53.520 --> 00:01:55.880
is released, where the release branch,

00:01:55.880 --> 00:01:58.120
which is called Emacs-30 usually,

00:01:59.060 --> 00:02:01.400
will be cut and then development will become

00:02:01.400 --> 00:02:03.960
frozen with only bug fixes going into that

00:02:03.960 --> 00:02:07.200
branch. That may take quite some time until

00:02:07.200 --> 00:02:09.639
it actually comes to fruition as a release,

00:02:09.639 --> 00:02:11.660
but at least it means that the release is

00:02:11.660 --> 00:02:13.860
going to start taking shape in that branch

00:02:13.860 --> 00:02:17.720
soon. So, for now, Emacs 30 looks like maybe

00:02:17.720 --> 00:02:19.540
it's going to be a little less interesting

00:02:19.600 --> 00:02:23.160
than Emacs 29 was, meaning not a huge number

00:02:23.160 --> 00:02:25.120
of changing features. But there are still

00:02:25.120 --> 00:02:26.820
some new things going in.

00:02:26.980 --> 00:02:29.960
So 1 of them is that Emacs 30 is going to

00:02:29.960 --> 00:02:32.300
have Android support. So you will be able to

00:02:32.300 --> 00:02:34.900
run Emacs 30 on your Android devices.

00:02:35.140 --> 00:02:37.120
So if you've ever wanted to have native Emacs

00:02:37.120 --> 00:02:39.500
on a tablet, which I know I've always wanted,

00:02:40.140 --> 00:02:42.940
that will become possible with Emacs 30.

00:02:43.140 --> 00:02:45.480
There's also going to be much better support

00:02:45.480 --> 00:02:46.780
for touchscreen devices,

00:02:47.440 --> 00:02:50.260
coincidentally, both laptops and tablets.

00:02:50.740 --> 00:02:53.240
So that'll enhance that Android support.

00:02:54.860 --> 00:02:57.180
There will be some recently gained support

00:02:57.240 --> 00:03:01.720
for LLDB in GUD.dl. So if you're on a Mac OS

00:03:01.720 --> 00:03:05.580
machine or a machine that uses just LLVM as

00:03:05.580 --> 00:03:07.060
part of the compilation process,

00:03:07.400 --> 00:03:10.180
then you probably are familiar with LLDB as

00:03:10.180 --> 00:03:11.380
the command line debugger.

00:03:11.720 --> 00:03:14.940
And that support for using LLDB through a GUD

00:03:15.040 --> 00:03:17.320
will become possible in Emacs 30.

00:03:17.320 --> 00:03:19.000
I'm looking forward to this actually quite a

00:03:19.000 --> 00:03:22.620
bit as well. C Perl mode is being deprecated,

00:03:23.200 --> 00:03:25.920
and all future work now is only being put

00:03:25.920 --> 00:03:30.660
towards C Perl mode. Another 1 is that there

00:03:30.660 --> 00:03:33.000
are going to be some new major modes based on

00:03:33.000 --> 00:03:35.780
TreeSitter. They will be for the languages

00:03:35.860 --> 00:03:38.160
Lua, Elixir, and HTML.

00:03:38.800 --> 00:03:39.980
And if you're not familiar,

00:03:40.160 --> 00:03:42.620
I think TreeSitter was introduced in Emacs

00:03:42.620 --> 00:03:46.660
29. It's a library that allows you to specify

00:03:47.460 --> 00:03:49.900
the grammar of a programming language as a

00:03:49.900 --> 00:03:53.300
BNF file, and I think using JavaScript,

00:03:53.540 --> 00:03:56.600
and then with that file as input to Emacs,

00:03:56.600 --> 00:03:59.840
it is then able to do syntax highlighting,

00:04:00.940 --> 00:04:03.080
syntax discovery, all of those things within

00:04:03.080 --> 00:04:06.140
Emacs without having to use elisp and regexps

00:04:06.460 --> 00:04:08.300
to discover the structure of the language.

00:04:08.300 --> 00:04:10.440
It defers the structure gathering to

00:04:10.440 --> 00:04:13.420
TreeSitter and then uses that information to

00:04:13.420 --> 00:04:14.560
navigate the language.

00:04:15.200 --> 00:04:17.300
So, As time goes on, you'll see more and more

00:04:17.300 --> 00:04:19.540
languages taking on TreeSetter support.

00:04:19.540 --> 00:04:21.160
So the next 3 coming up,

00:04:21.160 --> 00:04:22.660
Lua, Elixir, and HTML.

00:04:24.060 --> 00:04:26.680
And then the last feature for Emacs 30 is

00:04:26.680 --> 00:04:29.860
that the byte compiler will now detect and

00:04:29.860 --> 00:04:32.920
warn about many more questionable constructs.

00:04:33.340 --> 00:04:35.300
Things like empty macro bodies,

00:04:35.740 --> 00:04:37.160
missing lexical constructs,

00:04:37.580 --> 00:04:40.220
or say, condition case without any handlers.

