summaryrefslogtreecommitdiffstats
path: root/2023/captions
diff options
context:
space:
mode:
authorEmacsConf <emacsconf-org@gnu.org>2023-12-03 10:00:22 -0500
committerEmacsConf <emacsconf-org@gnu.org>2023-12-03 10:00:22 -0500
commit543807855cd595c6ceffbcc07bf4b7fab67e396a (patch)
tree511aed481e3be1580055f75a0673660e18348812 /2023/captions
parentd7fbb95609564e8df79a8874dba2fa70681667f4 (diff)
downloademacsconf-wiki-543807855cd595c6ceffbcc07bf4b7fab67e396a.tar.xz
emacsconf-wiki-543807855cd595c6ceffbcc07bf4b7fab67e396a.zip
Automated commit
Diffstat (limited to '2023/captions')
-rw-r--r--2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main--chapters.vtt68
-rw-r--r--2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main.vtt1044
2 files changed, 1112 insertions, 0 deletions
diff --git a/2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main--chapters.vtt b/2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main--chapters.vtt
new file mode 100644
index 00000000..15bb79f9
--- /dev/null
+++ b/2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main--chapters.vtt
@@ -0,0 +1,68 @@
+WEBVTT
+
+
+00:00:02.120 --> 00:00:23.279
+Introduction
+
+00:00:23.280 --> 00:01:18.179
+Interactive development
+
+00:01:18.180 --> 00:02:53.719
+REPL: Read Eval Print Loop
+
+00:02:53.720 --> 00:04:07.599
+Long-lasting loops
+
+00:04:07.600 --> 00:05:23.159
+Not interruptible
+
+00:05:23.160 --> 00:05:51.479
+No protocol
+
+00:05:51.480 --> 00:07:25.859
+Not scalable
+
+00:07:25.860 --> 00:09:01.739
+nREPL
+
+00:09:01.740 --> 00:10:34.179
+Arei, Ares, and how to try
+
+00:10:34.180 --> 00:11:27.639
+Demo
+
+00:11:27.640 --> 00:12:32.459
+Continuations
+
+00:12:32.460 --> 00:13:33.419
+Reading from stdin
+
+00:13:33.420 --> 00:15:13.159
+Fancy example with continuations
+
+00:15:13.160 --> 00:17:42.059
+Guix API
+
+00:17:42.060 --> 00:17:57.019
+Support
+
+00:17:57.020 --> 00:18:46.219
+Future steps - Multiple simultaneous evaluations in different contexts
+
+00:18:46.220 --> 00:18:56.879
+Tree-sitter integration
+
+00:18:56.880 --> 00:19:22.759
+Full-fledged debugger
+
+00:19:22.760 --> 00:19:58.379
+FAQ - Does it support other Scheme implementations?
+
+00:19:58.380 --> 00:20:22.120
+Is it possible to use it with other text editors?
+
+00:20:22.121 --> 00:20:45.879
+Conclusion
+
+00:20:45.880 --> 00:21:00.680
+Contacts
diff --git a/2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main.vtt b/2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main.vtt
new file mode 100644
index 00000000..8423ba94
--- /dev/null
+++ b/2023/captions/emacsconf-2023-scheme--bringing-joy-to-scheme-programming--andrew-tropin--main.vtt
@@ -0,0 +1,1044 @@
+WEBVTT captioned by sachac, checked by sachac
+
+NOTE Introduction
+
+00:00:02.120 --> 00:00:07.399
+Hello and welcome everyone on EmacsConf 2023.
+
+00:00:07.400 --> 00:00:08.719
+I'm Andrew Tropin.
+
+00:00:08.720 --> 00:00:11.919
+I work on operating systems and programming languages.
+
+00:00:11.920 --> 00:00:16.639
+Today, we discuss Lisps, Schemes, REPLs,
+
+00:00:16.640 --> 00:00:18.139
+interactive development,
+
+00:00:18.140 --> 00:00:23.279
+and how to make your own cozy development environment.
+
+NOTE Interactive development
+
+00:00:23.280 --> 00:00:26.319
+Let's start from interactive development.
+
+00:00:26.320 --> 00:00:29.519
+Lisps are famous for a nice
+
+00:00:29.520 --> 00:00:32.479
+Interactive Development Experience.
+
+00:00:32.480 --> 00:00:33.999
+They have REPLs.
+
+00:00:34.000 --> 00:00:40.119
+Emacs Lisp has its own Lisp machine,
+
+00:00:40.120 --> 00:00:44.719
+and a lot of cool IDE with different functionality
+
+00:00:44.720 --> 00:00:47.879
+is already here and providing
+
+00:00:47.880 --> 00:00:51.619
+a nice and pleasant experience.
+
+00:00:51.620 --> 00:00:56.839
+The question is, is it enough?
+
+00:00:56.840 --> 00:00:59.920
+In most cases, yes, but for some languages,
+
+00:00:59.921 --> 00:01:04.839
+we have some white spaces, some missing pieces.
+
+00:01:04.840 --> 00:01:08.299
+And for example, in Scheme world,
+
+00:01:08.300 --> 00:01:10.879
+we already have a few tools.
+
+00:01:10.880 --> 00:01:14.599
+We have REPL, we have integration for REPL in Emacs,
+
+00:01:14.600 --> 00:01:16.679
+but is it enough?
+
+00:01:16.680 --> 00:01:18.179
+Let's see.
+
+NOTE REPL: Read Eval Print Loop
+
+00:01:18.180 --> 00:01:22.839
+We know that Emacs is very good for Lisps and REPL.
+
+00:01:22.840 --> 00:01:26.039
+Lisp and Emacs should be a perfect setup.
+
+00:01:26.040 --> 00:01:30.079
+But let's see how REPL basically works.
+
+00:01:30.080 --> 00:01:34.799
+It's an event loop which does three things.
+
+00:01:34.800 --> 00:01:37.279
+It reads an expression, it evaluates the expression,
+
+00:01:37.280 --> 00:01:40.739
+and it prints the result.
+
+00:01:40.740 --> 00:01:47.279
+We can take a simple expression, input it into REPL,
+
+00:01:47.280 --> 00:01:48.959
+and evaluate it and see the result.
+
+00:01:48.960 --> 00:01:50.819
+Very nice, very convenient.
+
+00:01:50.820 --> 00:01:55.339
+You can experiment and see immediately what is happening.
+
+00:01:55.340 --> 00:01:57.759
+You can even run a long-running process
+
+00:01:57.760 --> 00:01:58.919
+which does something.
+
+00:01:58.920 --> 00:02:07.199
+You can interrupt it and everything will be okay.
+
+00:02:07.200 --> 00:02:08.639
+But the problem appears
+
+00:02:08.640 --> 00:02:11.659
+when you start to develop a bigger project.
+
+00:02:11.660 --> 00:02:14.039
+And in most cases, you don't do
+
+00:02:14.240 --> 00:02:16.399
+your whole development in REPL.
+
+00:02:16.400 --> 00:02:18.460
+You do only a small part of it.
+
+00:02:18.461 --> 00:02:20.679
+In most cases, you just write
+
+00:02:20.680 --> 00:02:22.919
+the source code in text files,
+
+00:02:22.920 --> 00:02:26.399
+and after that, you run those snippets of code
+
+00:02:26.400 --> 00:02:30.520
+from those text files, or run the whole project.
+
+00:02:30.721 --> 00:02:33.719
+It's not very convenient to copy and paste
+
+00:02:33.720 --> 00:02:36.039
+every time the snippets of code to the REPL,
+
+00:02:36.040 --> 00:02:38.879
+see the result, modify the snippet of code,
+
+00:02:38.880 --> 00:02:41.199
+copy it again, and so on.
+
+00:02:41.200 --> 00:02:44.039
+So people invented some integration
+
+00:02:44.040 --> 00:02:46.079
+between REPL and your text editor.
+
+00:02:46.080 --> 00:02:51.599
+So you can evaluate expressions inside your text editor
+
+00:02:51.600 --> 00:02:53.719
+and see the result here.
+
+NOTE Long-lasting loops
+
+00:02:53.720 --> 00:02:56.679
+Works good so far, but what happens
+
+00:02:56.680 --> 00:03:02.299
+if we run a long-lasting loop,
+
+00:03:02.300 --> 00:03:04.999
+which does a lot of operations.
