Why all of innovations come from elsewhere?

  • Thread starter Deleted member 63539
  • Start date
Maybe he just likes to argue. Some people like to argue for the sake of it.

And w.r.t. copying, before a fair in my industry profession the boss reminded us to check if the others copied our stuff. I replied that more worrying than seeing our stuff being copied would be to see nobody copying it. After thinking a bit he replied that he had to concur.
 
My friend has a very strong hatred for FreeBSD. The reason he gives me is that OS just a copy cat and didn't invented anything.
FreeBSD's roots are ATT UNIX. Not a copy. The real thing. Ask him what he thinks of Linux which originally copied how Unix works. Most of its utilities and base programs are all copies. No creativity. Strictly a copy.

ALL of your examples in your second post are copies. They are copies of the Linux operating system with different configurations and installed programs. Nothing original. Only copies.
 
to be honest, most innovation happens in Linux land, the portion in FreeBSD is tiny. Loads of apache projects are Linux only (lots of stuff coming from the Linux Foundation), Java is still big in business, software defined network projects as an example - many of them lack FreeBSD support.. If the discussion is about that you cannot hold up against it because that is a fact. FreeBSD has other strenghts, for example: it is well designed: in Linux you have fast innovation that pumps out a lot of code of questionable quality - short sighted/"immediate solutions" that tend to show its quirks after some time running. Look at the history of sound systems, the kernel message bus systems, or the ideas docker had with overlay filesystems/aufs: several buggy iterations. Also systemd is a great example: all my ~ 500 linux server have been quite stable until systemd came with the distributions: quite some distributions deliver bad unit files which leave the whole systemd ecosystem in a questionable, fragile state - there is a good reason why docker did not interact with systemd in the beginning and still cannot really cope with it without errors.

Look at the lines of code of the Linux kernel: no wonder they discover new vulnerabilities almost every week, the Linux Kernel Self Protection Project was a good idea but definitely needs more drive. Just ask your friend if he is happy that he has to install weekly Ubuntu/whatever kernel patches and has to reboot.
 
There are so many broken and unmaintained ports, let alone not updated ones.
Does this mean there are no broken or unmaintained packages on Linux?

Note: nginx, the server, was originally created on FreeBSD. It only moved development to Linux recently due to marketing pressure.
 
As a FreeBSD user, I'm pleased with it. But saying the Ports system is one of the most up to date repo out there is plain wrong. There are so many broken and unmaintained ports, let alone not updated ones. I'm still waiting for the maintainer to update CodeLite to the latest version.

I found the link I was after, here. 74.9% of packages are up to date. While "top 50" doesn't sounds amazing, it's sits above some of the big Linux players. That percentage is impressive when you consider FreeBSD Ports is in the top 25 when it comes to number of available packages.

>5000 outdated packages is a lot, but some careful usage decisions can leave you with a nicely maintained system.
By the way, the Makefile for CodeLite doesn't look hugely complicated, have you tried upgrading it and submitting a patch? If my experience is anything to go by, you'll find it easier that trying to patch some of the various Linux distributions packages!

I have never used launchd but to be honest, I like SystemD more than SMF.

I've not done a tonne of stuff in systemd and it is fine, but it doesn't feel as powerful as SMF - then again I learnt SMF when I worked at Sun, so I may be biased ;)
 
As others have mentioned, FreeBSD is a continuation of BSD, which was an extension of Research UNIX. It is clear that neither the OP nor his friend have spent any serious effort to understand what they are talking about. BSD folks (Kirk McKusick, Bill Joy and others) contributed major contributions to UNIX (look ‘em up). FreeBSD continues that work in the present. Read Bruno Latour, if you want to blow up the myth of invention on a philosophical basis. To take a Clintonism completely out of context, “It takes a village”. Another chestnut, “Let’s not reinvent the wheel”, and my personal favorite, where would we be if Reeses hadn’t dropped the chocolate into the peanut butter?
 
  • Like
Reactions: a6h
Does this mean there are no broken or unmaintained packages on Linux?

Note: nginx, the server, was originally created on FreeBSD. It only moved development to Linux recently due to marketing pressure.

Sooner or later all need to move to Linux as the first class development platform.
 
By the way, the Makefile for CodeLite doesn't look hugely complicated, have you tried upgrading it and submitting a patch? If my experience is anything to go by, you'll find it easier that trying to patch some of the various Linux distributions packages!

I could get it build, but the build failed with something about kmem and kvm. I have no idea how to troubleshoot it.

BTW, the name is misleading, it's a FreeBSD library and have nothing to do with Linux KVM: kvm()
 
I have moved on from Linux to BSD as my first class development platform.
Long ago, when I started my web dev company, my brother-in-law worked as a project manager for a large Windows shop. So I got a bunch of free development tools from him. That wound up being a disaster so he recommended moving to Linux but, after a few days, we realized FreeBSD was the only logical choice. We grew to a development team of 10 developers--all using FreeBSD workstations and servers and never skipped a beat.
 
He refuted it by stating the the Gentoo portage system has to remove many systemd-ism, too.
The ignorance in this statement is staggering. Gentoo portage started out as a re-implementation of Freebsd ports in Python. Systemdisms are creeping into Gentoo. Portage predates systemd by about a decade.

Gentoo is OpenRC based. I said regardless of SystemD or OpenRC, it's the same Linux system and GNU userland.
What the heck does this have to do with anything? The creator of Openrc was chased out of Gentoo by politics. He's a Netbsd dev now. He gave up Openrc to a systemd fanboi that's busy killing it by making it bad-design compatible with systemd.
 
I could get it build, but the build failed with something about kmem and kvm. I have no idea how to troubleshoot it.

BTW, the name is misleading, it's a FreeBSD library and have nothing to do with Linux KVM: kvm()

The FreeBSD KVM library predates Linux's Kernel-based Virtual Machine by around 13 years, maybe more...The man page you pointed to goes back to FreeBSD 2.0-RELEASE in 1993, whereas development for Linux's KVM started in 2006.

So the name is not misleading, not least because we are on FreeBSD and not Linux :)
 
Why do you have to choose ONE operating system? They all have strengths and weaknesses. I've yet to find the perfect OS for all tasks on all hardware on all budgets.

And it's not just the OS - it's the windows manager (Xfce, Gnome, KDE), the editor (emacs/vim), the database (PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, SQLite, etc.), the CPU (AMD, Intel, ARM), the hardware (spinning drives, SSDs), programming/scripting languages (Perl, Python, Tcl, C, PHP, C++, Java), shells, blah blah blah.

For task t, if you NEED x or y and they are ONLY available on OS MegaOS, then you gonna need MegaOS for that task.

Otherwise, shrug, it's up to you. People have different likes/dislikes, different thresholds of what annoys them, different things are of value to them.

This is like being back at school and arguing which 8-bit micro is "best" ... best for what?

In terms of innovation - for server use - I don't want a fast pace of innovation. I want something rock-solid and conservative. That has evolution rather than revolution. FreeBSD suits me fine because of that. It can't stay stuck in the 1980s, or 1990s or whatever so changes have to come. Some changes come from other places - that's OK. On the GUI side - things seem to change all the time regardless of OS ...
 
Sooner or later all need to move to Linux as the first class development platform.
One certitude, this assertion is false.
in the past I ear :
«Sooner or later all website will be in flash.»
«sooner or later all worksration will use Windows»
«Sooner or later, all web project will be developed in ruby»
...
 
In this particular case, as an individual, it doesn't matter what OS you use to develop as long as your software runs on Linux.
As a VC-funded start-up, it's a different story: when you live on others' money, you have to abide by their will.
And what they require from you is what they've read in Gartner's et al. reports, because their customers use the same references to make their investment decisions.
 
I don't understand why these pro GPL people obsessed too much with the fear of someone stealing their code and think the GPL is the only way to keep this from happening.

As I expressed several times in this forum, GPL community isn't really about "copyleft" but about parasitism. This is not about someone stealing their GPL code[1] these people are afraid[2] of but losing the ability of hijack (in very lazy terms) someone else code into GPL. If you develop an important code and use a custom license with the same characteristics of GPL but being incompatible with GPL (all copyleft licenses are incompatible with each other by nature) you will receive the same hate because they cannot pull your code in.

This is the typical "I'm saving the world" complex. These people think the GPL community is fundamentally important to the world, and the GPL licensed code is the most important code in the world when in reality, the most important code in the world (nuclear power plants, ICBM, avionics, railway; safety critical software in general) is running anything[3] but Linux, GPL, or open-source code in general.

The closest Linux/GPL/open-source software get of a safety-critical system is in the entertainment system of the passenger cabin in commercial jets, which is the most irrelevant code in the whole jet (not safety critical at all).

[1] mostly completely irrelevant anyway.
[2] yet, as proven, GPL code is often stole and they don't go anything against the thief.
[3] VxWorks and other certificated proprietary software, including the filesystems (Datalight come to mind), a lof of custom OSes and applications.
 
As I expressed several times in this forum, GPL community isn't really about "copyleft" but about parasitism. This is not about someone stealing their GPL code these people are afraid of but losing the ability of hijack (in very lazy terms) someone else code into GPL. If you develop an important code and use a custom license with the same characteristics of GPL but being incompatible with GPL (all copyleft licenses are incompatible with each other by nature) you will receive the same hate because they cannot pull your code in.
Whereas I largely agree with you, the interesting question to me is "are the GPL fanatics wrong?" I'm not sure. Quoting Adrian Chadd:

I used to bend over backwards to try and get stuff in to stable releases of the open source software I once worked on. And that was taken advantage of by a lot of people and companies who turned around to incorporate that work into successful commercial software releases without any useful financial contribution to either myself or the project as a whole.

...the most important code in the world (nuclear power plants, ICBM, avionics, railway; safety critical software in general) is running anything but Linux, GPL, or open-source code in general.
There's at least one important exception:
 
Whereas I largely agree with you, the interesting question to me is "are the GPL fanatics wrong?" I'm not sure.

In another life I could have been a religious GPL zealot. I very much like the idea of ensuring that free software remains free. However as we know, companies like Google, Apple, Microsoft, etc still find ways to sh*t on their users even with GPL'ed software.

Disappointingly, from what I see, the viral effect doesn't really work for the big guys and ends up just punishing other open communities. I.e the example of LibreOffice sucking dry OpenOffice is a little sad. Coupled with the fact that the ODF now have very little competition as a result and are feeling fairly untouchable right now. This is probably the main drive for them trying to monetise in the cloud (along with "enterprise" partners) with LibreOffice Online rather than actually improve things for the open-source community.

I would like to see a more weaponised GPL license however. For example "this software is not allowed to be run on a platform with known DRM schemes". I think free software also needs to be compiled to be more aggressive, like refusing to work if it detects Valve's Steam DRM service, etc. Embed this "feature" into something like Gtk+ and it will propagate fast!
I would even suggest go so far as to enforce the GPL license behind GCC to ensure that it cannot be used to build proprietary software. That would have pretty much eliminated Android (before it moved to clang) and many other locked down embedded crap.

The AGPL is good though. I wish more people would use it. It would be especially effective in the "web development" space where people drag in a load of random crap just to split a string. It would "infect" projects very quickly.

One of my favorites was the original Open-Motif license (before it moved to standard GPL). It was open-source but did not allow for building on commercial operating systems. So no Windows port was allowed, even with Cygwin.

I feel sorry for Richard Stallman. He has the right idea and has been fighting for free software for years. And yet the industry just keeps on getting worse and worse. I couldn't do it, I would become too bitter!
 
I think developers should have the liberty of doing whatever they want with their contributions; individual or corp. Coercion is never a good thing and there is no debt obligation with Free Software. If they leave it permissively open, great. If something is proprietised, so what? You still have your own code/user experience/whatever to live with. By the way, the GPL isn’t going to magically make something ’good’ or a great user experience. In fact, a lot of GPL’d software are managed by developers who’re completely inept (ie. GNOME). If X software is proprietary, and Y software is open source, but Y application is useless shit or functionally inferior; I will be using X software to get work done.
 
In another life I could have been a religious GPL zealot. I very much like the idea of ensuring that free software remains free. However as we know, companies like Google, Apple, Microsoft, etc still find ways to sh*t on their users even with GPL'ed software.
I think the original gangsta was Tivo. Hence why "Tivoization" was coined.

I would like to see a more weaponised GPL license however. For example "this software is not allowed to be run on a platform with known DRM schemes". I think free software also needs to be compiled to be more aggressive, like refusing to work if it detects Valve's Steam DRM service, etc. Embed this "feature" into something like Gtk+ and it will propagate fast!
I would even suggest go so far as to enforce the GPL license behind GCC to ensure that it cannot be used to build proprietary software. That would have pretty much eliminated Android (before it moved to clang) and many other locked down embedded crap.
I believe this is what the GPLv3 aims to be.

I think developers should have the liberty of doing whatever they want with their contributions; individual or corp. Coercion is never a good thing and there is no debt obligation with Free Software. If they leave it permissively open, great. If something is proprietised, so what? You still have your own code/user experience/whatever to live with...
Linus Torvalds agrees with you, and so do I, for whatever that's worth. I found it interesting that he admits he's not concerned with what's good for users of Linux. He actually liked Tivo because they helped make Linux on MIPS more robust.
 
Back
Top