|
|||
Quote:
The Hurd is not really his focus or concern, at least not these days: From his recent slashdot Q and A: "The GNU Hurd kernel (and the GNU/Hurd system, which is GNU/Linux with the Hurd instead of Linux) is not a high priority for us any more, because it would be a replacement for the free parts of Linux, and we don't need to replace those. Volunteers continue to work on the Hurd, because it is an interesting technical project. The parts of Linux we need to replace are the nonfree parts, the "binary blobs". But replacing those has nothing to do with the GNU Hurd. The main work necessary to replace the blobs is reverse engineering to determine the specs of the peripherals those blobs are used in." Whether he feels some kind of private wistfullness to see a GNU kernel replace Linux is anyone's guess, but publicly there's no evidence that that is his "dream" if it ever was. He makes few public statements "dreaming" about particular GNU technical roadmaps. I've seen posts of his on mailing lists recently where he seems to care a little about guile, e.g. in a recent guilemacs thread on emacs-devel he confirmed that he still considers guile scheme the official GNU extension language and encourages its use in emacs. Plus he expressed approval for Ludovic Courtes's efforts with guile to write guix, saying it looked cool or something like that. As far as I can tell (and I only occasionally watch Hurd mailing lists) he's not paying any attention whatsoever to the Hurd. As far as there being a GNU operating system, in his words above and in other places, you can see that he considers that done, with Linux as the kernel, except where there are proprietary parts mixed in. So, yes, there's gnusense that FSF sponsors and some other "pure" distros FSF vouches for as meeting all their criteria. But this is only one of many FSF campaigns you see on their website, not their main purpose. There are a handful of GNU developers, aside from Hurd people, who seem interested in making a linux distro named Guix, but it sounds like an individual pet project, not something brought about top down from FSF or rms. In fact, whether it's desirable or important to have an official GNU run linux distro seems to be a matter of debate among FSF people. The only places rms seems to me to care overly much whether people use GNU vs. competing free software is... 1. that one GNU project encourage use of other GNU projects (I can't think of an example off the top of my head, but I think I've seen this). I'd guess the motivation here is related to #2 below. 2. when the competing free software has a license that allows proprietary extensions, particularly where he sees real tactical advantage (in his effort to have all software under free licenses) being weakened. He's bothered quite a bit right now by the growing popularity of llvm vis a vis gcc, for instance, since he's claimed to see gcc's licensing terms push large companies to release software and feels that with llvm they wouldn't bother to send out their patches, and that more users would end up with a proprietary compiler. Last edited by thirdm; 20th September 2014 at 12:58 PM. Reason: more guessing at rms's motives. |
|
||||
I guess Gnu has responded to Systemd, with dmd.
It replaces Sysv-init. I haven't yet seen how much it resembles Systemd. It is supposedly intended for use on Gnu/Hurd. |
|
|||
Funny, previously I thought Ludovic Courtes (a GNU developer) created it for Guix, but it turns out it's way older, older than systemd by far, perhaps even older than pulseaudio (not sure the dates on that one):
https://lists.gnu.org/archive/html/g.../msg00019.html Since it's written in Guile and not very large or complicated even now I'd think the resemblance to systemd would be minimal. Looking at the manual perhaps there are some similarities maybe (I'm not seriously familiar with systemd, so I'm kind of guessing here). You communicate with dmd using a special command that talks to it over a socket to start up services. But it doesn't use dbus. It seems to keep dependencies of a service and start them automatically. Also when you stop a service it stops the services that depend on it. However, it does not use cgroups to identify related daemon processes. It seems to support automatically restarting a service if it dies, but can be configured not to keep trying if something keeps dieing. Systemd does something like that too, doesn't it? I don't see where it wants to be more than an init system, though. That would be a major, and pleasant, difference. |
|
||||
Systemd is a project. The init system is one part of it. Systemd integrates everything, replacing the UNIX modular design with a monolithic structure that resembles Windows more than UNIX. The purpose? Not difficult to figure out. That is why I am testing BSD. I am hoping to find an alternative before Linux is a locked-down system where choice and the terminal are frowned upon. (The latter is already a reality with the most popular distributions.)
|
|
||||
Quote:
|
|
||||
Quote:
I will have to learn the ins and outs of systemd, just like the developments incorporated into other Unix like and non-Unix like systems. For the lowly position I station in the service world, this is a must. I'd probably admire RMS more if his goal was to develop a structure of freedom based on freedom from the business world. The gnu license is largely targeted at competition with proprietary control. That license still allows for a successful, albeit possibly more honest, business model. This is more true will the OpenSource movement. It would be nice if the inspiration was to develop a system that didn't benefit anyone interested in making money. The technology at hand provides a powerful tool with regards to the exchange of knowledge, accumulation of personal education, networking and general communication. These things are important to people outside of the business world. The business world doesn't mind selling people tools for those needs. But, the existing free implementations of these tools work well if not better, often offering the user a chance to expand their comprehension of the technology they use. A tool like OpenBSD is really like hammering in a nail with a dump truck, for my personal needs. But, systems like it are the only ones that offer clean administration and advanced tools for my tinkering needs. But, even systems like OpenBSD are inevitably going to be swayed by the politics involved in commercial competition. Things like systemd are a result of that competition. It competes by maneuvering a whole structure under its influence and eventually eliminates any alternative. Like the Gnu/Hurd no business free system will likely gain enough investment to take off. No one will be interested in maintaining a system that does not include all the requirements needed to compete commercially. The only way I see RMS's position on this helpful to the private enthusiast, is that by some degree companies and interested parties are obligated to contribute back. This does give faster development to projects with lesser invested interest. But, it only supplements the need for people to actual develop outside of the models presented by business minded progress. This also helps prevent the investments into his model become usable by companies introducing proprietary control. This I imagine seems useful since it helps keep the free software at some level of actual competition with the more well known proprietary entities. Just maybe this allows people to see a value in moving away from what they already know and use. Again this is just a bandage on the actual problem. People are willing to pay and conform for their ability not to pay attention. The main attraction here is for the slightly more competently involved persons, like system administrators, developers, service techs, etc. There are many Linux users of all types opposed to the way systemd is being implemented. Paid developers aren't gaining anything by forking out of it without pay and the hobbyists aren't united enough to propose, organize, and develop a suitable code based revolt that would be luring enough to the larger groups. The status is conform, move elsewhere, or fade out. The idea of standardization and progress together make for a powerful means of control. If you lack the standard, you don't fit into progress. If you don't fit into progress, you lose attention. Once attention is lost, you fight to keep the lights on. This in no way means that the standard is good for everyone. But, it is usually good for the ones that can effectively define it. Well, this is my narrow, not nearly well enough researched, response devised by the mental patterns of todays current whims. |
|
||||
Quote:
Really, my intent isn't to turn this into a GPL vs ISC/BSD debate, but I can't stand the hypocrisy of the GPL camp. One need look no further than Linux to see that it is a complete pig-sty of proprietary control flying a flag of "freedom" (in RMS holy name Amen). Complete hypocrisy. Don't like systemd? Tough. Red Hat and Canonical are giving you the finger right now and you're going to take it and like it. That's the system the GPL and RMS promote.
__________________
Linux/Network-Security Engineer by Profession. OpenBSD user by choice. Last edited by rocket357; 20th October 2014 at 02:14 AM. |
|
||||
Quote:
(Service Management Facility) on Solaris which is functional since 2005 and SRC (System Resource Controller) which AIX had since God knows when. Also check SMIT (System Management Interface Tool) on AIX. The another great achievement of the most popular commercial distribution Red Hat 7.0 LXC is 20 years late reimplementation of Solaris Zones and FreeBSD Jails. Docker is cheap reimplementation of BSD Warden. Finally the grandest achievement of all in Red Hat 7 is XFS. The only problem that is the classical IRIX files system which I used in mid 90s and which Silicon Graphics ported to Linux 15 years ago. So 22 years and counting in the development of Linux the community could not come up with a native file system lets alone with something really original. /proc came out of Plan 9 for the record! Lets continue to invent hot water |
|
||||
Quote:
|
|
||||
I didn't mean to do that.
It all is hypocrisy, just like you've said. But, even following the line of thinking that it is served with, you still have ill logic. I don't think with him its just about money. I'm pretty sure he leverages for control too. He does get a little testy when OpenSource is stapled to his name and also has an issue with the blobs in the Linux Kernel. But, all in all the GPL is still competition. It just isn't freedom competing with non-freedom. Its leveraging which has nothing to do with freedom. During the systemd initiations <- bad pun - he didn't say a thing. Systemd is really close to the type freedom antagonist he has been complaining about. Its maybe not as bad as Adobe Flash, but it provides the way. Last edited by fn8t; 20th October 2014 at 02:20 PM. |
|
||||
Quote:
Quote:
|
|
||||
I'm going to divert back to stallman for my 2 cents.
Is he crazy? Absolutely, but he's the good kind of crazy. 1. Old code was an oddity, peak of research type stuff. Like higher maths, it was shared freely among fellows, yet retained a certain pecuniary value primarily through obscurity. 2. Proprietary closed software licensing lends itself rather well to obfuscating real computer knowledge and maximizes profits. Limiting competition to those in the "know", a la Bill Gates/Oracle. 3. One day, after living in item 1, world Stallman was confronted rather rudely with item number 2. A thousand odd geniuses before him had gathered round to work on this new marvel of calculation for near a hundred years sharing source code and algorithm freely as it advanced the state of the art and one day in an attempt to print something Stallman found his work held hostage by some narrow minded, shallow, myopic, obstructionist, chisling, money grubber. The rest of his days have been spent in a haze of nerd rage and idealism. I love him. I like the BSDs more in culture and substance, but he did a lot to make the ecosystem what it is. GCC, Emacs, etc are better than Linux imho. The problem with systemd as I understand it is that the "year of linux on the desktop" folks, really, really, really, really want that thing, like bad, like the only thing they want for Christmas kinda bad and they don't really care about esoteric *nix hackers that like to run things the old fashioned way. Enlightenment is the heavyweight contender in bsd licensed window managers. here's their systemd page: https://phab.enlightenment.org/w/e18..._user_session/ http://0pointer.de/blog/projects/socket-activation.html Personally, I don't much get the obsession with boot times. The face of power management has changed a bit. I put my laptop to sleep and it wakes up rather quickly and un-bothered. Uptime is 3 days 13 hours. I can't see how even cutting boot time in half would matter. |
|
||||
Quote:
http://marc.info/?l=freebsd-advocacy...3629713977&w=2 It is futile attempt to try to understand reasoning of people like Lennart Poettering whose success string include some of the worst code and ideas ever to come from Linux like avahi daemon, PulseAudio, HAL, systemd and counting. Last edited by Oko; 23rd November 2014 at 05:14 PM. |
|
||||
I'm sure that hal was developed by someone else rather than the same developer?
Poettering is just out to recreate windows desktop/server, probably at the behest of Red Hat, who stand to gain and expand their market. Red Hat employ him and the rest of the team, Red Hat use it in their testing distribution and earlier this year released it to enterprise users. Arguably that makes it a Red Hat project in everything except name. An integrated Linux base system, rather than the situation which has persisted for years would have gone a long way towards remedying this, ironically that's what Poettering et al are trying to achieve, but it's just not to the tastes of the people who thought they were running free *nix. The Linux world is full of ironies - many of it's users went there to escape windows and immediately demanded equivalent software and all of the same apps, working out of the box no less. These kind of people became the majority users. The Linux server market, competes directly with windows server and is therefore doomed to reinvent it in order to suck in the MS certified people (who will be completely lost with something like Debian Linux, never mind a *BSD). Is it any wonder that systemd was born? Supply and demand. |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Stallman: How Much Surveillance Can Democracy Withstand? | J65nko | News | 0 | 15th October 2013 12:27 AM |
Opinion for new SOHO Machine - Processor and Windows compatibility | IronForge | OpenBSD Installation and Upgrading | 4 | 1st January 2013 12:44 AM |
Stallman: Did I say Jobs was evil? I meant really evil. | J65nko | News | 1 | 30th October 2011 08:18 PM |