Hacker News new | past | comments | ask | show | jobs | submit | vurpo's comments login

Last time I went to CCC I was greeted with a giant Antifaschistische Aktion banner at the entrance. That made me feel pretty safe about the event (unironically).


Please read the beginning halves of sentences, not just the latter halves. As a European, this whole thread reeks _painfully_ American.

I honestly had no idea what the original commenter was even talking about before I started reading replies, which in hindsight was a mistake.


You sound like you may be the right person to ask about European feelings towards gypsies..thoughts? As an American, I don't understand the almost universal contempt held for them by Europeans.


Noto Emoji and Noto Color Emoji are two different fonts. Do you somehow already have Noto Emoji (the new one) installed on your system?


TLS or SSL never meant that kind of safety in the first place. Even before LE, there was no guarantee that HTTPS means it's not a scam, and the PKI system has never been meant to guarantee that anyway! Let's Encrypt didn't change anything here, and they're doing exactly what they or any other CA is supposed to do.


They didn't need to do that. They simply got their cert into browsers and OSes, and there you go.


That's a lot of text without giving any motivation for your main point. What makes an app package created by the app developer more "commercial" and anti-FOSS than an app package created by a distro-specific packager? What's the difference between an RPM package and a Flatpak package? Proprietary software and FOSS software both get distributed in both kinds of packages.

In short: what are you talking about?


> What makes an app package created by the app developer more "commercial" and anti-FOSS than an app package created by a distro-specific packager?

The process: no developer can interact with a vast community, there is too much noise. No developer can prune bad bugreports, poorly formed questions etc. The multi-level community can. No developer can alone design the best software, good interactions with other people expert enough to interact well can.

With the classic FLOSS model:

- developers just concentrate on their code, there is no noisy community pushing messages to them, there is no issues about packaging, distro-specific stuff etc to care about;

- packagers just grab code and package it for a distro they know so an easier task for them than for the developer that might know just another distro/OS;

- community interact users-with-users and users-with-packagers and packagers-with-packagers that's keep the noise low enough to all parties profit and makes ideas flow from a cohort to another, processed and adjusted at any passage.

The outcome is: all users get help, at a level no other system can give, all face easier tasks than others development models, developers get the best feedback possible, all involved interacts each others creating new ideas, resolving issues, improving.

In the "direct" model: devs do not care of distros, they give something to anyone, as a result community is so noisy that they can't get good feedback, patches, bugreports, they get some, but mixed with countless other bad to a point no one can keep up the information flow. Developers do not care much about deps they use, so, as you can easily see in the wild, most "packages" are full of old vulnerabilities no one care about. Various cohort of people involved are "isolated" between each other. There is no real community just a mechanical hierarchy unable to improve itself BUT able to be steered by big players who decide founding a developer or another what ideas will grow and what will be pushed into oblivion.

It's essentially the very same thing you see in our society, from the classic internet, with usenet, minitel, mailing lists and the modern platform era. In the modern era there are flock of sheep and some shepherd, in the classic era there were peoples, peer between peers. In the classic era innovation was a thing, nowadays nothing really new born and most things fall apart.

> What's the difference between an RPM package and a Flatpak package?

The level of crappiness. An RPM can't normally work alone, so for instance it should not run an old openssl vulnerable version just because the developer of a small chat client have no time nor interest to update it. A flatpak can and normally do: see actual repos, classic packaging systems tend to have all deps up to date, flatpak, appimages, snap tend to be abandonware from the day 0.

Also an rpm can package a kernel, the modern ones can't, but still need a distro to run. So the classic systems are complete, the new one can't exists alone, can't form a system.

They have exactly NO REASON to exists from the FLOSS or the end user point of view.

> In short: what are you talking about?

About the reality, just look around and see. If you can't see there are two options: you are inexperienced enough to not notice anything, perhaps inhabited to the commercial model "go to a website, download a program, install it" single app per single app with the system pre-installed by someone else or rarely installed by hand in a day-long manual and painful process. Or you have commercial interests. I see no other options.

Try to see NixOS or Arch repo, then see Snap, Flatpak, AppImage repos and compare them: what about the real freshness of them? What about the completeness and options? What's the outcome? Do you want a containerized chat client 158Mb instead of 2.5Mb due do the gazillion of copies of multiple version of same deps that when you exchange a file demand a long process to being able to see it with another application? Or do you prefer have no files at all, all on someone else computer, like Google Drive etc and you just run a WebVM "the endpoint must be just a dumb terminal of modern cloud-mainframe"? Because that's is.


I don't know what kind of hoops Snap require to access files. Flatpak, however, has a pretty simple system for a fully sandboxed app to open a file or a directory: it just opens a file or directory picker, you pick a file or directory, and the app gets access to that file or directory without getting access to anything outside what you picked. Behind the scenes this is done via the XDG Portal system, but that's irrelevant to the end user who only sees a normal file open dialog.

Then for apps that have dotfile-type directories where you can put config files and other stuff, those simply exist in app-specific directories under .var in your home directory, so they're not difficult to find either.


> Flatpak, however, has a pretty simple system for a fully sandboxed app to open a file or a directory: it just opens a file or directory picker, you pick a file or directory, and the app gets access to that file or directory without getting access to anything outside what you picked.

Unfortunately that still breaks any kind of multi-file format, where opening one file might implicitly require accessing additional other files, and in some cases those additional files might even be distributed all over the directory hierarchy.


Snaps use exactly the same system with the same underlying APIs. An app that uses XDG Portals will work seamlessly when packaged as a snap, too.


The 1541 was slow because the VIC-20 shipped with faulty hardware that couldn't move data on the serial port at full speed, so they patched their software to get it working at a much slower speed. The C64 didn't come with faulty hardware but still retained the same software routines for backwards-compatibility. Those fast loaders then replaced these slow compatible routines with full speed ones.


Putin said in his recent speech that he thinks Lenin made a mistake in breaking up the Russian Empire into independent nations (that then joined the USSR). I think Putin is more after restoring the old Russian Empire than restoring the USSR.


GPT-3 and these kind of natural language generators in general can already present opinions if prompted to do so. Depending on what you mean by "opinion", you might also require some memory to let it be consistent in what opinions it presents to you. (Not that humans are always consistent either.)

If you dig any further than this into the question, you quickly get back to the age-old question of "when is it 'real' consciousness and not just an automaton that acts and sounds conscious?"


True. I'm just thinking of what makes AI different from people. Could "opinions" be such a thing.

When we say somebody has an "opinion" we often mean they have some hidden agenda why they raise their "opinion". It is not a fact but opinion. It is frequently about what should be done, and to whom, and thus is most often self-serving.

Humans are intention driven creatures which continuously try to advance their own agendas. So I wonder are there AIs which would exhibit similar behavior, trying to influence the behavior of others with their "opinions". AAnd do we need such AIs? Are they not evil?


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: