WEBVTT captioned by rodrigo

NOTE Introduction

00:00:01.750 --> 00:00:11.969
Welcome everyone to my EmacsConf presentation,

00:00:15.370 --> 00:00:22.089
where we're going to look at Lisp's vdemo[??] app from 1978,

00:00:22.139 --> 00:00:25.659
but done in modern Elisp.

00:00:28.460 --> 00:00:33.039
Then we're going to look at getting inferior-lisp working,

00:00:33.390 --> 00:00:34.753
my dear Common Lisp.

00:00:37.003 --> 00:00:42.059
Then we're going to use the Common Lisp interface manager,

00:00:42.159 --> 00:00:45.663
the open source child of the Common Lisp interface manager,

00:00:47.813 --> 00:00:52.959
McCLIM, from inferior-lisp in Superior Lisp,

00:00:55.710 --> 00:00:58.325
which is the opposite order that I initially thought

00:00:58.375 --> 00:00:59.276
we were going to do things in,

00:00:59.326 --> 00:01:00.779
but this is how it's kind of turned out.

00:01:00.829 --> 00:01:01.795
So it's this article,

00:01:01.845 --> 00:01:05.199
I thought, from 1978, was interesting to us today.

00:01:05.999 --> 00:01:08.702
“Programming in an Interactive Environment:

00:01:08.752 --> 00:01:11.805
the ‘Lisp’ Experience.”

00:01:12.005 --> 00:01:17.678
Explaining what all these LISP programmers were doing to

00:01:19.478 --> 00:01:22.466
people unfamiliar with LISP programming.

00:01:22.566 --> 00:01:24.401
“LISP systems have been used for

00:01:24.451 --> 00:01:27.454
highly interactive programming for more than a decade.”

00:01:27.604 --> 00:01:29.790
This was from 1978.

00:01:29.990 --> 00:01:31.959
And we're just going to go and,

00:01:34.109 --> 00:01:39.717
almost verbatim, port the demo application for LISP

00:01:40.017 --> 00:01:46.119
out of 1978 Interlisp and into Emacs Lisp.

00:01:49.720 --> 00:01:51.428
And we're going to do it without Org Mode,

00:01:51.478 --> 00:01:54.231
because I think everyone is finding out

00:01:54.281 --> 00:01:56.950
how great Org Mode is already.

00:01:58.100 --> 00:02:00.787
For inferior Lisp, I thought instead of org-mode,

00:02:00.837 --> 00:02:02.823
we'll theme it around slime-mode and then it turned out

00:02:02.873 --> 00:02:06.059
most of this presentation happens in Emacs Lisp,

00:02:06.109 --> 00:02:08.929
because this is an Emacs Lisp conference.

00:02:09.379 --> 00:02:12.799
But I found out that there's ielm-mode,

00:02:12.849 --> 00:02:17.604
which is pretty similar to slime-mode,

00:02:19.054 --> 00:02:23.610
but in our hearts, we can imagine that this is

00:02:23.660 --> 00:02:27.448
an Emacs example, homed around slime-mode,

00:02:27.498 --> 00:02:33.353
Superior Lisp Interaction Mode for Emacs, instead of org-mode.

00:02:33.403 --> 00:02:36.390
Though, you know, I do secretly use org-mode as well,

00:02:36.440 --> 00:02:37.858
not very secretly.

00:02:37.958 --> 00:02:40.539
But that means we're kind of improvising,

00:02:46.790 --> 00:02:48.786
improvising a presentation mode.

00:02:49.986 --> 00:02:55.576
So we have this big buffer of Elisp, and let's just...

00:02:56.726 --> 00:02:59.164
This is me, not Sandewall, out of interest,

00:03:00.614 --> 00:03:02.616
let's make a keyboard macro,

00:03:04.216 --> 00:03:10.240
go get the S-expression wherever the cursor is,

00:03:10.290 --> 00:03:13.611
move it to the next S-expression, paste it in here,

00:03:13.661 --> 00:03:15.989
and the keyboard macro, alright.

00:03:20.190 --> 00:03:21.652
What was basically happening...

NOTE Calendar

00:03:21.702 --> 00:03:25.306
So the demo application in this case is

00:03:25.356 --> 00:03:27.842
a scheduling application.

00:03:28.792 --> 00:03:34.631
And so you can imagine improvising a calendar like this

00:03:34.731 --> 00:03:36.633
if you weren't existingly using an

00:03:36.683 --> 00:03:38.952
org-mode integrated calendar.

00:03:39.252 --> 00:03:42.739
So, you know, just have a list and maybe

00:03:43.989 --> 00:03:49.479
each list is of sublists, where the first element,

00:03:49.579 --> 00:03:51.531
the car of the sublist, I guess,

00:03:51.931 --> 00:03:55.669
is a sensible-sounding date format, like sat dec 7,

00:03:56.119 --> 00:03:59.589
this is clearly the first day of EmacsConf.

00:03:59.939 --> 00:04:06.230
Then after that, a list of basically from and to times,

00:04:06.280 --> 00:04:18.592
and then a quick written description of whatever event

00:04:18.642 --> 00:04:19.576
was at that time.

00:04:19.876 --> 00:04:23.130
So similar to what we were seeing recently

00:04:23.180 --> 00:04:25.449
as of me recording this in our Structure and

00:04:25.499 --> 00:04:28.902
Interpretation of Computer Programs study group,

00:04:29.902 --> 00:04:35.709
you have all these little methods, sorry, defuns.

00:04:35.759 --> 00:04:39.513
So hours of a time is actually just car of the time,

00:04:40.763 --> 00:04:42.966
minutes of the time, cadr of time.

00:04:43.216 --> 00:04:46.719
And so we all know these kind of little,

00:04:46.720 --> 00:04:50.173
basically turning car and cadr,

00:04:50.223 --> 00:04:53.093
which we do use into descriptive names like from and to.

00:04:53.143 --> 00:04:56.329
You can see how this lines up with the appointments.

00:04:56.379 --> 00:05:01.701
So for "reproducibly building emacs" described event,

00:05:01.751 --> 00:05:05.679
I guess from would be 16 25.

00:05:05.680 --> 00:05:07.658
Let's just grab that and literally use those,

00:05:07.708 --> 00:05:11.011
because we're in interactive programming, right?

00:05:11.711 --> 00:05:18.216
So we have this quoted form and then

00:05:18.266 --> 00:05:22.240
let's do from of that. Great.

00:05:22.290 --> 00:05:24.458
And let's do minutes.

00:05:25.308 --> 00:05:31.498
So from 16 25, 16 45 reproducibly building Emacs,

00:05:31.548 --> 00:05:34.801
so when I do from to that I get 16 25

00:05:34.851 --> 00:05:40.707
when I do minutes to 16 25, I get 25

00:05:42.057 --> 00:05:45.489
and a description of this number in octal and hexadecimal,

00:05:45.590 --> 00:05:47.147
out of interest.

00:05:48.047 --> 00:05:52.953
What was up next? Okay, getting nextday from a date.

00:05:53.553 --> 00:05:57.023
And so, what was it? Saturday, December 27th.

00:05:57.073 --> 00:05:59.245
So what happens? So this function,

00:06:01.345 --> 00:06:05.932
I guess this was part of Sandewall's personal exploration of

00:06:06.032 --> 00:06:12.188
atoms using symbols with property lists,

00:06:12.238 --> 00:06:15.492
with an attached property list to every symbol.

00:06:15.942 --> 00:06:18.059
And so you can kind of see how this works.

00:06:21.510 --> 00:06:32.799
I get whatever was keyed by the symbol nextweekday out of

00:06:32.800 --> 00:06:36.529
the day symbol, so I guess we would do...

00:06:36.579 --> 00:06:46.640
so if we had Sat December 6, I guess we get Sat

00:06:48.140 --> 00:06:54.379
Oh, it destructures that itself somehow, get nextweekday.

00:06:58.429 --> 00:07:02.279
Sorry, nextday.

00:07:02.280 --> 00:07:10.170
set December 6th

00:07:10.270 --> 00:07:12.732
But of course these weren't defined yet,

00:07:12.782 --> 00:07:23.593
so what if we (setf (get 'sat 'nextweekday) 'sun)

00:07:23.793 --> 00:07:25.239
and it's going to be sun right

00:07:28.990 --> 00:07:34.369
And then if we (nextday '(sat dec 6)),

00:07:36.720 --> 00:07:38.392
Sunday, December 7th.

00:07:38.942 --> 00:07:41.628
Yeah, so that kind of makes sense in an intuitive way.

00:07:41.778 --> 00:07:43.550
And you can see cond is being used.

00:07:43.850 --> 00:07:51.538
So if you're not a Lisp person, cond evaluates a condition.

00:07:51.588 --> 00:07:54.708
If the condition returns non-nil,

00:07:54.758 --> 00:07:57.160
it executes the code attached to that condition.

00:07:57.210 --> 00:08:00.247
It evaluates each of these conditions in terms

00:08:00.297 --> 00:08:04.851
and then stops when one returns non-nil,

00:08:04.901 --> 00:08:07.420
as you might expect as a Lisp programmer.

00:08:07.670 --> 00:08:09.622
Let's keep marching along.

00:08:09.672 --> 00:08:11.691
Okay, so we have to populate this.

00:08:11.741 --> 00:08:13.493
And this was an interesting one.

00:08:13.543 --> 00:08:15.562
As you can see, this was my attempt.

00:08:15.612 --> 00:08:19.199
let days mon tue wed, so forth.

00:08:19.249 --> 00:08:23.403
Then I used cl-loop from Emacs Lisp, you know,

00:08:23.453 --> 00:08:32.359
for day in days, for nex in cdr days, setf nextweekday.

00:08:32.610 --> 00:08:35.382
This I thought was surprisingly deep, was that

00:08:36.432 --> 00:08:38.435
Lisp programmers, Sandewall thought,

00:08:38.485 --> 00:08:43.123
are basically constantly trying to break new ground and they

00:08:43.723 --> 00:08:45.325
lose interest very quickly when

00:08:45.425 --> 00:08:46.743
that's not what they're doing.

00:08:46.843 --> 00:08:50.129
And he was making the point that breaking new ground

00:08:50.229 --> 00:08:52.832
involves solving a problem exactly once.

00:08:54.232 --> 00:08:55.903
And so you kind of should do it

00:08:56.003 --> 00:08:58.572
in the most expedient way possible

00:08:58.672 --> 00:09:00.692
to solve it kind of for the first time.

00:09:00.992 --> 00:09:07.113
And after that, it's not breaking new ground anymore.

00:09:07.663 --> 00:09:10.350
And so I thought that was a kind of fun point

00:09:10.400 --> 00:09:14.521
from Sandewall's paper that we could kind of chew on.

00:09:15.121 --> 00:09:19.359
Do something similar for months and month lengths.

00:09:21.059 --> 00:09:30.120
But as we can see, if I get Feb,

00:09:32.120 --> 00:09:41.982
nrdr days I think that characters in 1978,

00:09:42.082 --> 00:09:45.418
characters and symbols in 1978 were more precious

00:09:45.468 --> 00:09:47.821
which is why Sandewall was writing in this abbreviated way.

00:09:47.871 --> 00:09:51.341
We might write more verbosely these days.

00:09:51.391 --> 00:09:54.177
We get 28th but it is in fact a leap year,

00:09:54.277 --> 00:09:58.998
so let's just set that to be 29. Now, it's fine.

00:09:59.048 --> 00:10:05.638
Then I guess if we do, we were doing nextday.

00:10:06.038 --> 00:10:10.944
Sure, nextday. I don't know what day it was.

00:10:10.994 --> 00:10:16.716
let's assume it was Sunday Feb 28

00:10:22.716 --> 00:10:26.179
some kind of typo crept in there, but we can deal with it.

00:10:26.230 --> 00:10:43.844
Let's just setf get sun nextweekday month.

00:10:43.944 --> 00:10:48.648
So once we've solved it once, we should be fine, right?

00:10:48.698 --> 00:10:51.849
Okay, I think I have to speed up a bit.

00:10:52.000 --> 00:10:55.555
Yeah, there we go, Monday February 29th and then

00:10:55.605 --> 00:11:01.194
what if we nextday... Tuesday March 1st.

00:11:01.244 --> 00:11:03.700
Okay, it basically seems to be working.

00:11:03.750 --> 00:11:06.398
And I'd argue it's actually fun that

00:11:06.448 --> 00:11:08.535
we kind of had to do that.

00:11:08.585 --> 00:11:13.473
Then, just imagining you're kind of functioning with

00:11:13.523 --> 00:11:18.311
other people: day begins at 900, ends at 1700,

00:11:18.361 --> 00:11:22.415
so kind of a 9:00 to 5:00, imagining... though perhaps,

00:11:23.915 --> 00:11:25.318
shorter is more appropriate,

00:11:25.368 --> 00:11:27.320
but let's not quibble about that.

00:11:27.420 --> 00:11:29.789
What was happening next in this app?

00:11:29.839 --> 00:11:35.562
holesin. Oh, yeah, so get me the free slots

00:11:35.712 --> 00:11:40.060
in from start time, list of appointments,

00:11:40.110 --> 00:11:45.238
remembering characters, I guess, were more precious

00:11:45.288 --> 00:11:46.556
is my theory.

00:11:46.906 --> 00:11:58.399
And then an end time, so if I go holesin and then from 900

00:11:59.050 --> 00:12:02.689
to where I've written 9 space 0 0

00:12:03.389 --> 00:12:05.959
and what are my appointments?

00:12:07.459 --> 00:12:09.930
So I have a list of appointments that

00:12:09.980 --> 00:12:12.032
need to have start times and finish times.

00:12:12.132 --> 00:12:20.080
So let's go 12 59 to 13 01.

00:12:23.580 --> 00:12:25.379
The description, I guess, can be test.

00:12:25.429 --> 00:12:28.715
And then let's finish that at 1700,

00:12:29.115 --> 00:12:30.729
like we kind of thought.

00:12:37.830 --> 00:12:41.049
Oops, some of these were meant to be arguments.

00:12:54.800 --> 00:12:55.742
Okay, there we go.

00:12:56.042 --> 00:13:01.297
And so the available times between nine and five,

00:13:01.397 --> 00:13:04.429
if there's one appointment from 12:59 to 1:01,

00:13:05.580 --> 00:13:13.092
are from 9 to 1259 and from 1 past 1 to 5,

00:13:13.792 --> 00:13:15.528
which is kind of what we were expecting.

00:13:15.728 --> 00:13:18.631
And so our appointment app is kind of coming along.

00:13:18.931 --> 00:13:20.379
Then we have commonholes,

00:13:23.430 --> 00:13:25.189
kind of works in a sensible way.

00:13:33.890 --> 00:13:36.199
Do we have, what are some common holes from

00:13:36.249 --> 00:13:40.153
two lists of appointments, I guess? commonholes.

00:13:50.903 --> 00:13:53.800
I guess, let's have one person have an appointment

00:13:53.850 --> 00:13:56.419
from 9 to 10.

00:13:59.569 --> 00:14:02.508
That's going to be their only appointment.

00:14:02.558 --> 00:14:05.946
And let's have another person have an appointment

00:14:05.996 --> 00:14:10.399
from 9.30 to 10.30.

00:14:14.450 --> 00:14:18.391
Oh, we didn't do beforetime yet.

00:14:18.441 --> 00:14:22.295
This is the order Sandewall implemented things in there.

00:14:22.345 --> 00:14:25.465
Ok, there.

00:14:25.515 --> 00:14:32.972
So the commonholes are from 9.30 to 10.

00:14:34.272 --> 00:14:42.182
That's actually the overlapping time.

00:14:42.232 --> 00:14:43.850
I'll re-record this

00:14:43.900 --> 00:14:50.799
if some kind of gremlin has crept into my things.

00:15:04.150 --> 00:15:06.839
And then, what did we first start by doing?

00:15:07.389 --> 00:15:11.444
Emacsconf track 1 and Emacsconf track 2,

00:15:11.494 --> 00:15:15.139
and we'll do it on Saturday, December 7.

00:15:21.140 --> 00:15:23.756
commontime, and I'm being prompted, of course.

00:15:23.806 --> 00:15:27.226
emacsconf-track-1,

00:15:29.676 --> 00:15:32.369
emacsconf-track-2,

00:15:33.370 --> 00:15:35.485
the date that we grabbed.

00:15:36.585 --> 00:15:39.519
Let's say we want five minutes.

00:15:49.120 --> 00:15:52.418
And, as always, I'm forgetting to quote things.

00:15:52.468 --> 00:15:56.689
And so my opportunities for common times

00:15:56.739 --> 00:16:01.326
between emacsconf-track-1 to emacsconf-track-2

00:16:01.376 --> 00:16:02.819
on the first day of the conference,

00:16:02.869 --> 00:16:05.464
Saturday, December 7th, for five minutes.

00:16:05.864 --> 00:16:08.834
I have an opportunity between 9:20 and 9:40,

00:16:08.984 --> 00:16:13.639
11:40 and 13:00, 13:10, 13:25,

00:16:14.489 --> 00:16:19.612
13:45, 13:55, and 16:15 to 16:25.

00:16:19.662 --> 00:16:21.581
So I think that's all working.

00:16:23.881 --> 00:16:29.155
And it was done with almost exactly the same code,

00:16:31.355 --> 00:16:38.965
except for my loopy populating symbol plist bits earlier,

00:16:39.015 --> 00:16:46.489
in Interlisp in 1978 as Emacs Lisp in 2024.

00:16:46.539 --> 00:16:48.625
I thought that was pretty interesting that

00:16:48.675 --> 00:16:49.492
you could do that.

00:16:49.542 --> 00:16:57.049
I think setf get was put prop

00:16:57.099 --> 00:17:02.737
was the biggest difference in 1978 Interlisp and

00:17:02.787 --> 00:17:04.690
modern Emacs Lisp.

00:17:04.740 --> 00:17:09.178
And obviously modern Emacs Lisp has Common Lisp loop in it,

00:17:09.228 --> 00:17:11.514
kind of. It's another big difference.

00:17:11.914 --> 00:17:16.099
How are we doing for time? 17 minutes?

00:17:16.450 --> 00:17:19.105
Great, because now let's get a bit closer to what

00:17:19.805 --> 00:17:21.941
my talk title has promised.

00:17:21.991 --> 00:17:24.911
We're going to use Common Lisp from Emacs Lisp

00:17:25.211 --> 00:17:28.080
and we're going to use Common Lisp Interface Manager,

00:17:28.130 --> 00:17:30.116
the open source implementation of

00:17:30.166 --> 00:17:33.786
Common Lisp Interface Manager from

00:17:33.836 --> 00:17:36.999
Inferior Lisp from Emacs Lisp.

00:17:37.300 --> 00:17:42.695
And what we just went over in Sandewall's paper...

00:17:45.550 --> 00:17:47.683
This was the starting point, and Sandewall says,

00:17:47.783 --> 00:17:51.537
well, if you just kind of jammed these ideas using lists

00:17:51.587 --> 00:17:55.341
you just kind of made hour a synonym for car and

00:17:55.391 --> 00:17:58.945
minute a synonym for cadr and so forth,

00:17:59.795 --> 00:18:03.066
you can then make a whole bunch of improvements.

00:18:04.166 --> 00:18:06.689
Can we actually get a glimpse of the …

00:18:09.239 --> 00:18:10.423
No, I didn't really write this.

00:18:10.473 --> 00:18:14.244
Yeah, so you can see I basically just ported out of

00:18:14.294 --> 00:18:17.430
the Interlisp code in the text of this paper.

00:18:17.580 --> 00:18:21.300
Yeah, you see from lambda of appointment,

00:18:21.350 --> 00:18:24.904
car appointment, and that's defineq.

00:18:24.954 --> 00:18:27.857
I guess Interlisp in 1978 used defineq,

00:18:27.907 --> 00:18:32.311
whereas we used defun in Elisp,

00:18:33.361 --> 00:18:35.264
but I'd encourage you to explore that yourselves.

00:18:35.314 --> 00:18:38.551
But the extension that I wanted to develop today is,

00:18:38.601 --> 00:18:41.187
since Common Lisp interface manager is just Lisp,

00:18:41.237 --> 00:18:43.422
and Inferior Lisp is just Lisp,

00:18:43.472 --> 00:18:45.291
and Emacs Lisp is just Lisp,

00:18:45.341 --> 00:18:48.878
and Interlisp in 1978 is just Lisp,

00:18:48.928 --> 00:18:50.112
we could kind of use them all together

00:18:50.162 --> 00:18:51.430
in a kind of funky way.

00:18:51.480 --> 00:18:57.036
But a complaint you often hear, or a statement of confusion

00:18:57.086 --> 00:19:00.373
that you often hear as a Lisp person,

00:19:00.423 --> 00:19:06.712
people from other languages often can't kind of figure out

00:19:06.762 --> 00:19:09.715
how to get started with Lisp.

00:19:09.765 --> 00:19:12.752
So I'm just going to try and do everything here and now.

NOTE Inferior Lisp and McCLIM

00:19:12.802 --> 00:19:18.357
And this is the second piece of my three pieces,

00:19:18.407 --> 00:19:22.628
so let's just get inferior lisp and McCLIM happening.

00:19:22.928 --> 00:19:29.268
We normally use slime Superior Lisp Interaction Mode

00:19:29.318 --> 00:19:41.847
for Emacs, which we could get from Emacs's package manager.

00:19:41.897 --> 00:19:48.604
If I go package-list-packages and then we wait

00:19:48.954 --> 00:19:50.969
for however fast my network is,

00:19:54.270 --> 00:19:55.311
we're going to see how long it's been

00:19:55.361 --> 00:19:57.079
since I updated something.

00:19:58.629 --> 00:20:00.999
Yeah, but you see it's in MELPA, right?

00:20:15.750 --> 00:20:16.982
It's got to be somewhere here.

00:20:18.082 --> 00:20:18.501
Yeah, all right.

00:20:18.551 --> 00:20:21.420
So there's one slime available in non-GNU,

00:20:21.470 --> 00:20:25.958
one possibly more recent one available in

00:20:26.758 --> 00:20:28.678
the MELPA package repositories.

00:20:30.278 --> 00:20:30.963
I don't know about you,

00:20:31.013 --> 00:20:33.616
but I can sympathize with people who feel confused by this

00:20:33.666 --> 00:20:36.469
because I think if you look online,

00:20:36.519 --> 00:20:38.637
if you found a search engine result,

00:20:38.687 --> 00:20:44.110
it doesn't customize these like I customize them.

00:20:44.160 --> 00:20:45.945
So let's just do this.

00:20:45.995 --> 00:20:48.147
customize-variable.

00:20:48.197 --> 00:20:52.318
package-archives.

00:20:54.518 --> 00:21:00.392
Yeah and you can see here I've ins-ed gnu, non-gnu, and melpa.

00:21:00.542 --> 00:21:02.628
It's melpa, not melpa stable.

00:21:03.478 --> 00:21:04.359
Saved and set.

00:21:06.810 --> 00:21:08.334
So that's how I do that.

00:21:08.384 --> 00:21:13.279
We're just kind of totally get an inferior lisp working.

00:21:17.130 --> 00:21:20.599
And then, like we saw before,

00:21:20.949 --> 00:21:22.648
this might occur, buffer is still open.

00:21:23.148 --> 00:21:26.879
No. occur. slime.

00:21:35.330 --> 00:21:38.631
So we could install and execute that.

00:21:38.681 --> 00:21:41.734
I cloned slime since Common Lisp people

00:21:41.784 --> 00:21:48.874
might use slime and swank, like McCLIM does outside of Emacs.

00:21:48.924 --> 00:21:51.043
I think we can secretly see

00:21:51.093 --> 00:21:55.614
I've got it here already in my ~/.emacs.d/slime,

00:21:56.064 --> 00:22:00.119
but you could install it like that. The hotkey is I,

00:22:00.169 --> 00:22:02.021
and then execute the install.

00:22:02.071 --> 00:22:03.789
I realize you all already know this,

00:22:04.039 --> 00:22:06.759
but many of you might not yet be Common Lisp programmers,

00:22:06.809 --> 00:22:09.569
and you can use both Superior and Inferior Lisp.

00:22:10.370 --> 00:22:14.733
Instead of package archives in Emacs,

00:22:14.783 --> 00:22:19.705
in Inferior Lisp, and with great controversy,

00:22:19.755 --> 00:22:22.708
we often use QuickLisp.

00:22:22.758 --> 00:22:25.778
People sometimes complain QuickLisp releases

00:22:25.828 --> 00:22:27.780
get out of date quite quickly.

00:22:28.630 --> 00:22:31.717
But there was one recently, so they're pretty in date.

00:22:32.167 --> 00:22:35.321
And so there should be a …

00:22:35.371 --> 00:22:39.058
Yeah, so basically you download quicklisp.lisp,

00:22:39.158 --> 00:22:41.959
you check its SHA-256 sum.

00:22:44.910 --> 00:22:47.829
So I guess in, ooh, over here, if I press D,

00:22:51.229 --> 00:22:53.706
~screwtape/downloads/quicklisp.lisp.

00:22:53.756 --> 00:22:57.376
Great. If we open a shell, we realize that

00:22:57.426 --> 00:23:03.599
this is just riveting for all of our experts there.

00:23:06.450 --> 00:23:15.919
sha256 ~/Downloads/quicklisp.lisp

00:23:19.720 --> 00:23:22.001
Your SHA-256 might be different to mine.

00:23:22.601 --> 00:23:25.471
Somebody please tell me during the actual conference,

00:23:25.621 --> 00:23:29.125
the right Emacs idiom to do this.

00:23:29.225 --> 00:23:32.679
I guess it will be tools decrypt something.

00:23:33.930 --> 00:23:36.014
But then once you have QuickLisp,

00:23:36.214 --> 00:23:37.959
let's go back to our shell.

00:23:41.810 --> 00:23:46.625
I'm on OpenBSD or a BSD, so I type pkg_add ecl

00:23:46.675 --> 00:23:48.239
to get embeddable Common Lisp.

00:23:53.590 --> 00:23:55.090
However, I think I already have it.

00:23:58.540 --> 00:24:02.069
And then, let's start ecl [embeddable Common Lisp].

00:24:04.670 --> 00:24:07.039
Quicklisp inserts itself already,

00:24:11.290 --> 00:24:14.920
but just loading that quicklisp.lisp file

00:24:14.970 --> 00:24:16.989
kind of would get you here anyway.

00:24:17.039 --> 00:24:22.939
Now let's (ql:quickload :mcclim)

00:24:27.140 --> 00:24:28.769
I'm actually following my plan.

00:24:33.269 --> 00:24:38.294
McCLIM is kind of like Emacs in that

00:24:38.344 --> 00:24:40.713
it handles all sorts of graphical stuff

00:24:41.013 --> 00:24:42.748
like if you have GTK Emacs,

00:24:42.798 --> 00:24:44.333
you know, it can do really a lot of stuff.

00:24:44.383 --> 00:24:48.604
That's the reason why compiling and loading

00:24:50.504 --> 00:24:53.125
the McCLIM package takes a while

00:24:53.175 --> 00:24:56.962
and I do use my old computer laptop,

00:24:57.012 --> 00:25:00.299
my old computer challenge laptop all the time,

00:25:00.349 --> 00:25:02.685
so when you do this, your computer will be

00:25:02.735 --> 00:25:05.504
substantially faster or you will be feeling

00:25:06.404 --> 00:25:09.609
smug about how slow you're running your computer,

00:25:12.759 --> 00:25:15.481
and so there's no problem.

00:25:15.531 --> 00:25:17.316
Next, we're going to start Swank

00:25:17.366 --> 00:25:23.699
which is the Lisp part of Emacs's slime-mode.

00:25:27.649 --> 00:25:32.298
If I remember, we looked at this before

00:25:32.348 --> 00:25:34.700
in my .emacs.d,

00:25:35.050 --> 00:25:40.499
and then there should be slime/start-swank.lisp.

00:25:41.699 --> 00:25:44.643
Let's just grab that and load it.

00:25:46.293 --> 00:25:48.470
You don't have to type #p.

00:25:48.870 --> 00:25:51.150
The string will work as well as the path name,

00:25:51.200 --> 00:25:52.618
but it's kind of fun to do so.

00:25:52.668 --> 00:25:55.287
Now, slime-connect.

00:25:55.437 --> 00:25:56.859
You installed slime earlier.

00:25:57.559 --> 00:25:59.024
Default values.

00:25:59.074 --> 00:26:01.794
Oh, the reason we had to do this

00:26:01.844 --> 00:26:07.064
and not use Slime, start Slime normally was that

00:26:07.214 --> 00:26:10.736
CLIM, Common Lisp Interface Manager,

00:26:10.786 --> 00:26:12.204
McCLIM, implementation of the

00:26:12.254 --> 00:26:16.275
Common Lisp Interface Manager spec, also uses Swank

00:26:16.675 --> 00:26:19.912
and McCLIM really wants to be in the driver's seat.

00:26:20.162 --> 00:26:23.282
We can have our Emacs connect

00:26:23.332 --> 00:26:28.053
to the Swank that McCLIM was already running.

00:26:28.553 --> 00:26:30.823
So (in-package :clim-user)...

00:26:30.873 --> 00:26:33.575
I don't know if this is an Emacs Lisp idiom,

00:26:33.625 --> 00:26:37.112
so you normally have user packages to absorb user code

00:26:37.162 --> 00:26:38.809
just for fooling around.

00:26:42.110 --> 00:26:47.556
Then let's just use clim at all.

00:26:47.606 --> 00:26:52.544
And so clim does lots of things around presentations

00:26:53.094 --> 00:26:58.317
but one thing it does is automatically, to a large extent,

00:26:58.367 --> 00:27:03.956
generate graphical user interfaces.

00:27:04.006 --> 00:27:09.394
So there's a lot to it that's not about graphics,

00:27:09.444 --> 00:27:13.332
kind of richer interfaces than

00:27:14.082 --> 00:27:17.089
ANSI Common Lisp's character streams.

00:27:19.389 --> 00:27:23.158
But let's use it to grab a graphical pop-up

00:27:23.258 --> 00:27:24.393
that returns a value.

00:27:24.743 --> 00:27:27.413
So we have this accepting-values tool.

00:27:28.013 --> 00:27:31.950
t is just choose whatever you think the default stream is.

00:27:32.150 --> 00:27:36.088
We're going to want our own window, t as well,

00:27:36.688 --> 00:27:38.390
so just a keyword argument.

00:27:38.640 --> 00:27:43.328
After that, we can just write things like (accept 'string).

00:27:44.978 --> 00:27:47.883
And I'm just doing this in, for people only listening,

00:27:47.933 --> 00:27:50.936
I've just done this in my ECL Slime REPL.

00:27:50.986 --> 00:27:55.279
Enter a string: foo bar baz.

00:27:56.530 --> 00:27:58.849
Alt-Enter, I think, is the shortcut to return that.

00:27:59.150 --> 00:28:02.248
As you can see, its first return was "foo bar baz",

00:28:02.298 --> 00:28:04.316
and it had second and third returns,

00:28:04.366 --> 00:28:10.105
string being the type, presentation type that it got,

00:28:10.155 --> 00:28:13.739
and t, I think, to indicate that it exited successfully.

00:28:16.189 --> 00:28:20.916
We can also do expression, might be more interesting.

00:28:20.966 --> 00:28:22.189
There are a whole bunch of these.

00:28:26.140 --> 00:28:29.308
clim's examples folder is pretty good.

00:28:29.358 --> 00:28:32.294
And the documentation, everybody's documentation,

00:28:32.344 --> 00:28:34.713
McCLIM's documentation, you know,

00:28:35.413 --> 00:28:39.135
lispworks and franz's documentation will tell you about

00:28:39.385 --> 00:28:45.069
the different default presentation types you can accept.

00:28:47.969 --> 00:28:49.862
And they're pretty intuitive and

00:28:49.912 --> 00:28:52.464
they're kind of very Lisp compatible,

00:28:53.264 --> 00:28:54.266
which everything kind of is.

00:28:54.316 --> 00:29:00.399
We could accept an expression. foo bar baz.

00:29:05.350 --> 00:29:09.414
return "foo bar baz 123" because that's what I wrote.

NOTE Putting things together

00:29:10.964 --> 00:29:14.570
And now the kind of pièce de résistance

00:29:14.820 --> 00:29:17.055
of putting those things together.

00:29:17.555 --> 00:29:21.326
Let's just run over to our scratch buffer.

00:29:22.226 --> 00:29:24.096
Oh, or our ielm thing, right?

00:29:24.246 --> 00:29:25.564
That's what we're actually doing.

00:29:25.964 --> 00:29:28.951
So now, I think we're going to do, hang on,

00:29:30.251 --> 00:29:32.871
M-:.

00:29:32.921 --> 00:29:41.313
I think it's going to be slime-eval-save,

00:29:41.663 --> 00:29:44.549
because it'll eval asynchronously.

00:29:45.849 --> 00:29:49.121
There's a question of how to get the value out of

00:29:49.171 --> 00:29:52.124
the asynchronous slime evaluation.

00:29:52.524 --> 00:29:55.761
And so an obvious way is to yank it.

00:29:59.661 --> 00:30:02.234
I'm looking forward to learning from kickingvegas.

00:30:02.284 --> 00:30:06.638
He finally makes it stick in my brain how to...

00:30:07.138 --> 00:30:10.559
Options other than just starting to write a command

00:30:10.959 --> 00:30:14.013
in the minibuffer and then mashing Tab

00:30:14.063 --> 00:30:16.465
and looking at what the likely sounding results are.

00:30:16.765 --> 00:30:21.720
That's what I will be learning this conference, if anything.

00:30:22.070 --> 00:30:26.558
slime-eval-save and then, what was happening over here?

00:30:26.958 --> 00:30:31.769
(accepting-values (t :own-window t) (accept 'expression)).

00:30:33.369 --> 00:30:38.319
Kind of confusingly, slime-eval-save takes a string,

00:30:39.169 --> 00:30:41.840
but it works really well, I'd argue.

00:30:41.940 --> 00:30:43.475
I think it's about to work really well.

00:30:45.825 --> 00:30:46.712
Ok, this is great.

00:30:46.762 --> 00:30:50.182
So you can see I've used... Oh, please enter an expression.

00:30:50.882 --> 00:30:52.117
So we're doing an expression.

00:30:52.167 --> 00:30:59.057
And so let's write like (3 4/2 5/6).

00:30:59.357 --> 00:31:00.609
That's an expression, right?

00:31:03.609 --> 00:31:05.797
And now I've yanked it.

00:31:05.847 --> 00:31:07.379
I should have really been in *scratch*,

00:31:07.429 --> 00:31:11.770
because I just want to dump. Sorry, *scratch*.

00:31:13.770 --> 00:31:15.290
Dump whatever that did.

00:31:16.540 --> 00:31:19.480
Oh, yeah, and see, it kind of simplified that

00:31:19.530 --> 00:31:20.712
while it was reading it.

00:31:20.762 --> 00:31:27.486
So running in Elisp, (slime-eval-save "(accepting-values

00:31:27.536 --> 00:31:32.557
(t :own-window t), accept 'expression))").

00:31:33.157 --> 00:31:36.328
And I got the three returns all the way

00:31:36.378 --> 00:31:43.235
from McCLIM to inferior-lisp to Elisp.

00:31:43.885 --> 00:31:46.471
3, 2, 5/6.

00:31:46.521 --> 00:31:50.575
Second return, it was a presentation type of expression,

00:31:50.625 --> 00:31:52.811
returned successfully, key.

00:31:53.561 --> 00:31:56.679
Yeah, so that's pretty cool. How out of time are we?

00:31:58.530 --> 00:32:00.859
Oh, we did this in basically half an hour.

00:32:04.009 --> 00:32:07.209
Maybe we could stop and assume there will be a bit under

00:32:07.259 --> 00:32:10.796
half an hour of interactive discussion.

00:32:11.646 --> 00:32:15.634
And when I also watch this and also receive suggestions,

00:32:16.934 --> 00:32:20.071
there will be additions and changes to make, I think.

00:32:21.671 --> 00:32:32.617
But can we quickly defun accept-date?

00:32:34.667 --> 00:32:36.589
Let's not take any arguments.

00:32:40.239 --> 00:32:41.709
Let's do this.

00:32:48.409 --> 00:32:49.769
accept-date.

00:32:55.519 --> 00:32:58.099
Saturday, December 7th.

00:33:02.649 --> 00:33:04.015
Oh, it doesn't return it.

00:33:04.215 --> 00:33:05.518
But it gets pretty close to returning it.

00:33:05.568 --> 00:33:06.718
I guess it yanks it.

00:33:07.818 --> 00:33:18.359
I would have to call … I'm going to say it's slime-sync,

00:33:21.160 --> 00:33:25.370
at which point it will resolve the slime-eval-async.

00:33:25.670 --> 00:33:28.757
And then I could get the first value out of

00:33:28.807 --> 00:33:32.179
my kill-ring pop, right?

00:33:33.629 --> 00:33:35.046
Does kill ring pop what I want?

00:33:36.646 --> 00:33:39.918
You know, you tell me in the interactive feedback,

00:33:39.968 --> 00:33:41.720
which should be happening in LambdaMOO.

00:33:43.020 --> 00:33:45.559
Well, I'll record another draft of this later

00:33:45.609 --> 00:33:47.709
after getting some feedback.

00:33:48.609 --> 00:33:51.580
Or worse to worse, this is just what you'll hear.

00:33:52.180 --> 00:33:58.503
Hopefully we have some questions and constructive comments.

00:33:59.303 --> 00:34:00.255
All right, see everybody later.

00:34:00.305 --> 00:34:01.359
Thank you so much for being here.

00:34:04.470 --> 00:34:08.613
Please do drop in to my live show

00:34:08.663 --> 00:34:11.266
if you're so inclined on Wednesdays

00:34:11.316 --> 00:34:14.186
where you're not busy with cool Emacs conferences.

00:34:14.936 --> 00:34:16.169
The list be go for climate.

00:34:16.219 --> 00:34:17.572
Thank you to everyone who helped.

00:34:17.772 --> 00:34:18.640
Thank you to Sacha

00:34:18.690 --> 00:34:21.660
for really getting me into this conference

00:34:21.710 --> 00:34:23.428
and kind of getting me going and

00:34:23.678 --> 00:34:26.181
submitting this prerecorded talk.

00:34:27.981 --> 00:34:28.780
See everybody later.