Ask HN: Why are PDFs still heavily used?
PDFs are terrible to view on mobile. They have accessibility issues, navigation is painful, the content is static, and so on [1].
Web standards fix many of the problems. So why are PDFs still used, particularly for publishing reports?
1: https://www.nngroup.com/articles/pdf-unfit-for-human-consump...
45 comments
[ 3.8 ms ] story [ 107 ms ] threadThe lack of changing on mobile is technically a feature.
And once the old paper based process has been "made electronic" by using PDF's, they don't bother with trying to continue past a "sheet of paper" metaphor.
And ... printers work pretty well with it and while even I have to admit that digital documents are superior for most documents, paper still has a lot going for it: underlining, flipping back and forth, taking notes in the margins. I also find the added spatial dimension sometimes helps me remember (e.g. that was up on the top left of the page next to a chart) ...
Why are you saying this like it's a bad thing? Why would you want dynamic content in a document format?
You can't give someone an epub and tell them to read page 197. You can't look at a webpage and know how it will look when you print it out. You also can't just netcat an ebook to your printer and get useful output.
What alternatives do we really have? PDF is the most widely supported format outside of plain text (arguably more widely supported than text actually, with phones being the major computing device). I haven't seen an ebook format that can correctly and consistently deal with code sections, inline images, quotes, font styling, paging, etc (maybe more a software issue than format problem, but if nobody can make it work right, the format isn't blameless).
Stuff like, embedding text in a way that improves accessibility and copying text works nicely. Being able to right click copy images would be also nice.
If someone wants something to be private, they shouldn't publish it. Once you put something out into the world, you should expect that people will do whatever they want with it.
Can you 'just netcat' a PDF to a printer? Which port number? How could I control features such as printing to both sides of the page? I imagined there would be some wrapping protocol (or conversion to postscript?) Does the printer that receives just a PDF binary, and no other control signal, just decide to use a default mode (like always printing single sided only)?
Thus, PDF was a game changer. Basically, no matter what machine you opened the file on it would look exactly the same. It kinda sucks you can't edit it or have to use special tools to parse it but when it comes to readability across devices it cant be beat.
Posting papers from arXiv to HN I have a choice of pointing to the HTML abstract or the PDF and overall people express preference for the PDF and I believe I get more upvotes for the PDF.
I can see a phone being a little too small though. PDF does have features that make it possible to make a PDF which is reflowable like an HTML document for the sake of accessibility and better UX, but those features are complicated as hell and rarely used.
I did a deep dive into PDF because I was pitching and sketching out the ultimate test extraction system for corporate use and figured PDF was the most important format to support and I could say boy is it tricky to get the correct text out with the structure properly mapped unless some of those obscure features were in use and it would still be pretty hard then.
At the moment I'd like to add OCR text overlays onto scanned documents and just can get myself to deal with the format for longer than ten minutes until my brain gives up.
I also have a rule about reading PDFs. I almost universally print them off to actual paper if it's more than 2-3 pages worth of content. Printing websites is a non-starter.
Word/excel/powerpoint all do whatever they want.
https://en.m.wiktionary.org/wiki/Chesterton%27s_fence
I hate reading web pages on my phone. PDFs are far superior, especially on tablet.
Google Docs + MS Office has probably figured this out by now, but there's also ton of historical momentum keeping it in use.
As a financial auditor I remember being baffled how people would go out of their way to create paper processes. A spreadsheet that requires confirmation would be printed, signed, stapled to another piece of paper - all so I can remove the staples and scan it to track digitally in our audit software. At least I knew sending them a PDF would come back looking the same...
Web standards may have caught up, but nobody is sending anybody .mhtml files. I'm pretty sure most email providers would flag emails with such attachments as malicious.
The other use case are fillable forms that are legally binding. In larger business or the government where you have your identity tied to a private key, you can sign a PDF with it.
A bunch of them are just dumb assumptions, like vague complaints about accessibility, ignoring the fact that PDFs have supported screen readers, alt text, etc for longer than the web has.
I laughed out loud at the assertion that PDFs are "stuffed with fluff" and the web is a model of focused, concise writing.
"Jarring user experience" and "Cause disorientation" are the same complaint rewritten to highlight different aspects of one problem.
Web sites and PDFs are different tools serving different needs. The complaint that PDF content is "static" is, confoundingly, both untrue and entirely the point. PDFs have support Javascript, media embeds, advanced navigation tools (hyperlinks, cross-refs), and more, for years and years... but most people don't encounter much beyond formfills and hyperlinks, because the rest of that stuff is not what people want out of a document format.
In other words, all the consultants can write all the words about how PDF should be the web, but the market has spoken, and PDFs are the way they are because they meet real needs. Figuring out what those are might be a better use of time than shouting into the wind.
Now that I think about it, why should I have to throw away thirty years of PDF-compatible software because nobody bothered to make a decent PDF reader for your phone?
- They render consistently on any mainstream platform, without access to external resources.
- They load, scroll, and zoom relatively quickly on mainstream platforms.
I don't think PDFs are terrible to view on mobile. I think it works out pretty well. I prefer viewing them than the mess that is an HTML page with CSS. Usually whoever authored it ddin't do a good job of making it pinch/zoomable and accessible.
Sometimes I wonder how in touch Nielsen is with the real world.
Where I work we routinely send PDFs to print houses. They're large format, as in paper size. Honestly I can't imagine sending an html/css bundle to a factory. How would you even specify different varnish layers, or embossed areas?
For someone who talks about accessibility and usability, his personal website was always an absolute shitemare to read.
A PDF is, in the popular consciousness, a digital form of paper. Anything that could be on paper? PDF.
Think of it as an FFI for the paper-based OS that our civ still runs upon (and may always.)
Like, network-based civ is kind of like a VM, and numerated paper documents (laws, etc.) are libc, and when you want to call out to the underlying reality, you use PDF.
You can't use HTML, because it's totally inconsistent. You can't use Word docs, because they're totally inconsistent and not cross-platform.
What format can you think of that you can use to describe how each page looks, right down to fractions of a millimetre?