00:04:40.580 --> 00:04:43.340
Just silly stuff that might litter the code,

00:04:43.340 --> 00:04:45.180
but now you'll get a warning about it from

00:04:45.180 --> 00:04:46.920
the byte compiler to help you clean up the

00:04:46.920 --> 00:04:49.160
code and get rid of those potential sites of

00:04:49.160 --> 00:04:52.740
error. So this is the main thing that will be

00:04:52.740 --> 00:04:55.160
worked on for Emacs 30 and what's looked like

00:04:55.160 --> 00:04:56.400
shaping up for the release.

00:04:56.680 --> 00:04:58.940
And also, he wanted me to announce that

00:04:58.940 --> 00:05:01.340
Stefan Kongas is now a new co-maintainer.

00:05:01.980 --> 00:05:03.260
And Stefan is, I believe,

00:05:03.260 --> 00:05:05.500
here with us in the conference and he'll be

00:05:05.500 --> 00:05:07.940
able, I hope, to help me answer any questions

00:05:08.080 --> 00:05:10.120
about future Emacs development because I'm

00:05:10.120 --> 00:05:12.240
not in the heat of it and don't have all

00:05:12.240 --> 00:05:13.600
those answers at the moment.

00:05:14.580 --> 00:05:17.180
So That is all there is as far as a

00:05:17.180 --> 00:05:18.840
development update for now.

00:05:19.340 --> 00:05:21.980
And I am available to take any questions.

00:05:24.860 --> 00:05:26.200
All right. Thank you so much,

00:05:26.200 --> 00:05:29.340
Sean, for being the messenger of all this

00:05:29.340 --> 00:05:31.820
good news. I mean, you did start by saying

00:05:31.820 --> 00:05:33.480
this would not be as exciting,

00:05:33.600 --> 00:05:36.280
perhaps, as prior releases of Emacs,

00:05:36.280 --> 00:05:38.980
but you then proceeded to say a lot of stuff

00:05:38.980 --> 00:05:40.760
that it felt very exciting to me.

00:05:40.760 --> 00:05:43.620
So good, good. Glad to hear that.

00:05:44.540 --> 00:05:47.180
Right. So we do have questions coming in

00:05:47.180 --> 00:05:50.380
already and again people the link is on IRC

00:05:50.380 --> 00:05:52.120
and also on the talks page if you want to

00:05:52.120 --> 00:05:53.260
start asking questions.

00:05:53.480 --> 00:05:54.840
So John what I'm going to do I'm going to

00:05:54.840 --> 00:05:56.580
read you the questions and then you can

00:05:56.580 --> 00:05:58.300
answer them. Is that okay with you?

00:05:58.320 --> 00:06:01.620
Absolutely. So starting with the first

00:06:01.620 --> 00:06:04.440
question which changes in recent Emacs

00:06:04.440 --> 00:06:06.540
releases are you enjoying using?

00:06:08.360 --> 00:06:11.820
I have really liked the visual line mode.

00:06:13.140 --> 00:06:15.060
I'm not sure how recent that is.

00:06:15.060 --> 00:06:16.820
Some of these features I only discovered

00:06:16.840 --> 00:06:19.580
quite late, the new display line number

00:06:19.700 --> 00:06:21.380
functionality, where it's much,

00:06:21.380 --> 00:06:23.040
much, much faster, and of course,

00:06:23.040 --> 00:06:25.320
native compilation. Native compilation has

00:06:25.320 --> 00:06:27.360
been quite brilliant for some of the larger

00:06:27.360 --> 00:06:29.480
packages that I use. I do a lot of stuff in

00:06:29.480 --> 00:06:31.820
Emacs. I use GNU's, I use E-Shell,

00:06:31.820 --> 00:06:33.540
I use Org Mode quite a lot.

00:06:33.580 --> 00:06:36.100
So native compilation has brought the user

00:06:36.100 --> 00:06:39.960
experience much closer to a modern app than

00:06:39.960 --> 00:06:42.180
some of the lagging and slowness that I might

00:06:42.180 --> 00:06:43.580
have experienced in the past.

00:06:44.340 --> 00:06:47.180
Definitely. Moving on to the next question.

00:06:47.260 --> 00:06:49.200
What do you think the future in the area of

00:06:49.200 --> 00:06:51.040
artificial intelligence from the developer

00:06:51.060 --> 00:06:53.580
point of view? Could you say that 1 more

00:06:53.580 --> 00:06:54.876
time? Your voice broke up a little bit.

00:06:54.876 --> 00:06:55.025
Oh, sorry. What do you think the future in

00:06:55.025 --> 00:06:55.141
the area of artificial intelligence from the

00:06:55.141 --> 00:06:55.208
developer point of view?

00:06:55.208 --> 00:06:55.324
Could you say that 1 more time?

00:06:55.324 --> 00:06:55.440
Your voice broke up a little bit.

00:06:55.440 --> 00:06:58.100
Oh, sorry. What do you think the future in

00:06:58.100 --> 00:07:00.580
the area of artificial intelligence from the

00:07:00.580 --> 00:07:01.560
developer point of view?

00:07:01.560 --> 00:07:02.860
It's also a shaky question,

00:07:02.860 --> 00:07:04.500
I think, but you get the point.

00:07:04.960 --> 00:07:09.220
I do use chat-gpt-shell inside of Emacs quite

00:07:09.220 --> 00:07:10.920
a bit, actually, when doing development in

00:07:10.920 --> 00:07:12.440
other languages. Just the other day,

00:07:12.440 --> 00:07:15.200
I was working on my Ledger accounting

00:07:15.220 --> 00:07:18.080
program, and I haven't done a lot of C++ in

00:07:18.080 --> 00:07:21.180
recent years. So I had forgotten how to

00:07:21.180 --> 00:07:23.760
exactly compare 2 strings only up to the

00:07:23.760 --> 00:07:25.020
length of the shortest string.

00:07:25.020 --> 00:07:26.940
I know I could have cranked that out just

00:07:26.940 --> 00:07:28.740
writing it C style, but I didn't remember

00:07:28.740 --> 00:07:30.800
what the current state of the art is for C++

00:07:30.940 --> 00:07:33.460
and the STL. So I just asked chatGPT.

00:07:33.680 --> 00:07:36.140
I asked the exact question that I just said

00:07:36.140 --> 00:07:38.220
to you and sure enough it popped out the

00:07:38.220 --> 00:07:40.080
one-liner that was exactly what I needed.

00:07:40.080 --> 00:07:42.240
So I think in terms of developer assistance,

00:07:42.800 --> 00:07:45.820
not having to keep all of standard libraries

00:07:45.860 --> 00:07:47.460
or common idioms in memory.

00:07:47.520 --> 00:07:49.460
I don't know if other people are familiar

00:07:49.480 --> 00:07:51.080
with Rosetta Stone projects.

00:07:51.460 --> 00:07:53.680
They're projects where you have say a hundred

00:07:53.680 --> 00:07:55.780
different languages and there's a particular

00:07:55.860 --> 00:07:58.640
question, say, how do I read a file and copy

00:07:58.640 --> 00:07:59.720
it to another location?

00:07:59.820 --> 00:08:01.880
And then it has an instance of doing that

00:08:01.880 --> 00:08:03.820
activity for every 1 of those languages.

00:08:04.180 --> 00:08:05.140
That's a great database,

00:08:05.140 --> 00:08:07.160
and I've used them quite a bit in the past

00:08:07.160 --> 00:08:09.260
for remembering how to do certain things,

00:08:09.440 --> 00:08:12.540
say, converting a string to UTF-8.

00:08:13.280 --> 00:08:15.660
I think that AI does a great job of

00:08:15.660 --> 00:08:17.780
completely replacing the need for databases

00:08:17.900 --> 00:08:19.920
like that because you can just ask how do I

00:08:19.920 --> 00:08:21.980
copy a convert a string to UTF-8.

00:08:23.760 --> 00:08:27.440
Yeah exactly and you know especially with

00:08:27.440 --> 00:08:30.480
languages which are tried well tried you know

00:08:30.480 --> 00:08:32.360
it's very easy to get an answer that is

00:08:32.360 --> 00:08:34.940
correct. But sometimes what I find bothersome

00:08:34.940 --> 00:08:37.460
with this type of coding,

00:08:37.460 --> 00:08:39.840
I think it's AI-aided coding,

00:08:39.840 --> 00:08:40.820
but it's still coding,

00:08:41.120 --> 00:08:43.140
is that, especially with C languages,

00:08:43.140 --> 00:08:44.240
sometimes you're going to end up with

00:08:44.240 --> 00:08:45.860
undefined behaviors and stuff like this just

00:08:45.860 --> 00:08:47.900
because other people have been doing it,

00:08:47.900 --> 00:08:50.860
not because the algorithm or the model was

00:08:50.860 --> 00:08:53.680
trained with data that dates back to 10 years

00:08:53.680 --> 00:09:00.060
ago. At the time, C++ was a little different.

00:09:00.060 --> 00:09:01.400
Anyway, I'm not here to talk,

00:09:01.400 --> 00:09:03.960
you are here to talk. Moving on to the next

00:09:03.960 --> 00:09:06.560
question. People already get to hear my voice

00:09:06.560 --> 00:09:09.640
plenty, whereas yours are much sparser.

00:09:10.520 --> 00:09:13.580
All right. So, what is the future of Emacs on

00:09:13.580 --> 00:09:16.080
macOS? I understand that there are too few

00:09:16.080 --> 00:09:17.420
developers for the platform.

00:09:17.440 --> 00:09:21.600
Is that still true? That's a good question.

00:09:21.600 --> 00:09:23.620
I don't know what the current statistics are.

00:09:23.620 --> 00:09:27.040
I've been a user of Emacs on Mac OS for

00:09:27.040 --> 00:09:29.920
decades now. It feels like the,

00:09:30.140 --> 00:09:33.480
There's also that Mac port version of Emacs,

00:09:33.480 --> 00:09:35.920
which builds Emacs more directly using the

00:09:35.920 --> 00:09:37.860
GUI libraries on the platform.

00:09:38.300 --> 00:09:40.320
That continues to be updated with every

00:09:40.320 --> 00:09:42.040
single new release that comes out.

00:09:42.620 --> 00:09:45.220
So I'd say that the support may not be as

00:09:45.220 --> 00:09:47.540
great as it is on Linux and other platforms,

00:09:47.640 --> 00:09:50.500
but to this day I haven't suffered from being

00:09:50.500 --> 00:09:55.220
a Mac user. Great. The only thing I remember

00:09:55.240 --> 00:10:00.480
about Emacs on macOS was that emojis made it

00:10:00.480 --> 00:10:03.540
inside the GUI first before they did it

00:10:03.540 --> 00:10:05.820
anywhere else. That's the 1 anecdote that I

00:10:05.820 --> 00:10:09.520
have on MacOS. Right. And historically that

00:10:09.520 --> 00:10:12.720
feature was removed in order to prevent Mac

00:10:12.720 --> 00:10:14.840
from having features that Linux did not.

00:10:15.160 --> 00:10:16.860
I didn't want to go into that point.

00:10:16.860 --> 00:10:18.340
I just wanted to mention the beginning of the

00:10:18.340 --> 00:10:19.840
anecdote and people can find it out.

00:10:19.840 --> 00:10:22.040
But yes, that's also what it led to.

00:10:24.020 --> 00:10:25.340
Moving on to the next question.

00:10:25.640 --> 00:10:28.120
Why aren't you contributing to Emacs anymore?

00:10:28.140 --> 00:10:29.360
Lack of time, I guess?

00:10:30.240 --> 00:10:31.840
Lack of time, primarily.

00:10:32.300 --> 00:10:33.840
Work has been very consuming.

00:10:33.900 --> 00:10:36.020
There are a lot of other projects and things

00:10:36.020 --> 00:10:39.000
that I like doing. I still find Emacs Lisp

00:10:39.000 --> 00:10:40.460
very, very fun to write.

00:10:40.840 --> 00:10:43.260
Just the other day, I was hacking up some

00:10:43.260 --> 00:10:45.760
extension macros for myself for org mode.

00:10:45.920 --> 00:10:48.900
But to have the time needed to sit down and

00:10:48.900 --> 00:10:51.580
design a whole new mode and work on it.

00:10:51.580 --> 00:10:53.660
I've been spending a lot of my time now in

00:10:53.660 --> 00:10:55.580
functional languages, especially theorem

00:10:55.580 --> 00:10:57.840
provers. I just find that so intellectually

00:10:58.100 --> 00:10:59.620
satisfying and interesting.

00:11:00.400 --> 00:11:01.860
Plus it pays a lot better.

00:11:01.860 --> 00:11:03.680
Never had a paying job as an Emacs list

00:11:03.680 --> 00:11:06.420
developer. So when it comes to now just being

00:11:06.420 --> 00:11:08.380
a fun language or a hobby language,

00:11:08.420 --> 00:11:10.680
it is relegated to the time that I have free

00:11:10.680 --> 00:11:13.680
when it's available. Right.

00:11:13.700 --> 00:11:15.420
Well, the good thing is that it's kind of

00:11:15.420 --> 00:11:16.360
like riding a bicycle,

00:11:16.360 --> 00:11:17.980
you know, writing a major mode,

00:11:17.980 --> 00:11:20.160
it comes back relatively quickly and still

00:11:20.160 --> 00:11:22.600
enjoyable. You know, the other day,

00:11:22.600 --> 00:11:25.200
actually, I took notes on a mode that I

00:11:25.200 --> 00:11:28.100
wanted to write. There's an app I use on the

00:11:28.100 --> 00:11:30.060
Mac called drafts, and I really love it.

00:11:30.060 --> 00:11:31.460
I use it all the time.

00:11:31.720 --> 00:11:34.600
I wanted to mimic the interface of this app

00:11:34.600 --> 00:11:37.620
in Emacs. So I could use Emacs as my drafts

00:11:37.680 --> 00:11:40.060
application rather than this separate 1.

00:11:40.440 --> 00:11:42.720
So I noted down all the different user

00:11:42.720 --> 00:11:44.580
parameters and how it should function and

00:11:44.580 --> 00:11:47.240
everything to describe the app to myself as

00:11:47.240 --> 00:11:50.320
sort of notes to get me started on that work

00:11:50.320 --> 00:11:52.460
when I did have free time to work on it.

00:11:52.540 --> 00:11:54.760
Somebody out there on the internet just saw

00:11:54.760 --> 00:11:57.040
these notes, because I keep a lot of my stuff

00:11:57.040 --> 00:11:59.680
on GitHub. They fed it to chat GPT,

00:12:00.280 --> 00:12:01.960
going back to your AI question.

00:12:02.220 --> 00:12:04.540
And they actually sent back to me a mode that

00:12:04.540 --> 00:12:06.640
implemented everything that I had said,

00:12:06.820 --> 00:12:07.740
which was effectively,

00:12:08.200 --> 00:12:10.960
chat GPT, seeing that what I had described

00:12:10.960 --> 00:12:14.500
was clear enough for it to derive most of the

00:12:14.500 --> 00:12:16.560
code that I would have wanted to write.

