[joekrill]

Why a PDF? The PDF is just pictures of things that link to an external site (Thingverse, various blogs, etc) where you find the actual instructions. The PDF itself doesn't contain instructions at all - just an index/catalog, really.

[seemaze]

While I enjoy the content, I find myself surprisingly annoyed at choice of PDF as the format. It's the worst of both traditional and web publication modes; low accessibility AND low portability

[nlarion]

This isn't the first time I've read a comment criticizing PDF, but it is the first time I can recall the Portable Document Format (PDF) being accused of low portability.

[nox101]

I'd call it low-portability. They're designed for large screens yet they're often viewed on phones where you need to manually zoom in and view each page some tiny portion at a time.

Was at a restaurant with a QR code that led to a PDF if their paper menu. It was horrible. For reference it was the Ralph Lauren Cafe in Omotesando Tokyo.

[wpm]

I'd prefer a static PDF I can pinch, zoom, and move about on my own than some crappy, overwrought, slow, "mobile" website that forces me to jump around to different sections of the menu with links that it *hides* anytime I scroll, which of course I have to because only 3 menu items are going to be able to fit on the screen at a time.

All that, VS a PDF that shows the entire menu that I can navigate spatially, show as much as I want depending on light levels, interest, and screen size, and can all fit in one, static tab.

Of course, the real answer is that electronic menus at sit down restaurants opened with QR codes are trash. Paper never really had these problems unless the restaurant was bad at menu design. I dislike being forced to use my phone at all when I'm out with friends or family for a nice dinner. Growing up, if I used my phone at the dinner table I would've been scolded.

[musicale]

Covid probably had something to do with the PDF menu trend. I think we can go back to printed menus now.

[squarefoot]

PDFs were born to make real life documents manageable in digital form, that is, letter, legal, A4 etc; nothing that would (and could) have been used on a phone back then. If they don't display properly on a phone I'd say it's on the phone itself. They're not computers usability-wise, not even close: too many compromises with ergonomics, information density and readability fighting against each other in the name of portability. That restaurant had sloppy management/techies, as finding that the .pdf was the wrong size would probably take ten seconds to any of them, the correct the problem by having the QR point to a correctly sized .pdf generated automatically by reading the same db the bigger .pdf was created from. PC sees the link to the bigger one, phone reads the QR containing a link to the mobile one. not hard at all to solve, it's not a .pdf format fault.

[MajimasEyepatch]

Dynamically generating PDFs on the fly based on the user’s screen size sounds downright Kafkaesque. I feel a deep sense of dread any time I have to touch PDF-rendering code. Any halfway decent CMS (or just plain old HTML and CSS) can easily create a good-looking, responsive, accessible webpage with a fraction of the work. Leaning on PDFs in this case is a classic example of, “When all you have is a hammer, everything looks like a nail.”

The PDF format can’t necessarily be blamed for its inability to be responsive to screen size, but that’s one of several good reasons to discourage the use of PDFs for purely digital documents. Its design is linked to a specific historical context, and it should be limited to that context.

(Of course, I imagine the actual reason for the PDF is that it’s just an export of the InDesign/Photoshop/whatever file that was used to create the original paper menus. There probably is no “database.”)

[Supermancho]

> Any halfway decent CMS (or just plain old HTML and CSS) can easily create a good-looking, responsive, accessible webpage with a fraction of the work.

Easily? Because it's not a fraction of the work. Let's just assume you mean to use someone else's solution (of the literal thousands, each with their own quirks), to make your bespoke website.

PDFs have a standard. Libraries to resize are not difficult to find. Most people are able to run a PDF through their printer driver for a resize, because it's a solved problem.

I would say creating a job that outputs a bunch of PDFs is going to be easier than a build system for different devices (CMS or not), every time.

[sethammons]

A pdf should not reflow nor be responsive. Aside from zoom/pan, how else could you possibly interact? I think it works fantastic, like holding a magnifying glass and panning around a surface

[wittjeff]

https://helpx.adobe.com/acrobat/using/reading-pdfs-reflow-ac...

[lostlogin]

Oh wow, that’s awful. ‘Like the web, but on a pdf and done entirely by Adobe.’

[wccrawford]

That's exactly why they're saying it's "not responsive". It wasn't designed to be, and it's not.

[rad_gruchalski]

PDF is over 30 years old. I’m pretty sure people had different idea of „portable” back then. It’s still really good for printing…

[makeitdouble]

When the PDF is a version of the paper menu, we can understand they only produced one version, and the alternative was no digital version at all (which is the case for an awful lot of restaurant)

If it's PDF or nothing, I'll be glad to take the PDF.

[musicale]

Large screen phone + landscape mode if needed usually works for me.

[seanw444]

Yeah that is ironic regarding both the name, and the fact that it's one of the most viewable formats on any given system.

[deskr]

> most viewable

It is, but the UX trying to read a A4/letter sized PDF document on a phone is absolutely horrible.

[sophacles]

This is basically the same reason I think sports cars are horrible... trying to fit a 10 ton roll of steel in the backseat is nearly impossible.

[nox101]

No one wants to carry a 10 tons rol of steel in a sports car. The majority of people getting a PDF on their phone though, do want to be able to read the information in it easily and they can't.

So no, that's not the same type of complaint

[deskr]

It shouldn't be. 10 tons of steel takes up just about 1,300 litres. That should fit nicely in the back seat. Remember to secure your load.

[drekipus]

only a4 sized pdf.

You can make a phone-sized PDF and it'll be just fine.

it's not the file format.

[chrismorgan]

And then the experience on a larger screen will be bad.

[rixed]

It is, unless PDF supports reflow?

[esperent]

Shouldn't "portable" in the context of a document format mean "comfortably viewable on most systems" rather than "yeah you can view it but it's gonna suck and you'll get eye strain"?

[metadat]

I believe this is called "responsive design", a useful but differentiated concept.

It's pretty cool how PDFs don't typically load resources from an external website or CDN. They are self-contained, and thus demonstrably portable.

Gross analogy: Have you ever been to a multi-day music festival? A port-o-potty in the dark is disgusting, but still somewhat useable compared the alternative of going to the bathroom in public. PDF artifacts are in the same ballpark.

[seanw444]

Also, I think I'd personally dislike a "portable document format" that doesn't look identical on all systems. By making it visually identical everywhere, you of course sacrifice comfort on smaller screens by nature. But to me that's an acceptable tradeoff for knowing I'm looking at the document, as it was intended to be displayed. I don't think of PDFs the way I think of websites. My physical papers don't reflow text.

[SoftTalker]

> low accessibility AND low portability

The two main pain points everyone said PDF had solved for documents, when it was introduced.

[lostlogin]

A growing horror is how Windows handles them. On a Mac they open and work fine, but on Windows they grind the system to a halt. What’s going on?

[SamBam]

Seems like it would be a lot more accessible for a lot of people if it were a set of HTML pages.

I expect that whatever was used to create the PDF could be configured to produce a website...

[thecatspaw]

also on a webpage you would be able to filter it by available tools/knowledge

[TeMPOraL]

Or not. Maybe you'd get "infinite scroll". Or some other modern webdev bullshit that's strictly inferior to PDF (or plain HTML) and CTRL+F.

I'm actually surprised by the anti-PDF sentiment here (in general case, not necessarily this book). Modern web is so bad, that almost every day I end up on some page that would be strictly better if it were a PDF. So, to play devil's advocate, PDFs are cool because:

- The links may rot, but they remain, and so does surrounding content. Once you get a PDF, no one can take it away from you.

- It's self-contained. It can easily be transferred between devices and read without an Internet connection.

- It's a file. Yes, it's important to mention because in 2024, files cannot be taken for granted.

- Rich format without spurious dynamics and other web nonsense. Sure, PDFs technically can run arbitrary JavaScript, but hardly any reader supports that.

- Can't track you or spy on you (theoretically it could, in practice, see previous point).

I could come up with a few more. Point being, you could do worse, and modern web quite often is worse.

As for what could be an even better format, my mind is drawn towards CHM[0]. You know, like the help files in old Windows software. A self-contained file built of interlinked HTML pages, complete with index and internal search/xref. Kind of a better EPUB[1].

(Ironically, marketers should actually love PDF - total control over presentation is exactly what they've been trying to gain on the web all these years.)

--

[0] - https://en.wikipedia.org/wiki/Microsoft_Compiled_HTML_Help

[1] - Despite being a decade older.

[larodi]

I guess the sentiment is because

a) it cannot be automatically reflowed

b) it’s a complete mess of a file format with tech deb spanning 30 years

c) the linked ressources info

Everything else is fine with it and apparently lotta folks write plenty of software to read it.

Perhaps epubs (and .mobi etc) come close but they are not universally adopted

[pjerem]

I do agree, especially on

- It's self-contained. It can easily be transferred between devices and read without an Internet connection. - It's a file. Yes, it's important to mention because in 2024, files cannot be taken for granted.

(Btw it’s a sad state of affairs that we can less and less own our files)

But I do think in the post here, the issue is that the PDF in itself is pretty useless : it doesn’t contain any information and only links to … websites.

(Btw I wouldn’t criticize this a lot because I’m admirative of the indexation work, it’s pretty cool !)

[freefaler]

It's a strange choice to put a PDF with links on a website that contains just links. Why don't put the links on the web site, as the web was invented to be used for? What's the benefit?

[treefarmer]

Had the same thought - it's also annoying to update the PDF once links die, so I doubt that'll happen often. I guess it might be helpful if you want it as a coffee table book...

[saintrain]

Never done it before, but https://cloudconvert.com/pdf-to-html did a very nice job in about a minute

[deivid]

n-o-d-e publishes most (all?) projects as open source, I'm sure the book sources are/will be available somewhere

[Waterluvian]

Yeah it seems like a wrong choice. It’s useless in any context that cannot be linked back to the actual content. It’s a 220 page printable index.

[wutwutwat]

pdf is a common malware delivery mechanism

[Y_Y]

This sounds like a good job for a fancy LLM.

”For each entry in this PDF, follow the link and give a structured explanation of how to make the thing."

I don't know if any off-the-shelf tool is capable of that today though.

[croes]

A picture says more than 1000 words