A tool to create slides using Markdown easily for you | Hacker News Reader

A tool to create slides using Markdown easily for you

github.com

110 pointshiroppy41 comments

Very nice! I've been using Marp https://marp.app/ at work for the same purpose as well. One cool thing is that it comes with a VSCode extension, making it very convenient to use.

+1 for Marp, and +1 for VS code as a ecosystem

VS Code is great but I'm concerned about there being a single canonical "app store". It would be nice to see an equivalent "F-Droid" that's usable side-by-side with the MS marketplace. I believe code-server (VSCode in Docker over a Browser) does something like that for OSS extensions but it's my assumption that the VSCode codebase only supports one store at a time.

There's already OpenVSX (https://open-vsx.org/) maintained by the Eclipse foundation, and it contains almost all of the mostly used FOSS extensions (not the propietary Microsoft extensions though). It's used by default in VSCodium (FOSS version of VS code), and also Eclipse's own cloud editor Theia (https://theia-ide.org/).

I hadn't heard of this! Thank you! My VSCode experience is primarily "VSCode" (not Codium) w/ the MS Remote extensions and looking at code-server for its more unique capabilities. Definitely going to look into this and adopt it where I can.

FYI, it looks like the code-server devs have some concerns about # of extensions and matching the MS market: https://github.com/cdr/code-server/issues/1473 . I'm not too worried about that for myself though.

Slides that are text only distract the audience.

They read the text.

Slides that are non-self-explanatory images, _require_ the audience to listen to the speaker.

Try it. Put two circles on the screen and see what people do. They look at you for explanation.

I've got to reference Patrick Wilson's "How To Speak" course from MIT OpenCourseWare here[0]. An hour of life well spent!

[0] https://www.youtube.com/watch?v=Unzc731iCUY

Worth watching again!

David Ogilvy said 'You have to attack people through the eyes and the ears' or something along those lines when referring to text on flip overs and what you say. As in, the spoken and written text needs to be exactly the same in order to land the message. If you speak different words than are written, he says, you are doing it wrong.

I don't know if I agree with text vs no-text, but I do feel what Ogilvy says about this subject: it is so distracting when a speaker uses different words than his slides, even if ever so slightly. I also love the sticky image of 'attacking through the eyes and ears'. Again, not 100% sure these are his exact words.

Spot on about using little to no text.

TedX has a pretty good style-guide about slides [1]:

What goes in my slides?

•Images and photos: To help the audience remember a person, place or thing you mention, you might use images or photos.

- People will understand that the images represent what you’re saying, so there is no need to verbally describe the images onscreen.

•Graphs and infographics

- Keep graphs visually clear, even if the content is complex. Each graph should make only one point.

•No slide should support more than one point.

What should the slides look like?

•Use as little text as possible -- if your audience is reading, they are not listening.

•Avoid using bullet points. Consider putting different points on different slides.

[1] https://storage.ted.com/tedx/manuals/tedx_speaker_guide.pdf

Great point, and it works great for Ted, but Ted is only but one format. For academic purposes, I would go with all the opposite.

I had similar problems at work until I realized there were two competing agendas at play. One group wants the slides to be "here's a record of what I talked about so people who didn't make it can read it later", while another wants the slides to be "here's stuff that supports what I'm talking about now". Those are fundamentally incompatible, and it's a bad idea to try to use a deck generated for one purpose to satisfy the other.

One potential solution is two slide decks--one for the presentation and one as a "reading deck", but that's a pain in the neck. Another option we're currently using at work is to go with the Amazon 6-page-paper: write what you're going to talk about, and then the slides are only to support the actual presentation. The paper is the thing people read if they missed the meeting.

I came to the same realization and started putting all the text for the "reading version" in the presenters notes part. The first slide says "If you're reading this on your own, press 'p' for presenter mode" or whatever.

Smart! I just wish PowerPoint had better support for content formatting in the notes. It’s really barebones and hard to edit—more like working with text in an Excel cell than a proper document. (Which kind of makes sense, from their perspective.)

most slide formats support graphics and notes to cater for this.

Slides with images and no text are a nightmare for non visual person like myself. I see text: I remember text. I see image: I remember exactly nothing.

I understand what you're saying and agree that slides with more than a little text are distracting.

For many years, even before becoming aware that the approach had a name, I've always made slides for my presentations contain only a few words, and, for the most part, no images.

https://en.wikipedia.org/wiki/Takahashi_method

I don't take it quite as far as having one or two words: I normally limit it to six or seven.

Basically, something people can read instantly.

People have commented that my approach to presentations is quite unusual, but at the same time well received and fairly effective.

> They look at you for explanation.

That's exactly it!

Years ago, my dad's hospital had a rule that the head of the department had to review all slides before you gave a presentation at a conference. The department heads had one consistent rule: they had to be able to view your slide without putting it in the projector and tell what it said. Worked pretty well: you can imagine how big and limited the content would have to be to be able to view it on a slide with the naked eye, and it kept people from filling slides with bullet points.

I used to think that the best presentations are those where the slides are text only and the speaker just reads the text out aloud.

Just repeating the text of the slides verbatim

Because then I can download the slides at home, read them for myself, and do not miss anything. If the speaker would say anything important that is not written on the slides, I would actually have to go to the presentation, and am annoyed when they talk too slowly and angry that I cannot stay home.

Although my perspective is changing. Now I have glasses and cannot read as fast as I used to. And nowadays there are HD video recordings. With a video recording it does not matter how the presentation is done. In any case I could stay home and watch the recording at 2x speed

Even better: when you want them to listen to you, use the "blank" shortcut (B key on Macs using Keynote). The screen goes blank (black), and then they look at you for explanation.

Source: presented 600+ times in front of audiences of all types and size.

Just use asciidoc.

E.g.

- https://github.com/Mogztter/asciidoctor-web-pdf/tree/master/... - https://asciidoctor.org/docs/asciidoctor-revealjs/

I'm a huge fan of writing in markdown. I've been doing presentations in markdown for years with a different tool with fewer features (https://pkg.go.dev/golang.org/x/tools/present). I plan to switch to Fusuma next time I'm doing presentations though.

How is this different or better than reveal.js https://github.com/hakimel/reveal.js?

reveal.js is mostly aimed at creating slides using HTML, it has markdown support for writing the content, but this project seems to aim to create the entire presentation from markdown. I'd compare it to deckset - https://www.deckset.com/

You can create the entire presentation in markdown with reveal.js too https://github.com/rectalogic/gist-reveal

What's that advantage of using markdown? html is far more expansive in visual options.

Not Everything needs to be markdown.

Some things are just far less typing in markdown than in HTML, for example lists of bullet points and inline code.

Only “some” things?

...but then you're just using the wrong editor.

I largely prefer writing my slides in Markdown, for the same reason that I prefer writing (a first draft of) my letters in Markdown rather than in a word processor: this helps me focus on the content first, rather than the presentation.

YMMV

As a Deckset user, my only gripe (minor, and usually not a big deal) is not having an HTML version generated by it (this is why I wrote this CLI tool [1])

[1]: Haskset: https://github.com/rberenguel/haskset

I've been using this: https://github.com/webpro/reveal-md

which is only markdown on top of reveal.js, quite cool

I use https://play.presenta.cc/ and I like it a lot. Btw:I use it online, directly on that website...

Others have already mentioned RevealJS, which I've used via Org Mode export. I've also had a good experience with Remark, which also focuses on Markdown as the authoring language.

https://remarkjs.com/

Note that you can also use pandoc[0] to convert a markdown-like file to a reveal.js or PowerPoint presentation, among other formats.

[0]: https://pandoc.org/MANUAL.html#slide-shows

I love markdown. I hate making slide decks. I’m looking forward to comparing this to similar tools I’ve tried. In a perfect world I would be able to `brew install fusuma` and not need node, but that’s a small thing. Thanks!

Does this encourage people to make slides which are just walls of text? The examples are pages of text rather than visual slides.

https://github.com/jdan/cleaver

Here's a similar project that I've used in the past and liked a lot

http://platonio.herokuapp.com/

Love the idea in theory, just wish it would export to Google Slides

Reminds me of xaringan for R.

Reply on news.ycombinator.com