WEBVTT NOTE #+YOUTUBE_URL: https://youtube.com/live/aEnacESU-DU #+LANGUAGE: en Opening 00:00:00.031 --> 00:00:03.528 [Sacha]: But first, I will press the go live 00:00:03.568 --> 00:00:05.738 button because this time I'm going to remember. 00:00:09.042 --> 00:00:13.068 Okay, now I think we are live. 00:00:13.829 --> 00:00:17.014 Welcome to Emacs Chat 31. 00:00:17.034 --> 00:00:20.840 Today we have Pedro Aranda Gutierrez, who is here 00:00:20.900 --> 00:00:23.212 to talk about all the cool things 00:00:23.213 --> 00:00:24.706 with Org LaTeX Export. 00:00:25.527 --> 00:00:28.612 And Prot is here also because he knows a lot more 00:00:28.652 --> 00:00:30.795 about fonts and writing in multiple languages and 00:00:30.815 --> 00:00:31.356 all that stuff. 00:00:31.376 --> 00:00:33.559 So I'm counting on Prot to ask questions when my 00:00:33.579 --> 00:00:37.345 mind just goes like, ooh, cool demo. 00:00:37.848 --> 00:00:42.934 [Pedro]: Hello, here we are. 00:00:44.256 --> 00:00:49.803 So yes, I wanted to talk about what is news for all. 00:00:50.544 --> 00:00:55.269 At least this part, the integration of the basic 00:00:55.309 --> 00:01:01.937 font management with lualatex is in main 00:01:01.938 --> 00:01:03.079 right now in Org Mode. 00:01:04.358 --> 00:01:05.239 [Prot]: Very well, very well. 00:01:05.259 --> 00:01:10.968 [Pedro]: I was not expecting that to happen, 00:01:10.969 --> 00:01:13.072 but it happened. 00:01:13.092 --> 00:01:14.194 So that's the situation. 00:01:14.755 --> 00:01:20.004 And I would just like to go stepwise on what you 00:01:20.044 --> 00:01:22.468 have to do, what you can do, what you can't do 00:01:23.289 --> 00:01:24.932 with these things. NOTE Setting things up 00:01:24.912 --> 00:01:30.099 This is my Emacs 31 compiled a couple of days 00:01:30.139 --> 00:01:31.821 ago and I'm on devel. 00:01:32.081 --> 00:01:37.167 I'm on main here, on main for Org mode. 00:01:38.549 --> 00:01:42.054 I have a small Org file here which initially 00:01:42.154 --> 00:01:44.777 I'm going to be using pdflatex to compile, 00:01:46.073 --> 00:01:49.218 with a trick: I'm not compiling in the background, 00:01:49.279 --> 00:01:53.165 but I have a script here in the terminal to have 00:01:53.205 --> 00:01:57.413 a bit more of feedback or direct feedback of what 00:01:57.493 --> 00:01:58.054 is happening. 00:01:59.296 --> 00:02:02.901 If we go through this... 00:02:02.902 --> 00:02:07.550 I mean this is basically my template 00:02:07.830 --> 00:02:09.854 to start any article. It's going to be an 00:02:09.954 --> 00:02:10.495 article. 00:02:12.533 --> 00:02:14.296 we are going to use pdflatex. 00:02:15.358 --> 00:02:21.770 And the other thing is I have a very small thing 00:02:21.891 --> 00:02:24.596 with the default font. 00:02:24.676 --> 00:02:27.100 Then I have also the Sans Serif font. 00:02:27.701 --> 00:02:32.871 I also have some monospaced fonts with it. NOTE Local variables #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-03-16-167.jpg #+SCREENSHOT_TIME: 00:03:16.167 00:02:33.222 --> 00:02:35.404 The things that I have here that might be 00:02:35.444 --> 00:02:37.867 interesting for people is that another thing that 00:02:37.967 --> 00:02:41.911 is quite recent for Emacs is that we are starting 00:02:41.971 --> 00:02:49.339 to have a lot of the configuration variables as 00:02:49.459 --> 00:02:50.420 local variables. 00:02:50.440 --> 00:02:53.723 So you don't have to go and get in 00:02:53.724 --> 00:02:57.508 your init's .el to have anything configured. 00:02:57.725 --> 00:02:58.547 Why? 00:02:58.567 --> 00:03:02.474 This is another sort of fight against the 00:03:02.654 --> 00:03:03.696 windmills I've had. 00:03:04.618 --> 00:03:04.879 Why? 00:03:05.099 --> 00:03:09.287 Because with this I can make my Org files more 00:03:09.327 --> 00:03:11.251 distributable, so I can share them with more 00:03:11.291 --> 00:03:13.936 people and they don't have to touch anything in 00:03:13.996 --> 00:03:15.138 their configurations. 00:03:16.160 --> 00:03:18.805 That's the basic idea. 00:03:19.392 --> 00:03:22.795 Small little trick that I have is that I'm using 00:03:22.875 --> 00:03:26.258 fancyvrb for everything instead of listings. 00:03:27.139 --> 00:03:31.163 And there was somewhere in the documentation, you 00:03:31.183 --> 00:03:35.046 also have an explanation for how to use fancyvrb 00:03:35.507 --> 00:03:38.469 to substitute listings and get nice listings in 00:03:38.529 --> 00:03:45.796 your LaTeX exported by Org. 00:03:45.816 --> 00:03:49.399 And that's the only thing. 00:03:49.379 --> 00:03:51.663 Initially, pdflatex, 00:03:51.763 --> 00:03:52.844 Everything is plain. 00:03:53.726 --> 00:03:56.830 If I go and just go and export... NOTE Mini export menu #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-03-57-700.jpg #+SCREENSHOT_TIME: 00:03:57.700 00:03:56.831 --> 00:04:02.158 This is a mini export menu I have myself 00:04:02.159 --> 00:04:02.920 because I think it's 00:04:03.461 --> 00:04:09.350 much more handy than the export thing that we 00:04:09.410 --> 00:04:10.092 have in Org Mode. 00:04:10.112 --> 00:04:12.315 So I did a small one for me. 00:04:12.295 --> 00:04:17.280 And so I'm exporting to LaTeX and there you see the... NOTE Basic output with different fonts 00:04:17.281 --> 00:04:19.784 This was quick, of course, but what you 00:04:19.803 --> 00:04:25.311 see is I have my serif, my sans serif, and I have 00:04:25.431 --> 00:04:30.497 my monofont and that's nice, right? 00:04:30.517 --> 00:04:33.381 So that's one of the things that we always want 00:04:33.421 --> 00:04:33.741 to have. 00:04:34.622 --> 00:04:38.387 Now, the next thing is everything would be nice. 00:04:39.008 --> 00:04:45.759 And now I need my [??], full file where I 00:04:45.799 --> 00:04:47.842 have the rest of the things that I want to copy 00:04:47.882 --> 00:04:49.304 and paste into this presentation. NOTE Adding a code block as source 00:04:49.865 --> 00:04:56.215 Everything would be nice if I just were to go and 00:04:56.575 --> 00:05:04.253 add a small command to do things, but I'm only going 00:05:04.293 --> 00:05:05.756 to be listing the source. 00:05:06.157 --> 00:05:09.363 So let me just go and check that. 00:05:09.383 --> 00:05:12.689 If I do like this, so I have a small listing. 00:05:13.070 --> 00:05:14.492 Why do I need listings? 00:05:14.593 --> 00:05:17.378 This is partially for my lectures. 00:05:18.480 --> 00:05:22.347 I want to share code with my students and tell them 00:05:23.238 --> 00:05:24.820 how to program, right? 00:05:25.560 --> 00:05:28.724 Or they are in a lab and they need an example of 00:05:28.784 --> 00:05:32.508 a command that they need to achieve a search and 00:05:32.568 --> 00:05:33.849 output in the computer. 00:05:34.790 --> 00:05:38.094 So what I would do, first step, which would 00:05:38.314 --> 00:05:42.318 also work with pdflatex, is I'm going to go and 00:05:42.418 --> 00:05:44.140 export the source. 00:05:44.160 --> 00:05:46.663 If I now go and export the source, 00:05:48.010 --> 00:05:54.297 I got a second compile, and now you see that I 00:05:54.357 --> 00:05:55.422 included the command. 00:05:56.583 --> 00:06:01.571 Nice thing about using fancyvrb here is that as 00:06:01.651 --> 00:06:06.498 I have fancyvrb here configured, I'm using the 00:06:06.578 --> 00:06:09.242 listings stuff, but this is the configuration for 00:06:09.282 --> 00:06:09.863 fancyvrb. 00:06:10.204 --> 00:06:13.629 I can include things like a frame to separate 00:06:13.950 --> 00:06:18.777 code from other things in the document. NOTE The problem with exporting Unicode 00:06:18.757 --> 00:06:21.700 But now here problems started and this is this 00:06:21.741 --> 00:06:25.225 was practically what happened two years ago when 00:06:25.265 --> 00:06:28.669 someone asked in the list how can I do this and 00:06:28.709 --> 00:06:30.991 the thing is he also wanted to have something 00:06:31.071 --> 00:06:34.195 like I want to export both so I also want to 00:06:34.255 --> 00:06:34.956 export the results. 00:06:37.699 --> 00:06:39.041 Let's see what happens in this case. 00:06:39.341 --> 00:06:41.784 It's the tree command so it will be showing a 00:06:41.844 --> 00:06:44.747 couple of files that I have on my on my system 00:06:45.809 --> 00:06:47.751 and let's see what happens with pdflatex. 00:06:53.198 --> 00:06:58.327 It seems to be working, but then if you go to the 00:06:58.488 --> 00:07:00.852 log, and this is why I have the log here, you 00:07:00.872 --> 00:07:04.078 start seeing a lot of errors. 00:07:06.042 --> 00:07:06.342 Why? 00:07:06.402 --> 00:07:10.249 Because it has records that do not exist. 00:07:10.269 --> 00:07:11.151 [Prot]: Missing characters. 00:07:11.952 --> 00:07:13.656 [Pedro]: Missing characters and things like that. 00:07:13.796 --> 00:07:15.018 So this was the question. 00:07:15.058 --> 00:07:16.641 How can I solve this? 00:07:17.954 --> 00:07:21.740 The obvious answer is pdflatex is not as good 00:07:21.900 --> 00:07:27.208 with UTF-8 extended, so migrate to lualatex. 00:07:30.534 --> 00:07:31.615 That was one of the things. 00:07:32.637 --> 00:07:35.902 But you could actually do it. 00:07:36.122 --> 00:07:42.092 Before going to that, people said, let's go back 00:07:42.152 --> 00:07:43.033 and put source. NOTE Trying a different font 00:07:45.578 --> 00:07:47.781 You already have the 00:07:47.841 --> 00:07:52.989 different font packages in in pdflatex, so why not 00:07:53.390 --> 00:07:55.433 using one of the font packages that you have, for 00:07:55.493 --> 00:07:58.517 example, the Bera fonts. This Bera font here 00:07:58.557 --> 00:08:01.862 Bera, Bera Sans and Bera Mono should 00:08:02.604 --> 00:08:05.187 allow you to do what you want to do. I checked. 00:08:06.069 --> 00:08:08.793 What I'm doing here is I have 00:08:08.813 --> 00:08:13.560 my local variables. 00:08:14.198 --> 00:08:16.781 I've refreshed my local variables to include 00:08:17.262 --> 00:08:20.024 in the packages, the packages for Bera, 00:08:20.025 --> 00:08:21.767 Bera sans, and Bera mono. 00:08:23.049 --> 00:08:23.950 Let's see what happens. 00:08:24.070 --> 00:08:34.602 If I go just for the source, I get the Bera thing. 00:08:34.643 --> 00:08:38.607 So there should be some sort, I get the Bera font 00:08:38.628 --> 00:08:39.269 integrated. 00:08:40.429 --> 00:08:43.498 So there should be some hope with this. 00:08:43.938 --> 00:08:44.239 Fine. 00:08:44.820 --> 00:08:51.649 So let's go back into doing things with the 00:08:51.889 --> 00:08:52.890 results included. 00:08:56.615 --> 00:08:57.196 Bad luck. 00:08:58.558 --> 00:08:59.879 We continue to have this problem. 00:09:00.260 --> 00:09:02.303 This is inherent to pdflatex. 00:09:04.185 --> 00:09:04.425 Right? NOTE LuaLaTeX 00:09:05.026 --> 00:09:08.691 So we have to move to lualatex, right? 00:09:12.232 --> 00:09:16.878 And now to be able to do what I'm doing here with 00:09:16.938 --> 00:09:19.921 lualatex and with my script in the background 00:09:19.961 --> 00:09:22.585 to see what is happening, I just need to get out 00:09:23.666 --> 00:09:24.527 and get in again. 00:09:30.574 --> 00:09:34.619 So we were at this point, we had the Bera fonts. 00:09:35.300 --> 00:09:39.866 Let's go to see the Bera fonts just with a source 00:09:40.667 --> 00:09:41.648 on lualatex. 00:09:42.877 --> 00:09:44.881 This is something that some people ask. 00:09:53.256 --> 00:09:55.440 The Bera fonts don't have the mono font. 00:09:56.502 --> 00:09:58.405 So we have to move to something else. 00:10:00.494 --> 00:10:01.756 Fine. 00:10:01.776 --> 00:10:03.658 That's point number one. 00:10:04.379 --> 00:10:06.040 And since that does not work... 00:10:06.041 --> 00:10:08.484 [Prot]: Sans also, right? 00:10:08.504 --> 00:10:10.406 You should have a sans-serif line. 00:10:10.426 --> 00:10:12.869 I didn't see that. 00:10:13.050 --> 00:10:13.790 [Pedro]: Yes. Exactly. 00:10:14.511 --> 00:10:17.735 So the Bera fonts, as some people suggested, do 00:10:17.855 --> 00:10:21.540 not work with lualatex as we want. 00:10:22.820 --> 00:10:27.686 So if this doesn't work and the corresponding is 00:10:27.746 --> 00:10:32.452 not one-to-one, so why not go directly into using 00:10:32.492 --> 00:10:37.638 the full thing with lualatex, right? NOTE Switching to FreeSerif, FreeSans, and FreeMono 00:10:37.678 --> 00:10:40.762 The first thing I would have to do is I'm going 00:10:40.802 --> 00:10:42.964 to get rid of the Bera fonts here 00:10:42.965 --> 00:10:50.154 in my local variables. 00:10:50.174 --> 00:10:51.055 I put it at the end. 00:10:52.823 --> 00:10:56.528 So now with my local variables like this, I'm 00:10:56.608 --> 00:10:59.511 only going to have packages like enumitem and 00:10:59.591 --> 00:11:00.232 fancyvrb. 00:11:02.675 --> 00:11:07.361 I'm going to be using lualatex and now in a 00:11:07.401 --> 00:11:10.144 very big first thing, what I would have to do is 00:11:10.505 --> 00:11:15.811 to start with my LaTeX header and then start 00:11:15.851 --> 00:11:17.173 with my setmainfont. 00:11:17.573 --> 00:11:19.354 Let's say we use the simple ones, 00:11:19.355 --> 00:11:21.598 the setmainfont 00:11:22.979 --> 00:11:34.092 FreeSerif, FreeSans... As always, the fingers are... 00:11:38.096 --> 00:11:46.025 And here you go with the mono, and with the mono 00:11:46.406 --> 00:11:49.249 right. Let's try this at this point in time. 00:11:49.589 --> 00:11:51.210 What will happen? I'm going to be using 00:11:51.211 --> 00:11:53.908 the free fonts directly 00:11:53.909 --> 00:11:56.633 with lualatex and with lualatex commands. 00:11:58.315 --> 00:11:59.437 Let's see what happens here. 00:12:02.401 --> 00:12:04.524 Starts compiling, starts compiling. NOTE LuaLaTeX is slower 00:12:04.925 --> 00:12:06.507 lualatex is slower, yes. 00:12:07.529 --> 00:12:10.473 That's the main drawback, mainly for people who 00:12:10.553 --> 00:12:17.083 want to have equations, LaTeX preview. 00:12:19.055 --> 00:12:22.182 In Org, when they want to have the preview for 00:12:22.242 --> 00:12:25.689 equations and maths and so on in Org Mode, that 00:12:26.370 --> 00:12:30.860 is the main challenge because lualatex gets 00:12:30.880 --> 00:12:34.588 slower, and that will of course make them not be 00:12:34.688 --> 00:12:41.060 able to get a flashy auto-refreshing display and so on. NOTE The fonts are rendered 00:12:42.482 --> 00:12:46.428 But the thing for me and for my manuals, if you 00:12:46.488 --> 00:12:49.693 want to say it like this, is that first of all we 00:12:49.753 --> 00:12:53.158 have here the serif fonts, you have the sans 00:12:53.178 --> 00:12:57.064 fonts, you have the mono font with the free fonts, right? 00:12:58.146 --> 00:13:02.012 So the next thing that came in was of course, 00:13:02.453 --> 00:13:06.038 let's see what happens with the listing. 00:13:18.893 --> 00:13:24.664 Takes longer, but that is something that I can 00:13:24.864 --> 00:13:28.731 use to show my students how things will look like 00:13:28.832 --> 00:13:29.353 in the screen. 00:13:31.717 --> 00:13:35.424 And this was a question that was raised beginning 00:13:35.444 --> 00:13:37.628 of 2024 in the list. 00:13:40.392 --> 00:13:41.694 [Sacha]: I appreciate your persistence. NOTE The farmer on the donkey in front of the train 00:13:42.896 --> 00:13:44.558 [Pedro]: Yes. 00:13:45.560 --> 00:13:50.086 There's this film, Spanish film from the 30s, 00:13:50.126 --> 00:13:51.188 from beginning of the 30s. 00:13:51.288 --> 00:13:53.211 It was still Republic before the Civil War. 00:13:54.092 --> 00:14:00.882 And there's a man dressed as a farmer in Aragon 00:14:01.443 --> 00:14:05.809 (they are known to be very stubborn) on his donkey 00:14:05.789 --> 00:14:07.752 on a train track, 00:14:08.493 --> 00:14:11.778 and the train behind him, and he doesn't move 00:14:11.798 --> 00:14:12.799 from the track. 00:14:12.819 --> 00:14:16.885 The only thing he's saying, "You can toot as much 00:14:16.945 --> 00:14:18.467 as you want, I'm not going to move. 00:14:18.808 --> 00:14:20.450 If you want to move, move yourself." 00:14:20.871 --> 00:14:21.532 So he continues. 00:14:21.592 --> 00:14:25.377 And the train, and the people: "Why don't you move?" 00:14:25.898 --> 00:14:27.761 At the end, they come to a joint, he 00:14:27.821 --> 00:14:30.084 continues on the road, and the train takes the 00:14:30.144 --> 00:14:31.165 other track. 00:14:31.186 --> 00:14:34.170 When the train goes by, he says, I told you, 00:14:34.250 --> 00:14:35.712 you had to go another way. 00:14:37.548 --> 00:14:40.974 So this is more or less, this in Spanish has 00:14:41.054 --> 00:14:45.422 become a saying. 00:14:45.622 --> 00:14:47.045 It's more than a phrase. 00:14:47.145 --> 00:14:49.269 It's a chufla chufla. 00:14:50.311 --> 00:14:54.017 And I'm a bit of a chufla chufla. 00:14:54.197 --> 00:14:57.263 When I think that there's something that needs to 00:14:57.303 --> 00:14:59.607 be done, I can be very stubborn. 00:15:01.713 --> 00:15:05.779 [Prot]: And of course you need to be stubborn 00:15:06.119 --> 00:15:09.704 because these sorts of things are tricky and they 00:15:09.784 --> 00:15:11.606 require time. Especially on 00:15:11.607 --> 00:15:14.069 a mailing-list-driven workflow, 00:15:14.070 --> 00:15:15.952 you have to listen to what users want. 00:15:15.953 --> 00:15:18.696 You have to take in feedback. It is a slow process. 00:15:18.697 --> 00:15:20.279 You really have to be patient and 00:15:20.339 --> 00:15:21.640 stubborn, for sure. 00:15:21.641 --> 00:15:29.492 [Pedro]: Yes. Basically, my motivation was I know that I need it. 00:15:30.485 --> 00:15:33.089 Maybe other people don't want to say, 00:15:33.129 --> 00:15:34.310 but they need it also. NOTE Unicode and LuaLaTeX 00:15:35.772 --> 00:15:37.134 [Prot]: Can I ask a comment? 00:15:37.214 --> 00:15:39.898 Can I ask a question here or will you continue? 00:15:39.918 --> 00:15:43.844 Because I want to clarify the Unicode characters. 00:15:44.104 --> 00:15:45.406 Earlier we were getting an error. 00:15:46.187 --> 00:15:47.909 Now we see the tree. 00:15:49.151 --> 00:15:51.734 Is this because of FreeFont? 