Notices by LisPi (lispi314@udongein.xyz)

Did they get rid of the patent?

A patent as well as a design and model registration have been made to protect the technical architecture and design of Openprinter.

This is completely unnecessary if the information is publicly published into the Creative Commons. Prior work is already amply demonstrated.

while preserving the project's spirit of openness.

By ensuring a monopoly on manufacture and improvement. Right... pull the other one.

@toxi

@anemone @icedquinn It is a statement of community intent and it is enforceable to an extent, particularly with modern code versioning tools allowing for retroactive correction when an individual is found to be compromised.

The very same thing would apply if a dev was found to have been compromised by feds (or blackbagged and their computer taken over) or something similar.

@phnt

How many applications on your system actually use these? Exactly zero on my system. How many applications in the world actually use these?

Popularity in application is irrelevant in judging the novelty or esotericity aspect of a programming language or construct.

Scheme is well-known and documented, the most esoteric thing in Scheme is probably still the continuations and that is pretty old news (with ample computer science history anyway).

I don't have time to tinker for hours to fix a single small issue at this point, where the added "knowledge" would be a language I'm not gonna use again anyway. I'm past that in my life.

Then you could just leave an issue describing it and the problem to the best of your understanding (possibly with pseudocode) and move on.

There are some signing lines for git for that use-case, too.

I have done so in the past and it is a fairly common approach.

But it also doesn't make it a non-esoteric package manager that is hardly used by anyone excepts its developers and some FSF approved distro users. To be clear, even Nix is esoteric for me, despite its somewhat wider usage compared to guix.

I did omit that part as I am uncertain where they sit on the novelty/esotericity scale. They are novel, but at the same time it's been over a decade already since the first of their kind became stable.

@phnt

then it never gets fixed until it bothers someone with time to fix it.

I've been that someone, a few times. And yes, it doesn't really help anyone in the intervening timespan, but I do consider that such private software is less harmful than allowing slop, its implications & externalities.

@phnt Guile and Scheme are hardly esoteric. (If they are to you, then whatever institute you studied at failed you.)

And it solves a real problem, I already mentioned the issue of Gentoo and power stability where I live.

Immutable distributions do solve the issue of interrupted upgrades. (UPSes also fail eventually. Designing the software so it's always tolerant is the way to go.)

Functional package managers are also more convenient to mess around in multiple namespaces & ephemeral environments with. (A bit like Qubes without the massive overhead.)

And Guix might need forking to crawl back out of slop.

Unclear if the original project will rollback and fix things or not.

Chat, I'm tired. Why can't we just eat the boosters already and move on?

I don't really get what made Linus go corrupt.

Let's assume yearly net earnings of a million.

Basic needs are long covered. Rent is covered.

Finland is just not that expensive a place to live in.

Hobbies are long covered. Artist commissions are probably covered. What's left? Dump the rest on mutual aid & donations?

With the caveat that in USA basic needs aren't covered, that million needs to go into savings for emergency legal or medical fees. But that's because the USA isn't a functional country.

Why would you even personally want what the corposcum are offering to corrupt you?

@PeterLudemann @dalias @lauren Emails get blocked as spam trivially.

Ads often don't get blocked and ad targeting helps establish a probabilistic match for the specific filter you set on the ad campaign.

Yes, you do need Google or some other targeting ad network for maximal effect.

Firehose HTML+JavaScript email is considerably less useful (and pixel-size remote resource images were the common way to leak viewing status & some request data instead of JavaScript in HTML emails anyway).

but I suppose you'd see that as Google trying to stop a competitor

It is. Especially when considered in conjunction with how many spam operators operate off gmail's own servers that see little to no enforcement against them even when actively reported by humans.

@PeterLudemann @dalias @lauren

Only someone who hase never worked on anti-spam software would say that.

I have, though most of my experience with them is as a user/administrator, not a programmer.

Enough gets blocked that it is nontrivial to get through lastingly (LLM have made this worse), which means that it is very much not worth spending the labor & time (both of these cost money, even if the workers are slaves) making it so, instead of using a deliberately cooperative network that accomplishes the same purpose better.

