Remix.run Logo
Remix clone Hacker News
new | show | ask | jobsGithub
▲ sylware 3 days ago

Yeah yeah, virtue signaling, and most of EU online services are now gated by the use of one of the whatng cartel web engines (IRL, google blink), namely EU web sites are broken favoring web apps.

They have to restore interop with noscript/basic html web engines (past/present/and future).

Then, they have to be carefull with their file formats, for instance you never give "carte blanche" to such a disgusting format like PDF, you are very careful at defining a, as simple a possible, subset of it (with some internal software for validation).

▲Mashimo 3 days ago | parent | next [-]

Is ending endless scrolling really virtue signaling? Don't you think it will have a measurable effect?

▲sylware 13 hours ago | parent [-]

It feels like some distraction to avoid to deal with the much more important (stategically) issue which is the interop of EU online services with 'small tech'.

BTW, with noscript/basic HTML, no infinite scroll anyway. One Rock Two Birds. Ironically.

▲nanapipirara 3 days ago | parent | prev [-]

Yeah yeah, whataboutism.

I'm very happy they're taking a stance. I've seen too many messed up kids and there's no doubt the addictive design plays a big role in the problem.

▲soco 3 days ago | parent | next [-]

I must notice that every time, but really every time, EU moves a pinky finger against tech industry, a sizeable chunk of comments here will be like the one above. I wonder, is it about a general sentiment against EU? Or a general sentiment against restricting technology? Or a general sentiment against humans? Or what?

▲palata 2 days ago | parent | next [-]

I think it's easier and safer to complain about everything than to actually have a nuanced and informed stance.

Look at age verification: it's very easy and very safe to say "nobody sane would think that it is a good idea to force people to show their ID to every website they want to access, it will obviously leak the IDs, that is very bad!". While it is not wrong, it is manipulative: that is not the only way to implement age verification. In fact, there is technology that exists that would allow age verification in a privacy-preserving manner: some service that already have access to your ID can give you a token that proves your age, and you can then use this token to access a website. The service cannot know where you use the token, the website cannot know your ID, and they cannot collude.

So the constructive debate around age verification is this: assuming we implement it properly (i.e. in a privacy-preserving manner), is that something that we want or not? Does it solve a problem, or at least does it help?

But we cannot ever elevate the debate to that level, because nobody can't be arsed to get informed about it.

▲soco 2 days ago | parent [-]

But ironically exactly the people in the better place to influence that half-arsed implementation are those who comment here - instead of influencing said implementation. Somehow what we do ourselves at work is always good and nice and perfect (while what others do is brain-dead and pointless). Now okay you could say an engineer is too small to change politics, but how about 100 engineers? Thousands? If the whole practice would say loud "whoa this doesn't fly" I bet the decision makers would listen. But no, most of the techies will say "okay whatever let's do it" then (maybe) come online to complain.

▲palata a day ago | parent [-]

The thing, I believe, is that engineers are just people. They also have strong, uninformed opinions. It's not like every engineer is an expert in cryptography, right?

For instance, there is a large consensus against ChatControl, agreeing on the fact that it's impossible to let the good guys decrypt arbitrary data without opening it to the bad guys. As a result, I don't think it's still a debate: the EU won't ask to weaken the encryption. Now an alternative to that is stuff like scanning the messages on the device (which technically does not break the encryption) and report them when it makes sense. There is a problem with that, too: in order to detect "illegal" content, someone has to decide what is illegal. And because it evolves, probably it's a list that can be updated remotely. And whoever controls that list has some surveillance power we may not want to give anyone (I know I don't).

It seems to me that this second issue took more time to get traction, but now engineers generally understand it. So if a politician asks, they can say "you can technically do that, but you are building a dangerous surveillance apparatus". It doesn't mean that the politicians won't want it, but at least they should understand the technical stance on it.

So for ChatControl, engineers seem to understand the problem, and the consequence of that is that ChatControl hasn't passed. At least not yet, and now I think those in favour have to convince enough people that building those surveillance tools is a good idea.

For age verification it's a lot different, IMO. There is a website dedicated to it: https://ageverification.dev/, explaining how they want it to be privacy preserving, which is technically possible. But still, there doesn't seem to be a large consensus about this (see the comments on HN). Many, many engineers keep saying "it is not technically possible", which is wrong. Maybe it is not desirable, which is a different debate. But it is technically possible.

What does that mean? Well for age verification, from what I see the "engineers community" is not credible at all. I can't get anything from it "as a community" other than "people yell many things, many of them being factually wrong". Many will say "not only it's not technically possible, but ON TOP OF THAT it is not desirable", but why would I listen to their second point when I can verify that the first one is already wrong?

It's always the same problem: not every opinion is worth the same. The opinion of an expert is worth more in their domain of expertise. So you can't just run a poll on HN and go with its result, right? In the case of age verification, for instance, it seems like the EU has been talking to actual experts, because https://ageverification.dev/ seems pretty good.

I feel like many engineers yell against it because they don't like the idea of age verification, not because it's technically bad. Which is a fair stance to have. But the honest way to defend it is by actually talking about why they think privacy-preserving age verification is a bad thing. Claiming that it is not technically possible is either uninformed (and therefore not credible) or manipulative, IMHO.

▲ToucanLoucan 3 days ago | parent | prev | next [-]

Boiling kid's (and adult's) brains probably makes them a decent chunk of money, either directly via salary or indirectly via stocks. Ensuring kids remain healthy makes no money. An unfortunately large slice of the tech sector doesn't give the tiniest shit about the health of our broader society or any group in it if it means their lines stop going up, or even go up slightly less fast.

▲watwut 2 days ago | parent | prev | next [-]

Imo, both. The more right wing people started to have aggressively anti-EU stance once Vance openly stood on the side of Orban and against EU and democracies in general.

And some people see tech companies as worship worthy and trying to restrict them is kind of a blasphemy.

▲modo_mario 2 days ago | parent [-]

The Vance thing is far too recent and inconsequential across europe?

The sentiment precedes all that and mostly stems from the EU being in some ways originally lib left dominated and still being seen as facilitating non-eu migration

Regular right wing people (aka not one of the many parties potentially receiving donations) don't tend to love giant webtech companies. Especially since they feel like they're often used as a tool against them and aren't a local thing that draws nationalists either.

A focus on privacy also isn't a very left-right defined thing tho i have noticed that the most far reaching expressions of it come a bit more from the further ends of that spectrum. (you'll see some very left leaning people at fosdems privacy focused/related stands for example)

▲dgellow 2 days ago | parent [-]

> don't tend to love giant webtech companie

That’s a bit outdated since musk bought twitter

▲eowln 3 days ago | parent | prev [-]

[flagged]

▲palata 2 days ago | parent [-]

> The sentiment that having to present our ID to use tiktok gives us the heebie-jeebies, and for good reason.

So push for privacy-preserving age verification, such that you don't need to leak your ID to anyone but TikTok can still prevent kids from accessing it?

▲eowln 2 days ago | parent [-]

>privacy-preserving age verification

No such thing.

▲palata 2 days ago | parent [-]

That's my problem with the debate: people like you seem very proud to be uninformed. It exists as much as end-to-end encryption exists. It's cryptography, it's not up to debate.

But people who have no clue are very vocal about their belief that it does not exist.

▲eowln 2 days ago | parent | next [-]

There are no active implementations that do not suffer from severe issues.

▲soco 2 days ago | parent [-]

There are two ongoing implementations: one weaker in the EU boooo and one good in Switzerland. None having severe issues. Questions?

▲eowln 2 days ago | parent [-]

I have none, I already am aware of the functioning of the systems that have been deployed, hence my statements of fact.

▲palata 2 days ago | parent [-]

You obviously have no idea, and this really looks like a new account made for trolling (negative karma).

▲NewsaHackO 2 days ago | parent | prev [-]

I also am not aware of any, do you have any examples?

▲sylware 13 hours ago | parent | prev [-]

It feels like they use such topic such as this one as a decoy to dodge to deal with the core and real EU online services interop issue.

EU online service must interop with web sites and not be _ONLY_ web apps which are hard dependent on the massive and ultra complex web engines of the whatng cartel (this word fits their behavior).

The EU online services which cannot be provided via a web site (noscript/basic HTML), which are very few to none, should use well defined, as simple as possible, stable in time, open and published protocols/file formats. That can be 'subset' of broadly available protocols/file formats. The benchmark is writing real-life code to handle such protocols/file formats should be reasonable for a very small team of devs or a solo average dev.

I wonder where all the money to deal with web apps is actually going, you know where web sites were carefully broken...