HNHacker News
TopNewBestAskShowJobs

pveierland

1,496 karma · joined November 27, 2014

submissionscomments
pveierland··on EU Parliamentary Research Service calls VPNs "a loophole that needs closing"
That is the exact point. A "best intentions" approach to e.g. solve age verification through a "privacy respecting" mechanism leads to the government mandating that only physical devices with the built in security verification mechanisms, as allowed via government approved signatures, would be able to run these age verification apps (because otherwise you could cheat the mechanisms). This is a direct attack on general purpose computing and facilitates digital fascism by requiring that all software is signed in a manner approved by the government (directly or indirectly).

This is not hyperbole - Android will be locked down in only 115 days [0]. Further, in order to enforce such an age verification mechanism, you will end up requiring all software to either account for why it would not need to be integrated, or integrate with such a government mandated mechanism. This introduces accounting, surveillance, and approvals for every possible use-case. How would an online forum dedicated to discussing political topics survive? They would have to prove that they are complying with such measures that increase the barrier to operate and ensures that any forum or other arena of interaction would have to be accountable and justify how they are verifying interactions within the bounds of these laws and mechanisms.

Further - beyond locking down devices you would clearly need to lock down networks and communication in order to enforce such restrictions - which leads to deeper and broader filtering / scanning / monitoring - and preventing workarounds such as VPNs to ensure that all thoughts and actions are within these government set bounds.

Further - it is essential to realize that the outcome of this is not the best case of a single measure - the outcome in many cases will be the combination of measures taken by many different governments across the globe that each cut away certain freedoms.

Again - this is not only about age verification or digital wallets - it is the continuous pattern shown clearly through a range of actions made over time by the EU to introduce client-side scanning, age verification mechanisms, locking down devices. It's not all in place yet, but it is certainly being aggressively worked towards. All of these mechanisms will directly facilitate digital fascism as it will literally become illegal to effectively read, think, communicate without it being inside the remaining allowed bubble set by the government.

Another post on the front page of HN just now is about France seeking to dismantle end-to-end encryption. It is then not about the best possible view of a single measure in isolation, it is that these measures will lock in digital fascism broadly unless you are able to see what they facilitate when combined.

[0] https://keepandroidopen.org/en/

[1] https://reclaimthenet.org/france-moves-to-break-encrypted-me...

pveierland··on EU Parliamentary Research Service calls VPNs "a loophole that needs closing"
I don't care for a limited and selective best-possible interpretation of a subset of measures viewed in isolation. The point is that a broader set of vectors are being used continuously to gradually ensnare and limit digital freedom.

This is not a misleading headline, this is a document from the European Parliamentary Research Service that calls out VPNs as a technology that may need to be moderated in order to enforce restrictions such as age verification.

https://www.europarl.europa.eu/RegData/etudes/ATAG/2026/7826...

As you are calling me out - specifically answer how restricting access to VPNs would benefit the freedom of thought, communication, and information within Europe, and not be something that - together with other measures - can help facilitate digital fascism.

pveierland··on EU calls VPNs "a loophole that needs closing" in age verification push
Speaking from the POV of my country, you absolutely have prime minister + minister level understandings that seem to plainly be based on issues such as "we need to stop children from bullying each other on social media", "we need to help police surveillance to stop crime", "we need to protect people from internet porn" etc, and it seems to be that the political capital and will to create these measures comes from short-term attempts to solve certain problems, without being able to understand how a broader set of these measures will together create digital fascism.

Beyond that I fully believe there are intelligence agencies, advertising agencies, military interests, IP control interests etc that are all working very diligently and in more targeted ways to each achieve their goals better by pushing for specific measures and helping to amplify moral panics to build the necessary political capital.

pveierland··on EU Parliamentary Research Service calls VPNs "a loophole that needs closing"
Age restrictions + VPN bans + encryption restrictions + client-side monitoring + restricting general purpose computing.. It's just rapid descent into digital fascism set up by people who have no ability to see how the dots will end up connecting.
pveierland··on AI uses less water than the public thinks
Annoying to see slop like this on the front page taking up people's attention. Resource utilization connected to new infrastructure that is planned to be scaled exponentially is clearly important to analyze. Instead of research and insight this post just facilitates ignorance on the topic.
pveierland··on EU Age Control: The trojan horse for digital IDs
It's frustrating to see how shortsighted and tech-illiterate politicians are on these topics. This article from Norway today presents the attitude of the minister of digitization as simply "Social media companies are making billions and we expect them to adequately implement age verification systems with solutions that respect privacy and we will fine those who fail at doing this".

The fantastic irony is that in some weak attempt to protect against the "evil big tech companies" they directly facilitate increased mass surveillance and removal of individual rights, instead of choosing more scalable and robust answers such as funding and promoting the development of protocols and open standards that can be applied voluntarily and in a decentralized manner to help mitigate these problems.

I have computers side by side on my desktop running Linux, and it is amazing to me how I can call `wormhole send --message hello` and receive it on the machine next to me, knowing that only I can receive this message, without it running through an age approval mechanism, without it being client-side scanned, and without being logged in some government database.

This is the century of AI and robotics - technologies which can facilitate great concentration of power and wealth. Gradually introducing mechanisms that facilitate digital fascism seems like a really bad way to guard us against this.

https://www.nrk.no/norge/datatilsynet-bekymret-for-personver...

pveierland··on Framework Laptop 13 Pro
Agreed - $255 sounded too good to be true. Fingers crossed it doesn't end up too crazy, because that chassis looks good!
pveierland··on Framework Laptop 13 Pro
I'm amazed looking at the upgrade kit. It's everything you need to go from the old 13" to the Pro; including chassis, display, battery, keyboard etc - listed at $255. That's an instant buy from me.

https://frame.work/sg/en/products/framework-laptop-13-pro-ch...

pveierland··on I just want simple S3
Garage has worked well for me and gives a good sense of stability. They provide helm charts for deployment and a CLI. There's also very few concepts to learn to start to use it, while e.g. for SeaweedFS I feel like you need to parse a lot of docs and understand more specific terminology.
pveierland··on All elementary functions from a single binary operator
Got curious to see whether SymPy could be used to evaluate the expressions, so I used Claude Code to build a quick evaluator. Numeric and symbolic results appear to agree:

    nix run github:pveierland/eml-eval
    EML Evaluator — eml(x, y) = exp(x) - ln(y)
    Based on arXiv:2603.21852v2 by A. Odrzywołek
    
    Constants
    ------------------------------------------------------------------------------
      1        K=1    d=0    got 1                    expected 1                    sym=ok   num=ok   [simplify]
      e        K=3    d=1    got 2.718281828          expected 2.718281828          sym=ok   num=ok   [simplify]
      0        K=7    d=3    got 0                    expected 0                    sym=ok   num=ok   [simplify]
      -1       K=17   d=7    got -1                   expected -1                   sym=ok   num=ok   [simplify]
      2        K=27   d=9    got 2                    expected 2                    sym=ok   num=ok   [simplify]
      -2       K=43   d=11   got -2                   expected -2                   sym=ok   num=ok   [simplify]
      1/2      K=51   d=15   got 0.5                  expected 0.5                  sym=ok   num=ok   [simplify]
      -1/2     K=67   d=17   got -0.5                 expected -0.5                 sym=ok   num=ok   [simplify]
      2/3      K=103  d=19   got 0.6666666667         expected 0.6666666667         sym=ok   num=ok   [simplify]
      -2/3     K=119  d=21   got -0.6666666667        expected -0.6666666667        sym=ok   num=ok   [simplify]
      sqrt2    K=85   d=21   got 1.414213562          expected 1.414213562          sym=ok   num=ok   [simplify]
      i        K=75   d=19   got i                    expected i                    sym=ok   num=ok   [i²=-1, simplify]
      pi       K=153  d=29   got 3.141592654          expected 3.141592654          sym=ok   num=ok   [simplify]
    
    Unary functions  (x = 7/3)
    ------------------------------------------------------------------------------
      exp(x)   K=3    d=1    got 10.3122585           expected 10.3122585           sym=ok   num=ok   [simplify]
      ln(x)    K=7    d=3    got 0.8472978604         expected 0.8472978604         sym=ok   num=ok   [simplify]
      -x       K=17   d=7    got -2.333333333         expected -2.333333333         sym=ok   num=ok   [simplify]
      1/x      K=25   d=8    got 0.4285714286         expected 0.4285714286         sym=ok   num=ok   [simplify]
      x - 1    K=11   d=4    got 1.333333333          expected 1.333333333          sym=ok   num=ok   [simplify]
      x + 1    K=27   d=9    got 3.333333333          expected 3.333333333          sym=ok   num=ok   [simplify]
      2x       K=67   d=17   got 4.666666667          expected 4.666666667          sym=ok   num=ok   [simplify]
      x/2      K=51   d=15   got 1.166666667          expected 1.166666667          sym=ok   num=ok   [simplify]
      x^2      K=41   d=10   got 5.444444444          expected 5.444444444          sym=ok   num=ok   [simplify]
      sqrt(x)  K=59   d=16   got 1.527525232          expected 1.527525232          sym=ok   num=ok   [simplify]
    
    Binary operations  (x = 7/3, y = 5/2)
    ------------------------------------------------------------------------------
      x + y    K=27   d=9    got 4.833333333          expected 4.833333333          sym=ok   num=ok   [simplify]
      x - y    K=11   d=4    got -0.1666666667        expected -0.1666666667        sym=ok   num=ok   [simplify]
      x * y    K=41   d=10   got 5.833333333          expected 5.833333333          sym=ok   num=ok   [simplify]
      x / y    K=25   d=8    got 0.9333333333         expected 0.9333333333         sym=ok   num=ok   [simplify]
      x ^ y    K=49   d=12   got 8.316526261          expected 8.316526261          sym=ok   num=ok   [simplify]
pveierland··on The Claude Code Source Leak: fake tools, frustration regexes, undercover mode
They have an annoying sandbox issue which pollutes your repository root with a set of empty files. Not the cleanest tool, but the paradigm is a big upgrade to previous AI coding.

    .bash_profile .bashrc .claude .env .gitconfig .gitmodules .idea
    .mcp.json .profile .ripgreprc .vscode .zprofile .zshrc config
https://github.com/anthropic-experimental/sandbox-runtime/is...
pveierland··on Index, Count, Offset, Size
When correctness is important I much prefer having strong types for most primitives, such that the name is focused on describing semantics of the use, and the type on how it is represented:

    struct FileNode {
        parent: NodeIndex<FileNode>,
        content_header_offset: ByteOffset,
        file_size: ByteCount,
    }
Where `parent` can then only be used to index a container of `FileNode` values via the `std::ops::Index` trait.

Strong typing of primitives also help prevent bugs like mixing up parameter ordering etc.

pveierland··on NPMX – a fast, modern browser for the NPM registry
- Certain pages load but are not able to load content, e.g. https://npmx.dev/package/@storybook/addon-docs fails to load content with:

> `[nuxt] Cannot load payload /package/@storybook/addon-docs/_payload.json?c459501f-8eb7-49c9-be9c-4a197fa35a39 Error: Invalid input`

- Scrolling fast on Firefox + Chrome is broken and resets the search results page to start.

- Pressing up/down arrows should navigate search item results instead of focusing individual tag elements.

pveierland··on OpenAI has deleted the word 'safely' from its mission
This is something I noticed in the xAI All Hands hiring promotion this week as well. None of the 9 teams presented is a safety team - and safety was mentioned 0 times in the presentation. "Immense economic prosperity" got 2 shout-outs though. Personally I'm doubtful that truthmaxxing alone will provide sufficient guidance.

https://www.youtube.com/watch?v=aOVnB88Cd1A

pveierland··on Outsourcing thinking
Yep! However the problem will increase by many orders of magnitude as the volume of generated content far surpasses the content created by autocorrect mechanisms, in addition to autocorrect being a far more local modification that does not generate entire paragraphs or segments of content, making it harder to excuse large changes in meaning.

I agree that they make for poor excuses - but as generative content seeps into everything I fear it will become more commonly invoked.

pveierland··on Outsourcing thinking
One bothersome aspect of generative assistance for personal and public communication not mentioned is that it introduces a lazy hedge, where a person can always claim that "Oh, but that was not really what I meant" or "Oh, but I would not express myself in that way" - and use it as a tool to later modify or undo their positions - effectively reducing honesty instead of increasing it.
pveierland··on Europe wants to end its dangerous reliance on US internet technology
Why do you think this is substantial? The software built in these repos is not tied to GitHub.

Git is decentralized and using a self-hosted instance of Gitea / Forgejo will give you a replacement for the essential parts of GitHub. GitHub is absolutely replaceable.

pveierland··on Europe wants to end its dangerous reliance on US internet technology
They have 1.5k public repositories here at least: https://github.com/orgs/alphagov/repositories?type=all

Collection of RFCs: https://github.com/alphagov/govuk-rfcs

Open design system: https://design-system.service.gov.uk

pveierland··on Europe wants to end its dangerous reliance on US internet technology
I'm hopeful that it can work if:

- The top-down mandate is very general: e.g. "default to using or contributing to open standards, protocols, file formats, and interoperability".

- It's applied across many nations and organizations that can themselves choose how they wish to allocate their resources to achieve their specific objective. Meaning that the tax authority in Norway can contribute to a specific tax-reporting software project and collaborate with nations X + Y + Z on this specific project as long as it is fit for their specific purpose and mandate.

Ideally this helps incentivize a diverse ecosystem of projects that all contribute to maximize public utility, without forcing specific solutions at the highest level.

One example of a recent French software project is Garage which is an open-source object storage service. It's received funding from multiple EU entities and provides excellent public utility: https://garagehq.deuxfleurs.fr/

pveierland··on Europe wants to end its dangerous reliance on US internet technology
The tax authority in Norway alone employs 500 full-time software developers. If all of Europe followed France's example to adopt the UN Open Source Principles for all publicly funded development - and prioritized open formats + protocols + interoperability - it would within only a few years be possible to greatly improve software reliability for all nations.
pveierland··on Bye Bye Gmail
Just disabled the "Smart features" setting in Gmail. Disabling this feature force-disables these other settings:

- Grammar

- Spelling

- Auto-correct

Pretty dark UX to force users into an all AI or nothing situation.

PS1: Yes, I've paid for Google One for years and I'm not just a free user.

PS2: Yes, these features are entirely possible to provide without training on your specific user data.

pveierland··on Nova Launcher added Facebook and Google Ads tracking
That sucks. I've been running the paid version for years - however it's clear that it hasn't been properly maintained for a while and it suffers from sporadic crashes.

Any recommendations for launchers that are functionally similar? The launchers mentioned in this thread so far are quite different.

pveierland··on Retreating from EVs could be hazardous for Western carmakers
You could also flip that and talk about the risks of when your gasoline supply get shut down due to some event. With an EV stack you can generate your power locally and add resilience that way.
pveierland··on Retreating from EVs could be hazardous for Western carmakers
One of the big risks for manufacturers seem to be that EVs are fundamentally more compatible with automated production and allows simplifications to the car stack. It would seem that the costs and risks of the keeping the ICE stack alive will keep increasing over time as it loses relevance to EVs.
pveierland··on Jimmy Wales trusts the process
I believe this is wrong for many topics. The news media is strongly incentivized to sensationalize and continuously produce content for their readers and viewers. Wikipedia is able to cover many topics that are less contested in a slower and more tempered manner, as the content does not need to be marketable or immediately available. As an example, for STEM topics I'd trust Wikipedia far more than any news media.
pveierland··on Ask HN: What are the best engineering blogs with real-world depth?
Tweag has many interesting entries with good technical depth:

https://www.tweag.io/blog

pveierland··on OpenTelemetry Distribution Builder
When using the OpenTelemetry Collector you might need a set of modules not supported by a standard distribution, e.g. the otelcol-k8s variant but with some extra exporter from otelcol-contrib. I found building such a custom distribution a bit cumbersome but that it was made much easier by this project.
pveierland··on It’s time to free JavaScript (2024)
Rust runs quite well today via WebAssembly. Continuing to improve interop between Web API / WASM / language runtimes seems like a good route that allows people to use the language they prefer against shared Web APIs.
pveierland··on Project Euler
What's really neat is that most of the problems are so compact that you can just read about the full problem, then spend hours and days thinking about possible solutions.

I created a PDF version that I keep on my Remarkable for puzzling: https://github.com/pveierland/project_euler_offline

pveierland··on Wikipedia says traffic is falling due to AI search summaries and social video
> The forcing factors aren't what they are supposed to be though.

Is it clear what they should rather be - and are there any examples of mechanisms that have worked better at a scale like this? How are you judging that they are not what they are supposed to be?

If the resulting body of work, which is the totality of Wikipedia, is able to be a curated and high signal collection of knowledge as a result of these mechanisms, how can it be said that they are not working? Having forcing factors, even if they are not ideally aligned or executed, which pushes contributors to increase the quality of their edits to pass, seems overall like a good thing. I'm not saying that its processes and mechanisms cannot be improved, I'm saying I believe it is incorrect to say that they are not working as a whole.

> "Credible" sources and citations are exclusively up to the article moderators personal tastes which are very rarely objective.

Overall I believe Wikipedia to be curated by a large group of people which coordinate through various rules and consensus mechanisms, where I don't believe it is correct to state that sources and citations are exclusively up to any specific article moderators, as they need to be able to build consensus and co-exist with other moderation.

Exactly because Wikipedia is such a large body of work it seems more resistant to corruption to have a large number of curators with different tastes and motivations. How would you determine that their selection of sources and citations is very rarely objective - especially when objectiveness itself seems quite hard to agree upon for many of the topics covered?

From my perspective it seems far more important to consider the quality and value of the totality of Wikipedia, which is massive and signs that many things are working, rather than insisting that it is not working, especially in times where knowledge is being broadly attacked, and where Wikipedia is one of the targets.

← PreviousPage 2 of 9Next →