00:12:16.560 --> 00:12:19.200
So maybe, maybe another thing that AI can do

00:12:19.200 --> 00:12:20.880
is it can increase the value,

00:12:21.340 --> 00:12:23.400
the efficiency of my free time.

00:12:24.360 --> 00:12:26.820
Exactly. I think that's a wonderful point.

00:12:27.120 --> 00:12:29.540
And phrasing it as efficiency of free time is

00:12:29.540 --> 00:12:31.360
great because you still have the expertise,

00:12:31.360 --> 00:12:33.420
obviously, that you're mobilizing into the

00:12:33.420 --> 00:12:35.560
design that you're formulating to charge DPT,

00:12:35.860 --> 00:12:37.680
but then this expertise is turned into

00:12:37.680 --> 00:12:39.060
something that actually works.

00:12:40.080 --> 00:12:41.780
Perhaps we're all going to become software

00:12:41.780 --> 00:12:42.980
architects at some point,

00:12:42.980 --> 00:12:45.800
and then the busy work of actually coding the

00:12:45.800 --> 00:12:48.760
library and the software will be relegated to

00:12:48.760 --> 00:12:51.000
AI. That's an interesting future where we

00:12:51.000 --> 00:12:55.240
still, however, need to acquire the skills to

00:12:55.240 --> 00:12:56.660
know what is code, I suppose.

00:12:56.660 --> 00:12:58.880
But that's an interesting future to think of.

00:13:00.940 --> 00:13:02.060
A fairly long question.

00:13:02.060 --> 00:13:04.280
So 1 of the tricky things about running Emacs

00:13:04.280 --> 00:13:06.900
on Android is do you use anything that

00:13:06.900 --> 00:13:08.320
requires extra packages?

00:13:08.420 --> 00:13:11.600
Example like PDF tools with new PDF or going

00:13:11.600 --> 00:13:13.820
with a database, playing music or video with

00:13:13.820 --> 00:13:15.840
MPD or MPV on Bonga, LFeed.

00:13:16.320 --> 00:13:17.900
Do you run Emacs Termex,

00:13:18.040 --> 00:13:20.200
Emacs APK, Emacs in virtual machine?

00:13:20.200 --> 00:13:22.420
This is also the case on Emacs for Windows to

00:13:22.420 --> 00:13:23.980
a lesser degree. So summarizing,

00:13:24.280 --> 00:13:27.240
how do you make Emacs work on Android if you

00:13:27.240 --> 00:13:30.040
do not have the synergy of stuff that you

00:13:30.040 --> 00:13:32.560
usually find on Linux systems like MPV and

00:13:32.560 --> 00:13:34.420
all the fancy applications like this?

00:13:35.340 --> 00:13:38.380
It's a good question. Since I'm not an

00:13:38.380 --> 00:13:40.460
Android user and I've never tried running

00:13:40.460 --> 00:13:42.020
Emacs on Android platforms,

00:13:42.100 --> 00:13:44.920
I'm not sure what's available out there to

00:13:44.920 --> 00:13:46.440
plug Emacs into. I mean,

00:13:46.440 --> 00:13:48.160
effectively, that question comes down to

00:13:48.160 --> 00:13:50.420
external dependencies and system support.

00:13:50.940 --> 00:13:53.320
That would be a great question for Stefan or

00:13:53.320 --> 00:13:55.740
somebody who has tried using Emacs,

00:13:56.200 --> 00:13:58.600
the development version of Emacs on Android.

00:14:00.660 --> 00:14:03.980
Great. We'll put a pin in this for Stéphane

00:14:04.540 --> 00:14:07.580
afterwards. Great, so moving on to the next

00:14:07.580 --> 00:14:09.860
question. Will Org Tech someday become the

00:14:09.860 --> 00:14:11.320
default tech mode in Emacs?

00:14:11.320 --> 00:14:14.880
And if so, when? Will Org what become?

00:14:15.240 --> 00:14:17.980
Org Tech, you know, the LaTeX mode.

00:14:19.120 --> 00:14:24.840
I do not know. It's been a while since I've

00:14:24.840 --> 00:14:26.880
done LaTeX. It must have been like 4 years,

00:14:26.880 --> 00:14:30.280
but it was a pretty, the major mode for

00:14:30.280 --> 00:14:32.560
editing documents, like the state of the art

00:14:32.560 --> 00:14:34.840
for editing latex documents in Emacs.

00:14:34.940 --> 00:14:36.320
And apparently it's not default.

00:14:36.340 --> 00:14:38.360
I assume there's latex mode or something that

00:14:38.360 --> 00:14:41.580
is doing it. So were you saying octech,

00:14:41.760 --> 00:14:45.020
like A-U-C tech? Oh, did I not pronounce the

00:14:45.020 --> 00:14:48.140
C? Octech, yes. I thought you said org tech.

00:14:48.140 --> 00:14:49.620
I wasn't familiar with that.

00:14:50.280 --> 00:14:52.540
Octech is the only 1 I've ever used.

00:14:52.640 --> 00:14:55.360
I know there is a built-in LaTeX mode,

00:14:55.580 --> 00:14:57.240
but I've never used it.

00:14:57.240 --> 00:14:59.140
I always just download whatever the latest

00:14:59.140 --> 00:15:01.160
version of Org Tech is and use that.

00:15:01.280 --> 00:15:03.840
I don't know why it's not a standard package.

00:15:03.840 --> 00:15:07.360
Becoming a standard package has its own costs

00:15:07.360 --> 00:15:09.860
for the development cycle because it slows

00:15:09.860 --> 00:15:11.740
down release cycle quite a bit.

00:15:12.120 --> 00:15:14.680
It's now you have to create PRs that are

00:15:14.680 --> 00:15:17.240
reviewed by the Emacs Devel mailing list.

00:15:17.620 --> 00:15:19.340
It is a little more inertia.

00:15:19.340 --> 00:15:21.680
Of course, it gets you more distribution

00:15:21.940 --> 00:15:24.020
because it's a default package now,

00:15:24.020 --> 00:15:26.100
and everybody can be using that.

00:15:26.140 --> 00:15:28.600
But it's not something every developer

00:15:28.660 --> 00:15:31.400
decides to do. It took a few years,

00:15:31.400 --> 00:15:34.620
in fact, to get usePackage into Emacs core.

00:15:34.660 --> 00:15:37.200
And that only happened after it was so stable

00:15:37.200 --> 00:15:39.360
that it really wasn't receiving many changes

00:15:39.360 --> 00:15:44.760
anymore. Yeah, yeah it's it's the thing when

00:15:44.760 --> 00:15:46.720
you move into core you lose a lot of your

00:15:46.720 --> 00:15:49.280
agility in terms of how you're writing the

00:15:49.280 --> 00:15:50.940
code or how you expand code.

00:15:51.300 --> 00:15:54.220
That's why you have this vibrant community on

00:15:54.220 --> 00:15:57.680
Melpa compared to core but you know it

00:15:57.740 --> 00:15:59.480
doesn't necessarily ought to be this way it

00:15:59.480 --> 00:16:02.320
could be a little different you know And it

00:16:02.320 --> 00:16:04.580
feels like there's this repetition between

00:16:04.760 --> 00:16:06.820
repartition, sorry, between people developing

00:16:06.820 --> 00:16:08.600
for the core of Emacs and people developing

00:16:08.600 --> 00:16:10.760
on Melpa, but at the end of the day those 2

00:16:10.760 --> 00:16:13.020
groups are constantly talking to 1 another

00:16:13.180 --> 00:16:15.240
and taking cues from 1 another as well.

00:16:15.240 --> 00:16:17.540
So that's great. And there's of course...

00:16:18.820 --> 00:16:21.880
May I jump in about this particular question

00:16:21.900 --> 00:16:25.080
because I think I mean You know,

00:16:25.080 --> 00:16:27.540
Org Mode doesn't really have any problems

00:16:27.540 --> 00:16:29.380
with releases Correct.

00:16:29.380 --> 00:16:32.820
Just because it's distributed with Emacs so

00:16:32.860 --> 00:16:35.020
there is a difference between being in the

00:16:35.020 --> 00:16:37.960
core proper and being distributed with Emacs.

00:16:38.520 --> 00:16:40.080
And for something like use packages,

00:16:40.080 --> 00:16:42.280
it's really necessary to be in the core.

00:16:42.600 --> 00:16:44.840
But for something like major mode,

00:16:45.060 --> 00:16:48.740
it's a bit easier. That's a very very good

00:16:48.740 --> 00:16:50.400
point. Yeah, I'd forgotten about that

00:16:50.400 --> 00:16:53.000
distinction. Org mode does advance pretty

00:16:53.000 --> 00:16:55.900
rapidly and then it makes releases into the

00:16:55.900 --> 00:16:59.640
core distribution. Gianni,

00:16:59.640 --> 00:17:01.480
I believe you also wanted to say something

00:17:01.480 --> 00:17:03.840
before someone started jumped in with a

00:17:03.840 --> 00:17:08.700
question do you happen to remember okay

00:17:08.700 --> 00:17:12.020
that's fine I lost her to lost to their time

00:17:12.560 --> 00:17:14.560
I'll be moving on to the next question then

00:17:15.700 --> 00:17:17.460
and by the way feel free to interrupt us you

00:17:17.460 --> 00:17:19.619
know The whole point of this discussion is

00:17:19.619 --> 00:17:21.140
for you to ask questions to John Wheatley.

00:17:21.140 --> 00:17:23.520
So whether it be via the other pad or via

00:17:23.520 --> 00:17:26.180
BBB, choose your weapon.

00:17:27.260 --> 00:17:28.820
All right, moving on to the next question in

00:17:28.820 --> 00:17:30.520
the meantime. And we have about 7 minutes

00:17:30.520 --> 00:17:32.760
left of Q&A and then we'll be moving on to

00:17:32.760 --> 00:17:35.780
Stéphane. So, do you use other IDEs for

00:17:35.780 --> 00:17:38.040
theorem proving work, notably VS Code for

00:17:38.040 --> 00:17:42.340
Lean? Which languages and provers can or do

00:17:42.340 --> 00:17:45.920
you use Emacs for? I've only used Emacs.

00:17:46.120 --> 00:17:49.840
I've used Emacs for working with ACL 2,

00:17:49.960 --> 00:17:53.480
Coq, Agda, and Lean, and I really love

00:17:53.480 --> 00:17:55.680
Proof-General. Coq is my favorite language to

00:17:55.680 --> 00:17:57.740
be working in. Agda has really great support

00:17:57.740 --> 00:17:59.540
as well, has a very nice Emacs mode.

00:17:59.540 --> 00:18:02.920
I'm only just now starting to get into Lean

00:18:02.920 --> 00:18:05.140
4. So I have everything installed,

00:18:05.280 --> 00:18:07.240
but I haven't really started coding in

00:18:07.240 --> 00:18:08.800
earnest. I'm still reading a lot of the

00:18:08.800 --> 00:18:10.920
tutorials and learning a bit about the

00:18:10.920 --> 00:18:13.440
language. There was a while there where I

00:18:13.440 --> 00:18:17.540
used a IDE for ACL 2 that was outside of

00:18:17.540 --> 00:18:19.920
Emacs, only because it was the same IDE all

00:18:19.920 --> 00:18:22.040
my co-workers were using and it was easier to

00:18:22.040 --> 00:18:24.240
share tips and tricks with them.

00:18:24.520 --> 00:18:28.240
But yeah, no, I found Emacs to be a great

00:18:28.240 --> 00:18:29.860
home for doing theorem proving.

00:18:32.120 --> 00:18:36.580
Right. Next question. Can we see that AI

00:18:36.580 --> 00:18:40.760
generated draft? You know what you mentioned

00:18:40.760 --> 00:18:43.420
before about the draft that you then fed into

00:18:43.420 --> 00:18:45.480
ChargPT? Do you happen to have this draft

00:18:45.480 --> 00:18:51.440
anywhere? Let me see if it's still on GitHub.

00:18:51.760 --> 00:18:54.300
Just take me 1 second to take a look here.

00:18:55.260 --> 00:18:59.820
Take your time. The problem is I don't quite

00:18:59.820 --> 00:19:01.820
remember where I made the note.

00:19:04.700 --> 00:19:07.260
But no, I don't see it on GitHub,

00:19:07.340 --> 00:19:09.840
so I don't have it readily at hand.

00:19:10.580 --> 00:19:13.240
Well, that's fine. We'll be able to...

00:19:13.520 --> 00:19:15.060
Well, if you happen to find it,

00:19:15.060 --> 00:19:17.280
we'll make sure to add it on the pad and then

00:19:17.280 --> 00:19:20.940
on the talks page. And I think we would all

00:19:20.940 --> 00:19:22.440
be interested to see what this design

00:19:22.440 --> 00:19:27.160
document that actually made something work

00:19:27.160 --> 00:19:28.980
afterwards in JudgeDPT with Elisp.

00:19:28.980 --> 00:19:30.720
I'm very interested to see what it would do

00:19:30.720 --> 00:19:32.720
because I tend to be very interested about

00:19:32.720 --> 00:19:34.780
this type of stuff I had generated but I

00:19:34.780 --> 00:19:36.860
never thought about doing it with Elisp

00:19:37.040 --> 00:19:39.000
because somehow it feels like 2 different

00:19:39.000 --> 00:19:41.020
worlds, like Elisp is kind of from the past,

00:19:41.260 --> 00:19:44.040
don't going me wrong, I love it and I use it

00:19:44.140 --> 00:19:46.920
every day But it's 2 different parts of my

00:19:46.920 --> 00:19:48.820
brain that I didn't think about linking.

00:19:48.820 --> 00:19:51.100
So I'd be very excited to see this as well.

00:19:53.040 --> 00:19:54.640
Moving on to the next question.

00:19:54.640 --> 00:19:56.980
Oh, go on, please. I did find it.

00:19:57.260 --> 00:20:01.080
I'm gonna have to give it to you as a link

00:20:01.080 --> 00:20:03.380
here. Sure, you can do it on the blue button

00:20:03.380 --> 00:20:05.900
and I'll put it on the pad.

00:20:10.160 --> 00:20:12.540
I put it into the public chat for Bibi.

00:20:12.960 --> 00:20:15.920
Yes. So if anyone is interested,

00:20:15.920 --> 00:20:18.280
I'm putting it right in the answer to the

00:20:18.280 --> 00:20:20.200
question right here on my screen.

00:20:20.200 --> 00:20:22.120
So feel free to click on it and explore it.

00:20:22.120 --> 00:20:23.680
I'm kind of curious, so I'm gonna...

00:20:23.680 --> 00:20:25.560
Can I click it on stream and can we look at

00:20:25.560 --> 00:20:26.620
it a little bit together?

00:20:26.740 --> 00:20:29.020
Sure, sure. I haven't tried running it,

00:20:29.020 --> 00:20:31.680
I can't say for its fitness,

00:20:31.680 --> 00:20:34.220
but it's definitely enough of the groundwork

00:20:34.400 --> 00:20:36.660
done that it's absolutely an assistance.

00:20:38.100 --> 00:20:40.260
Right. Okay, so it's loading up right now?

00:20:41.540 --> 00:20:42.620
You can see my webcam,

00:20:42.620 --> 00:20:45.680
right? I can see your browser attempting to

00:20:45.680 --> 00:20:47.360
load. There we go. Okay,

00:20:47.360 --> 00:20:50.060
cool. So I'm not sure what GitHub is doing.

00:20:50.060 --> 00:20:51.460
Let me give it a little more room.

00:20:51.460 --> 00:20:54.360
The reactive setup is not working too well.

00:20:57.260 --> 00:20:58.840
Oh, I see. You're viewing the...

00:20:59.540 --> 00:21:02.220
I see. Can I see the file then?

00:21:02.220 --> 00:21:04.020
I should be able to see the file.

00:21:06.460 --> 00:21:09.060
I think he just mentions the code in that

00:21:09.060 --> 00:21:11.640
comment. So if there's a way to view only the

00:21:11.640 --> 00:21:14.897
comment it would make it clear.

00:21:14.897 --> 00:21:16.660
Right. Okay. I'm loading the file separately.

00:21:17.320 --> 00:21:18.500
I'm checking the time.

00:21:18.740 --> 00:21:21.340
We have about 3 minutes left and I think we

00:21:21.340 --> 00:21:22.900
have a question. In the meantime,

00:21:22.900 --> 00:21:25.760
whilst I show this, I'm gonna launch another

00:21:25.760 --> 00:21:28.820
question, which was about Drafts.

00:21:29.440 --> 00:21:30.740
You are carrying, you're talking about

00:21:30.740 --> 00:21:32.480
Drafts, but does that mean you're not using

00:21:32.480 --> 00:21:35.920
org anymore? Oh, no, I use org all the time.

00:21:35.920 --> 00:21:38.100
In fact, the way that I've configured drafts

00:21:38.160 --> 00:21:40.580
is that after I type the thing in the note

00:21:40.580 --> 00:21:43.260
into drafts, I hit a key and it creates an

00:21:43.260 --> 00:21:45.420
org mode capture item for it.

00:21:45.540 --> 00:21:48.560
The reason why I use Drafts instead of Emacs

00:21:48.560 --> 00:21:50.660
is because it's always available.

00:21:51.100 --> 00:21:53.960
If Emacs is currently doing some job for me,

00:21:53.960 --> 00:21:56.340
say I'm running some long-running subshell

00:21:57.060 --> 00:21:59.440
and the UI is frozen up whatnot,

00:21:59.700 --> 00:22:02.360
Drafts is always 100% of the time instantly

00:22:02.360 --> 00:22:04.900
available. So that's why I tend to then lean

00:22:04.900 --> 00:22:07.920
on it a bit, but all of the destination of

00:22:07.920 --> 00:22:10.240
that data is still Org Mode and everything

00:22:10.240 --> 00:22:12.460
that I do gets tracked through Org Mode.

00:22:13.080 --> 00:22:16.160
That's also why I wanted to implement the UI

00:22:16.160 --> 00:22:18.800
scheme of drafts in Emacs so that I could

00:22:18.800 --> 00:22:20.860
drop the use of this external application.

00:22:21.540 --> 00:22:23.160
And then, I mean, I would still have the

00:22:23.160 --> 00:22:25.580
problem of sometimes Emacs being unavailable,

00:22:26.060 --> 00:22:30.040
but I would pay that price in order to have

00:22:30.040 --> 00:22:32.700
that good UI of drafts inside Emacs.

00:22:34.640 --> 00:22:37.700
Great. I think we have,

00:22:38.420 --> 00:22:39.720
we might be too tight on time.

00:22:39.720 --> 00:22:41.640
We only have about 2 minutes and I need to

00:22:41.640 --> 00:22:44.180
jump room to go into Stephane's room as well.

00:22:44.440 --> 00:22:47.380
So John, where I get to thank you so much for

00:22:47.380 --> 00:22:53.160
taking the time to answer our questions,

00:22:53.260 --> 00:22:55.040
but also doing a little bit of reporting on

00:22:55.040 --> 00:22:58.380
the state of Emacs. And now we'll get to

00:22:58.380 --> 00:22:59.280
continue this with Stéphane.

00:22:59.280 --> 00:23:01.300
So do you have any last words for everyone,

00:23:01.300 --> 00:23:03.660
John? No, no. I look forward to hearing

00:23:03.660 --> 00:23:06.420
Stéphane speak. Okay, great.

00:23:06.420 --> 00:23:07.360
And we'll look forward,

00:23:07.360 --> 00:23:09.300
potentially, to having you again next year,

00:23:10.520 --> 00:23:12.680
potentially still doing news like this,

00:23:12.980 --> 00:23:15.920
and, fingers crossed, maybe having you live

00:23:15.920 --> 00:23:18.660
again. Maybe, maybe let's see what happens.

00:23:19.700 --> 00:23:20.860
All right, John. Thank you so much.

00:23:20.860 --> 00:23:22.280
Bye bye. Bye bye.