Hacker News Reader

Comment by camgunz | Hacker News Reader

You got nerd sniped by the HTML vs. PDF format thing and missed the entire point of TA:

> Isn’t it a good thing that we enjoy rapid progress? To the extent that we get to enjoy things like YouTube and sandspiel, yes! But to the extent that we want the internet to be a place where we can work and live and think and communicate free of malware, surveillance, dark patterns and the insidious influence of advertising, the answer is, empirically, sadly, no. The web has become ad-corrupted hand-in-hand with growth in technological capability, and the symbiotic relationship between web and browser means they feed on each others’ churn. Ads demand new sources of novelty to put themselves on, so the web expands continually, the specs grow in complexity, the browsers grow in sophistication, the barrier to entry grows ever higher, the vast cost of it all demands more ad revenue to fund it... and thus the perpetual motion machine is complete.

View on HN

The author does identify a problem, and so you want to focus on that. That's fine. There is the issue of triviality, however.

The problem described is widely felt, and also widely discussed. We already know this stuff to be a problem. For the piece to be worthwhile, then, it should do something that is not present in the other instances where the topic has been raised. It should articulate (or at the very least exhibit, without necessarily articulating) a solution for us. It doesn't. A bad remedy to a genuine problem does not yield a solved problem.

The article is called "Deurbanising the Web", and its thesis is:

- Publish in static file formats.

- Date and hash your work.

- Stop spying on your users.

HN is a discussion forum, not project planning software. Not everything has to "yield a solved problem". Are you really setting the bar at "design a technology stack for replacing HTML/CSS/JS"? That's way, way too high.

You say that its thesis is (in part) to generally publish in static file formats, but that's not quite accurate. The piece specifically touts PDF/A as the best format and makes several arguments against the use of html/css. I agree that they're making a broader point than just "use pdf," but "use pdf" is definitely a large part of it.

Those points can be trivially met with static HTML and something like IPFS, and you can still download HTML for local storage and viewing. You can even print to PDF if you really want to do so. Meanwhile, PDFs also allow dynamic files, don't require dating and hashing, and can be used to spy on users or deliver malware.

EDIT: Oh, yeah, and static file formats doesn't necessarily have to mean static document formatting when viewing -- unless you're using PDFs, which tends to break useful stuff like reflowing for paginated documents (one of the worst things about even simple PDFs).

ipfs solve this well.

The author brings a solution, it is to publish documents in PDF instead of HTML.

"A bad remedy to a genuine problem does not yield a solved problem."

PDF is a great way to publish documents. which is what the web originally was.

The web has become a bad remedy to some distributed software problems.

Why do you feel PDFs are a bad remedy? PDFs are the usual way I absorb information.

No, the entire point of the article is to convince people to use PDF/A. Which I find comical since you have to go out of your way to check if a PDF is PDF/A compliant. If the web was run by PDF's, there's no reason why any big corporations would abide by those rules, and it'd be just as messy as HTML is today.

You've also been nerd sniped. TA goes on and on about surveillance capitalism and the attention economy. Weird, for an article that's supposedly convincing engineers of the merits of one file format over another.

Did you read beyond the "How did it come to this?" section? TA goes on and on about web standards and the need for PDF/A.

Edit: If the article _was_ all about surveillance capitalism, then it wouldn't be worth upvoting as actionable solutions are much more valuable than preaching to the choir.

If you don't think it's clear that the author's advocacy of PDF is a means to an end, subservient to their desire to dismantle surveillance capitalism and the duopoly that Google/Apple have on the web, I don't know where to go from here.

why don't we have both?

I think you're the one who got nerd-sniped here. 1.5 of the 13 pages in this PDF are about surveillance capitalism. The rest's about web standards.

What in the nine hells is nerd sniping?

It's when you trick a technically-minded person into jumping down a rabbit hole of a technical problem/controversy. Here it's PDF vs. HTML, but other classic nerd snipes are UTF-8 vs. anything else, "fixing" election tech, etc.

I tackled the premise. I think addressing the premise is the logical place to dismantle an argument.

But, again, the premise is not that "as a file format, PDF is better than HTML". The premise is: because HTML is two-way, it enables surveillance capitalism and allows bad actors to monopolize the attention economy. The author wrote it thus:

> Sure, you can write good HTML. I won’t argue with that. And if you’re writing good HTML, good for you. But HTML is a dual-use technology, the bad guys are dual-using it an awful lot, and I feel that the stone age still has a part to play in the progression of the information age.

The part where you engage with this is where you write:

> I'm sorry, the more I think about this the dumber I feel. The web is useful because it's 2-way. I am excited by the web because I can interact with other people. I come to hacker news to engage with thinkers, not to just read a published article from one single author. I want to read ad-hoc opinions and user submitted content. PDF web, really?

Which is interesting! Do you have thoughts on creating peer-to-peer systems that don't enable surveillance capitalism?

> > Sure, you can write good HTML.

A key here is that it's easier to write good HTML docs than good PDF docs, and much harder to deal with the harmful aspects of PDF docs given present technology.

> Which is interesting! Do you have thoughts on creating peer-to-peer systems that don't enable surveillance capitalism?

I don't know about the other person's ideas, but decentralization plus better anonymization and pseudonimization, with always-on strongest-reasonably-posible encryption, seems like the direction to go.

> A key here is that it's easier to write good HTML docs than good PDF docs, and much harder to deal with the harmful aspects of PDF docs given present technology.

Oh, yeah I'm not on the PDF train. That's wild. I'm more of a Markdown or Gemtext advocate, or even LaTeX.

> I don't know about the other person's ideas, but decentralization plus better anonymization and pseudonimization, with always-on strongest-reasonably-posible encryption, seems like the direction to go.

Yeah, projects like IPFS (which you reference above) are working towards this, but JavaScript still works over IPFS. Plus, fingerprinting techniques are pretty bonkers. Most of it comes down to JS and various state you keep on your local machine (cookies, flash cookies, etc.), but I think you need that. How do you maintain a session with a peer without some kind of token/cookie?

> Do you have thoughts on creating peer-to-peer systems that don't enable surveillance capitalism?

Yes, it's call TOR. However, legislation is where we should start. Crippling/abandoning an incredibly useful technology which works very well just because it's often used nefariously seems to be a bit of an overreaction.

Until then, stop using social platforms, use an ad blocker, and use VPN if you really care about "surveillance capitalism".

The classic mistaking the example for the topic.

Reply on news.ycombinator.com