+
+00:03:05.000 --> 00:03:07.839
+As you can see here with a simple example,
+
+00:03:07.840 --> 00:03:13.599
+the output of the function,
+
+00:03:13.600 --> 00:03:16.759
+stdout of the function is presented here,
+
+00:03:16.760 --> 00:03:18.799
+and the resulting value is here.
+
+00:03:18.800 --> 00:03:22.359
+If you run a long-running process,
+
+00:03:22.360 --> 00:03:24.639
+you don't see anything happening.
+
+00:03:24.640 --> 00:03:29.259
+And you see there's a watch instead of my cursor.
+
+00:03:29.260 --> 00:03:33.719
+Maybe you don't see it, but nothing actually happens,
+
+00:03:33.720 --> 00:03:36.379
+at least from the point of view of the user.
+
+00:03:36.380 --> 00:03:38.399
+But if we interrupt the evaluation,
+
+00:03:38.400 --> 00:03:41.439
+we will see that some process in the background
+
+00:03:41.440 --> 00:03:44.239
+was launched, but we didn't see anything.
+
+00:03:44.240 --> 00:03:51.039
+Because the REPL is a single-threaded blocking process,
+
+00:03:51.040 --> 00:03:54.319
+which reads stdin and prints stdout,
+
+00:03:54.320 --> 00:03:55.679
+make the integration
+
+00:03:55.680 --> 00:03:58.540
+between the REPL and your text editor
+
+00:03:58.541 --> 00:04:02.919
+is not an easy task.
+
+00:04:02.920 --> 00:04:04.320
+And even if you do it,
+
+00:04:04.321 --> 00:04:07.599
+you have a lot of downsides, usually.
+
+NOTE Not interruptible
+
+00:04:07.600 --> 00:04:13.679
+First of all, the process is not interruptible.
+
+00:04:13.680 --> 00:04:18.479
+If you have a remote process which listens on the socket
+
+00:04:18.480 --> 00:04:21.939
+to which you connect from your development environment,
+
+00:04:21.940 --> 00:04:25.479
+and you run some infinite loop, for example,
+
+00:04:25.480 --> 00:04:28.299
+you can't interrupt it.
+
+00:04:28.300 --> 00:04:31.239
+Because interruption is done via signals,
+
+00:04:31.240 --> 00:04:35.039
+and signals to remote processes are not usually
+
+00:04:35.040 --> 00:04:38.759
+the thing in such integrations.
+
+NOT Output is not interactive
+
+00:04:38.760 --> 00:04:41.159
+Output is also not interactive.
+
+00:04:41.160 --> 00:04:45.319
+Usually, for example, here you can see
+
+00:04:45.320 --> 00:04:47.799
+when I evaluate the expression,
+
+00:04:47.800 --> 00:04:51.119
+the output is captured on the evaluation side,
+
+00:04:51.120 --> 00:04:53.719
+and after that, after the whole evaluation
+
+00:04:53.720 --> 00:04:56.179
+of the whole expression finished,
+
+00:04:56.180 --> 00:05:06.759
+I get the result, all the stdout at once.
+
+00:05:06.760 --> 00:05:09.919
+And if I run the process which evaluates for 5 seconds,
+
+00:05:09.920 --> 00:05:13.780
+I will see the first signs of the life
+
+00:05:13.781 --> 00:05:17.039
+only after 5 seconds of evaluation.
+
+00:05:17.040 --> 00:05:23.159
+Okay, what else?
+
+NOTE No protocol
+
+00:05:23.160 --> 00:05:26.119
+When you do such integrations, you have no protocol,
+
+00:05:26.120 --> 00:05:29.759
+you have just stdin and stdout.
+
+00:05:29.760 --> 00:05:32.919
+You print to stdin from your text editor.
+
+00:05:32.920 --> 00:05:36.679
+You read from stdout of the process.
+
+00:05:36.680 --> 00:05:40.339
+It's hard to tell if evaluation is finished,
+
+00:05:40.340 --> 00:05:47.319
+if it requires stdin, and how to extend the REPL
+
+00:05:47.320 --> 00:05:51.479
+to make it more featureful, and so on.
+
+NOTE Not scalable
+
+00:05:51.480 --> 00:05:57.359
+And also, such integrations are usually not very scalable.
+
+00:05:57.360 --> 00:06:14.699
+For example, if you want to have a completion,
+
+00:06:14.700 --> 00:06:17.460
+you type something, you have the completion. Cool.
+
+00:06:17.461 --> 00:06:22.039
+But if you run the process and at the same time
+
+00:06:22.040 --> 00:06:24.620
+try to have a completion, you don't have it,
+
+00:06:24.621 --> 00:06:29.799
+because the evaluation is in progress,
+
+00:06:29.800 --> 00:06:33.279
+and you can't calculate the completion candidates
+
+00:06:33.280 --> 00:06:35.519
+at the same time. To make it more obvious,
+
+00:06:35.520 --> 00:06:41.019
+I will start a completion here.
+
+00:06:41.020 --> 00:06:43.279
+You see the completion pop-ups.
+
+00:06:43.280 --> 00:06:46.159
+I start the evaluation process,
+
+00:06:46.160 --> 00:06:49.859
+and when I try to complete something,
+
+00:06:49.860 --> 00:06:53.119
+the evaluation freezes and there is no completion.
+
+00:06:53.120 --> 00:06:55.479
+Not very convenient.
+
+00:06:55.480 --> 00:06:58.119
+Usually, you have some long-running processes
+
+00:06:58.120 --> 00:07:01.399
+and you want them to continue while you have
+
+00:07:01.400 --> 00:07:08.579
+your go to definition, completion, and other things.
+
+00:07:08.580 --> 00:07:13.659
+Overall, those issues make it quite inconvenient
+
+00:07:13.660 --> 00:07:18.419
+to integrate REPL in text editors or development environments,
+
+00:07:18.420 --> 00:07:21.379
+so you need something else
+
+00:07:21.380 --> 00:07:25.859
+to make the work comfortable.
+
+NOTE nREPL
+
+00:07:25.860 --> 00:07:28.979
+There is already a solution called nREPL.
+
+00:07:28.980 --> 00:07:31.119
+It's a synchronous protocol which allows
+
+00:07:31.120 --> 00:07:34.019
+to send operations to the server
+
+00:07:34.020 --> 00:07:37.759
+and receive responses in a synchronous manner.
+
+00:07:37.760 --> 00:07:42.159
+And here is a simple example of a few operations.
+
+00:07:42.160 --> 00:07:45.079
+First one is cloning the existing session,
+
+00:07:45.080 --> 00:07:49.240
+and as a response you will get a new session.
+
+00:07:49.241 --> 00:07:52.099
+Also you send the evaluation request with code
+
+00:07:52.100 --> 00:07:55.639
+that you want to evaluate, and you get two responses.
+
+00:07:55.640 --> 00:08:00.600
+First one says that output is captured
+
+00:08:00.601 --> 00:08:02.839
+and it's equal to "hi\n",
+
+00:08:02.840 --> 00:08:06.560
+and after that, you receive an "Evaluation completed",
+
+00:08:06.561 --> 00:08:12.439
+the value of this expression.
+
+00:08:12.440 --> 00:08:14.079
+This protocol was developed
+
+00:08:14.080 --> 00:08:15.879
+for CIDER development environment.
+
+00:08:15.880 --> 00:08:18.759
+It's a Clojure development environment for Emacs.
+
+00:08:18.760 --> 00:08:22.859
+It's very cool, featureful, reliable,
+
+00:08:22.860 --> 00:08:26.899
+and I would say production-ready.
+
+00:08:26.900 --> 00:08:31.499
+A lot of professional Clojure developers use it.
+
+00:08:31.500 --> 00:08:33.239
+The nREPL protocol is very simple.
+
+00:08:33.240 --> 00:08:38.219
+It has a few operations out of the box,
+
+00:08:38.220 --> 00:08:46.479
+and you can extend it with any arbitrary operation you want.
+
+00:08:46.480 --> 00:08:53.819
+I work a lot on Guix codebase and other Scheme projects,
+
+00:08:53.820 --> 00:08:57.299
+so the experience I had previously with nREPL
+
+00:08:57.300 --> 00:08:59.399
+was not satisfying. I decided
+
+00:08:59.400 --> 00:09:01.739
+to just implement nREPL protocol.
+
+NOTE Arei, Ares, and how to try
+
+00:09:01.740 --> 00:09:05.719
+First of all, I implemented nREPL server in Guile.
+
+00:09:05.720 --> 00:09:11.339
+I called it `guile-ares-rs`, and used it
+
+00:09:11.340 --> 00:09:13.959
+with a generic nREPL client for Emacs.
+
+00:09:13.960 --> 00:09:14.719
+It worked.
+
+00:09:14.720 --> 00:09:18.639
+It had some rough edges, but overall it was okay.
+
+00:09:18.640 --> 00:09:21.639
+And after that, to add more features
+
+00:09:21.640 --> 00:09:25.079
+to make the implementation more complete,
+
+00:09:25.080 --> 00:09:33.219
+I wrote my own nREPL client for Emacs and called it `arei`.
+
+00:09:33.220 --> 00:09:40.179
+And I got almost complete Guile IDE in two months.
+
+00:09:40.180 --> 00:09:45.319
+So `ares-rs` is nREPL server implementation.
+
+00:09:45.320 --> 00:09:49.679
+`arei` is Emacs client, which uses the same nREPL protocol.
+
+00:09:49.680 --> 00:09:54.439
+It utilizes `sesman` package for managing sessions,
+
+00:09:54.440 --> 00:10:00.079
+the association of buffers with nREPL connection.
+
+00:10:00.080 --> 00:10:04.379
+It has some roots.
+
+00:10:04.380 --> 00:10:06.639
+The implementation has some roots
+
+00:10:06.640 --> 00:10:09.979
+in Geiser, CIDER, Monroe, and Rail.
+
+00:10:09.980 --> 00:10:15.279
+I took small snippets for some parts of functionality.
+
+00:10:15.280 --> 00:10:19.479
+I used the CAPF and xref infrastructure
+
+00:10:19.480 --> 00:10:23.079
+for completion at point and cross-reference capabilities.
+
+00:10:23.080 --> 00:10:27.679
+And by the time of conference, I hope
+
+00:10:27.680 --> 00:10:30.199
+that README will be complete enough
+
+00:10:30.200 --> 00:10:34.179
+so you will be able to try it yourself.
+
+NOTE Demo
+
+00:10:34.180 --> 00:10:42.679
+Let's see what is possible with it already.
+
+00:10:42.680 --> 00:10:46.719
+Let's connect to nREPL server.
+
+00:10:51.900 --> 00:10:56.280
+After that, you can evaluate the expression.
+
+00:10:56.281 --> 00:11:02.319
+And you see the stdout and the result.
+
+00:11:02.320 --> 00:11:04.719
+Very nice, very convenient.
+
+00:11:04.720 --> 00:11:08.659
+You have different expression, you evaluate it,
+
+00:11:08.660 --> 00:11:10.359
+you get the value of the evaluation.
+
+00:11:10.360 --> 00:11:12.279
+You can run an infinite loop
+
+00:11:12.280 --> 00:11:15.639
+which prints to stderr and stdout
+
+00:11:15.640 --> 00:11:18.599
+and you see all necessary stuff.
+
+00:11:18.600 --> 00:11:19.299
+Very cool.
+
+00:11:19.300 --> 00:11:21.959
+But also, you can interrupt the evaluation,
+
+00:11:21.960 --> 00:11:25.159
+which is very convenient if you accidentally
+
+00:11:25.160 --> 00:11:27.639
+run an infinite loop.
+
+NOTE Continuations
+
+00:11:27.640 --> 00:11:32.939
+Also, do you remember here we have a few more examples
+
+00:11:32.940 --> 00:11:34.079
+that we didn't try yet?
+
+00:11:34.080 --> 00:11:39.159
+For example, on usual REPL implementation,
+
+00:11:39.160 --> 00:11:47.599
+if I evaluate this expression, I get return value.
+
+00:11:47.600 --> 00:11:50.759
+I make a continuation and save it to this variable
+
+00:11:50.760 --> 00:11:52.859
+and I try to call this evaluation
+
+00:11:52.860 --> 00:11:55.339
+and I get an exception,
+
+00:11:55.340 --> 00:11:58.399
+because the environment in which this continuation
+
+00:11:58.400 --> 00:12:03.479
+was created was different and it has redefined
+
+00:12:03.480 --> 00:12:06.159
+stdout and stderr to capture it.
+
+00:12:06.160 --> 00:12:08.979
+But when I run it one more time,
+
+00:12:08.980 --> 00:12:12.199
+when I resume the continuation,
+
+00:12:12.200 --> 00:12:15.799
+the environment changed and it doesn't work.
+
+00:12:15.800 --> 00:12:17.419
+What happens in `arei`?
+
+00:12:17.420 --> 00:12:21.759
+I define continuation, I save the continuation
+
+00:12:21.760 --> 00:12:23.479
+for the simple expression
+
+00:12:23.480 --> 00:12:27.279
+and I resume the continuation with a new argument,
+
+00:12:27.280 --> 00:12:30.139
+and you can see at the top of the screen
+
+00:12:30.140 --> 00:12:32.459
+that it works perfectly fine.
+
+NOTE Reading from stdin
+
+00:12:32.460 --> 00:12:35.559
+Also, with a usual REPL implementation,
+
+00:12:35.560 --> 00:12:40.319
+let's see what happens when we have a process
+
+00:12:40.320 --> 00:12:41.919
+which reads from stdin.
+
+00:12:41.920 --> 00:12:48.099
+I evaluate the expression and nothing visible happens.
+
+00:12:48.100 --> 00:12:52.999
+I can try to type `C-g`, `C-c`,
+
+00:12:53.000 --> 00:12:56.559
+and after some time it will say user interrupt.
+
+00:12:56.560 --> 00:13:00.439
+What actually I expect in such a case
+
+00:13:00.440 --> 00:13:04.679
+to have a minibuffer which prompts me for the input.
+
+00:13:04.680 --> 00:13:10.019
+When I evaluate the same expression in the `arei`,
+
+00:13:10.020 --> 00:13:12.199
+you see the prompt at the minibuffer
+
+00:13:12.200 --> 00:13:21.899
+and here I can tell, "Hello I'm a message from minibuffer".
+
+00:13:21.900 --> 00:13:26.099
+Cool. You will see that this message is printed to stdout,
+
+00:13:26.100 --> 00:13:28.679
+and unspecified was returned
+
+00:13:28.680 --> 00:13:33.419
+as a result of this expression.
+
+NOTE Fancy example with continuations
+
+00:13:33.420 --> 00:13:37.319
+Let's make some fancy example with continuations.
+
+00:13:37.320 --> 00:13:45.079
+Continuations is a very cool mechanism
+
+00:13:45.080 --> 00:13:47.999
+which is not the topic of today's talk,
+
+00:13:48.000 --> 00:13:50.999
+but you can find a lot of interesting information
+
+00:13:51.000 --> 00:13:54.439
+in Scheme documentation or in related books,
+
+00:13:54.440 --> 00:13:58.339
+and I advise you to do it because it's really nice thing
+
+00:13:58.340 --> 00:14:00.119
+that is actually applicable
+
+00:14:00.120 --> 00:14:03.519
+in many different programming languages.
+
+00:14:03.520 --> 00:14:05.199
+Here you can see the infinite loop
+
+00:14:05.200 --> 00:14:09.159
+which just prints values increasing one by one.
+
+00:14:09.160 --> 00:14:13.299
+And here we save a continuation on each iteration.
+
+00:14:13.300 --> 00:14:18.059
+I can call the continuation
+
+00:14:18.060 --> 00:14:21.939
+and it will resume from the previous saved step.
+
+00:14:21.940 --> 00:14:27.679
+And you can see, it resumed from the same step
+
+00:14:27.680 --> 00:14:31.640
+we interrupted earlier, but we provided a new value for it.
+another value for it.
+
+00:14:31.641 --> 00:14:33.920
+We can provide another value
+
+00:14:33.921 --> 00:14:39.199
+and it resumed from the same spot it was saved earlier.
+
+00:14:39.200 --> 00:14:42.579
+But I also can provide a `read-i` value
+
+00:14:42.580 --> 00:14:45.199
+and if I provide `read-i` value,
+
+00:14:45.200 --> 00:14:50.779
+the infinite loop will read the input from stdin
+
+00:14:50.780 --> 00:14:53.319
+and will continue the evaluation
+
+00:14:53.320 --> 00:14:56.679
+with a different `i` provided in this input.
+
+00:14:56.680 --> 00:15:03.039
+So let's try to type some arbitrary value
+
+00:15:03.040 --> 00:15:07.519
+and you see that the loop continued with this value.
+
+00:15:07.520 --> 00:15:08.039
+Very nice.
+
+00:15:08.040 --> 00:15:13.159
+And every time we could easily interrupt it.
+
+NOTE Guix API
+
+00:15:13.160 --> 00:15:17.319
+Okay, what most annoying thing that I had previously
+
+00:15:17.320 --> 00:15:19.339
+with the usual REPL implementation
+
+00:15:19.340 --> 00:15:22.759
+that I have a quite nice Guix API
+
+00:15:22.760 --> 00:15:27.579
+where I can build packages, systems and other stuff.
+
+00:15:27.580 --> 00:15:35.359
+But if I evaluate this expression, I will get an error.
+
+00:15:35.360 --> 00:15:38.039
+Okay. I will get an error
+
+00:15:38.040 --> 00:15:44.479
+because I don't have an appropriate environment.
+
+00:15:44.480 --> 00:15:51.579
+But what I can do, I can connect to the remote REPL
+
+00:15:51.580 --> 00:15:55.059
+by creating a server with `guix repl --listen` command
+
+00:15:55.060 --> 00:15:58.619
+and connecting to it with `geiser-connect` command.
+
+00:15:58.620 --> 00:16:01.819
+And now I can evaluate this expression.
+
+00:16:01.820 --> 00:16:03.359
+Right?
+
+00:16:03.360 --> 00:16:10.479
+Wow.
+
+00:16:10.480 --> 00:16:14.339
+Okay.
+
+00:16:14.340 --> 00:16:19.039
+It actually doesn't matter for my example.
+
+00:16:19.040 --> 00:16:22.879
+I will explain how it doesn't work easily.
+
+00:16:22.880 --> 00:16:26.519
+This is a long-running process which prints something
+
+00:16:26.520 --> 00:16:29.579
+and it can take up to a few minutes.
+
+00:16:29.580 --> 00:16:33.359
+And for the whole few minutes I don't see any results,
+
+00:16:33.360 --> 00:16:38.719
+the same as with this infinite loop which prints to stdout
+
+00:16:38.720 --> 00:16:42.199
+but I don't see anything interactively.
+
+00:16:42.200 --> 00:16:45.619
+With `arei`, I can run
+
+00:16:45.620 --> 00:16:47.920
+the evaluation of the same expression,
+
+00:16:51.440 --> 00:16:54.119
+and you will see instantly
+
+00:16:54.120 --> 00:17:00.200
+that stdout is presented here in slightly yellowish color.
+
+00:17:00.201 --> 00:17:02.920
+I can interrupt the evaluation
+
+00:17:02.921 --> 00:17:06.039
+if I don't want to wait until it's finished,
+
+00:17:06.040 --> 00:17:15.779
+and just after that, I can evaluate another value.
+
+00:17:15.780 --> 00:17:23.359
+So that's cool.
+
+00:17:23.360 --> 00:17:25.959
+And let's see one more thing.
+
+00:17:25.960 --> 00:17:30.339
+We have an infinite loop and we have some completion here.
+
+00:17:30.340 --> 00:17:32.579
+And completion still works,
+
+00:17:32.580 --> 00:17:33.659
+very nice,
+
+00:17:33.660 --> 00:17:40.259
+while the infinite loop is running.
+
+00:17:40.260 --> 00:17:42.059
+Okay.
+
+NOTE Support
+
+00:17:42.060 --> 00:17:44.919
+Actually it took me around two months
+
+00:17:44.920 --> 00:17:48.039
+of full-time work funded by my own savings,
+
+00:17:48.040 --> 00:17:51.599
+and you can support and help to the project
+
+00:17:51.600 --> 00:17:57.019
+using OpenCollective or by contributing on SourceHut.
+
+NOTE Future steps - Multiple simultaneous evaluations in different contexts
+
+00:17:57.020 --> 00:17:58.699
+The future steps for the project
+
+00:17:58.700 --> 00:18:03.674
+include an experimental workflow where you have
+
+00:18:03.675 --> 00:18:07.539
+multiple simultaneous evaluation in different contexts.
+
+00:18:07.540 --> 00:18:11.959
+For example, you have Fibers, you have Goblins,
+
+00:18:11.960 --> 00:18:16.919
+you have some HTTP server or some other thing,
+
+00:18:16.920 --> 00:18:22.119
+and you want to run all of them independently
+
+00:18:22.120 --> 00:18:25.319
+in slightly isolated sessions,
+
+00:18:25.320 --> 00:18:29.799
+and you want to have the ability
+
+00:18:29.800 --> 00:18:30.959
+to still interact with them.
+
+00:18:30.960 --> 00:18:33.979
+For example, if they require standard input
+
+00:18:33.980 --> 00:18:39.239
+or something else, you want to be able to provide it.
+
+00:18:39.240 --> 00:18:42.519
+You want to see the stderr and stdout
+
+00:18:42.520 --> 00:18:46.219
+of those long-running processes and so on.
+
+NOTE Tree-sitter integration
+
+00:18:46.220 --> 00:18:50.239
+The second thing is tree-sitter integration
+
+00:18:50.240 --> 00:18:53.399
+for better syntax highlighting, code navigation,
+
+00:18:53.400 --> 00:18:56.879
+and other features.
+
+NOTE Full-fledged debugger
+
+00:18:56.880 --> 00:19:01.399
+And after that, probably we will do a full-fledged debugger
+
+00:19:01.400 --> 00:19:06.239
+so you can jump expressions one by one
+
+00:19:06.240 --> 00:19:10.779
+and see the results and see some intermediate values
+
+00:19:10.780 --> 00:19:13.079
+during the evaluation.
+
+00:19:13.080 --> 00:19:14.479
+And it's very possible
+
+00:19:14.480 --> 00:19:17.079
+because nREPL is a very extensible protocol
+
+00:19:17.080 --> 00:19:18.199
+and you can implement
+
+00:19:18.200 --> 00:19:22.759
+whatever you want on top of it.
+
+NOTE FAQ - Does it support other Scheme implementations?
+
+00:19:22.760 --> 00:19:27.079
+I will answer two probably very frequent questions.
+
+00:19:27.080 --> 00:19:30.499
+Does it support other Scheme implementations?
+
+00:19:30.500 --> 00:19:32.279
+At the moment, it doesn't,
+
+00:19:32.280 --> 00:19:36.519
+but the Scheme implementation is not restricted.
+
+00:19:36.520 --> 00:19:40.639
+You have a server which is implemented in your language
+
+00:19:40.640 --> 00:19:43.974
+and you have a client--in our case, `arei`--
+
+00:19:43.975 --> 00:19:48.319
+which communicates with this protocol.
+
+00:19:48.320 --> 00:19:52.359
+So if you implement nREPL server in a different language,
+
+00:19:52.360 --> 00:19:58.379
+it should work with already implemented `arei` client.
+
+NOTE Is it possible to use it with other text editors?
+
+00:19:58.380 --> 00:20:04.079
+And is it possible to use the same functionality
+
+00:20:04.080 --> 00:20:06.999
+in other text editors, for example in VS Code,
+
+00:20:07.000 --> 00:20:08.679
+Vim, whatever?
+
+00:20:08.680 --> 00:20:13.799
+Yes, it's possible and the case is similar here.
+
+00:20:13.800 --> 00:20:16.599
+You have already implemented nREPL server
+
+00:20:16.600 --> 00:20:19.359
+and you can write your own nREPL client
+
+00:20:19.360 --> 00:20:22.120
+in a different text editor and it will work.
+
+NOTE Conclusion
+
+00:20:22.121 --> 00:20:26.759
+I would like to thank the authors and maintainers
+
+00:20:26.760 --> 00:20:30.439
+and contributors of Guile, Geiser, CIDER, Clojure,
+
+00:20:30.440 --> 00:20:33.359
+and Emacs, and all other people
+
+00:20:33.360 --> 00:20:38.779
+who are somehow related to the work on those projects
+
+00:20:38.780 --> 00:20:42.079
+involved in this talk.
+
+00:20:42.080 --> 00:20:45.879
+And I hope the Scheme programming will be enjoyable.
+
+NOTE Contacts
+
+00:20:45.880 --> 00:20:47.239
+If you want to contact me,
+
+00:20:47.240 --> 00:20:49.799
+join #tropin IRC channel at libera.chat,
+
+00:20:49.800 --> 00:20:53.039
+or drop me a message via email or feediverse
+
+00:20:53.040 --> 00:20:55.879
+using `andrew@trop.in` handle.
+
+00:20:55.880 --> 00:21:00.680
+I will see you in a bit in Q&A session.