Mail spam is an expense with an uncertain return.

Surveillance capitalism through ads is a guaranteed return and an additional possible return. For an end-user company that sells the user data they get and hopes to get a use sale too, the surveillance return may not be more than a minor mitigation to the cost of running the ads, though, but it is still considerably cheaper than the firehose approach.

I suppose next you'll be castigating Google because some of its Cloud customers send spam.

The effective impossibility to contact a human at Google in anything resembling a timely fashion without the use of a lawyer contributes a lot, alongside reports being widely ignored.

You can shove that flame, consequently.

@PeterLudemann @dalias @lauren

In other words, you have no idea how difficult it is to block spam, either incoming or outgoing.

That is logically inconsistent with the notion that I had to personally handle how much I received.

Enough gets blocked that it is nontrivial to get through lastingly

Have you any idea of the volume of spam that major providers such as Gmail, Hotmail, Yahoo receive and mostly block?

I thought not.

That is irrelevant to the point of view of a company considering ads vs spamming in hopes of sales. The odds of it being blocked or requiring constant labor & costs to avoid it get worse the better adversaries are at blocking.

That others don't get blocked or evaded a block doesn't matter. What matters is purely themselves.

It doesn't make sense to even consider the costs of spamming when ad networks are around.

Of course this is besides the fact that socially, products advertised by spam are typically expected to be scams (much like products that are too aggressively advertised). So spamming is quite possibly actively detrimental.

So getting enough marks that buy into spam to actively provide a signal back with identifiable information that makes it back to the original master for surveillance purposes? That kind of operation with physical mail would have the kind of costs and inefficiency that would make many parties give up on the notion entirely.

With email just opening it for checking would possibly provide a signal, but it is common now for "Unknown senders" to have most automatic signal-sending methods disabled in Email clients & web clients. So one is back to needing some level of victim buy-in/cooperation.

If this is true, then why is there so much spam? (See above - most of it gets blocked ... at one point ~90% of email traffic was spam but nowadays I think it's ~50%)

People are idiots. But also a lot spammers truly are scams and have the same lack of qualms as scammers usually display.

That is then not for surveillance, but because one mark out of thousands or millions that buys it and gets taken for all they're worth can more than offset the cost of getting to them, even with limited information value/observation (not necessarily true since exploiting someone this way can also involve stealing their ID for resale) and a lack of expectation of repeat business.

It is an entirely different business model.

What is your definition of "surveillance"? Showing adverts based on search queries is hardly "surveillance".

Deliberate information and data gathering or processing for purposes which intentionally lend themselves to things resembling intelligence work, whether on a personalized or mass scale, for any purpose.

That would be a workable definition.

Incidental or accidental mass gathering & processing of the same is something that can be called negligence.

@lauren @dalias It's not just streaming though.

Practically everything has ads, on purpose. And then there's the sale of user data on top of that for those who block ads.

The entire aggregated mass of commercial mass surveillance (the term surveillance capitalism is apropos) is then just bought for a token sum, as if it were any other buyer.

(This has been reported on and can be counted on as a fact. We're not even reaching very-likely federal agency shenanigans.)

Given how fairly often that includes a lot of PII and even what should be protected medical information, it can be quite enough to target someone off it.

Especially if, as most fascistic regimes, one doesn't actually care all that much about false positives.

#Gentoo wiki is wrong about btrfs & VMs and it annoys me.

No, one does not disable CoW on btrfs to improve performance. That nukes not only the CoW properly but also the integrity. The durability of data is also entirely ruined by doing that, btrfs' model of atomicity is entirely built on copy-on-write.

(swap is effectively the only-use case that makes sense for that file attribute.)

What one does is adjust commit frequency, dirty page cache ratios and in-VM workloads that unnecessarily request synchronous writes (get rid of systemd-journald persistent/on-disk IO, or the entire program, among others).

(One could just entirely disable the VMs' ability to do synchronous writes, but then they may truly need them sometimes. So it's best to actually understand one's workload properly.)

If one wants to use a faster filesystem that doesn't do integrity checking, don't mangle btrfs, configure it properly or just use xfs instead.