LinuxQuestions.org
Go Job Hunting at the LQ Job Marketplace
Go Back   LinuxQuestions.org > Forums > Non-*NIX Forums > General
User Name
Password
General This forum is for non-technical general discussion which can include both Linux and non-Linux topics. Have fun!

Notices

Reply
 
Search this Thread
Old 09-08-2014, 05:34 AM   #31
Germany_chris
Member
 
Registered: Jun 2011
Location: Stuttgart, Germany
Distribution: Arch
Posts: 615

Rep: Reputation: 227Reputation: 227Reputation: 227

If it hadn't been Lennart and Red Hat would it have been such a big deal?
 
Old 09-08-2014, 06:28 AM   #32
enorbet
Member
 
Registered: Jun 2003
Location: Virginia
Distribution: Slackware 14 is Main OpSys on Main PC, 2ndary are OpenSuSe 13 and SolydK
Posts: 725

Rep: Reputation: 328Reputation: 328Reputation: 328Reputation: 328
Quote:
Originally Posted by Germany_chris View Post
If it hadn't been Lennart and Red Hat would it have been such a big deal?
I think that's a really good question but not for the more obvious heaping of yet more vitriol on those two but for the wider view. It is my opinion that systemd or something like it was/is inevitable. Linux is just too good and too easy to take, like low hanging fruit, for Big Biz to ignore and since IBM, to name one, has invested $Billions, they haven't. The reasons behind "too good and too easy" are the same - Real Freedom and Labor of Love. Commercial stuff only has to be "good enough" and preferably, cheap. When all or the main gain is reputation among your peers, a person will not settle for just "good enough" and in addition has access to an immense amount of collaboration from very diverse sources. "With enough eyes, all bugs are shallow" is just one observation among many by Eric S. Raymond that was true in 1999 and still true today.

Compare and contrast the rate of deep changes in Windows, or even Mac, compared to Linux. How different, on more than a facial makeup level, is Windows 8 from XP versus how Linux was in 2003 compared to now?

I think it is just way too tempting to businessmen to see all those creative man/hours and product "go to waste". If they can take a big bite and lock it down, they surely will. Whether by design or by accident, systemd just lends itself to that process.

Hopefully there are enough "hackers, amateurs and hobbyists" (and people willing to make donations to those capable people) left to keep the original paradigm going. On bad days, I recall that Kit Stereo/Audiophile equipment and "pooging" (and just about every form of user controllable gear and "hot-rodding") are largely a thing of the past.
 
Old 09-10-2014, 08:55 PM   #33
theKbStockpiler
Member
 
Registered: Sep 2009
Location: Central New York
Distribution: Fedora14,Scientific 6.1?, Mandriva 2010 ;GO MAGEIA!!!Next up Gentoo
Posts: 836

Original Poster
Rep: Reputation: 51
Quote:
Back to the original question: "Is there an effort to challenge systemd?"
To answer this question you first have to define if you mean "systemd as init system and process supervisor" or "systemd as a project to have the building blocks of a Linux OS in one source tree".
At first I meant the challenge Of systemd as a init program but the second has got me thinking.

There must have been pressure put on Debian ; for example , to implement systemd because of the possibility of having to take on more than they could handle if they went against Redhat as far as coding goes. Redhat must have had the support of Gnome and others. How they got this support is the real story behind the acceptance of systemd. If not, systemd would have stayed as source code. As a purist ,I'm surprised that most of the distros are using it instead of just Redhat/Fedora, the rpm tag alongs of Mandrake forks and Suse. Take on a massive amount of coding ,sell it as better , progress is inevitable and you become the kingpin of Linux.

Seeing as this is not the culture that spawned Linux ,I'm also surprised that the battle lines where not drawn when systemd was (init)iated with fedora.They may have been but the silent war is now occurring. I actually see a shift to BSD, Gentoo and Slackware as inevitable. I like rpm and all the K stuff but Redhat can just be a fork of Linux as far as I'm concerned ,all by themselves. I want to follow the GPL/LINUX ideology and not economic rules.

Last edited by theKbStockpiler; 09-10-2014 at 09:08 PM.
 
Old 09-11-2014, 01:26 AM   #34
Germany_chris
Member
 
Registered: Jun 2011
Location: Stuttgart, Germany
Distribution: Arch
Posts: 615

Rep: Reputation: 227Reputation: 227Reputation: 227
Because an organization make a decision you disagree with doesn't necessarily mean they were pressured.
 
Old 09-11-2014, 09:21 AM   #35
TobiSGD
Moderator
 
Registered: Dec 2009
Location: Hanover, Germany
Distribution: Main: Gentoo Others: What fits the task
Posts: 15,617
Blog Entries: 2

Rep: Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073Reputation: 4073
Quote:
Originally Posted by theKbStockpiler View Post
I like rpm and all the K stuff but Redhat can just be a fork of Linux as far as I'm concerned ,all by themselves. I want to follow the GPL/LINUX ideology and not economic rules.
Red Hat is a major player in Linux, a player that large that many of the stuff you now use and appreciate actually does come from Red Hat or at least had their developers involved, especially when it comes to kernel stuff. I fail to see why that is good, but their involvement in systemd is somehow bad.
 
Old 09-11-2014, 12:08 PM   #36
enorbet
Member
 
Registered: Jun 2003
Location: Virginia
Distribution: Slackware 14 is Main OpSys on Main PC, 2ndary are OpenSuSe 13 and SolydK
Posts: 725

Rep: Reputation: 328Reputation: 328Reputation: 328Reputation: 328
Quote:
Originally Posted by TobiSGD View Post
Red Hat is a major player in Linux, a player that large that many of the stuff you now use and appreciate actually does come from Red Hat or at least had their developers involved, especially when it comes to kernel stuff. I fail to see why that is good, but their involvement in systemd is somehow bad.
Sorry TobiSGD but that statement is a non sequitur. Almost nothing, and certainly no one is so simple as to be all good or all bad. It is not at all uncommon for large corporations to have departments at odds with each other, often working against each other. It is almost routine for corporations to be humanitarian in some efforts (even if only for Public Relations) but utterly ruthless and lacking in any principles in others.

As this applies to systemd, it is further complicated by point of view, in this case whether you embrace a certain amount of loss of control in the quest for convenience, or worry that they threaten to marginalize what you have learned and loved, and grown used to.

"To the worm, the robin's song doesn't say 'CheerUp'"
 
Old 09-11-2014, 11:18 PM   #37
ReaperX7
Senior Member
 
Registered: Jul 2011
Location: California
Distribution: LFS-7.6, Slackware 14.1, FreeBSD 10.1
Posts: 3,660
Blog Entries: 15

Rep: Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128
Good or bad can come from anywhere. It's the motivations, reasoning, and ethics behind it that make it good or bad.

There are no real counter projects to systemd as Tobi said, but there are still a few viable alternatives from the older projects that still work. The OpenBSD shim is more or less just attempts to segregate code from the project, rework it, and make it work without the core systemd dependencies. I wish them success though, as it's needed.

Maybe we'll have one day what eudev amounts to for networkd, logind, etc. one day, but my concern is what is going to be done about netlink vs kdbus. If kdbus gets into the kernel are we going to get a libkdbus to handle communication, will d-bus itself work with kdbus, or will eudev be forced to fully fork?

Last edited by ReaperX7; 09-11-2014 at 11:28 PM.
 
Old 09-12-2014, 09:21 AM   #38
sundialsvcs
Guru
 
Registered: Feb 2004
Location: SE Tennessee, USA
Distribution: Gentoo, LFS
Posts: 5,425

Rep: Reputation: 1159Reputation: 1159Reputation: 1159Reputation: 1159Reputation: 1159Reputation: 1159Reputation: 1159Reputation: 1159Reputation: 1159
Linux is, of course, a poster-child for the notion of "cooperative software development," a.k.a. "open source," versus the proprietary model used by other vendors ... notably Microsoft.

As an industry, we've very clearly found that "a rising tide" not only "lifts all boats," but that it also lifts those boats higher than they otherwise could ever have done. What has proven to be much harder to develop, however, is consensus.

Linux is "big business." When the majority of the world's Internet servers run on this platform, "it's big business." Or, rather, it is the foundation technology of "big business."

The operating system, you see, never is "the real, be-all end-all, point." It's the engine that causes the car to move, but ... well, except for summer Saturday nights in certain shopping-center parking lots ... "'the engine' is never the point." Everything else that surrounds it ... is.

For instance, "Microsoft Windows" is really just a platform upon which you can run all kinds of Microsoft software in a tight, vertically-integrated system that includes Sharepoint, Lynx, Exchange, SQL Server, MSMQ, and "the list goes on." There's a huge amount of plumbing under there that your grandmother ... and maybe, you ... never knew about. But it gives corporations the thing that they're willing to pay millions of dollars for: a tightly integrated software system that actually works together which can be paid-for with one seven- or eight-figure invoice.

If a group of people want to build "more plumbing" and part of that effort is a replacement for the "init" daemon, and if it remains open-source, then I personally don't have a problem with that. I'm actually (tangentally ...) interested in what they're doing. I don't have a love-interest in configuration-files.

A very big issue in "the corporate world," where you might have thousands of computers to take care of, is "configuration management," which must be automated. Microsoft Windows is very, very good at that. So, it's only natural that you would see Linux-based initiatives that are moving in similar directions and doing so in a similar way. But this isn't "the sly Collective onslaught of The Borg." It's a response to genuine customer demand. If you're going to push out tens of thousands of Linux-based boxes, the effort required to reliably maintain (and to secure!) all of those boxes must be pushed down as close to "zero" as you can get it.
 
Old 09-12-2014, 04:48 PM   #39
ReaperX7
Senior Member
 
Registered: Jul 2011
Location: California
Distribution: LFS-7.6, Slackware 14.1, FreeBSD 10.1
Posts: 3,660
Blog Entries: 15

Rep: Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128Reputation: 1128
And with servers stability can not be compromised. Either it works right or someone might get a pink slip. I've seen it first hand where a corporate file server went offline during a routine maintenance, the administrator unintentionally updated software not on the list, and when the server was put back online, it crashed after about an hour up. The file server had to be taken offline during business hours, checked, rolled back, tested, and put back up, but it took several hours of work costing the company hundreds of thousands of dollars. Afterward, the admin was no longer with a job.

Critical system components have to work right, and when you deal with components of an OS that are modular in design, it's easier to diagnose problems, find fixes, keep one component while updating others, and ensuring the system works.

One of many problems I see with systemd is the fact it is so monolithic, if one part breaks, you have to diagnose a whole package. Example: If you're using logind, and it has an issue, you can't just keep the same revision of systemd or rollback logind. It's not like rolling back ConsolKit. You have to get a new or older version, find a patch if one exists, and rebuild the whole systemd package, test it, and pray it works.

This was a topic we learned in IT Software Administration in college about being careful with large monolithic inclusive projects, and what was hinted at in The Cathedral and The Bazaar.

If part of the "plumbing" breaks, you have a choice. Replace the faulty pipe itself, or replace a whole plumbing system.

It would be nice if the OpenBSD project does get the shim built because if it's backported to GNU/Linux it gives a more choices of software, not just for users but corporate level IT as well.

Last edited by ReaperX7; 09-12-2014 at 06:33 PM.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
[SOLVED] Is VLC Worth the Effort? tronayne Slackware 13 08-09-2013 06:02 AM
Boot Delay 30min: systemd-analyze blame systemd-tmpfiles-setup.service BGHolmes Fedora 0 07-27-2011 10:02 AM
Last ditch effort to keep using linux tothemax6 Linux - Newbie 33 12-14-2008 08:39 PM
last ditch effort at debian race Debian 11 03-22-2004 11:20 PM
Help, is it worth the effort? Jackcnd Mandriva 1 03-06-2004 12:24 PM


All times are GMT -5. The time now is 10:28 AM.

Main Menu
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
identi.ca: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration