One init to rule them all,
One init to find them,
One init to bring them all,
and to systemd bind them.

1 members found this post helpful.

I thought that GKH will not post incorrect information on the LKML, so I took this for true:http://lkml.iu.edu//hypermail/linux/...4.0/01945.html
Posted on 3rd of April. If that indeed is not correct than I apologize and stand corrected.

Please don't do that. And by "that" I mean start an argument with a nonsensical ad hominem attack.

Feel free to disagree with me or anyone else for any reason, and by all means join the discussion if you feel you have something to contribute. But the keyword is "contribute", which means actually presenting a coherent argument.

So now I'm going to ask that you explain yourself. Pray tell us, what exactly was wrong with that analogy? Admittedly I'm an amateur when it comes to cars, and I haven't been working on cars for as long as I've been doing computer-related stuff like assembly language programming, but I've still done my fair share of engine repairs in the last 30 years.

I don't mind being proven wrong and I'm always eager to learn, so I'm looking forward to reading your reply.

1 members found this post helpful.

In the beginning this attitude was productive, because the large base of stable working code did not exist! Lots of problems, lots to fix.

Perfect attitude for way back then. Over the years the code did evolve, but the attitude? Not so much. For the want of a nail...

Easy...to demonstrate the problem, or more importantly, the solution is with the "rock", not the "boat". Sure we could go "ironclad" on the "boat" but that creates more problems. Unless the "rock" is a commonly useful "mooring point", the simplest solution with the least side effects is to remove the "rock", or at least alter it or move it out of an area likely to cause damage.

It seemed to me no words even approaching "ludicrous" were hurled at Kay and Lennart. Since all I did was state an opinion regarding where the responsibility lies (and considering their history of deflecting, not a huge stretch) and since I didn't actually break something or try to commandeer something precious to you, yet you chose that judgment for me, I considered that by comparison, apologetic to K & L.


That's not what I said. I said WE (that includes me) are ALL judged by our choices, especially when they are made public. BTW while we're tossing dictionaries around, look up "Obfuscation". Not only does it apply to everything after "People are dying..." but it applies to coding as well, and certainly to the concept of "threading" considering this is not a thread about the ills of the entire planet, but rather just a discussion about a computer operating system and where it is headed when massive egos, possibly with moneyed agendas, are allowed on the "bridge" or in the boardroom.

1 members found this post helpful.

If I may interject this GTK+ vs Gnome3 nonsense: http://redmine.audacious-media-playe.../1/topics/1135

Seems that for reasons that astound me, server-side window decorations got changed to ad-hoc client-side window decorations. QT does indeed present itself as a viable alternative to GTK, since object-oriented C code seems insane, particularly sans documentation. C++ is designed for object-orient programming, not C.

Please delete this before you read it. It could be off-topic. Yet, scribitur ad narrandum, non ad probandum

Please remain on topic. The analogies don't suit and I'm not going to be held responsible for derailing a thread with material that is simply wrong. If you want to use an analogy at least have some understanding of what you are using (in this case cars) so you do not lead people who do know about the thing you are talking about astray on the thing you apparently know more about (in this case the kernel).

I'm sure a few of you just want to argue analogies but lets get this back onto systemd and the kernel. If you are so experienced in these things then use that experience and teach those of us who aren't but for the sake of a learning experience for those of us who don't know as much as you do about the kernel please refrain from comparing apples and oranges and thereby totally confusing the learning experience.

All I did was state an opinion to and you're still going on. Lets agree to disagree. I'm not worried about the personalities involved and that is why I have refrained from "choosing sides" I want to understand the kernel and systemd not your personal opinion of two individuals.

Um you said

and then said we will all be judged. Like I said I agreed with Tobi, about the kernel obviously having a problem that in my opinion should be fixed, and now you're getting all extremist about choosing sides and being judged. I really don't have the time for that.

Pot meet kettle (I'll explain that a little later). Everything after "people are dying" was to enforce the point that you have expressed an extremist view and an equally extreme response was justified to show that. If this thread is about a computer operating system let's stop using car and boat analogies so that people who actually have knowledge about cars and boats don't get side tracked by misinformation and actually learn about the kernel. Also let's stop the personality discussion, it is a side track that does nothing for the discussion about the kernel itself. It is, as you say, an "Obfuscation". Your use of this tactic is designed to get people to dislike systemd, not for the merits or lack of of systemd but rather because you want people to dislike it because of who is developing it. That is the ultimate obfuscation.
Lastly, to all those who want to keep going off-topic, do us a favour and take this particular off-topic discussion to PM with me if you want to continue it. I'll happily devote a few minutes each day to you in PMs but I'm not continuing with this off-topic banter in this thread any longer.

I present what I consider a perfectly accurate analogy directly relevant to the topic of this thread. You, on the other hand, make unfounded accusations and for the second time resort to personal attacks.

And you are accusing others of being "off-topic"? That's rich.

Now, I'm asking you again, please explain what's wrong with my analogy (an explanation that would certainly be on-topic), and please refrain from derailing this thread any further. Time to put up, or shut up.

I will attempt to remain exactly on topic since all of the personal stuff is now a matter of record in full, not cherry-picked and out of context, and simply respond to try to clarify my position on this thread.

The thread is entitled "Linus Torvalds vs/ Kay Seivers", not "Systemd vs/ The Kernel", and personality and character are often reflected in people's work. In this case, as Linus has pointed out, Kay is responsible for both shoddy code and a prima donna attitude and it has been going on for years.

However this was not my tack. (Please show where my argument was even loosely about "who is developing it" let alone entirely so). The laughable thing is that is exactly why I used an analogy - no personal references It is you k3lt01, who keeps resorting to personal attacks and persists on denying the validity of anyone's analogy.

My point, and what the analogy was for, was to show a common situation, where something, in the case of my analogy, boats and channels, are designed to work a certain way. I'll even extend the analogy if you like and add Kay Seivers comes along and dumps a rock in the channel and for no apparent reason. Nobody knows if it was malice (some deep, hidden conspiratorial Master Plan) or incompetence. That only matters a little and only after more evidence unfolds. For now, it, the rock, serves no useful purpose to him or anyone else, or any project. It does create a problem for boats to use the channel as it was designed and engineered. How is this even now either personal or a bad analogy?


Not good enough? Don't like analog-y? OK , here's some binary for ya.

Now to get away from the analogy and more specific.... Ideally, rather than simply casting blame, if we look at this constructively, there are only two possible resolutions

1) Limit the kernel debug feature to accommodate systemd and lose some actual, open functionality at kernel level

------------ OR ---------------

2) Limit systemd's verbosity of useless userspace noise at no loss to systemd or anyone else.

Which is more reasonable?

After answering that, we can move on to whether or not this was some honest mistake, promptly rectified, or more of the same prima donna crap in an ongoing pattern of behavior, which appears to be Linus's opinion. I'm siding with Linus.

1 members found this post helpful.

The most sensible thing would be to do #2. There's nothing wrong with the Linux kernel in any regards. It was never broken by any other multi-spectrum, multi-system, and POSIX compatible/compliant init system. Just because Kay and Lennart wanted to whine and cry doesn't mean Linus needs to cave in to them, nor any other kernel developer.

Every init system that has ever graced Linux has always worked with the kernel as-is without any problems. It was only systemd that caused the problem to begin with, therefore, it's not the kernel's fault, it's systemd's fault for being put together by a bunch of shrieking butt scratching monkeys that didn't want to abide by the established rules and bases of developing broad-spectrum UNIX software.

Isn't there an old common sense statement that goes something like:

"When you break the rules, there are consequences"... or something along those lines?

5 members found this post helpful.

I remember in another thread someone saying my response was childish. Wasn't that a personal attack? A thread which someone twisted my words saying I said something I had not, is doing that not something personal? A thread where someone said something is in the "bloody" paper that I hadn't read yet and denied what I said, which is plastered on the official website was true. Once another member showed you were wrong, as I had already done, you shut up. No oh gee I was wrong sorry about that so you can wait till the world burns before I respond to this demand. If you want to discuss automotive material join any automotive forum I am a member of and I'll gladly point you in the right direction. PM me for details.

I am aware of this but I was not, personally, discussing the attitude rather I was discussing the actual issue within the kernel and systemd, do you remember I said I agreed with Tobi with regards to the kernel, so I wonder why I was accused of having an apologetic attitude.

I'm resorting to personal attacks! Linus, as far as I am aware, works with, though his control of the kernel, and knows the people involved, do you? What gives you or anyone the right to say someone who hasn't got into the debate about personality has an apologetic attitude? If trying to stay away from discussing personalities of people, I don't personally know is being apologetic then I'm guilty as charged. Please slap me with a couple of hundred demerit points or ban me forever for my transgression.

Systemd is a rock and the kernel is a channel and that someone dumped a rock in the channel. Problem with this is nothing goes into the kernel without Linus authorising it so if it got in there to a stable release Linus has allowed it to get in without fully testing it, yet a meteorological event or even someone who just wants to have "fun" can dump a real rock in a real channel. Do you see the difference?

Ah here we have something which people who are interested in computer operating systems can learn from. If the result listed in 1 was the actual end result then of course it would not be good and 2 would be the better option. Having said that and this is my point, wouldn't it be prudent to fix the issue in the kernel that allowed systemd to create this problem and by doing so probably stop this ever happening again with other code? Don't limit the debug feature just limit what can use it.

As I have already said I am not interested in the personal debate so I never went there and am not going there, yet I was still accused of being apologetic for 2 people you obviously have an issue with.

As for "choosing sides" which I don't see a need for, considering Linus controls the kernel the decision is his to make and regardless of whether you or I decide to "side" with him and be judged by the community it neither here nor there. Because of this I will not "side" with anyone, as a personality, but I will trust that Linus knows, much better than we do, what he is doing with regards to the kernel. Is that good enough for you?

Ah now I see. You thought "the channel" was analogous to the kernel! No. The Channel, like systemd, is in Userspace but it can sink "the boat/kernel" when irresponsible behavior is allowed. Do you see the difference? and maybe more importantly, do you see that it is not a stretch to interpret "ludicrous" and "Do you see the difference?" as condescending if not downright denigrating?


But you did get personal when you attacked other peoples' opinions (and integrity of character), and not ones pulled from thin air but regarded, researched conclusions all the while giving Kay and Lennart a pass. Again I ask you - Where in this thread did I seem to base my opinion solely on personality of these two individuals? I clearly stated it is not their personalities but rather their behavior (and tried to actively avoid it by creating an analogy!), what they create and foist on others, with which I have an issue. Perhaps you might consider offering the tolerance and effort of understanding to people here that you offer to Kay & Lennart.

This thread reminds me of the Godwin's law.

Could a moderator kindly move it elsewhere, or close it before we reach there?

Thanks in advance.

2 members found this post helpful.

Since it's an issue that affects all distros/users, maybe it would be better moved to Linux General.

1 members found this post helpful.

Ah this is why confusion reigns with analogies that are not only badly put, such as this one, but also neglect to inform what is what. Could you kindly explain, preferably by PM, what the rock is in Linux speak and not nautical speak, and also explain how a channel can sink, as you are now saying in the quote so its your words not mine, a boat. You see in the real world the boat crashing into the channel could sink it but that's not the channels fault. If the boat sinks because of the channel it is most probably because the "irresponsible behaviour" of the captain who drove the boat (in your analogy the kernel) into the immovable channel (which has possibly been around for alot longer than the boat). It's like the Costa Concordia accident, The channel is the coastline, the boat is, well the boat. The captain was irresponsible and took the boat to close to the channel and it sank. This is an analysis based on the latest version of your analogy. Let's move away from flawed analogies, what do you think of that proposal?

Methinks some of you are way to sensitive and read attitude into everything that you don't agree with. Maybe read the words as they are written and not as you want to hear them. If I personally attack you it will be in private and you'll most certainly receive words that you never knew existed.

I never gave Kay and Lennart anything and this is the fundamental flaw here. I merely agreed with Tobi, I honesty don't know why this is such an issue for you, that it is obvious the kernel had a problem and systemd showed it to be there. Tobi posted a link in post 97 that points to this being a kernel issue and the fix is mentioned
which is what I have been talking about. Fix the problem that has been found in the kernel. Is this not relevant? Or because you don't agree with me does this mean the author of this quote is wrong and knows less than you do about the kernel? Just so you don't feel this is a personal attack I am trying to understand your level of experience in this regard. Are you more experienced that Theodore Tso? Is his opinion irrelevant and therefore should we just ignore him?

I'll take you back to the "choose a side" issue. If you think a side needs to be chosen then you really don't understand the position Linus has. He makes the decisions of what happens in the kernel not K or L and individuals who know this don't need to take "sides" and don't need to be judged by the community. I'm the type of person who trusts that Linus knows what should be in the kernel and while that, to you and many others, may seem to be a simplistic stance it is one that shows I have faith in Linus. If on the other hand you really believe a "side" must be taken then feel free to do so but don't expect others to follow you on what to me seems to be a fruitless exercise because, as I said Linus controls the kernel.