Some more news about WHY there is no good outcome from GNU/Linux (GNU/Linux-libre) in the near future:

First the news perhaps many have already read as Red Hat has announced their Enterprise-oriented system-distribution no longer being "open-source". Attention therefore also: If you want to post only single-sided criticism about "Red Hat" as company or "why they are the inbound evil in its pure form", please don't. Yes, this is not a fine announcement and wind for sailing in the definite wrong direction. But as always: It is one decision of many and we are here not bound to it.

Second is a bit more of problem as ClamAV announced also vesion 0.105 will definitely need Rust for compiling. To quote:

But there is even more to acknowledge:

You look close enough? I underline the term again: Database downloads allowed until: Sept. 14, 2025
This is includes also: Currently, every version from ClamAV 0.102 and down, including all patch versions, are unsupported, and are actively blocked from downloading new updates. (source: https://docs.clamav.net/faq/faq-eol.html)

So the elementary question: Where is the "free as in freedom" here? Answer: There is none or at minimum we would need to port back the databases to come for older releases. You can clearly see where the decision is and it is not for "stable development", more towards "progress at any cost". But also to link their argumentation for including Rust as it just fair to do and they have for sure a reasoning: https://docs.clamav.net/faq/faq-rust.html

Yes, the more services are focussed only towards the "global network" (internet) in a whole, the more security-issues will occur and the more complex solutions to oppose them have to be. The reasoning for the change towards Rust is nevertheless a major issue as neither the licensing-problems are solved nor the the further to include dependencies are cleared up. So no fork is helping and also no rewrite as the issues persist.

And I repeat again why we have decided against Rust: Not only because of its questionable licensing. For compilation there is always cargo needed as there is no way around. Who makes sure that every further dependency downloaded at build-time is licensed free, libre and permissive? Exactly: No one. So at a point we will have decide if we can further safe include clamav or need to remove it fully. Nevertheless: For the moment the decision is not needed, but surely to be discussed.

And this all demonstrates one elemental point very well: We have lost control as we have given up and handed out essential parts of the system-infrastructure towards outside parties, we once thought being trustworthy. But the question would be: Are they for real trustworthy as they decide for us towards for our "well-being"? We are not talking about some health-situation, but about keeping data and information in hands. So the situation is also crucial and exactly the point why HyperbolaBSD is coming to existance. ClamAV is in hands and development of Cisco Systems, so this company is surely interested that people will always use their newest release. And how can we call those dependencies "free and libre" compatible? How can we even dare to ignore those parts for so long? Our goal as project is to emancipate every user hosting own services, being independent for the whole lifespan in regards to the own decision either to upgrade or not.

Yes, that "internet" is big, full with services and different protocols and not easy in many parts. But do we really want to hand out elementals of our data? For what pricing? To be seen? To get in "touch and contact"? I vote for that we step back and look at those crude picture for a moment. For me speaking: I don't want all of this at once and want to decide on my own, when something works ... why do I always need the newest update? Just because someone tells me? The own decision is important, not someone else positioning.

Not to mention I get this when I go to clamav:

docs.clamav.net
Warning iconYour browser is out of date!
Update your browser to view this website correctly. More Information.

When I use iceweasel-uxp... that's probably a good sign they have become junk...

Btw, I am currently using 2023-06-18 version of iceweasel-uxp.

I built it from other sources.

Just thought I should add, yet another free software project is presuming as if rust is free software when it really isn't.  And as I have said, I hate it when these free software projects presume something to be something worthwhile when it is complete hogwash and/or unworthy of usage.

Corporateware as Throgh might say is also a problem though. Redhat, is another example of corporateware. If you want to know nicknames I have for redhat, message me on tox.

Oh yes, not to forget:

https://gitlab.gnome.org/vstanek/gnome-info-collect
https://archlinux.org/packages/extra/an … o-collect/

The year of the linux desktop as they call it may very well happen someday, but even if it does, I think its good to have something that is completely libre and is built for immense security and stability like Hyperbola.

I would agree that open source is not a great term though. Merely for one specific reason, some consider that horrible search giant corporation, Google to be open source. 

The difference between open source and libre is that LIBRE has to be fully free. Where as open source, does not have to be fully free.  OpenBSD is the best example of open source probably. But most open source ideologies fall way short of that mark.

Missing the mark is putting it mildly in fact.

I definitely support getting users away from fully proprietary systems, even if linux gets a bit more bloated. But at the same time, only if there is an alternative that is actually libre that actually has what I and others need at that exact time.

As of now, this doesn't seem to be the trend.

But HyperbolaBSD will at least be a giant shelter from this situation.

Have you ever thought about the question on how systems can get updates on years? Especially in regards of Debian for example? No, this is not meant to blame the project. But it is not ALL free and libre, especially not fully independent: Just have a look at the sponsoring for the DebConf 23 (as an example): https://debconf23.debconf.org/

And there is this company named freexian: https://www.freexian.com/
Where you can have a maintainer-contract and a price-listing for packaging: https://www.freexian.com/services/debian-packaging/

So ask again: Free as in freedom with capitalism in the background? Yes, people need to sleep, eat and live. But there it is: The big question on how to sustain a system and have a working support. And the idea many people like to describe like uncountable developers are working in the back just for them to have a free and libre system. Ask again and please respect especially that point: Free software is free, but also not possible to be done fully "gratis". Support is needed for a working system and to be honest going for a community-driven way forward. But on that level with uncountable packages? It is not to blame anyone here, but please deal with facts and information, not with "dreamcastles" building. Systems using Debian and its package-repositories have a big background to build on top of. Systems being independent have this not. It depends on the community and the people seeing that. So I ask foremost to go for the questions instead for the "dreams" to build. They are motivation for something, absolutely. But they are not the base to ask questions and approve the solid state of free and libre culture. We need to stabilize exactly that and there is only that: Facts and full transparent handle. Not slogans, not commercial handle, but the full idea of community-driven efforts.

The programming-language is not the ssential point: The licensing is and there we have even more candidates, likewise applications, libraries and frameworks. The problem is that the idea of free, libre software sounds nice and is for sure also. But the outcome is not going good the more people take it on the lightweight shoulders. Like "open-source" is not solving the issue also but people think those projects are either driven from thin air without anything or abandoned and someone takes over. Nothing is going in that exact way and even more projects have chosen to use Rust (and others) or integrate dependencies based on them. And the choice for waiting on integration of Rust within gcc is also going not that fast as it seem.

So the spheres of GNU and GNU/Linux are not going in the direction of freedom.

Sure thing both, but on different levels. In the generic we are not even at the point of sizing or complexity before, that was the reasoning as I was and be shocked about the learned details for PHP also. So the message is more that we should not take something as self-evident given: We should look at all details as for the moment it looks alike PHP is not going to be a possible part for HyperbolaBSD. And that's the main issue: The details are very important, so yours given the same as the license and complex situation of freedom-issues the same.

But we have another third level also: The enforced usage when more and more projects transfer towards those problematic languages, frameworks and dependencies. So the situation stays the same as before, but also critical.

Please read again and not use the wording "bloat" within as it is NOT about that. It is the same issue as others alike Rust have: The point not allowing modifications without asking for permissions, not allowing redistribution without clear distance and renaming. So why should we make the work then? The answer is easy: We don't and leave PHP when HyperbolaBSD is coming.

Free software includes the four freedoms, or it is not free software. Restrictions made are not getting free software and the corresponding license work. So says also the FSF about PHP: https://www.gnu.org/licenses/license-li … n#PHP-3.01

You may include PHP or better not as we already done that for others like zstd where the reasoning is then another one: A free license, but not really free given software as there is no further choice or alternative given instead.

It is about being honest in the end also: If the FSF would look clear enough, they would never identify PHP as "free and libre software" and their licensing the same. It is an illusion, a nice meant compromise, failing at that special point where those projects and resulting software is therefore no longer free and libre and only "open-source". Because that is PHP in the reality, nothing more. Yes, this means also that some parts of the web-infrastructure needs a rebuild therefore. But better we do that and better we throw PHP and all parts needing out as it is NOT free and libre software. Just some "open-source", where the need to ask for permission when doing more is demanded. We don't need to lie to ourselves for an illusion as this won't hold up any longer: PHP changed its course with making the release no longer possible under permissive licensing. They did their own long before Rust and it is kind a joke they come through with it. We thought also different, but as said: We do license-audit whenever possible and learning more of the details.

The answer we included PHP nevertheless: Hyperbola is used for the web-infrastructure of Hyperbola and our small team has no time to go on different parts. So we focus main on HyperbolaBSD, do updating on Hyperbola GNU/Linux-libre. And when we package for HyperbolaBSD PHP is exactly no longer on our listing and the web-infrastructure will be redone with Python and Tauthon.

Interesting how Hyperbola is seen when especially the GPL is mentioned in combination. And it goes that far, that people reject this point while claiming to have too much corporate influence in the ecosystem of GNU/Linux. That point is coming directly from some BSD-spheres also. So again to underline that GPL is sharing code as a MUST and not as an option alike. The problem is not only the license but also the vendor-lockin or in some cases the technology-lockin. Sharing code is a good option, not a bad one.

Our technology-stacks are nowadays more restricted than ever before and we think it is somewhat sufficient. But looking on the details for H.264 should teach the opposite also. Do we have enough choices? The clear answer is: No, we do not have that. And if everyone thinks only about own needs, not everyone else is also being thought of and get help. Society is breaking when we commercialize it as we are doing now. So we need to speak this point out clearly as commercializing is not only an issue towards GNU/Linux.

When no one is thinking about dreams for the opposite: Do we have dreams for the better? Or do we just accept the common "Status Quo"? The reality is just like this: We have to ask kind and have to await for sure nothing or just a guess to get. Emulation? Only granted throughout some slight laws. Free and libre gaming? Only granted because they have not that big attention. Does anyone think companies and corporations would like to grant more attention for supertuxkart or supertux, to be used, modifed and installed for free? Oh come on, that's a NO again. And especially the codecs for video and audio are nevertheless protected throughout patents. So do we need to view something alike the "few against the big"? Personal I state the opposite: Let's look for the niche and stop asking kind, begging even. A system granting rights only at the point where someone has to give endless back and nowadays only renting software? Not a good vision of the future for free information and facts.

We need to face the facts: Linux (without GNU) is no longer on the side of economic usage. It is going more and more only about convinience, about the next big thing and fast implemented parts without having backwards compatible interfaces. More and more the so-called "industry" is taking over and the idea of distributing software for a complete system is failing same way. You think that big distributions are community-driven? Think again: They are not. There are enough companies and corporations behind to give financials and support, by the same time also to provide more and more projects for packaging. And the GNU-project? Is standing besides and accepting the "Status Quo". Please, think again: It is time for change.

Everything is political, also software.
Making older systems again running fine? Have a small environment being capable to run most tasks without problems? Having a system in own hands, not on other hands or combined with other computers or services without choice. Convinience is coming because the user is defining it, not someone else for the user. Make a difference!

Simple question for wayland: Is it portable for all other systems? Answer is: No, it is not. And it is foremost a Linux-only project. Also here:

https://gitlab.gnome.org/GNOME/gnome-se … equests/98
https://gitlab.gnome.org/GNOME/gnome-se … equests/99

So before we get again to the discusion: Hyperbola is NOT implementing wayland exactly for this reasoning. It is NOT portable planned and the most perspective for Linux. So it is for Hyperbola completely out of scope, not for now and not for later.

The problem arises alone from that much different approaches and implementations, where some implement further or different functionalitities. So there is no real default and also no guarantee the projects are working long. What "wayland" is doing nevertheless: It is adding another abstraction layer towards the system for X-only based parts. So even more complexity, more dependencies and that is way not working for Hyperbola.

Our main interest with technical emencipation would disappear very fast: More abstraction layers, more to look through for a working system to learn. It is the same with SDL1 implemented through SDL2 again to make it "compatible". While that there is even work the same for SDL3 and having SDL2 working through this. At a point people would need to look through endless lists of services, libraries and dependencies. And the interest for a small system is noted, but not really longer working.