00:15:51.754 --> 00:15:53.417 Is this because of lualatex? 00:15:53.917 --> 00:15:54.739 Is it a combination? 00:15:54.879 --> 00:15:57.943 What exactly helped us fix this? 00:15:58.160 --> 00:16:01.404 [Pedro]: Okay, it's a very, very good question. 00:16:01.444 --> 00:16:02.545 I was forgetting that. 00:16:02.946 --> 00:16:04.528 So here we have a combination. 00:16:06.490 --> 00:16:11.576 lualatex is handling the UTF characters very nicely. 00:16:11.596 --> 00:16:15.541 Something that pdflatex did not or was very 00:16:15.581 --> 00:16:18.685 difficult to have them done, doing it. 00:16:19.686 --> 00:16:24.212 The second thing is that Free Mono has these 00:16:25.013 --> 00:16:26.074 display characters. 00:16:26.763 --> 00:16:28.307 which is also important. 00:16:29.010 --> 00:16:31.096 And of course, if you want to use free mono, you 00:16:31.156 --> 00:16:35.568 need to use lualatex to drive the thing. 00:16:35.589 --> 00:16:37.554 So it's the three things. 00:16:37.574 --> 00:16:40.663 You have the right font, the right compiler and 00:16:40.703 --> 00:16:41.987 the right environment. 00:16:43.486 --> 00:16:49.173 So the next thing is, of course, also on the list. 00:16:49.193 --> 00:16:52.838 Once again, I need my full because it's going to 00:16:52.878 --> 00:16:55.982 be a bit quicker than if I go writing everything 00:16:56.062 --> 00:16:56.583 on my own. 00:16:57.284 --> 00:17:00.427 If I go down... NOTE Emojis 00:17:00.428 --> 00:17:01.449 There was someone saying, 00:17:01.569 --> 00:17:04.313 "Hey, I want my emojis." 00:17:07.056 --> 00:17:08.278 [Sacha]: They're very important, you know. 00:17:09.490 --> 00:17:11.352 [Pedro]: Yeah, I know. 00:17:11.412 --> 00:17:15.217 For me, it's as important as for anyone. 00:17:16.238 --> 00:17:22.185 Because maybe you want this, or if you're just... 00:17:22.305 --> 00:17:27.431 If I'm on a manual, I might want to point out 00:17:27.491 --> 00:17:33.378 things like... And then I go with the hand, with 00:17:33.418 --> 00:17:37.843 the body, with the hand, with the... 00:17:39.072 --> 00:17:42.256 single finger which is the S and then I just want 00:17:42.297 --> 00:17:46.342 to point at things like this for example and to 00:17:46.402 --> 00:17:50.648 make it... I might want to highlight something 00:17:50.688 --> 00:17:55.014 that people have to do with this. Let's try 00:17:55.235 --> 00:17:57.778 with this current configuration. If I try with 00:17:57.798 --> 00:18:00.422 this current configuration... And there we are going 00:18:00.462 --> 00:18:06.661 to see the second part of the answer to Prot's question: 00:18:06.681 --> 00:18:09.804 we start compiling, we start compiling, we start 00:18:09.864 --> 00:18:13.728 compiling, and then suddenly we have like two 00:18:13.828 --> 00:18:17.452 characters which are not part of the thing. 00:18:18.312 --> 00:18:18.533 Why? 00:18:18.553 --> 00:18:21.315 Because they are not part of Free Font. 00:18:21.956 --> 00:18:27.582 Or in this case they are not part of FreeSerif. 00:18:28.482 --> 00:18:30.244 They are not included in FreeSerif. 00:18:30.304 --> 00:18:32.046 I need something else. 00:18:32.768 --> 00:18:36.953 There's where the full thing came in. NOTE Fallback fonts #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-20-01-900.jpg #+SCREENSHOT_TIME: 00:20:01.900 00:18:37.654 --> 00:18:43.061 I was very much inspired by what Pandoc is doing. 00:18:43.101 --> 00:18:45.484 In Pandoc, you can say, "I have my serif font, which 00:18:45.545 --> 00:18:49.970 is this font. For this font, I want 00:18:50.010 --> 00:18:54.656 these fallbacks because that's..." Fallback is, first 00:18:54.696 --> 00:18:55.638 of all, fallback fonts. 00:18:55.798 --> 00:19:01.425 Fallback is a term used by lualatex to 00:19:01.540 --> 00:19:04.744 say, if you can't find a character in a font, 00:19:04.745 --> 00:19:07.709 try to find it in another font. 00:19:08.030 --> 00:19:11.756 And then, if you provide lualatex with a list of 00:19:11.816 --> 00:19:15.161 fonts, it's going to be looking up the fonts 00:19:15.501 --> 00:19:16.723 to get your characters. 00:19:18.686 --> 00:19:20.148 That's the main idea. 00:19:21.330 --> 00:19:24.715 [Prot]: Let me ask something here, because if you 00:19:24.855 --> 00:19:26.518 notice the output... 00:19:26.735 --> 00:19:29.939 Of course, it is not what we want, but it is also 00:19:30.099 --> 00:19:34.245 not just the tofu, you know, like the missing 00:19:34.285 --> 00:19:34.865 character. 00:19:35.406 --> 00:19:39.652 If you notice, there is some kind of a symbol 00:19:39.732 --> 00:19:41.354 there going on. 00:19:41.594 --> 00:19:45.079 From what I recall, we have characters and 00:19:45.119 --> 00:19:46.861 then we have something that is called the 00:19:46.961 --> 00:19:48.742 grapheme or grapheme cluster, 00:19:48.743 --> 00:19:49.764 something like that, 00:19:49.765 --> 00:19:51.647 which is basically characters laid on top 00:19:51.707 --> 00:19:53.890 of each other to produce the emoji. 00:19:54.359 --> 00:19:57.908 It seems here that Free Font has the 00:19:58.028 --> 00:19:59.993 underlying characters, but maybe not the 00:20:00.073 --> 00:20:01.896 composition codes or something like that. 00:20:01.897 --> 00:20:05.966 [Pedro]: No. It only has some characters. 00:20:05.967 --> 00:20:07.671 It only has some some underlying 00:20:07.731 --> 00:20:10.157 characters. For example it only has the... 00:20:10.424 --> 00:20:13.407 It has the finger, but it doesn't have 00:20:13.987 --> 00:20:16.310 all the compositions that you need for that. 00:20:16.330 --> 00:20:16.950 [Prot]: Exactly. 00:20:16.990 --> 00:20:19.293 [Sacha]: That one in particular, I think, changes 00:20:19.333 --> 00:20:20.494 the color of it. 00:20:20.514 --> 00:20:21.415 So that's why. 00:20:21.435 --> 00:20:22.576 [Pedro]: Yes. 00:20:22.816 --> 00:20:26.460 And if you look at the, if we go back here, it's 00:20:26.520 --> 00:20:33.086 saying there's no character at all for the face. 00:20:34.087 --> 00:20:36.169 But if you go to the beginning, you will also 00:20:36.289 --> 00:20:39.452 notice that there was some, where was it? 00:20:39.512 --> 00:20:40.313 I've seen it. 00:20:44.393 --> 00:20:48.918 At some point, they are also telling you that the 00:20:48.978 --> 00:20:50.120 thing... where is it? 00:20:50.420 --> 00:20:50.981 Okay, fine. 00:20:51.622 --> 00:20:52.843 It's going to take too long to find. 00:20:53.304 --> 00:20:54.325 [Sacha]: It's fine. 00:20:54.345 --> 00:20:56.307 Besides, you fixed the problem, so you don't have 00:20:56.327 --> 00:20:57.929 to dig into the old way of doing things. 00:20:59.231 --> 00:21:03.276 [Pedro]: Yes, so that's my main problem right now. 00:21:04.477 --> 00:21:09.743 So I go and I'm going to go to the... I keep 00:21:09.783 --> 00:21:15.336 trying to use the teleconference, the conference screen, 00:21:15.396 --> 00:21:17.199 the shared screen instead of the red one. 00:21:17.580 --> 00:21:21.827 This is why sometimes I sort of get mixed up. 00:21:22.708 --> 00:21:24.091 We had this. 00:21:27.136 --> 00:21:31.623 And I'm just going to go and comment this out 00:21:32.144 --> 00:21:34.127 because I'm going to go for the new one. NOTE Fallback configuration #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-22-31-167.jpg #+SCREENSHOT_TIME: 00:22:31.167 #+SCREENSHOT_CROP: 442 323 1160 648 00:21:39.389 --> 00:21:42.132 For the new one, I have a full example here. 00:21:42.592 --> 00:21:46.456 Here we go with the fontspec configuration, 00:21:49.639 --> 00:21:53.462 which is at the beginning. You say, what are you 00:21:53.502 --> 00:22:00.769 doing there, right? I know it might 00:22:00.809 --> 00:22:03.372 look a bit scary at the beginning but then if you 00:22:03.552 --> 00:22:05.774 if you're familiar with what Pandoc is doing it's 00:22:05.854 --> 00:22:06.915 not that different 00:22:08.701 --> 00:22:09.302 [Sacha]: Nonsense. 00:22:09.482 --> 00:22:10.162 It's not scary at all. 00:22:10.182 --> 00:22:11.584 We'll just copy whatever you post. 00:22:15.007 --> 00:22:15.828 [Prot]: Life is simple. 00:22:16.809 --> 00:22:20.212 [Pedro]: Yeah, no, but I mean, that's the basic idea. 00:22:20.913 --> 00:22:23.636 You have something very similar to this in the 00:22:24.036 --> 00:22:29.261 actual documentation, both in the docstring and 00:22:29.441 --> 00:22:30.583 in the manual. 00:22:31.143 --> 00:22:34.066 So all this process that I'm telling you is also 00:22:34.166 --> 00:22:37.189 part of the manual, step by step. 00:22:37.928 --> 00:22:41.414 So let's go for this and let's try to... First of 00:22:41.475 --> 00:22:43.478 all, I'm going to load my new variables. 00:22:43.498 --> 00:22:45.001 I'm going to refresh the variables. 00:22:46.143 --> 00:22:47.225 Let's go. NOTE Selecting fonts 00:22:47.285 --> 00:22:49.810 What I'm saying here is I have a main font, 00:22:51.072 --> 00:22:52.194 I have a sans font, 00:22:53.096 --> 00:22:55.140 and I have a mono font, right? 00:22:55.921 --> 00:22:57.744 That's something that I also had here... 00:22:57.745 --> 00:22:59.548 about here in this part. 00:23:02.043 --> 00:23:04.126 The next thing that I'm saying is that the font 00:23:04.146 --> 00:23:07.770 is free serif, which is this mapping, but saying 00:23:07.810 --> 00:23:10.012 that for main I'm going to use free serif, 00:23:10.013 --> 00:23:12.335 for sans I'm going to use free sans, 00:23:12.336 --> 00:23:14.298 and for mono I'm going to use Noto Sans Mono. 00:23:14.599 --> 00:23:15.680 Why Noto Sans Mono? 00:23:15.700 --> 00:23:16.381 Because I like it. 00:23:17.702 --> 00:23:20.806 It's a bit of a change too. So you can also 00:23:20.846 --> 00:23:25.112 see that you can mix font families and so on. 00:23:25.132 --> 00:23:27.915 Basically I like it because it's very readable. 00:23:29.060 --> 00:23:30.181 And it's very accessible. 00:23:30.762 --> 00:23:33.525 So for this, everything would be what I had 00:23:33.585 --> 00:23:34.306 before, right? NOTE #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-23-39-033.jpg #+SCREENSHOT_TIME: 00:23:39.033 #+SCREENSHOT_CROP: 197 494 837 586 00:23:35.287 --> 00:23:38.151 If you have noticed before I compile, you see 00:23:38.231 --> 00:23:41.214 that each font has its own size. 00:23:42.636 --> 00:23:46.140 That is sometimes sort of not very nice, right? 00:23:46.741 --> 00:23:49.224 So what I would do... The next is I'm going to 00:23:49.484 --> 00:23:53.188 add a property to both the sans and the monophon 00:23:53.228 --> 00:23:56.913 to say, try to scale them so that they match the 00:23:56.973 --> 00:23:58.915 lowercase of the main font. 00:24:00.768 --> 00:24:05.294 That sort of makes them more proportional, 00:24:05.775 --> 00:24:09.480 better look and feel, right? NOTE Fallbacks 00:24:10.642 --> 00:24:14.547 Now comes the really nice thing, the fallbacks. 00:24:15.569 --> 00:24:17.632 We've said, fallback is what? 00:24:17.672 --> 00:24:22.218 Fallback is an alternative font I want to add to 00:24:22.699 --> 00:24:27.445 the document when a character is missing. 00:24:27.465 --> 00:24:30.049 I could just have a list of different fonts 00:24:31.210 --> 00:24:36.796 and go with it and have lualatex decide which 00:24:36.976 --> 00:24:37.857 characters to use. 00:24:39.059 --> 00:24:42.563 But that would make the whole thing very, very, very slow. 00:24:43.564 --> 00:24:50.011 So there's one extra step here, which is look at the document, 00:24:51.693 --> 00:24:55.497 find what kind of Emacs scripts you have in the document. 00:24:56.776 --> 00:24:59.539 Then, if you need to add fallbacks, add them 00:24:59.599 --> 00:25:01.702 for the scripts that are present in the document. 00:25:03.664 --> 00:25:05.907 For example, in this case, we have the emojis. 00:25:07.789 --> 00:25:11.654 These are the ones that I want to change. 00:25:12.655 --> 00:25:17.581 But I don't have any simplified Chinese here, 00:25:18.803 --> 00:25:22.547 so there's not going to be any Han character 00:25:23.860 --> 00:25:26.684 in the document, so it's not going to be included 00:25:28.126 --> 00:25:30.710 in the produced LaTeX. 00:25:33.434 --> 00:25:35.177 That's the whole idea of the fallbacks. 00:25:37.720 --> 00:25:40.945 You have fallbacks for the fonts that need them. 00:25:40.965 --> 00:25:43.509 You might want to have fallbacks 00:25:43.629 --> 00:25:47.134 for sans serif, maybe, if you need it. 00:25:48.751 --> 00:25:50.196 If not, you can omit it. 00:25:50.216 --> 00:25:53.024 In this case, we are only having fallbacks here 00:25:53.125 --> 00:25:57.880 in the serif font, so I'm only going to be adding 00:25:58.321 --> 00:26:01.832 the serif, the fallbacks for serif. 00:26:03.263 --> 00:26:05.006 So that's the basic idea. 00:26:05.827 --> 00:26:08.051 If you go to Pandoc and you read the 00:26:08.111 --> 00:26:11.016 documentation, you will notice that most of the 00:26:11.076 --> 00:26:12.999 things are okay. 00:26:13.199 --> 00:26:21.573 It's not an Emacs Lisp list here or a plist because 00:26:21.633 --> 00:26:25.339 it's YAML, but it's very similar. 00:26:26.045 --> 00:26:30.710 I did some state-of-the-art research before 00:26:31.430 --> 00:26:34.113 actually coming at the end to this definition. 00:26:35.034 --> 00:26:38.778 So now, before I compile and we see that we have 00:26:38.978 --> 00:26:42.081 everything we want, I'm also going to show you 00:26:42.181 --> 00:26:44.704 what is produced. 00:26:44.724 --> 00:26:45.104 [Sacha]: Let's see. NOTE The LaTeX produced #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-26-51-067.jpg #+SCREENSHOT_TIME: 00:26:51.067 #+SCREENSHOT_CROP: 462 345 1010 522 00:26:46.485 --> 00:26:49.869 [Pedro]: This is the LaTeX that is produced. 00:26:49.889 --> 00:26:51.791 The important part is here. 00:26:52.932 --> 00:26:53.993 That's the fallbacks. 00:26:55.306 --> 00:26:59.452 If you see, I'm only having the Noto Color 00:26:59.472 --> 00:27:02.116 emoji font being included. 00:27:02.236 --> 00:27:04.579 Why? Because what I told you before, there is no Han, 00:27:05.080 --> 00:27:08.845 so there's no character being added there. 00:27:09.526 --> 00:27:12.510 I don't add the Han font because I don't need it. 00:27:15.094 --> 00:27:18.078 That's sort of a mini optimization that you may 00:27:18.158 --> 00:27:18.799 want to have. 00:27:20.399 --> 00:27:21.520 Next thing... NOTE Character families should be in the Messages buffer 00:27:21.521 --> 00:27:23.644 [Sacha]: Sorry, how do people find out what keys 00:27:23.704 --> 00:27:26.087 they need to put into that alist? Can they 00:27:26.147 --> 00:27:28.371 describe-char for example to see whether 00:27:28.731 --> 00:27:31.235 they need to put in emoji or some other...? 00:27:32.336 --> 00:27:37.163 [Pedro]: That's a very good question. For 00:27:37.203 --> 00:27:39.987 that, I need to get to the Messages, 00:27:43.693 --> 00:27:47.318 or... Oops. I have Warnings... 00:27:52.006 --> 00:27:52.687 Where is it? 00:27:54.028 --> 00:28:07.020 You should normally have, in your Messages, 00:28:07.021 --> 00:28:10.283 a message from Org Mode 00:28:10.284 --> 00:28:13.687 telling you that it found the characters. 00:28:13.727 --> 00:28:16.110 I don't know why I'm not seeing it here. 00:28:16.130 --> 00:28:17.912 [Sacha]: Is it maybe in that Tilix window that 00:28:17.952 --> 00:28:19.313 you have there with a make edit? 00:28:20.778 --> 00:28:21.419 [Pedro]: No, no, no. 00:28:22.161 --> 00:28:24.286 It's inside the Emacs. 00:28:24.587 --> 00:28:31.783 And I'm not finding it. 00:28:35.105 --> 00:28:35.966 I'm not finding it. 00:28:36.026 --> 00:28:38.390 I don't know why, but it should be there. 00:28:39.151 --> 00:28:45.240 Normally, Org Mode would tell you that you have 00:28:46.222 --> 00:28:49.547 these and those characters, that you have the 00:28:49.587 --> 00:28:50.388 different families. 00:28:50.849 --> 00:29:01.324 You will also see it if I were to, say... I'm going 00:29:01.365 --> 00:29:04.369 to move this out so I don't have fallbacks. 00:29:06.475 --> 00:29:09.499 This would be a basic configuration. 00:29:10.260 --> 00:29:18.292 So now that I have my fallbacks out, and I go and 00:29:18.312 --> 00:29:26.783 I... So you see that I don't have fallbacks. 00:29:27.184 --> 00:29:30.829 And then if I go to the warrings, where is it? 00:29:33.020 --> 00:29:34.590 [Prot]: Maybe messages? 00:29:35.435 --> 00:29:37.005 [Pedro]: To the messages, exactly. 00:29:44.984 --> 00:29:46.225 I'm sorry, I don't find it. 00:29:46.766 --> 00:29:47.287 [Sacha]: It's okay. 00:29:47.487 --> 00:29:48.528 [Pedro]: But there is a message, 00:29:49.009 --> 00:29:51.911 I'm sure there is a message 00:29:51.912 --> 00:29:56.797 which tells you which families or which scripts 00:29:56.798 --> 00:29:59.601 you have in your document. 00:30:00.362 --> 00:30:00.562 [Sacha]: All right. 00:30:00.582 --> 00:30:01.403 No problem. 00:30:01.423 --> 00:30:03.826 I'm sure all of this will come with wonderful 00:30:03.886 --> 00:30:09.633 documentation because the Org project is very good at that. NOTE Behind the scenes 00:30:09.653 --> 00:30:12.076 [Pedro]: Yes. We spend a lot of time going through the 00:30:12.116 --> 00:30:13.337 documentation. 00:30:14.009 --> 00:30:19.398 The current documentation currently has undergone 00:30:19.658 --> 00:30:23.444 a lot of scrutiny both by Ihor and by myself. 00:30:23.645 --> 00:30:25.608 I've rewritten it two or three times to make it 00:30:25.708 --> 00:30:30.656 clearer as I gained insights into what I wanted 00:30:30.716 --> 00:30:33.621 to tell people and 00:30:33.601 --> 00:30:36.985 At one point in time I isolated myself completely. 00:30:37.005 --> 00:30:38.326 I rewrote everything. 00:30:38.887 --> 00:30:42.471 I forgot about all the exchanges I had with Ihor 00:30:42.791 --> 00:30:47.957 and I had my own ideas of what I wanted to say. 00:30:48.337 --> 00:30:50.639 I rewrote everything from scratch and suddenly 00:30:50.800 --> 00:30:52.181 Ihor said, this is what I wanted. 00:30:53.883 --> 00:30:58.508 Currently the documentation is at least a bit 00:30:58.548 --> 00:30:59.809 more than just usable. NOTE Refreshing local variables 00:31:00.410 --> 00:31:02.292 Let's go back to this. 00:31:03.082 --> 00:31:09.171 As you see... This is something stupid. 00:31:09.191 --> 00:31:12.797 I mean, you see that this refresh local variables 00:31:13.258 --> 00:31:15.100 is a small function that I've written myself. 00:31:15.842 --> 00:31:20.889 Since I'm working a lot with local variables, I'm 00:31:21.270 --> 00:31:24.315 also trying to have shortcuts for all this kind 00:31:24.375 --> 00:31:24.695 of things. 00:31:25.036 --> 00:31:27.179 [Sacha]: Oh, I thought C-c C-c did that in the 00:31:27.219 --> 00:31:27.840 directive. 00:31:29.288 --> 00:31:34.155 [Pedro]: Sometimes it does not, or at least... 00:31:34.175 --> 00:31:39.682 the case is... I did it myself. 00:31:39.963 --> 00:31:42.246 I rewrote everything to be sure that I got my 00:31:42.486 --> 00:31:44.509 local variables, my directory variables, and 00:31:44.569 --> 00:31:49.115 everything nice as I wanted. Once I have this... 00:31:49.395 --> 00:31:53.241 Once again... There. If I now go and compile 00:31:53.261 --> 00:31:55.684 for the last time... 00:31:58.313 --> 00:31:59.334 What is missing? 00:31:59.375 --> 00:32:01.778 What have I done wrong? 00:32:15.958 --> 00:32:19.764 [Sacha]: Do you want a series of undoes? 00:32:20.384 --> 00:32:22.648 [Pedro]: I'm going to go the quick way. 00:32:23.329 --> 00:32:25.231 [Prot]: I think there was a missing parenthesis in that. 00:32:25.492 --> 00:32:26.413 [Pedro]: Yeah, somewhere. 00:32:27.322 --> 00:32:31.927 [Prot]: Yeah, yeah, yeah, there it is. 00:32:37.073 --> 00:32:39.576 [Pedro]: No, it looks right, actually. 00:32:39.596 --> 00:32:44.682 No, yes, because I've copied it from my notebook. 00:32:49.948 --> 00:32:50.889 [Sacha]: It's not doing the thing yet. 00:32:53.312 --> 00:32:53.532 [Pedro]: Yeah, yeah. 00:32:55.165 --> 00:32:59.346 And that comes... That's something that I... 00:32:59.347 --> 00:33:01.176 So at least if you do it like this... 00:33:03.215 --> 00:33:06.560 I don't know why this hand does not work 00:33:06.600 --> 00:33:09.264 correctly with Noto Color Emoji. 00:33:09.765 --> 00:33:11.167 Another takeaway. 00:33:11.188 --> 00:33:14.773 You have to find your own wonderful emoji file. 00:33:15.094 --> 00:33:16.115 The other one is Apple. 00:33:16.296 --> 00:33:21.343 Apple Color Emoji has a different [??] for emojis. 00:33:21.964 --> 00:33:23.807 That's something that I also wanted to highlight. 00:33:23.887 --> 00:33:27.433 Sometimes you don't get the full... 00:33:27.413 --> 00:33:31.458 It's important you have to do a lot of look and find. 00:33:32.459 --> 00:33:35.661 But at least with this you get your... 00:33:35.662 --> 00:33:37.705 At least with this, what do I get? NOTE Scaling fonts to match lowercase #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-33-39-067.jpg #+SCREENSHOT_TIME: 00:33:39.067 #+SCREENSHOT_CROP: 175 429 841 934 00:33:37.885 --> 00:33:43.331 I get my fonts scaled so that they look more or 00:33:43.351 --> 00:33:44.733 less the same size, 00:33:46.194 --> 00:33:47.376 which is a nice appearance. 00:33:47.456 --> 00:33:52.281 I have my listings and my outputs. 00:33:53.963 --> 00:33:54.984 And I have my emojis. 00:33:56.646 --> 00:33:57.387 [Prot]: This looks good. 00:33:58.396 --> 00:33:59.579 [Pedro]: This is basically what I need. 00:33:59.580 --> 00:34:02.883 [Prot]: Can I ask something? 00:34:03.484 --> 00:34:04.746 Two questions, actually. 00:34:05.708 --> 00:34:10.535 One is, I see the props parameter you have for the... 00:34:10.536 --> 00:34:11.838 if you switch to your Emacs buffer... 00:34:13.360 --> 00:34:13.981 [Pedro]: Just a second. 00:34:14.001 --> 00:34:16.205 I was on the wrong stream. 00:34:16.344 --> 00:34:18.007 [Prot]: Okay. NOTE Specifying the fallbacks 00:34:17.987 --> 00:34:22.375 So you see you have that props, and then you are 00:34:22.435 --> 00:34:27.963 passing it a string. In the fallbacks, you have a 00:34:28.023 --> 00:34:32.268 slightly different expression where you have the 00:34:32.329 --> 00:34:35.554 font set, and then you pass it the font, but for 00:34:35.594 --> 00:34:37.838 example, with Noto Color Emoji, you are not just 00:34:37.878 --> 00:34:40.021 passing Noto Color Emoji, you are also passing a 00:34:40.081 --> 00:34:44.788 parameter to me that looks like the pango markup, 00:34:45.292 --> 00:34:48.518 whereas the other one looks like something else, 00:34:48.539 --> 00:34:50.322 another markup. 00:34:50.342 --> 00:34:52.708 I just want to ask if you know, off the top of 00:34:52.748 --> 00:34:56.034 your head, if the documentation for Pango and for 00:34:56.074 --> 00:34:57.758 whatever else that is, 00:34:57.759 --> 00:35:02.000 is clearly available somewhere so that we can follow it because 00:35:02.001 --> 00:35:03.933 one issue I always find with 00:35:03.934 --> 00:35:07.833 LaTeX or with anything that has to do with these complex things 00:35:07.834 --> 00:35:10.600 is that they assume you are a domain expert. 00:35:10.601 --> 00:35:12.167 And it's like, okay, just do fonts 00:35:12.168 --> 00:35:13.733 the way you do it in LaTeX. 00:35:13.734 --> 00:35:15.833 And I'm like, okay, how do you do it... 00:35:15.834 --> 00:35:16.567 [Pedro]: "No, but how?" 00:35:16.568 --> 00:35:23.955 Yeah, I know. Exactly. That part is extremely complex. 00:35:25.117 --> 00:35:28.621 Because in this case, what you're saying 00:35:28.741 --> 00:35:31.303 is do whatever you do with your normal fonts 00:35:31.304 --> 00:35:32.345 if you don't have the colon. NOTE HarfBuzz 00:35:33.006 --> 00:35:36.350 If you have the colon, then what they say is you 00:35:36.410 --> 00:35:39.394 need to have, in this case, you need to have the 00:35:39.434 --> 00:35:46.582 HarfBuzz rasterizer or however it is called. 00:35:47.563 --> 00:35:49.986 And that... Yes, you're completely right. 00:35:50.306 --> 00:35:52.569 This is a problem with the documentation. 00:35:52.589 --> 00:35:56.256 I have... To the best of my knowledge, 00:35:56.257 --> 00:36:03.225 the only ones that really need harf are the emojis. 00:36:05.360 --> 00:36:07.690 That's the example. 00:36:07.710 --> 00:36:10.233 For the rest of the font, at least at this point 00:36:10.313 --> 00:36:14.117 in time, you don't need, or with this feature 00:36:14.137 --> 00:36:17.261 set, you don't need any sort of extra things. NOTE Upcoming work on Babel (the LaTeX package) 00:36:17.821 --> 00:36:20.284 The extra things will come in the future, 00:36:20.804 --> 00:36:26.211 when you try to integrate things like Babel, 00:36:26.212 --> 00:36:29.255 with controlling the fonts in Babel, 00:36:29.256 --> 00:36:31.418 which I personally already have running, 00:36:31.419 --> 00:36:35.045 but it's not yet integrated into Org Mode. 00:36:35.766 --> 00:36:37.408 That's going to take another year, I think. 00:36:39.751 --> 00:36:42.275 [Prot]: But you are persistent, so yeah, you will do it. 00:36:42.295 --> 00:36:45.239 [Pedro]: Yeah, no, no. I mean, I want to have it. 00:36:45.660 --> 00:36:46.782 I need it also. 00:36:46.802 --> 00:36:49.866 At least I would like to see it officially 00:36:50.183 --> 00:36:54.096 sanctioned by whoever needs to sanction it before 00:36:54.176 --> 00:36:55.461 I retire in two years, 00:36:57.869 --> 00:37:00.598 so that people can use it once I retire. NOTE More about HarfBuzz and ligatures? 00:37:02.401 --> 00:37:05.364 [Prot]: Good, good. The other question I have related to this and 00:37:05.384 --> 00:37:09.048 emoji is I remember when HarfBuzz was discussed 00:37:09.068 --> 00:37:11.130 in the context of integrating it with Emacs. 00:37:11.170 --> 00:37:13.813 Maybe that was Emacs 28 or something along those lines. 00:37:14.734 --> 00:37:15.915 There was this point. 00:37:15.995 --> 00:37:18.397 Now, maybe I am conflating things, but there was 00:37:18.417 --> 00:37:21.861 this point that with this included, like you 00:37:21.921 --> 00:37:25.304 build Emacs with support for HarfBuzz and now it 00:37:25.385 --> 00:37:27.767 also supports ligatures. 00:37:27.747 --> 00:37:31.375 Now, a ligature is something that is fancy. Like, for 00:37:31.435 --> 00:37:34.021 example, you take three characters, like 00:37:34.061 --> 00:37:36.025 the equal sign in JavaScript, 00:37:36.026 --> 00:37:37.147 three equal signs, 00:37:37.148 --> 00:37:39.894 and you make it look like one character. 00:37:40.155 --> 00:37:42.280 I don't know if this is part of what you are 00:37:42.360 --> 00:37:42.921 doing here. 00:37:43.778 --> 00:37:47.864 [Pedro]: Well, a very good question. 00:37:50.287 --> 00:37:55.274 This discussion that there was on Emacs 28 for 00:37:55.594 --> 00:37:58.298 HarfBuzz (yes or no) and ligatures and all these 00:37:58.338 --> 00:38:02.343 kind of things has not helped. 00:38:03.365 --> 00:38:04.666 Let's say it this way... 00:38:06.269 --> 00:38:08.031 It adds confusion. 00:38:08.011 --> 00:38:13.178 In this context, sorry to be that clear. 00:38:13.419 --> 00:38:16.723 Why? Because you have to think it as two different, 00:38:17.264 --> 00:38:19.868 this is two different, completely different pairs 00:38:19.928 --> 00:38:20.248 of shoes. 00:38:20.869 --> 00:38:24.154 [Sacha]: Okay, because we're saying it's not LaTeX export. 00:38:24.174 --> 00:38:28.099 [Pedro]: Yes, so I'm delegating as much as I can 00:38:28.800 --> 00:38:32.345 to lualatex and to all these, so ligatures and 00:38:32.405 --> 00:38:35.790 things like that are part of LaTeX. 00:38:36.479 --> 00:38:39.505 I don't want to treat them at all. 00:38:40.126 --> 00:38:44.575 The only point where I need to sort of touch 00:38:45.997 --> 00:38:48.242 lower level, which would be like HarfBuzz, for 00:38:48.302 --> 00:38:52.430 example, is telling lualatex to use HarfBuzz to 00:38:53.351 --> 00:38:54.313 include some characters. 00:38:56.097 --> 00:38:58.842 So that's the full question. 00:38:59.041 --> 00:39:00.463 [Sacha]: So far you've shown us how to include 00:39:00.523 --> 00:39:04.027 Unicode in the output and also how to specify 00:39:04.067 --> 00:39:06.870 fallback fonts for things like emojis and other 00:39:06.910 --> 00:39:08.992 things that are not included in your main font. 00:39:09.493 --> 00:39:11.435 Were there other parts of your very large patch? 00:39:13.617 --> 00:39:18.203 [Pedro]: Oh, the other parts, which are not... 00:39:18.223 --> 00:39:22.127 Well, I mean, the current state is that we have 00:39:22.387 --> 00:39:23.108 all this. NOTE Upcoming talk on accessible presentations 00:39:23.982 --> 00:39:28.612 We also have support for accessible presentations, 00:39:29.193 --> 00:39:33.082 and there... In EmacsConf 26, we're going to have a 00:39:34.645 --> 00:39:37.391 presentation on how this came to life by the 00:39:37.451 --> 00:39:38.593 person who inspired it, 00:39:38.594 --> 00:39:42.362 and by the person who wrote it, yours truly. 00:39:42.629 --> 00:39:46.593 The third thing that we have for that, 00:39:46.594 --> 00:39:51.201 we also have a lot of things related with metadata 00:39:52.382 --> 00:39:56.468 management for LaTeX and for PDF documents. 00:39:57.369 --> 00:40:01.635 That's also something where I directly call on 00:40:01.755 --> 00:40:07.082 experts to help us because I know what I know, 00:40:07.342 --> 00:40:10.206 which is at this point I know I don't know. 00:40:10.608 --> 00:40:12.890 That's my current situation. 00:40:14.652 --> 00:40:16.734 I know very little about all this document 00:40:17.535 --> 00:40:19.237 metadata stuff. 00:40:19.257 --> 00:40:23.401 I'm learning in the process of writing things 00:40:23.461 --> 00:40:26.865 for accessible presentations. 00:40:27.285 --> 00:40:31.369 I'm also learning things related with tagging 00:40:31.409 --> 00:40:34.152 PDFs and the tag PDF and all this kind of 00:40:35.554 --> 00:40:38.016 packages which are emerging right now and latest. 00:40:39.160 --> 00:40:40.122 And all this kind of things. 00:40:40.202 --> 00:40:44.730 This integration, which is starting to be fruitful. 00:40:45.672 --> 00:40:50.140 I already have... I'm using it with a couple of 00:40:50.200 --> 00:40:51.442 students who need that. 00:40:52.003 --> 00:40:54.688 They can access my presentations right now. 00:40:57.753 --> 00:41:03.143 I'm also sort of using them as guinea pigs, 00:41:03.123 --> 00:41:07.827 as willing guinea pigs. I'm going beyond 00:41:08.308 --> 00:41:10.610 presentations to providing them with documents 00:41:10.670 --> 00:41:13.592 and so on, and they provide feedback about how 00:41:13.693 --> 00:41:15.554 accessible those documents are and so on. 00:41:16.615 --> 00:41:18.937 For that, we need help. NOTE Upcoming work on Polyglossia and Babel 00:41:19.838 --> 00:41:22.741 Then the other big thing which is still part 00:41:22.781 --> 00:41:29.667 of our conversation right now is the Polyglossia 00:41:30.928 --> 00:41:32.910 and Babel packages. 00:41:33.346 --> 00:41:40.535 Polyglossian Babel for internationalization. 00:41:42.898 --> 00:41:46.061 For example, Babel is very nice in managing 00:41:46.062 --> 00:41:50.007 left to right and right to left and things like that. 00:41:51.389 --> 00:41:53.211 [Sacha]: I don't suppose you have a quick demonstration? 00:41:54.558 --> 00:41:55.560 [Pedro]: Not yet. 00:41:57.042 --> 00:42:00.327 That would be another long presentation to show 00:42:00.368 --> 00:42:02.892 you all the kinds of things that we can do. NOTE Chinese, Japanese, and Korean #+SCREENSHOT: emacs-chat-31-pedro-a-aranda-gutierrez-00-43-47-100.jpg #+SCREENSHOT_TIME: 00:43:47.100 #+SCREENSHOT_CROP: 432 623 1027 689 00:42:02.932 --> 00:42:09.583 And just before we close this, I've been showing 00:42:10.023 --> 00:42:14.190 what I mostly hear dominate, which is English, 00:42:14.571 --> 00:42:18.878 Spanish, or Latin alphabet. 00:42:18.858 --> 00:42:23.224 But with this package, we also have initial 00:42:23.264 --> 00:42:28.390 support for when you're using xelatex. 00:42:29.011 --> 00:42:33.336 There's a package which is called xeCJK to 00:42:33.477 --> 00:42:36.559 provide simplified Chinese, Japanese, 00:42:36.560 --> 00:42:37.962 and Korean documents. 00:42:38.643 --> 00:42:43.689 So this, with lualatex, raw, without any Babel or 00:42:43.770 --> 00:42:46.732 anything else, raw and monolingual, 00:42:46.733 --> 00:42:48.055 is also supported. 00:42:48.794 --> 00:42:53.861 What I basically do is I map what is on 00:42:54.101 --> 00:42:58.307 Overleaf describing how to do this with xelatex 00:42:58.427 --> 00:43:01.992 and this package which is called xeCJK. 00:43:03.575 --> 00:43:07.761 I map that with the same kind of header. 00:43:08.021 --> 00:43:11.946 You can use this kind of variables to do the same 00:43:12.547 --> 00:43:17.434 for this Chinese, Japanese, and Korean document. 00:43:18.022 --> 00:43:20.888 [Sacha]: Would you happen to have an example so 00:43:21.650 --> 00:43:23.754 that people can see what the markup looks like? 00:43:25.738 --> 00:43:28.564 [Pedro]: The markup, I mean, the idea would be 00:43:28.724 --> 00:43:32.993 instead of using main... I don't have an example here.. 00:43:34.289 --> 00:43:36.895 But the idea is that instead of calling the fonts 00:43:37.155 --> 00:43:41.464 main, sans, and mono, you have CJKmain. 00:43:41.945 --> 00:43:43.910 So you would have something like, the font would 00:43:43.950 --> 00:43:45.132 be like CJKMain. 00:43:45.774 --> 00:43:50.203 C, J, K. And the other thing is that you would 00:43:50.243 --> 00:43:54.272 have to change your compiler to xelatex. 00:43:55.973 --> 00:44:02.502 That's also very, very well documented in current 00:44:03.082 --> 00:44:04.364 main Org. 00:44:05.165 --> 00:44:06.206 So we have it. 00:44:10.672 --> 00:44:13.977 That would be more or less the effort of the last 00:44:14.137 --> 00:44:20.265 couple of years and how it has crystallized into 00:44:20.305 --> 00:44:22.588 things that are already in master. NOTE The development process 00:44:22.973 --> 00:44:26.497 [Sacha]: I'm quite curious about the development process. 00:44:26.577 --> 00:44:29.641 Since it's a large change and it's taking a while 00:44:29.661 --> 00:44:32.144 to develop, what was it like for you to keep it 00:44:32.204 --> 00:44:35.989 going despite all the developments in Org 00:44:36.069 --> 00:44:39.113 underneath? 00:44:39.133 --> 00:44:41.055 [Pedro]: Well, it's not that difficult. 00:44:41.756 --> 00:44:46.382 At the end, I needed to learn to use VC. 00:44:47.948 --> 00:44:51.653 Last year in Emacs, I'm doing everything from the 00:44:51.673 --> 00:44:53.996 VC package in Emacs. 00:44:54.016 --> 00:44:58.361 So I have a nice workflow there to go and get 00:44:58.461 --> 00:45:03.948 my... I have two... On one side, I'm, of course, 00:45:05.150 --> 00:45:08.034 going back and putting everything into the main 00:45:08.314 --> 00:45:11.458 Org git. 00:45:11.438 --> 00:45:16.086 But I'm also having my own Codeberg with most 00:45:16.146 --> 00:45:18.290 things that I'm using. 00:45:19.993 --> 00:45:23.379 With Emacs it's quite easy to go, either modify 00:45:23.519 --> 00:45:26.885 one and have the things pushed up, or generating 00:45:26.925 --> 00:45:31.793 the patches so that Ihor blesses them, 00:45:31.773 --> 00:45:33.315 puts them into main. 00:45:33.735 --> 00:45:38.520 Or when I'm doing lots of things that I need for 00:45:38.600 --> 00:45:40.803 a certain point, I go to my Codeberg. 00:45:40.843 --> 00:45:42.725 There I have my FontSpec 3 branch. 00:45:43.305 --> 00:45:45.428 There I do my own changes. 00:45:45.448 --> 00:45:46.108 I check them. 00:45:46.168 --> 00:45:46.809 I test them. 00:45:46.989 --> 00:45:49.051 I test drive them. 00:45:49.692 --> 00:45:52.715 Because, yes, it's been a lot of work to have 00:45:52.795 --> 00:45:53.917 everything running. 00:45:55.138 --> 00:45:59.322 But it's always based on real life. 00:45:59.342 --> 00:46:04.119 And how would... I try to say 00:46:04.120 --> 00:46:06.703 how would someone do it, who knows a 00:46:06.763 --> 00:46:12.151 bit about Emacs and a bit about LaTeX and wants 00:46:12.651 --> 00:46:14.133 to write as little as possible. 00:46:15.736 --> 00:46:17.799 [Sacha]: This is so much better than the 00:46:17.879 --> 00:46:21.403 other way of specifying a new LaTeX command that 00:46:21.424 --> 00:46:23.326 just changes the font for that one specific 00:46:23.366 --> 00:46:24.908 character and then you have to use it for all the 00:46:24.968 --> 00:46:26.430 emojis that you have in your document. 00:46:26.891 --> 00:46:29.775 So having it automatically fall back is so much nicer. 00:46:31.308 --> 00:46:34.452 [Pedro]: Yeah, no, I mean, I was not going to go 00:46:34.512 --> 00:46:37.856 the PDF way of having every emoji that I want, 00:46:38.557 --> 00:46:41.300 recoded, and then inserted into the document. 00:46:42.041 --> 00:46:44.965 I tried some. 00:46:46.887 --> 00:46:47.528 What was it? 00:46:47.949 --> 00:46:48.810 When did we start? 00:46:50.652 --> 00:46:53.115 I tried it like eight years ago, and I decided 00:46:53.155 --> 00:46:57.154 once I have emojis in a nice way, I will use them. 00:46:57.795 --> 00:47:00.340 So my first manuals, if you go back, the first 00:47:00.380 --> 00:47:03.285 manuals that I distributed for that specific 00:47:03.345 --> 00:47:05.189 course did not have any emojis. 00:47:06.571 --> 00:47:12.482 So I was able to do basically this, which was 00:47:12.783 --> 00:47:16.890 what I needed for... I started with a header like this. 00:47:19.334 --> 00:47:23.639 And then when things came and I also needed to go 00:47:23.739 --> 00:47:28.765 like the scaling, what I did was scale. 00:47:31.969 --> 00:47:35.093 So this was my beginning in all this. 00:47:37.496 --> 00:47:39.939 And these were the commands that I always had in 00:47:39.959 --> 00:47:42.802 the beginning with my lualatex. 00:47:43.062 --> 00:47:45.325 The moment I moved to lualatex 00:47:45.777 --> 00:47:52.248 which was when I needed to do things like this. 00:47:53.730 --> 00:47:58.017 Then once I wanted to go further, then I 00:47:58.057 --> 00:48:00.742 started. When this came to the mailing list, 00:48:00.982 --> 00:48:03.246 my first answer was I'm doing this. 00:48:04.660 --> 00:48:06.523 And then the next question was, 00:48:06.524 --> 00:48:07.507 what about the emojis? 00:48:07.527 --> 00:48:12.499 That sort of triggered the rest. NOTE Retirement 00:48:12.519 --> 00:48:14.583 [Sacha]: You mentioned wanting to retire in two years. 00:48:15.024 --> 00:48:17.751 I very much hope that you will retire and do even 00:48:17.811 --> 00:48:18.653 more Emacs. 00:48:19.814 --> 00:48:26.965 [Pedro]: I want to continue because it's a nice pastime. 00:48:28.587 --> 00:48:34.396 I want to transition from my tool at work, or in 00:48:34.456 --> 00:48:38.762 this case, my tool at the university, to this is 00:48:38.843 --> 00:48:44.291 a nice pastime to keep this up and running. 00:48:44.812 --> 00:48:47.155 It would be a real pity to lose all this. 00:48:47.860 --> 00:48:49.544 [Sacha]: I'm sure we can find you volunteer 00:48:49.584 --> 00:48:50.106 guinea pigs. 00:48:52.652 --> 00:48:56.341 [Pedro]: I hope I'll get more and more, because 00:48:56.762 --> 00:49:01.313 the more people test it, the better it will get. 00:49:01.968 --> 00:49:04.411 And I'm really looking forward to... I mean, I 00:49:04.471 --> 00:49:08.716 already showed all the things that you can do in 00:49:08.736 --> 00:49:15.383 the last big... In the last Emacs conference last winter. 00:49:16.444 --> 00:49:19.368 So I'm not going to repeat this here. 00:49:20.509 --> 00:49:24.413 So the talk, because I think it's more important 00:49:24.474 --> 00:49:27.377 to go move on to the next big thing and have 00:49:27.437 --> 00:49:30.220 people doing accessible documents. 00:49:31.634 --> 00:49:35.826 I'll keep on top of my donkey with the train 00:49:35.866 --> 00:49:39.357 behind me until I get Babel and Polyglossia also 00:49:39.437 --> 00:49:41.122 integrated. NOTE Fallback fonts 00:49:41.142 --> 00:49:42.285 [Prot]: Nice, nice, nice. 00:49:43.902 --> 00:49:46.484 Let me ask you the other question I had, 00:49:46.504 --> 00:49:48.146 which maybe it is simple. 00:49:48.666 --> 00:49:50.428 Because you mentioned that you want to have 00:49:50.908 --> 00:49:53.571 local variables so that the documents are 00:49:53.651 --> 00:49:57.474 portable, which brings me to the question about 00:49:58.075 --> 00:49:59.916 specifying generic fonts. 00:49:59.936 --> 00:50:04.320 Like, can you do something like use whatever the 00:50:04.380 --> 00:50:07.864 system has for Serif, Sans, Mono? 00:50:07.884 --> 00:50:11.467 Or you always have to specify a font family, like 00:50:11.547 --> 00:50:13.909 Free Serif, Noto Sans, etc. 00:50:13.889 --> 00:50:19.198 [Pedro]: At this point, this was a very very long 00:50:19.238 --> 00:50:20.360 discussion with Ihor. 00:50:20.901 --> 00:50:23.605 Do you want to have generics or do you want to 00:50:23.665 --> 00:50:24.607 have specifics? 00:50:26.510 --> 00:50:29.956 I was looking more towards specifics. 00:50:30.317 --> 00:50:35.911 Why? Because in my experience, I mean, before I 00:50:35.931 --> 00:50:39.216 started lecturing, I was for a long time in a 00:50:39.316 --> 00:50:41.079 research and development company. 00:50:41.981 --> 00:50:43.483 We had European projects. 00:50:43.744 --> 00:50:49.754 There you had your own LaTeX templates. 00:50:50.294 --> 00:50:53.059 There you had to specify each and every font. 00:50:54.052 --> 00:50:57.578 Then in parallel with that, in the company they 00:50:57.618 --> 00:51:01.784 had their own templates for PowerPoint, for Word 00:51:01.844 --> 00:51:05.510 and so on, and you had to specify each and every 00:51:05.570 --> 00:51:07.854 font that you wanted to use. 00:51:08.695 --> 00:51:13.723 So I've been mapping my experience in my prior 00:51:13.763 --> 00:51:16.547 life into this. 00:51:17.409 --> 00:51:19.452 Having font families 00:51:20.090 --> 00:51:22.592 might be nice. For example, Bera, I said at the 00:51:22.632 --> 00:51:24.494 beginning, that's one of the 00:51:24.554 --> 00:51:27.577 things that might be interesting, right? But as you 00:51:27.617 --> 00:51:31.720 see, even in pdflatex, with Bera, you have to 00:51:31.921 --> 00:51:35.384 include three different packages. So what's 00:51:35.444 --> 00:51:37.306 the difference between including three different 00:51:37.346 --> 00:51:41.870 packages, or directly saying, here, you have 00:51:41.890 --> 00:51:44.492 three different fonts, one for main, one for... 00:51:44.552 --> 00:51:47.495 it's only one for main, one 00:51:47.535 --> 00:51:50.097 for sans and one for... 00:51:50.077 --> 00:51:52.083 That's the only thing that you have in LaTeX. 00:51:52.705 --> 00:51:55.454 [Sacha]: So, not like the CSS way of specifying 00:51:55.474 --> 00:51:59.146 multiple fonts just in case one of them is not there? 00:51:59.166 --> 00:52:04.232 [Pedro]: No. I mean, in this case, 00:52:04.233 --> 00:52:09.400 you are always going to go with what the system has. 00:52:09.420 --> 00:52:12.424 Normally, most of the fonts that, at least 00:52:12.484 --> 00:52:15.608 most of the fonts that I recommend to use, are 00:52:15.669 --> 00:52:19.333 fonts that are included because you have your 00:52:19.554 --> 00:52:22.718 LaTeX installed underneath. 00:52:24.160 --> 00:52:26.843 It's not that I'm going to be there with a, I 00:52:26.883 --> 00:52:30.428 don't know, what was this, what was it? 00:52:31.505 --> 00:52:36.614 Well, there was this Telefonica font family. 00:52:40.120 --> 00:52:41.583 No one is going to have it in real life. 00:52:43.987 --> 00:52:49.016 But the Free fonts are part of LaTeX. 00:52:49.537 --> 00:52:51.560 Another font that I really like a lot, because 00:52:51.620 --> 00:52:56.128 it's very, very readable, is the TeX Gyre family. 00:52:56.969 --> 00:52:58.813 TeX Gyre is very, very readable. 00:52:59.174 --> 00:53:00.436 They have all kinds of fonts. 00:53:00.937 --> 00:53:01.879 I'm using two. 00:53:02.140 --> 00:53:02.260 Why? 00:53:02.300 --> 00:53:04.505 Because it's part of LaTeX. NOTE LaTeX versions 00:53:06.429 --> 00:53:09.254 And by the way, all this that I've shown today 00:53:09.295 --> 00:53:15.788 can be done with a LaTeX which is basically 2024. 00:53:17.202 --> 00:53:20.570 What I'm going to be showing in December needs a 00:53:20.650 --> 00:53:22.274 very, very modern LaTeX. 00:53:22.935 --> 00:53:23.116 Why? 00:53:23.136 --> 00:53:27.025 Because accessibility has not been in the wire 00:53:27.165 --> 00:53:30.954 for the last... Well, it has only been in the 00:53:30.994 --> 00:53:32.157 wire for the last months. 00:53:33.352 --> 00:53:36.295 So, the real support for accessibility things 00:53:36.415 --> 00:53:39.718 comes with very, very modern LaTeX stuff. 00:53:39.738 --> 00:53:42.681 But this is something you can do with, I mean, 00:53:42.901 --> 00:53:49.307 Ubuntu 2024, I'm sure, and Ubuntu even earlier, 00:53:49.668 --> 00:53:50.769 you could also do. 00:53:51.690 --> 00:53:55.393 Because the LaTeX is not that fancy. 00:53:55.413 --> 00:53:58.016 I think fallback fonts are already a 00:53:58.076 --> 00:54:01.779 well-established mechanism in lualatex for a lot 00:54:01.819 --> 00:54:02.380 of time. 00:54:03.608 --> 00:54:05.210 [Sacha]: Would you happen to know if the widely 00:54:05.250 --> 00:54:08.234 released LaTeX 2024, for example, might have 00:54:08.314 --> 00:54:10.837 support for multiple fallback fonts? 00:54:13.040 --> 00:54:16.445 [Pedro]: That should be starting 2020. 00:54:16.485 --> 00:54:26.378 [Sacha]: Okay. So it's just in the Org side 00:54:26.498 --> 00:54:29.182 where you might say main, this font is just Free 00:54:29.202 --> 00:54:32.586 Serif, and not Free Serif, something, something. 00:54:33.022 --> 00:54:37.770 [Pedro]: Yeah exactly all right because... 00:54:37.771 --> 00:54:40.235 [Sacha]: I do actually use custom fonts. When I export my 00:54:40.295 --> 00:54:42.760 configuration, I want to use my handwriting for 00:54:42.840 --> 00:54:45.465 the titles, but of course nobody else is going 00:54:45.485 --> 00:54:47.508 to have my handwriting... I mean, it's 00:54:47.548 --> 00:54:49.392 public but people are not going to go to the 00:54:49.412 --> 00:54:52.337 trouble of downloading it, so yes, it would be nice 00:54:52.377 --> 00:54:55.142 to have that fallback. Not that I'm saying, "Please 00:54:55.162 --> 00:54:56.785 get on your donkey and go in that direction..." 00:54:58.233 --> 00:55:01.397 [Pedro]: No, of course not. But in this case, 00:55:01.817 --> 00:55:05.682 good news for you, if you have this as a ttf or as 00:55:05.742 --> 00:55:10.587 an otf font, then you can include that as your 00:55:10.627 --> 00:55:14.492 main font, and the PDF that you post at the end 00:55:15.413 --> 00:55:19.157 will have that already embedded, so you can 00:55:19.378 --> 00:55:21.540 produce your PDFs with your fonts. 00:55:21.858 --> 00:55:24.843 The thing is, if you want to share the font, if 00:55:24.863 --> 00:55:27.688 you want to share the source and then have other 00:55:27.728 --> 00:55:30.633 people compile it, then you have to tell them to 00:55:30.693 --> 00:55:35.782 install the fonts that are missing. 00:55:36.523 --> 00:55:38.026 But I think that's something that happens always. 00:55:39.148 --> 00:55:40.871 You enter a company, they have their own fonts. 00:55:40.911 --> 00:55:43.195 The first thing that you have to do, if you don't 00:55:43.235 --> 00:55:48.918 get a pre-installed computer: download the fonts. 00:55:49.018 --> 00:55:51.061 If you want to produce documents for the company, 00:55:51.081 --> 00:55:53.084 you have to download the fonts and install them 00:55:53.124 --> 00:55:54.406 so that you can produce. 00:55:55.247 --> 00:55:56.949 You don't have a boss telling you, no, 00:55:56.989 --> 00:55:58.672 the fonts, the fonts, the fonts. 00:55:59.593 --> 00:56:01.375 [Prot]: It's part of the brand, of course, yeah. 00:56:01.716 --> 00:56:03.879 [Pedro]: Yeah, no, no, it's a very important part 00:56:03.919 --> 00:56:04.420 of the brand. 00:56:04.660 --> 00:56:08.926 [Sacha]: And of course, LaTeX is very good for 00:56:08.986 --> 00:56:11.710 controlling all these matters of presentation. 00:56:12.702 --> 00:56:15.549 [Pedro]: Exactly. I mean one of the things that I 00:56:15.910 --> 00:56:21.663 I like most... Once again, here I've tried a 00:56:22.385 --> 00:56:24.670 couple of times to have 00:56:25.696 --> 00:56:29.141 a movement in that direction, a bit more 00:56:29.241 --> 00:56:35.229 control of fonts within a LaTeX document or 00:56:35.350 --> 00:56:37.032 within an Org document. NOTE Sans serif fonts are better for accessibility 00:56:38.714 --> 00:56:42.640 What you can do now is in LaTeX you can go and 00:56:42.700 --> 00:56:47.767 say no, the main font is SF family. 00:56:48.888 --> 00:56:51.152 That's the default font so that you have 00:56:51.212 --> 00:56:52.834 everything in some series. 00:56:53.017 --> 00:56:56.180 That's something that, by the way, enhances 00:56:56.561 --> 00:56:57.262 accessibility. 00:56:59.324 --> 00:57:03.268 Because there are some people who will just 00:57:03.869 --> 00:57:09.755 collapse with serif fonts and they need sans serif. NOTE Wanting markup for sans serif 00:57:09.775 --> 00:57:12.518 The other thing, the step that I wanted to move 00:57:12.558 --> 00:57:15.641 forward is I want to highlight some words with 00:57:15.942 --> 00:57:17.083 Sans Serif. 00:57:17.148 --> 00:57:21.393 Instead of bold or with the emphasis with the 00:57:21.453 --> 00:57:26.298 italics and so on, why not make some words stand 00:57:26.399 --> 00:57:27.960 out with sans serif? 00:57:28.381 --> 00:57:30.804 That's something that I also discussed in the 00:57:30.844 --> 00:57:32.826 Org list and no one wanted that. 00:57:34.308 --> 00:57:37.131 That's a pity. 00:57:38.232 --> 00:57:40.314 [Prot]: Because it might introduce new markup, of 00:57:40.375 --> 00:57:40.635 course. 00:57:41.055 --> 00:57:46.281 [Pedro]: Yes. I would need just one more markup. 00:57:46.301 --> 00:57:48.535 And I was like, "Another one, please?" "No." 00:57:50.217 --> 00:57:50.657 [Sacha]: I see. 00:57:52.379 --> 00:57:54.402 I've been just overriding the standard one 00:57:54.422 --> 00:57:57.305 because, for example, I won't actually underline things. 00:57:57.846 --> 00:58:00.930 So I just change my underline to be something else. 00:58:00.950 --> 00:58:03.313 But of course, as a general practice, it's not 00:58:03.353 --> 00:58:04.995 going to be that useful. 00:58:05.436 --> 00:58:07.438 You only have so many emphasis characters. 00:58:08.532 --> 00:58:09.474 [Pedro]: Yeah, no, of course. 00:58:09.554 --> 00:58:12.900 But if you don't want to underline another thing 00:58:12.920 --> 00:58:17.809 that many people find disturbing or really 00:58:17.869 --> 00:58:21.756 challenging, this is why I'm saying, let's try to 00:58:21.816 --> 00:58:24.962 also have the possibility of having a specific 00:58:25.022 --> 00:58:27.486 markup just for Sans Serif for that. 00:58:30.284 --> 00:58:32.667 It's all accessibility. 00:58:33.307 --> 00:58:38.053 I'm quite, how can I say, I'm very sensible to 00:58:38.113 --> 00:58:38.853 this kind of things. 00:58:39.995 --> 00:58:41.656 I'm very sensitive to this kind of things. 00:58:44.179 --> 00:58:48.344 Mainly because I've been from very, very young, 00:58:48.644 --> 00:58:56.713 I've been very involved in the disability world. 00:58:57.436 --> 00:59:00.421 [Sacha]: I'm looking forward to hearing more 00:59:00.461 --> 00:59:05.008 about your accessibility improvements in December 00:59:05.188 --> 00:59:06.490 at EmacsConf. 00:59:08.053 --> 00:59:09.956 Thank you so much for working on this. 00:59:11.518 --> 00:59:12.480 [Pedro]: You're welcome. 00:59:12.500 --> 00:59:16.947 Thank you, yes. 00:59:17.433 --> 00:59:21.685 [Sacha]: You showed us Unicode, you showed us 00:59:21.785 --> 00:59:24.894 emojis, you showed us theoretically how you would 00:59:25.356 --> 00:59:28.625 do it for non-Latin character sets. 00:59:29.146 --> 00:59:31.172 It will all be wonderfully documented. 00:59:31.490 --> 00:59:34.535 And it is now in Org Devel, right? 00:59:35.156 --> 00:59:37.399 [Pedro]: It's in Org Main. 00:59:38.220 --> 00:59:38.701 [Sacha]: Oh, sorry, yes. 00:59:38.801 --> 00:59:39.242 Org Main. 00:59:39.262 --> 00:59:40.924 So it'll make it into its release. 00:59:41.946 --> 00:59:44.670 I'm looking forward to seeing all of that. 00:59:45.351 --> 00:59:52.142 [Pedro]: It should be like Org 10 when it comes into being. 00:59:52.162 --> 00:59:56.128 People will still need to download stuff and 00:59:56.548 --> 00:59:58.932 compile it on their own. 00:59:58.912 --> 01:00:01.514 That's the main problem. 01:00:01.814 --> 01:00:04.116 That's the only thing that I would like to see 01:00:04.156 --> 01:00:06.659 advance at some point. 01:00:07.119 --> 01:00:09.340 But maybe that's with the next Emacs, 01:00:09.341 --> 01:00:10.182 with Emacs 32. 01:00:11.082 --> 01:00:12.203 [Sacha]: Okay. 01:00:12.604 --> 01:00:14.365 Looking forward to seeing how that works out. 01:00:14.445 --> 01:00:16.587 If people are curious, they can find you on the 01:00:16.687 --> 01:00:19.449 Org Mode mailing list and volunteer to be guinea 01:00:19.469 --> 01:00:20.871 pigs for various things. 01:00:21.671 --> 01:00:24.073 Your code is already in main, so it's easy enough 01:00:24.113 --> 01:00:24.834 to try out. 01:00:25.474 --> 01:00:27.596 And thank you very much for sharing the 01:00:27.636 --> 01:00:28.497 demonstration today. 01:00:29.877 --> 01:00:30.678 [Pedro]: Okay, fine. 01:00:30.898 --> 01:00:31.679 Thank you very much. 01:00:32.100 --> 01:00:33.862 Just one small comment. NOTE Contacting Pedro 01:00:33.922 --> 01:00:38.668 I'm also maintaining LaTeX and Beamer in Org. 01:00:38.688 --> 01:00:42.453 So any bugs, any questions, anything through the 01:00:42.493 --> 01:00:46.038 mailing list and I'll try to answer as soon as 01:00:46.078 --> 01:00:49.122 possible. 01:00:49.142 --> 01:00:50.443 [Sacha]: I will end the stream there so people 01:00:50.483 --> 01:00:52.486 have an hour of wonderful stuff to listen to. 01:00:52.726 --> 01:00:54.028 Thanks everyone for joining. 01:00:55.358 --> 01:00:56.767 [Pedro]: Thank you. 01:00:56.768 --> 01:00:58.205 [Prot]: Thank you.