LinuxQuestions.org
View the Most Wanted LQ Wiki articles.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Distributions > Slackware
User Name
Password
Slackware This Forum is for the discussion of Slackware Linux.

Notices



Closed Thread
 
Search this Thread
Old 07-13-2012, 01:51 PM   #61
ottavio
Member
 
Registered: Nov 2007
Posts: 312

Rep: Reputation: 46

Quote:
Originally Posted by Mercury305 View Post
So why not Linux Kernel + BSD Userspace?
One could build a very stripped down distribution based on the Linux kernel and a BSD package manager such as pkgscr and retrieve all sources from a central repository. Then for Slackware mantainers all they had to do would be to optimize/repackage the resulting binaries for Slackware. It would probably take less time and would bring more rewards that constantly catching up with all that garbage from Fedora/Ubuntu.
 
Old 07-13-2012, 02:25 PM   #62
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Quote:
Originally Posted by volkerdi View Post
Not even close.

"This is the Unix philosophy: Write programs that do one thing and do it well. Write programs to work together. Write programs to handle text streams, because that is a universal interface." -- Doug McIlroy, the inventor of Unix pipes and one of the founders of the Unix tradition
Pipe Creator:

Doug McIlroy, the inventor of Unix pipes and one of the founders of the Unix tradition, had this to say at the time [McIlroy78]:

(i) Make each program do one thing well. To do a new job, build afresh rather than complicate old programs by adding new features. (isnt that what he did by getting rid of sysV system by centralizing the startup task?)

(ii) Expect the output of every program to become the input to another, as yet unknown, program. Don't clutter output with extraneous information. Avoid stringently columnar or binary input formats. Don't insist on interactive input. (doesnt the init sys create more clutter?)

(iii) Design and build software, even operating systems, to be tried early, ideally within weeks. Don't hesitate to throw away the clumsy parts and rebuild them. (isnt he throwing away some clumsy parts?)

(iv) Use tools in preference to unskilled help to lighten a programming task, even if you have to detour to build the tools and expect to throw some of them out after you've finished using them. (exactly what he is doing?)
 
Old 07-13-2012, 02:51 PM   #63
volkerdi
Slackware Maintainer
 
Registered: Dec 2002
Location: Minnesota
Distribution: Slackware! :-)
Posts: 877

Rep: Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827Reputation: 1827
Quote:
Originally Posted by Mercury305 View Post
Pipe Creator:

Doug McIlroy, the inventor of Unix pipes and one of the founders of the Unix tradition, had this to say at the time [McIlroy78]:

(i) Make each program do one thing well. To do a new job, build afresh rather than complicate old programs by adding new features. (isnt that what he did by getting rid of sysV system by centralizing the startup task?)
This isn't a new job, and furthermore appears to be trying to combine several different programs. This is the exact opposite of (i).

Quote:
(ii) Expect the output of every program to become the input to another, as yet unknown, program. Don't clutter output with extraneous information. Avoid stringently columnar or binary input formats. Don't insist on interactive input. (doesnt the init sys create more clutter?)
Clearly you haven't actually looked at this. McIlroy is talking about making programs pipe-friendly, not D-Bus friendly. Again, the exact opposite.

Quote:
(iii) Design and build software, even operating systems, to be tried early, ideally within weeks. Don't hesitate to throw away the clumsy parts and rebuild them. (isnt he throwing away some clumsy parts?)
No, he is pushing a clumsy replacement.

Quote:
(iv) Use tools in preference to unskilled help to lighten a programming task, even if you have to detour to build the tools and expect to throw some of them out after you've finished using them. (exactly what he is doing?)
I fail to see how this one is relevant to the discussion at all, unless there's some magical tool to convert init scripts to systemd units that I'm not aware of. If there isn't, then it's a fail on point (iv) as well.
 
11 members found this post helpful.
Old 07-13-2012, 03:22 PM   #64
Martinus2u
Member
 
Registered: Apr 2010
Distribution: Slackware
Posts: 353

Rep: Reputation: 56
Rarely have I read such collection of conspiracy theories and, dare i say it, fud.

1.

The unix nature of an OS does not depend on the nature of system startup. The fundamental nature of unix architecture is the monolithic block:
* kernel address space is part of every process address space
* context switch only swaps user space and the "current" pointer
* kernel services are provided without context switch
* kernel ABI follows "everything is a file" - well, almost everything

There is also a fundamental principle of the unix user space, as quoted by Pat :
* do one thing well
* integrate
* use text streams

You can argue that systemd breaks the user space principle of doing just one thing, although others would argue it does the job better than the current bunch of applications (hal, udev, dbus, hotplug and all the other shit).

You can argue about its quality, usefulness, simplicity etc. But it certainly does not change the unix nature of the OS, and certainly does not turn anything into windows.

Come to think of it, there is one more fundamental principle of user space, maybe not so much of traditional unix but definitely of the open source world: CHOICE. And yes, this principle seems to be negatively impacted by current developments. This is not so much a technical problem of systemd but a social issue surrounding Redhat.

2.

Why do people keep arguing "windows makes things simple for the user, so everything that makes things simple to the user must be like windows". This is wrong on so many levels. for starters, the premise is wrong. windows (including the additional software required to keep it alive) is a convoluted complicated system that is hard to maintain in a healthy state and requires lots of resources. windows can be beaten on technical grounds, and i wish Canonical put their minds to that goal rather than confusing their user base with every new release.

And the logical conclusion of the statement above is so dumb that any refutation is a waste of time.

My opinion: making things easy for the user is a Good Thing TM. Even if windows did that (which they don't), i wouldn't hold this aspect against them. What makes windows evil is something totally different, namely the business ethics and market dominance of the supplier, and the implicit war against free software, free works of art and free information.

3.

don't like the attitude of a developer? well, that's your right. That's why we need choice.

4.

not every user space program can be easily compiled for a new kernel or user space environment. don't make it sound like it's a piece of cake. Even if there are already code segments compiled conditionally (#ifdef __linux__ etc.)

5.

just because something was right 30 years ago it doesn't mean it is the best solution for today's problems. it may be. it may be not. just leave the dogma out of it. there is no eternal truth. everything flows.


You step in the stream,
but the water has moved on.
Init is not here.

(after a haiku by Cass Whittington)
 
1 members found this post helpful.
Old 07-13-2012, 03:25 PM   #65
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Quote:
Originally Posted by volkerdi View Post
This isn't a new job, and furthermore appears to be trying to combine several different programs. This is the exact opposite of (i).



Clearly you haven't actually looked at this. McIlroy is talking about making programs pipe-friendly, not D-Bus friendly. Again, the exact opposite.



No, he is pushing a clumsy replacement.



I fail to see how this one is relevant to the discussion at all, unless there's some magical tool to convert init scripts to systemd units that I'm not aware of. If there isn't, then it's a fail on point (iv) as well.
Upon hearing this the novice linux user was enlightened.
 
1 members found this post helpful.
Old 07-13-2012, 03:37 PM   #66
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Quote:
Originally Posted by Martinus2u View Post
Rarely have I read such collection of conspiracy theories and, dare i say it, fud.

1.

The unix nature of an OS does not depend on the nature of system startup. The fundamental nature of unix architecture is the monolithic block:
* kernel address space is part of every process address space
* context switch only swaps user space and the "current" pointer
* kernel services are provided without context switch
* kernel ABI follows "everything is a file" - well, almost everything

There is also a fundamental principle of the unix user space, as quoted by Pat :
* do one thing well
* integrate
* use text streams

You can argue that systemd breaks the user space principle of doing just one thing, although others would argue it does the job better than the current bunch of applications (hal, udev, dbus, hotplug and all the other shit).

You can argue about its quality, usefulness, simplicity etc. But it certainly does not change the unix nature of the OS, and certainly does not turn anything into windows.

Come to think of it, there is one more fundamental principle of user space, maybe not so much of traditional unix but definitely of the open source world: CHOICE. And yes, this principle seems to be negatively impacted by current developments. This is not so much a technical problem of systemd but a social issue surrounding Redhat.

2.

Why do people keep arguing "windows makes things simple for the user, so everything that makes things simple to the user must be like windows". This is wrong on so many levels. for starters, the premise is wrong. windows (including the additional software required to keep it alive) is a convoluted complicated system that is hard to maintain in a healthy state and requires lots of resources. windows can be beaten on technical grounds, and i wish Canonical put their minds to that goal rather than confusing their user base with every new release.

And the logical conclusion of the statement above is so dumb that any refutation is a waste of time.

My opinion: making things easy for the user is a Good Thing TM. Even if windows did that (which they don't), i wouldn't hold this aspect against them. What makes windows evil is something totally different, namely the business ethics and market dominance of the supplier, and the implicit war against free software, free works of art and free information.

3.

don't like the attitude of a developer? well, that's your right. That's why we need choice.

4.

not every user space program can be easily compiled for a new kernel or user space environment. don't make it sound like it's a piece of cake. Even if there are already code segments compiled conditionally (#ifdef __linux__ etc.)

5.

just because something was right 30 years ago it doesn't mean it is the best solution for today's problems. it may be. it may be not. just leave the dogma out of it. there is no eternal truth. everything flows.


You step in the stream,
but the water has moved on.
Init is not here.

(after a haiku by Cass Whittington)
Upon hearing this the Novice linux user grew more confused then ever... Who should he trust? Perhaps both? Perhaps Slackware should try both with and without systemd versions? Similar to a 32bit and 64bit isos?

Information = Clarity.

The more this Topic is discussed the more Light shall be drawn upon the subject matter and shall reveal all evils and goods of the new "systemd" architecture.

Last edited by Mercury305; 07-13-2012 at 03:39 PM.
 
Old 07-13-2012, 03:57 PM   #67
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Quote:
Originally Posted by Martinus2u View Post
Rarely have I read such collection of conspiracy theories and, dare i say it, fud.

1.

The unix nature of an OS does not depend on the nature of system startup. The fundamental nature of unix architecture is the monolithic block:
* kernel address space is part of every process address space
* context switch only swaps user space and the "current" pointer
* kernel services are provided without context switch
* kernel ABI follows "everything is a file" - well, almost everything

There is also a fundamental principle of the unix user space, as quoted by Pat :
* do one thing well
* integrate
* use text streams

You can argue that systemd breaks the user space principle of doing just one thing, although others would argue it does the job better than the current bunch of applications (hal, udev, dbus, hotplug and all the other shit).

You can argue about its quality, usefulness, simplicity etc. But it certainly does not change the unix nature of the OS, and certainly does not turn anything into windows.

Come to think of it, there is one more fundamental principle of user space, maybe not so much of traditional unix but definitely of the open source world: CHOICE. And yes, this principle seems to be negatively impacted by current developments. This is not so much a technical problem of systemd but a social issue surrounding Redhat.

2.

Why do people keep arguing "windows makes things simple for the user, so everything that makes things simple to the user must be like windows". This is wrong on so many levels. for starters, the premise is wrong. windows (including the additional software required to keep it alive) is a convoluted complicated system that is hard to maintain in a healthy state and requires lots of resources. windows can be beaten on technical grounds, and i wish Canonical put their minds to that goal rather than confusing their user base with every new release.

And the logical conclusion of the statement above is so dumb that any refutation is a waste of time.

My opinion: making things easy for the user is a Good Thing TM. Even if windows did that (which they don't), i wouldn't hold this aspect against them. What makes windows evil is something totally different, namely the business ethics and market dominance of the supplier, and the implicit war against free software, free works of art and free information.

3.

don't like the attitude of a developer? well, that's your right. That's why we need choice.

4.

not every user space program can be easily compiled for a new kernel or user space environment. don't make it sound like it's a piece of cake. Even if there are already code segments compiled conditionally (#ifdef __linux__ etc.)

5.

just because something was right 30 years ago it doesn't mean it is the best solution for today's problems. it may be. it may be not. just leave the dogma out of it. there is no eternal truth. everything flows.


You step in the stream,
but the water has moved on.
Init is not here.

(after a haiku by Cass Whittington)
Upon meditating on what hath been said. The novice linux user started to favor Patriarch Martinus2u's opinion on systemd.

I believe that it is CHOICE that will give us the answer. Why not have both? Just as I have both centOS and Slackware on my other laptop? Why say one is worse then the other? Why not try both out until we learn to Love it or Hate it or not even care about it?

Unix Philosophy can be interpreted in many ways... it depends on what angle we look at it.
 
Old 07-13-2012, 04:08 PM   #68
astrogeek
Senior Member
 
Registered: Oct 2008
Distribution: Slackware [64]X{.0|.1|.2|-current} ::X>=12<=14, FreeBSD_10{.0|.1}
Posts: 2,193

Rep: Reputation: 862Reputation: 862Reputation: 862Reputation: 862Reputation: 862Reputation: 862Reputation: 862
Too much Orthoptera > Caelifera... not enough Dennis Ritchie.

Last edited by astrogeek; 07-13-2012 at 04:14 PM.
 
1 members found this post helpful.
Old 07-13-2012, 04:12 PM   #69
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Quote:
Originally Posted by astrogeek View Post
Too much orthopetra... not enough Dennis Ritchie.
Hey, we still got Ken Thompson alive and he is writing a new language called Go.

He states that C has some obsolete edges that need to be fixed using garbage collection etc.

We should call him up and ask him about systemd.
 
Old 07-13-2012, 04:26 PM   #70
brianL
LQ 5k Club
 
Registered: Jan 2006
Location: Oldham, Lancs, England
Distribution: Slackware & Slackware64 14.1
Posts: 7,140
Blog Entries: 52

Rep: Reputation: Disabled
Yaaaaaaawn.
 
Old 07-13-2012, 04:31 PM   #71
ReaperX7
Senior Member
 
Registered: Jul 2011
Location: California
Distribution: LFS-7.6, Slackware 14.1, FreeBSD 10.1
Posts: 3,850
Blog Entries: 15

Rep: Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191
No need, I have a statement about SystemD we can probably agree on...

"SystemD... the INIT system that currently breaks your Linux distribution."

Patrick, if we need to go with BSD Userspace, then just do it. If the rest of the Linux world wants to damn itself to Lennart's personal hell, then let them go and do it. We already use BSD style INIT scripts for Slackware, so why not push Slackware deeper into being a more traditional UNIX system with a Linux kernel and use more BSD style tools.
 
3 members found this post helpful.
Old 07-13-2012, 04:35 PM   #72
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Quote:
Originally Posted by ReaperX7 View Post
No need, I have a statement about SystemD we can probably agree on...

"SystemD... the INIT system that currently breaks your Linux distribution."

Patrick, if we need to go with BSD Userspace, then just do it. If the rest of the Linux world wants to damn itself to Lennart's personal hell, then let them go and do it. We already use BSD style INIT scripts for Slackware, so why not push Slackware deeper into being a more traditional UNIX system with a Linux kernel and use more BSD style tools.
Reaper,

Check this out!

http://www.linuxtoday.com/developer/1999050701605NWSM
 
Old 07-13-2012, 04:45 PM   #73
ReaperX7
Senior Member
 
Registered: Jul 2011
Location: California
Distribution: LFS-7.6, Slackware 14.1, FreeBSD 10.1
Posts: 3,850
Blog Entries: 15

Rep: Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191Reputation: 1191
Thanks Merc.

The Anything-but-Microsoft trend has stemmed from the proposed release of Windows 8 which developers and people already hate with a passion. However, while I do agree Linux could position itself to be a marketable viable alternative to Windows, most of the major distributions are going about it haphazardly by trying to implement tools and services that not only mimic Windows, but recreate it. This is bad for Linux because it pushes Linux away from what it is, a kernel for a GNU based operating system, and tries to turn Linux into an actual OS.
 
1 members found this post helpful.
Old 07-13-2012, 04:49 PM   #74
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Quote:
Originally Posted by ReaperX7 View Post
Thanks Merc.

The Anything-but-Microsoft trend has stemmed from the proposed release of Windows 8 which developers and people already hate with a passion. However, while I do agree Linux could position itself to be a marketable viable alternative to Windows, most of the major distributions are going about it haphazardly by trying to implement tools and services that not only mimic Windows, but recreate it. This is bad for Linux because it pushes Linux away from what it is, a kernel for a GNU based operating system, and tries to turn Linux into an actual OS.
Linux userland is like another Microsoft. Its mutating the UNIX Code... Ubuntu is a perfect example in my opinion no matter how many people disasgree with me. It is just tat a non Microsoft Windows.

Redhat is also kind of like a non Microsoft Windows NT...

Meanwhile...

UNIX is still UNIX

Last edited by Mercury305; 07-13-2012 at 04:57 PM.
 
Old 07-13-2012, 05:09 PM   #75
Mercury305
Member
 
Registered: Jul 2012
Location: Rockville, MD
Distribution: CrunchBang / Ubuntu
Posts: 540

Rep: Reputation: Disabled
Ken Thompson in his own words

"Reflections on Trusting Trust", 1983 Turing Award Lecture[2], Communications of the ACM 27 (8), August 1984, pp. 761-763.
I view Linux as something that's not Microsoft—a backlash against Microsoft, no more and no less. I don't think it will be very successful in the long run. I've looked at the source and there are pieces that are good and pieces that are not. A whole bunch of random people have contributed to this source, and the quality varies drastically. My experience and some of my friends' experience is that Linux is quite unreliable. Microsoft is really unreliable but Linux is worse."

A lot has changed since then. So making a philosophy about someone that is wrong may not be the right idea?

There are some principles I can agree with like Simplicity is key.

But maybe he is not right about everything. Is Linux unsuccessful on the longrun?

OK... So lets say he is right?

Then what is the common sense of going for UNIX philosophy on a Linux System?

After all the man behind the philosophy say Linux will be unsuccessful in the longrun. So what is the point?

I mean I think its time to put away these "Demigod" forms... and start to really look at things the way they really are.

Do I personally think Linux Kernel is unsuccessful? No. Do I think Linux should be just like UNIX or anti-Windows? No.

Lets stop saying things should be like this or that... because he or she said so... Instead lets look at the truth behind the rock.

Lets look at what works. What works in a simple and efficient and concise way. Something that makes our life easier instead of having to read endless textbooks worth of garbage just to say I know Linux.

Or lets not just patch up everything and go for something that really works... Heck maybe even build a new system from scratch?

thats my 2 cents.

Last edited by Mercury305; 07-13-2012 at 05:12 PM.
 
  


Closed Thread


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] slackware and systemd fl0 Slackware 512 08-29-2012 12:07 PM
Boot Delay 30min: systemd-analyze blame systemd-tmpfiles-setup.service BGHolmes Fedora 0 07-27-2011 10:02 AM


All times are GMT -5. The time now is 11:03 PM.

Main Menu
Advertisement
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