Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
The Politics of systemd
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, 4, 5  Next  
Reply to topic    Gentoo Forums Forum Index Gentoo Chat
View previous topic :: View next topic  
Author Message
djdunn
l33t
l33t


Joined: 26 Dec 2004
Posts: 709
Location: Arrakis

PostPosted: Fri Apr 04, 2014 7:24 pm    Post subject: Reply with quote

http://lkml.iu.edu//hypermail/linux/kernel/1404.0/01331.html

Linus Torvalds, was so upset over the debug kernel line bug in systemd that he said he would not be accepting any more patches from systemd developer Kay Sievers specifically concerning kdbus merging into the kernel. even though kay didn't submit the patch and systemd was not mentioned linus picked up on this quick. Some execs at Redhat need to think long and hard about lennart and kay, and how their constant behavior reflects on red hat.

>
> The response is:
>
> "Generic terms are generic, not the first user owns them."

And by "their" you mean Kay Sievers.

Key, I'm f*cking tired of the fact that you don't fix problems in the
code *you* write, so that the kernel then has to work around the
problems you cause.

Greg - just for your information, I will *not* be merging any code
from Kay into the kernel until this constant pattern is fixed.

This has been going on for *years*, and doesn't seem to be getting any
better. This is relevant to you because I have seen you talk about the
kdbus patches, and this is a heads-up that you need to keep them
separate from other work. Let distributions merge it as they need to
and maybe we can merge it once it has been proven to be stable by
whatever distro that was willing to play games with the developers.

But I'm not willing to merge something where the maintainer is known
to not care about bugs and regressions and then forces people in other
projects to fix their project. Because I am *not* willing to take
patches from people who don't clean up after their problems, and don't
admit that it's their problem to fix.

Kay - one more time: you caused the problem, you need to fix it. None
of this "I can do whatever I want, others have to clean up after me"
crap.

Linus
_________________
A process cannot be understood by stopping it. Understanding must move with the flow of the process, must join it and flow with it.

-The First Law of Mentat
Back to top
View user's profile Send private message
FastTurtle
Guru
Guru


Joined: 03 Sep 2002
Posts: 409
Location: Shake & Bake

PostPosted: Fri Apr 04, 2014 7:32 pm    Post subject: Reply with quote

Something interesting on Slashdot today about SystemD
Edit: Added the Link
http://linux.slashdot.org/story/14/04/04/1523231/linus-torvalds-suspends-key-linux-developer

Linus Torvalds has told the primary SystemD dev that until he fixes some bugs, no further patches will be accepted into the kernel and from the quotes posted, he was actually pretty polite about the whole problem.

In my investigations, I've seen no benefit to us as end users for SystemD - in fact, the only ones it benefits are Redhat and Suse because they've gone the same route as Microsoft with the inpenetrable boot system that only they can fix and the problem is, that as with Microsoft, they can't even do that 100 percent of the time and with it being inpenatrable and mysterious (you pixies get back to work or I'll assign you to correcting 2nd grade engrish papers).

With Debian making the change and looking into my cracked crystal ball, all I can see is that in the future, I wont be using Linux if SystemD manages to take over the entire user-space as they're aiming for. This means either switching to BSD (Berkley Software Distribution - University of California at Berkely) or accepting what ever Microsoft wants to shove down my throat as I am not a programmer.


Last edited by FastTurtle on Sat Apr 05, 2014 2:49 pm; edited 1 time in total
Back to top
View user's profile Send private message
mv
Advocate
Advocate


Joined: 20 Apr 2005
Posts: 4316

PostPosted: Fri Apr 04, 2014 8:01 pm    Post subject: Reply with quote

NeddySeagoon wrote:
For recent kernels, PARTUUID works on MSDOS partitions too.

NeddySeagoon, your post is the best news to me since sliced bread. :D
Just tested, and it seems to work like a charm. 8)

However, obviously the magic string is not stored in the partition itself: It seems that some magic is used to label the disk, and then "just" the partition number is appended. The latter is fine, but what worries me is how reliable is the magic to label the disk: Probably you remember that some people complained that the new "permanent" interface labeling of udev is not as permanent as it appears at a first glance (e.g. some machines relabel after certain hardware changes). Could msdos' PARTUUID suffer from the same problem, or do they user a more clever method? In the former case, one might be better off to stay with /dev/sdxy partition names in the grub.cfg ...
Back to top
View user's profile Send private message
The Doctor
Veteran
Veteran


Joined: 27 Jul 2010
Posts: 1523

PostPosted: Fri Apr 04, 2014 8:34 pm    Post subject: Reply with quote

Quote:

http://lkml.iu.edu//hypermail/linux/kernel/1404.0/01331.html

Linus Torvalds, was so upset over the debug kernel line bug in systemd that he said he would not be accepting any more patches from systemd developer Kay Sievers ...


Now that is an interesting flame war. As I have said before it's the politics, not necessarily the code, of systemd I object to. Hopefully, other devs will follow Linus in demanding the systemd guys get their house in order.

I just wish I had the time and the experience to put some code where my mouth is.
_________________
First things first, but not necessarily in that order.
Back to top
View user's profile Send private message
broken_chaos
Guru
Guru


Joined: 18 Jan 2006
Posts: 370
Location: Ontario, Canada

PostPosted: Fri Apr 04, 2014 10:04 pm    Post subject: Reply with quote

mv wrote:
It seems that some magic is used to label the disk, and then "just" the partition number is appended.

It's not magic, it's the MBR identifier. It's stored in the MBR and should never change unless you redo the partition table from scratch -- typically anything that would affect that value in a negative way could also trash the partition table itself.
Back to top
View user's profile Send private message
djdunn
l33t
l33t


Joined: 26 Dec 2004
Posts: 709
Location: Arrakis

PostPosted: Sat Apr 05, 2014 5:10 am    Post subject: Reply with quote

The Doctor wrote:
Quote:

http://lkml.iu.edu//hypermail/linux/kernel/1404.0/01331.html

Linus Torvalds, was so upset over the debug kernel line bug in systemd that he said he would not be accepting any more patches from systemd developer Kay Sievers ...


Now that is an interesting flame war. As I have said before it's the politics, not necessarily the code, of systemd I object to. Hopefully, other devs will follow Linus in demanding the systemd guys get their house in order.

I just wish I had the time and the experience to put some code where my mouth is.


I object to the buggy poorly designed code, one example being this bug where debug in the kernel command line causing systemd to fail to boot, which was caused by systemd abusing the kernel ring buffer, yes it did in fact expose a weakness in the kernel, but systemd was dumping logs where it never should have been to begin with. Not only that but the immature and unprofessional conduct of the systemd developers doing like always, declaring their lazy and buggy code perfect and demanding that other projects own and fix these problems that they (systemd developers) caused via their lazyness and buggy code.
_________________
A process cannot be understood by stopping it. Understanding must move with the flow of the process, must join it and flow with it.

-The First Law of Mentat
Back to top
View user's profile Send private message
FastTurtle
Guru
Guru


Joined: 03 Sep 2002
Posts: 409
Location: Shake & Bake

PostPosted: Sat Apr 05, 2014 2:57 pm    Post subject: Reply with quote

broken_chaos wrote:
mv wrote:
It seems that some magic is used to label the disk, and then "just" the partition number is appended.

It's not magic, it's the MBR identifier. It's stored in the MBR and should never change unless you redo the partition table from scratch -- typically anything that would affect that value in a negative way could also trash the partition table itself.


AFIRC: If using a pure GPT partitioning, there's no MBR to be read. Not absolutely sure as I haven't used Gparted very often as I haven't needed it. No drives large enough to really take advantage of the features though I am getting a 3TB Adv. Format drive to play with.
Back to top
View user's profile Send private message
djdunn
l33t
l33t


Joined: 26 Dec 2004
Posts: 709
Location: Arrakis

PostPosted: Sat Apr 05, 2014 4:12 pm    Post subject: Reply with quote

FastTurtle wrote:
broken_chaos wrote:
mv wrote:
It seems that some magic is used to label the disk, and then "just" the partition number is appended.

It's not magic, it's the MBR identifier. It's stored in the MBR and should never change unless you redo the partition table from scratch -- typically anything that would affect that value in a negative way could also trash the partition table itself.


AFIRC: If using a pure GPT partitioning, there's no MBR to be read. Not absolutely sure as I haven't used Gparted very often as I haven't needed it. No drives large enough to really take advantage of the features though I am getting a 3TB Adv. Format drive to play with.


any sane disk management tool creates gpt disks with LBA0 as a protective MBR for compatability reasons, and to keep non gpt aware tools from overwriting LBA1 the gpt partition table header.
_________________
A process cannot be understood by stopping it. Understanding must move with the flow of the process, must join it and flow with it.

-The First Law of Mentat
Back to top
View user's profile Send private message
depontius
Advocate
Advocate


Joined: 05 May 2004
Posts: 2652

PostPosted: Sat Apr 05, 2014 4:43 pm    Post subject: Reply with quote

djdunn wrote:
FastTurtle wrote:
broken_chaos wrote:
mv wrote:
It seems that some magic is used to label the disk, and then "just" the partition number is appended.

It's not magic, it's the MBR identifier. It's stored in the MBR and should never change unless you redo the partition table from scratch -- typically anything that would affect that value in a negative way could also trash the partition table itself.


AFIRC: If using a pure GPT partitioning, there's no MBR to be read. Not absolutely sure as I haven't used Gparted very often as I haven't needed it. No drives large enough to really take advantage of the features though I am getting a 3TB Adv. Format drive to play with.


any sane disk management tool creates gpt disks with LBA0 as a protective MBR for compatability reasons, and to keep non gpt aware tools from overwriting LBA1 the gpt partition table header.


Is Gparted a sane disk management tool? I have a few disks that I created as GPT using Gparted, so I'm curious. Would I check by doing "fdisk -l /dev/sdx"? (I presume fdisk only works with BIOS format??)
_________________
.sigs waste space and bandwidth
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 32338
Location: 56N 3W

PostPosted: Sat Apr 05, 2014 6:40 pm    Post subject: Reply with quote

depontius,

Brain dead BIOSes check for a bootable flag in the MSDOS partition table at the end of LBA 0. Even on GPT disks.
This means that you need a valid MSDOS partition table' with one partition or more and the signature word in the last bytes of LBA 0.
This is a BIOS thing, not a GPT thing.

fdisk has never changed LBA1. Traditionally, boot loaders us use LBA1..LBA62, or at least a part of it. However The primary GPT copy starts at LBA1, so this is a bad idea.
legacy grub and grub2 are GPT aware.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
miket
Apprentice
Apprentice


Joined: 28 Apr 2007
Posts: 228
Location: Gainesville, FL, USA

PostPosted: Fri Apr 11, 2014 4:24 am    Post subject: Reply with quote

Shamus397 wrote:
Linking my post in this thread, as it is relevant to the political dimension. Also, link to ridiculous bug by ulenrich, one of the people responsible for the label-them-then-attack-them-with-that-label style of "debate". Seems the rhetoric of those pushing SystemD has become more and more shrill these days, at least around these parts. :)

Gracious. A bug in the bugzilla complaining about discussions in the forums and written in the tone of "mommy, he hit me". Ah, the voice of inexperience.

Just in the last two days we have a huge example of what can happen when programmers in charge of critical software make a dumb mistake. The OpenSSL project has been around for more than 15 years, yet its maintainers managed to put a huge security hole in it. Here are some of the comments I read today on Slashdot.
Quote:
I've been through the bowels of OpenSSL, and there are parts of it that frighten me.

Quote:
The issue is that the code allocates a buffer of a size specified by the user, without validating it, and doesn't zero the allocated memory.

Quote:
The real issue here is that the development process did not detect the mistake and correct it in a timely manner. Code that is as security-critical as OpenSSL should really be code-reviewed and tested out the wahzoo before it is released to the public, so either that didn't happen, or it did happen and the process didn't detect this fault; either way a process-failure analysis and process improvements are called for.

And all this is for software that has been around for ages and considered solid, mature and reliable. The bug went undetected for two years and now affects millions of people.

Systemd may have a really nice feature set, but it is an order of magnitude more complicated than OpenSSH and was put together in much more of a hurry. This really ought to raise a lot of red flags for anyone who has been around software development. This whole OpenSSL crisis has pointed out the need for diversity: it has certainly raised people's interest in GnuTLS, NSS, and others. Will those who want to complain to some higher authority about systemd in the forums draw some lesson from this episode?
Back to top
View user's profile Send private message
TomWij
Developer
Developer


Joined: 04 Jul 2012
Posts: 1551

PostPosted: Fri Apr 11, 2014 10:18 am    Post subject: Reply with quote

miket wrote:
And all this is for software that has been around for ages and considered solid, mature and reliable. The bug went undetected for two years and now affects millions of people.


These pages are interesting:

OpenSSL is not developed by a responsible team.
heartbleed vs malloc.conf
analysis of openssl freelist reuse (links to earlier reports here and here, edit: login with username "guest" and password "guest")

They explain a large part of why it has been undetected as well as not so solid, mature and reliable. Long story short; if they weren't using an improper malloc/free wrapper like that, they could have had this crash at an earlier security audit or when someone actually tried to abuse it. This, alongside your quotes, is what makes OpenSSL's security far worse than most things one would compare it against; because of that, security comparisons with other projects are unfair. Indeed, there are multiple lessons to be learned from this entire event; in various areas, more than just detecting the security issue.


Last edited by TomWij on Fri Apr 11, 2014 4:27 pm; edited 1 time in total
Back to top
View user's profile Send private message
steveL
Advocate
Advocate


Joined: 13 Sep 2006
Posts: 2960
Location: The Peanut Gallery

PostPosted: Fri Apr 11, 2014 12:07 pm    Post subject: Reply with quote

miket wrote:
Quote:
The real issue here is that the development process did not detect the mistake and correct it in a timely manner. Code that is as security-critical as OpenSSL should really be code-reviewed and tested out the wahzoo before it is released to the public, so either that didn't happen, or it did happen and the process didn't detect this fault; either way a process-failure analysis and process improvements are called for.

And all this is for software that has been around for ages and considered solid, mature and reliable. The bug went undetected for two years and now affects millions of people.

Systemd may have a really nice feature set, but it is an order of magnitude more complicated than OpenSSH and was put together in much more of a hurry. This really ought to raise a lot of red flags for anyone who has been around software development.

Indeed, and it does. That's why more experienced members of the community have not been so enthusiastic about it. The point about development process is very relevant here: you don't put people who can't yet even see the need for disciplined development in charge of core system software. They're still wet behind the ears.

BTW I have it on very good authority that openssh is not affected by heartbleed; see /topic #openssh on freenode.
Quote:
This whole OpenSSL crisis has pointed out the need for diversity: it has certainly raised people's interest in GnuTLS, NSS, and others. Will those who want to complain to some higher authority about systemd in the forums draw some lesson from this episode?

Evidently not: they'll avoid any such question and sidetrack into whatever they can; in this case the bug itself, and not the lessons around it that you wanted to discuss.
Back to top
View user's profile Send private message
miket
Apprentice
Apprentice


Joined: 28 Apr 2007
Posts: 228
Location: Gainesville, FL, USA

PostPosted: Fri Apr 11, 2014 1:55 pm    Post subject: Reply with quote

TomWij wrote:
Indeed, there are multiple lessons to be learned from this entire event; in various areas, more than just detecting the security issue.

Yes there are. That's precisely why I raise the point. I have been a professional programmer for many years. Yes, I've read a lot about the dangers of complexity, but more than that, I know through a lot of bitter experience that as complexity increases, the bug rate goes up--as well as the difficulty in diagnosing bugs--as well as the difficulty in fixing them so that the fixes don't break other things. Too often have I had the nasty task of cleaning up after some hotshot young programmer who worked quickly to make something over-complicated and buggy.

You do a lot of work in Gentoo kernel maintenance. I salute you for that as well as for your other work in Gentoo. Quite contrary to what I sometimes find among young programmers, you exhibit a great degree of care in your work. I'd be pleased to have you as a coworker.

I don't have anything like that level of confidence in the maintainers of systemd. They are of just the kind I have learned to associate with disaster: very ambitious, fast working, fast churning, slow to accept advice or correction, very prideful, and quick to force others to adapt to their new ways. It's their way or the highway. Believe me: this leads to years of headaches for those who have to live with the aftermath.

The bug in OpenSSL was astounding in its boneheadedness. Well before anyone should have been able to get to the stage of detecting the error with a tool like Valgrind, Robin Seggelmann should have hit himself over the head when thinking that he could depend on the contents of a buffer returned by malloc() to contain just the data he was looking for. He had been working on OpenSSL for more than two years before making that fateful commit 4817504d069b4c5082161b02a22116ad75f822b1. I've also seen the "fix". Guess what: they introduced only a bounds check; they still depend on that LIFO behavior in their handmade memory allocator.

The sheer size and scope of systemd, the haste in which it was made, the "we can do no wrong" attitude of its programmers, and the sharp rebuke that Linus dealt to one of its principals all make me wary of hidden catastropic bugs in it that have the flavor of the Heartbleed bug.
Back to top
View user's profile Send private message
TomWij
Developer
Developer


Joined: 04 Jul 2012
Posts: 1551

PostPosted: Fri Apr 11, 2014 2:50 pm    Post subject: Reply with quote

miket wrote:
TomWij wrote:
Indeed, there are multiple lessons to be learned from this entire event; in various areas, more than just detecting the security issue.

Yes there are. That's precisely why I raise the point. I have been a professional programmer for many years. Yes, I've read a lot about the dangers of complexity, but more than that, I know through a lot of bitter experience that as complexity increases, the bug rate goes up--as well as the difficulty in diagnosing bugs--as well as the difficulty in fixing them so that the fixes don't break other things. Too often have I had the nasty task of cleaning up after some hotshot young programmer who worked quickly to make something over-complicated and buggy.


Yes, they want to get the job done with the least amount of discussion, work, meetings; up to the point that they only look at the result and not at the many *ity's (eg. reliability) that lead to it, which then leads to having to iron out various bugs as a result of that towards releasing the result or being surprised that there are these kinds of bugs after release.

miket wrote:
You do a lot of work in Gentoo kernel maintenance. I salute you for that as well as for your other work in Gentoo. Quite contrary to what I sometimes find among young programmers, you exhibit a great degree of care in your work. I'd be pleased to have you as a coworker.


Yes, the care put into that is based on being scared to lose the ability to contribute due to one or another stupid mistake; some small mistakes can be forgiven, but not much beyond that. It is again not only the result that matters here, but also the contribution; as good contribution contributes towards good results, and it only gets to be a happy and rewarding experience if you pursue those two.

miket wrote:
I don't have anything like that level of confidence in the maintainers of systemd. They are of just the kind I have learned to associate with disaster: very ambitious, fast working, fast churning, slow to accept advice or correction, very prideful, and quick to force others to adapt to their new ways. It's their way or the highway. Believe me: this leads to years of headaches for those who have to live with the aftermath.


Me neither; while I use systemd myself as it works out for me, I wouldn't suggest this to others or use this for servers that need to be quite stable and/or secure. The way things are done are often concerning; take for example the recent "debug" bug, the first responses "that is the expected current behaviour" and further statements that "debug" is for more than just the kernel violently ignore that "debug [KNL] Enable kernel debugging (events log level)." clearly is for "kernel debugging" where "KNL is a kernel start-up parameter". At which point "move discussions to the mailing list" should be met with "keep bugs on bugzilla".

miket wrote:
The bug in OpenSSL was astounding in its boneheadedness. Well before anyone should have been able to get to the stage of detecting the error with a tool like Valgrind, Robin Seggelmann should have hit himself over the head when thinking that he could depend on the contents of a buffer returned by malloc() to contain just the data he was looking for. He had been working on OpenSSL for more than two years before making that fateful commit 4817504d069b4c5082161b02a22116ad75f822b1. I've also seen the "fix". Guess what: they introduced only a bounds check; they still depend on that LIFO behavior in their handmade memory allocator.


Yes; given that, it is definitely still possible for another part of their code to have a problem of similar size. Those bugs remain unfixed; you would've think they've learned, but I doubt this will be the last we've seen...

miket wrote:
The sheer size and scope of systemd, the haste in which it was made, the "we can do no wrong" attitude of its programmers, and the sharp rebuke that Linus dealt to one of its principals all make me wary of hidden catastropic bugs in it that have the flavor of the Heartbleed bug.


In the previous answer I wanted to point out that; that for that to be the case, there would need to be something of equal concern to the OpenSSL handmade memory allocater in systemd; without it the security bugs are way more visible when doing security audits. That allows them to fix the worst bugs; or well, at least if they do fix them. Is there a good overview of their security bugs? On their bug tracker; nope, only a general component. Perhaps a CVE search? Though, that only shows the recorded ones; it gives the impression they don't really track all security bugs in a separate category. That says something about their importance, not about whether they get fixed; it gives the impression that they don't consider security to be important...
Back to top
View user's profile Send private message
Yamakuzure
Veteran
Veteran


Joined: 21 Jun 2006
Posts: 1448
Location: Bardowick, Germany

PostPosted: Mon Apr 14, 2014 8:45 am    Post subject: Reply with quote

krinn wrote:
It depend on Wayland to answer that : as wayland faq doesn't gave me answer to
Is Wayland systemd dependent ?
Is Wayland portable ? (if previous answer is no)

So to judge if an issue is raised, you must know answer to those two questions as :
- If wayland is portable : adding need to start it with systemd prevent linux users without systemd to run kde/wayland, but also BSD users. There's an issue as those users cannot run wayland with kde then.
- If wayland is not portable but not systemd dependent : Same issue but limited to linux kde users not using systemd.
- If wayland depend on systemd : only systemd users will be able to run kde/wayland. No issue raised by him.
Note that i don't know Wayland enough to be able to tell you if kde users will suffer from a big or small lost from not being able to use kde with wayland.
You did it again. Could you please read the rest, too? It says that systemd provides socket activation for some extra features. This has nothing to do with X11+kwin or Wayland+kwin in general.

So to your questions:
  1. No. At least nobody said that.
  2. Must be, or an X11 substitution would be impossible, wouldn't it?
  3. Not needed, it is (said to be planned to be) an option
  4. It is (or at least must be)
  5. It doesn't.

_________________
systemd - The biggest fallacies
Back to top
View user's profile Send private message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 169
Location: Ur-th

PostPosted: Thu Apr 24, 2014 4:58 pm    Post subject: Reply with quote

Found this nugget on the Debian mailing list. The best way to counter the political methodology used by those pushing SystemD is to start asking the right questions, as Mr. Toppins points out. :)

On that note, it seems that even Mr. Poettering doesn't know what SystemD is supposed be (gotten from a conversation in #docker):
Quote:
<poettering> jpetazzo: well, systemd is an init system
<poettering> jpetazzo: if you run an init system, you need to grant it privs to do so...

<jpetazzo> an init system doesn't have any legit reason to create device nodes
<jpetazzo> (especially when running in a virtual environment where expectations are well-defined in advance)

<poettering> jpetazzo: well, i disagree, a good service manager provides options like PrivateDevices= and PrivateNetwork= and PrivateTmp= to sandbox system services

So, according to Mr. Poettering, it's an init system *and* a service manager; I wonder what else it's supposed to be? :P
Back to top
View user's profile Send private message
TomWij
Developer
Developer


Joined: 04 Jul 2012
Posts: 1551

PostPosted: Thu Apr 24, 2014 7:26 pm    Post subject: Reply with quote

Shamus397 wrote:
Found this nugget on the Debian mailing list. The best way to counter the political methodology used by those pushing SystemD is to start asking the right questions, as Mr. Toppins points out. :)


It is much about nothing, quoting talk irrelevant to his central point; an ineffective demonstration. It claims "systemd and Linux are *fundamentally incompatible* -> and I can prove it", but has no proof or foundations. Instead the lack of inspection is demonstrated multiple times, as replies show. He expects simple answers, but for the simple "basic building blocks" answer he claims it misses detail; contradiction.

Such attempts to eulogize or tear apart products are tiresome; he has done it 2 years ago, does it again now and if he continues this way his next attempt will be the same. It serves as unintended free advertisement, because this form of bad advertisement acts as good; people will notice the central point misses a proof, thus being compatible which is further backed by Debian being able to run it.

There are better ways to achieve what he wants.

Shamus397 wrote:
On that note, it seems that even Mr. Poettering doesn't know what SystemD is supposed be


It takes a few iterations to get an idea of that, and this is a first iteration near the end of the waterfall model; innovation, forks, successes and/or failures are around the corner.
Back to top
View user's profile Send private message
John R. Graham
Administrator
Administrator


Joined: 08 Mar 2005
Posts: 7889
Location: Somewhere over Atlanta, Georgia

PostPosted: Thu Apr 24, 2014 7:36 pm    Post subject: Reply with quote

Eulogize? "I come to bury systemd, not to praise it." ;)

- John
_________________
I can confirm that I have received between 0 and 999 National Security Letters.
Back to top
View user's profile Send private message
TomWij
Developer
Developer


Joined: 04 Jul 2012
Posts: 1551

PostPosted: Thu Apr 24, 2014 7:45 pm    Post subject: Reply with quote

Yes, eulogize. Products get praised, in a similar way to getting buried; the combined efforts end up in a net result, the basis of that neutral statement.
Back to top
View user's profile Send private message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 169
Location: Ur-th

PostPosted: Thu Apr 24, 2014 8:02 pm    Post subject: Reply with quote

I think you missed the point Tom, he's asking very valid questions and nobody has been able to answer them. The main question being:
Quote:
What role is systemd designed to facilitate?

By apologizing for Mr. Poettering's ineptitude (saying it's a "waterfall" development model--what a load of bollocks!) shows you have no idea how to answer that question either.

And there are lots of questions that absolutely should be asked about SystemD (Mr. Toppins starts with these), why can't they be answered? They are simple questions.
Back to top
View user's profile Send private message
TomWij
Developer
Developer


Joined: 04 Jul 2012
Posts: 1551

PostPosted: Thu Apr 24, 2014 8:18 pm    Post subject: Reply with quote

His begging of the simple question "What role is systemd designed to facilitate?" is a fallacy of petitio principii that isn't right to (dis)prove the main central point "systemd and Linux are *fundamentally incompatible*".
Back to top
View user's profile Send private message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 169
Location: Ur-th

PostPosted: Thu Apr 24, 2014 9:09 pm    Post subject: Reply with quote

You're connecting things that aren't connected. What he's saying, to wit,
Quote:
There is a well known saying that runs in every engineering discipline,
including software engineering...

-> if you can't put it in writing, then you don't understand it well enough

is perfectly reasonable and the question that arises from that, "What role is systemd designed to facilitate?" is a perfectly reasonable one to ask.

To say that it's 'begging the question' is, quite frankly, a load of crap. He asked Poettering the question directly and the answer he received was telling:

Quote:
I even went to Lennart Poettering's google+ page and ....

...

I said -> I have a question for you. If you can answer it with one, and
ONLY one, concept that describes fully what systemd is I will consider I
might have misjudged this.

He replied...

-> systemd is a suite of basic building blocks to build an OS from


Okay -> right there he gives two important pieces of information...

1 -> there is nothing about how it works with linux

2 -> his answer is so vague, it should tell you he hasn't really thought
this out


systemd will wreck linux, I am certain of it.

Without some kind of design blueprint of some sort -> systemd ended up
being built by programming blindly in the dark.

There is no boundary where systemd stops and linux begins.

They will keep on absorbing pieces of linux until systemd is the entire
operating system -> and there is no coherent design to how it does / should
work.

It seems pretty clear to me that this is what's happening, that SystemD is morphing Linux into something that is Not-Linux, and so to say that "systemd and Linux are *fundamentally incompatible*" is not that far of a stretch. Poettering's answer also puts the lie to his claim that SystemD is just an init system, it's obvious, even to the meanest of intelligence, that it is not.

You can seize on things like he hasn't proven his point to your satisfaction, but you can't dodge one of the main questions being asked: "What role is systemd designed to facilitate?"

It's a simple question. It should be answerable, and *not* with vague non-answers.
Back to top
View user's profile Send private message
TomWij
Developer
Developer


Joined: 04 Jul 2012
Posts: 1551

PostPosted: Thu Apr 24, 2014 10:01 pm    Post subject: Reply with quote

Shamus397 wrote:
You're connecting things that aren't connected.


No, the main question is disconnected from the main central point "systemd and Linux are *fundamentally incompatible*".

Shamus397 wrote:
He asked Poettering the question directly and the answer he received was telling:

Quote:
[...] If you can answer it with one, and ONLY one, concept that describes fully what systemd is I will consider I might have misjudged this. [...]


A simple question does not ask for a full description; disguised as a consideration, it is designed to point out a lack of details in the answer.

Shamus397 wrote:
It seems pretty clear to me that this is what's happening, that SystemD is morphing Linux into something that is Not-Linux, and so to say that "systemd and Linux are *fundamentally incompatible*" is not that far of a stretch.


That stretch is a disconnection; products can indeed morph other products, but that two products are in a(n in)compatible state is orthogonal to that.

Shamus397 wrote:
Poettering's answer also puts the lie to his claim that SystemD is just an init system, it's obvious, even to the meanest of intelligence, that it is not.


The claim is that it is an init system, nowhere he claims that it is just an init system; that A is X still allows A to be Y too, with X and Y either different or the same.

Shamus397 wrote:
You can seize on things like he hasn't proven his point to your satisfaction, but you can't dodge one of the main questions being asked: "What role is systemd designed to facilitate?" It's a simple question. It should be answerable, and *not* with vague non-answers.


It has been answered, with a simple answer; perhaps not to his satisfaction, but that's the most he can get out of begging the question.
Back to top
View user's profile Send private message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 169
Location: Ur-th

PostPosted: Thu Apr 24, 2014 10:53 pm    Post subject: Reply with quote

Again, you're ignoring the point. He wasn't begging the question, he didn't have a preconceived answer to the question, he wasn't arguing in circles (which is, I have seen, a favorite tactic of yours).

He asked Poettering, "I have a question for you. If you can answer it with one, and ONLY one, concept that describes fully what systemd is I will consider I might have misjudged this." And Poettering replied, "systemd is a suite of basic building blocks to build an OS from."

And again, the point raised by Mr. Toppins being that the answer Poetteing gave has nothing to do with how it works with Linux and is so vague as to be meaningless.

The long and the short of it is, the main question was never answered. Again, the question is: "What role is systemd designed to facilitate?" Nobody from the SystemD camp has ever answered that question, and indeed, I seriously doubt that they can.

Furthermore, I'm thinking that you don't have a really good grasp of argumentation, as you don't seem to understand English that well:

TomWij wrote:

Shamus397 wrote:
You're connecting things that aren't connected.

No, the main question is disconnected from the main central point "systemd and Linux are *fundamentally incompatible*".

You are agreeing with me that the main point and the question isn't connected, but telling me "no", that I'm wrong!

Don't bother to parse it out, this is the one and only time I am going to point this out to you--I am not going to get bogged down in responding to your scrutiny of pointless minutæ. If you're going to ignore the main points being brought up by seizing on trivial points, then I'm going to ask you to kindly stop polluting this thread with your nonsense and get the hell out. Thanks.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Gentoo Chat All times are GMT
Goto page Previous  1, 2, 3, 4, 5  Next
Page 2 of 5

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum