Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Systemd headsup of L.P. - reads like at first of april
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next  
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Gentoo Chat
View previous topic :: View next topic  
Author Message
grey_dot
Tux's lil' helper
Tux's lil' helper


Joined: 15 Jul 2012
Posts: 142

PostPosted: Sun Oct 27, 2013 6:53 pm    Post subject: Reply with quote

TomWij wrote:
it depends on what you define as systemd


Systemd function is really hard to define, yeah. From my experience, systemd can be defined as somewhat "one man orchestra" - it does many separate things and sucks at mostly all of them, but many people shit all over themselves from excitement and demand more and more of that stuff.
Back to top
View user's profile Send private message
grey_dot
Tux's lil' helper
Tux's lil' helper


Joined: 15 Jul 2012
Posts: 142

PostPosted: Sun Oct 27, 2013 6:54 pm    Post subject: Reply with quote

steveL wrote:
crappy initscripts written by amateur Linux "developers"?


This implies that there are mature Linux developers. Given the code quality at freedesktop.org I'd say there are none.
Back to top
View user's profile Send private message
Dr.Willy
Guru
Guru


Joined: 15 Jul 2007
Posts: 547
Location: NRW, Germany

PostPosted: Sun Oct 27, 2013 6:56 pm    Post subject: Reply with quote

TomWij wrote:
And even if you get everyone to write in POSIX shell

I don't care what anyone writes their stuff in.
But if the first line of that script says #!/bin/sh it better be POSIX.
Back to top
View user's profile Send private message
Anon-E-moose
Watchman
Watchman


Joined: 23 May 2008
Posts: 6095
Location: Dallas area

PostPosted: Sun Oct 27, 2013 7:28 pm    Post subject: Reply with quote

steveL wrote:
Why are we labelled haters, when we simply do not care about a new piece of software


Some people get awfully upset with others that don't find said piece of new software is the holy grail.

I don't care if someone wants to write a new init system. More power to them.
But I don't like it being forced on my system whether overtly, as in you have no other choice,
or leveraged in by taking something like udev, and making it part of the the new software
and then to keep on changing things until no one can use it without the new software itself.

I realize we're not there yet, with systemd, but it seems that those of us who don't like the above scenario,
whatever our reasons are, are considered to be haters and conspiracy minded nogoodniks.

As far as redhat, and all the other distros, they can do as they wish.
They pretty much sell a canned package, with some selectability in it.
Gentoo isn't supposed to be that way.

Thank you but I like choice.
_________________
PRIME x570-pro, 3700x, 6.1 zen kernel
gcc 13, profile 17.0 (custom bare multilib), openrc, wayland
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Sun Oct 27, 2013 11:23 pm    Post subject: Reply with quote

steveL wrote:
TomWij wrote:
The whole point about systemd is not having to do extra configuration in the system daemon for matters that the service itself allow you to configure.

Huh? We configure our services somewhere, whether in systemd or openrc. That you don't like the name conf.d, does not change that. At least with openrc we're not restricted to an insane setup foisted upon us by red-hat's inability to etc-update, which sticks configuration data in /usr/lib of all places, plus various other apparently random choices we have to dig through to see what is happening.


With systemd, you configure services separately; taking the frequent example NetworkManager, I have that configured without touching a systemd directory or file once; which does happen in /etc.

steveL wrote:
Quote:
It's a paradigm shift; so, there's no need and no point in rewriting it manually (and no, you're not forced to do that as in that paradigm it is not a sane thing to do).

No it's not; it's a regression. Instead of mostly-declarative config, plus custom initscripts should the admin require them and the existing scripts do not fit the bill, and dynamic configuration in line with everything else an admin does, we have a purely-declarative config, or use javascript (what a corker of an idea!); afaict that appears to be the direction posited.


Exactly, it may very well be that way for you; because not one size fits all. So, one person's regression is another person's improvement. Regardless, it is supported by both; just in slightly different ways where it indeed depends on the way you prefer to do it...

steveL wrote:
As discussed above, this "paradigm shift" actually leads to more work. Keep it thanks. ;-)


It only does if you work against the shift; but that's is a fairly normal thing that happens, as people get used to the way things have been done for years it is hard to adapt to new principles.

steveL wrote:
Now, can we have our Gentoo back without being labelled haters?


What do you mean by "have our Gentoo back"? Which change do you aim to get undone?

steveL wrote:
We don't mind what others want to do, we just want to make the traditional setup work, exactly as it always has, and indeed is required to within an initramfs.


Nothing sits in the way of keeping that working (except for perhaps the recent initramfs news item); and if it does, please let me know so we can keep Gentoo about choice.

steveL wrote:
Why are we labelled haters


Who labels you a hater?

steveL wrote:
when we simply do not care about a new piece of software


Yes, I understand that you want to keep the traditional setup working.

steveL wrote:
and we have proven we can keep the traditional setup running, in line with upstream requirements? (Well, as stated by the one member of the udev "inner-circle" who isn't apparently off his head, Greg K-H.)


There are no plans to change the traditional setup (except initramfs, see above); so, I think it is not at all in danger in the near future.

How are upstream requirements limiting it? Or do you mean GNOME and/or separate /usr initramfs here?

steveL wrote:
Answer that question to your own satisfaction, and you will understand a lot more about group-dynamics, groupthink, and peer-pressure, as well as marketing, propaganda and "fashion".


Those pressure forms can work in both ways (not necessarily both directions); while its intention might intend one way or the other, the effect of it can cause the opposite to happen or perhaps neither of both.

steveL wrote:
TomWij wrote:
Exactly. The bash scripts I usually write don't reach end-users, but circle around my home directory and use Bash; so, the only cases where it is necessary (eg. init scripts) I switch to POSIX compliance.

And what does that have to do with "the bad old days" of crappy initscripts written by amateur Linux "developers"?


No idea, you tell me. And what does that have to do with this systemd thread?
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Mon Oct 28, 2013 12:59 am    Post subject: Reply with quote

TomWij wrote:
Quote:
Huh? We configure our services somewhere, whether in systemd or openrc. That you don't like the name conf.d, does not change that. At least with openrc we're not restricted to an insane setup foisted upon us by red-hat's inability to etc-update, which sticks configuration data in /usr/lib of all places, plus various other apparently random choices we have to dig through to see what is happening.

With systemd, you configure services separately; taking the frequent example NetworkManager, I have that configured without touching a systemd directory or file once; which does happen in /etc.

I see; so are you claiming there are no config files in /usr/lib, then some other place, and then /etc, such that an admin must be aware of all them, if s/he wants to track down all relevant settings? Or that systemd has no service configuration, only what the service itself provides?
If so, you're just plain wrong, to the best of my knowledge, which is as full of holes as anyone else's.
And if not, then there is no difference between conf.d and the several directories that systemd uses, except that the former is a lot easier to work with.
Quote:
steveL wrote:
Quote:
It's a paradigm shift

No it's not; it's a regression. Instead of mostly-declarative config, plus custom initscripts should the admin require them and the existing scripts do not fit the bill, and dynamic configuration in line with everything else an admin does, we have a purely-declarative config, or use javascript (what a corker of an idea!); afaict that appears to be the direction posited.

Exactly, it may very well be that way for you; because not one size fits all. So, one person's regression is another person's improvement.

Nonsense: taking away functionality that is streamlined (if you're not a nub) and fits in with everything else the admin does, to give less capability and a completely different syntax to everything else the admin does, in a language notorious for being even easier to mess up in than sh, is simply a regression. No ifs nor buts about it: you may prefer the more limited project, but don't for an instant pretend it has not taken anything away, even if the "reason" for that is that "users simply can't be trusted", which as shown is in fact more a case of "amateur developers need proper supervision, code review and QA, and are likely to blame their shortcomings on anything and everyone but themselves."
Quote:
Regardless, it is supported by both; just in slightly different ways where it indeed depends on the way you prefer to do it...

What is supported by both? Using a shellscript as a service? Amazing, especially when you consider that every executable may be a script, thanks to the traditional Unix shebang.
Quote:
steveL wrote:
As discussed above, this "paradigm shift" actually leads to more work. Keep it thanks. ;-)

It only does if you work against the shift; but that's is a fairly normal thing that happens, as people get used to the way things have been done for years it is hard to adapt to new principles.

No no no: I pointed out how an admin now has to be aware of 3 possible locations at least for the configuration. That is more work, and the only reason for it is that red-hat doesn't have etc-update, at least according to bonsaikitten, who administers cross-distro for a living. Certainly nothing which applies on Gentoo requires us to use those 3 directories, apart from "conformance with upstream" which is just another way of saying "groupthink and fashion told us to." Distributions used to pride themselves on value-add: if you are just going to blindly follow upstream even to the detriment of your users, you might as well just run mirrors of upstream sources, and get out of the way.

Or are you claiming that awareness of those directories is "working against" systemd, when it is systemd which introduced them in the first place?
Quote:
steveL wrote:
Now, can we have our Gentoo back without being labelled haters?

What do you mean by "have our Gentoo back"? Which change do you aim to get undone?

If you read a bit further you'd see. *sigh* In essence: stop labelling people haters, and indulging in groupthink, and get on with helping us, even if we don't do things how you'd prefer. Welcome to the real world of imperfect circumstance and "unreasonable" users who usually pay your wages. In this case, you don't get paid, so presumably you have an/other motive/s. It does not matter what that is: without users you aren't a distro: you're a clique.
Quote:
steveL wrote:
Why are we labelled haters

Who labels you a hater?

I've been labelled that directly on IRC by a gentoo developer, who got an /ignore for his troubles, especially since he was way off-topic as well as wrong. I've seen the same label being used on mailing-lists, as if it were an astute summary of the situation, and not simply groupthink leading to flipping of the bozo bit, a well-known phenomenon I suggest you read up on. Note some of this was in discussions where we were simply trying to ascertain the upstream requirement for udev, and indeed where I suggested that since the problem of early init is exactly the same whether you are in an initramfs or not, both "sides" should collaborate. OFC no-one took that up, since they're not really interested in the best result for everyone, only in validating their own choices by foisting them on others.

Further, I must again point out that it is only systemd zealots who were even interested in taking sides in the first place, as if the whole world were against them, and they needed to deliver unto us a steaming pile of turd, for our own good (after all "we can push a new version later.") Even when we said "no thanks".
Quote:
steveL wrote:
when we simply do not care about a new piece of software

Yes, I understand that you want to keep the traditional setup working.

That is not my main goal, thanks. However, if I am not interested in a piece of software, it should not be forced upon me, and certainly not in such an underhand, and frankly obnoxious manner when it comes to Gentoo "developers" flinging poo in their sandpit.
Quote:
steveL wrote:
and we have proven we can keep the traditional setup running, in line with upstream requirements? (Well, as stated by the one member of the udev "inner-circle" who isn't apparently off his head, Greg K-H.)

There are no plans to change the traditional setup (except initramfs, see above); so, I think it is not at all in danger in the near future.

No I know it's not: that was my point. You will never get away from that initial setup, the most you will do is pretend it doesn't exist, by hiding it within initramfs, and labelling it "a really hard problem to be handled by scripts", while more experienced people laugh, or simply turn away in disgust.
Quote:
How are upstream requirements limiting it? Or do you mean GNOME and/or separate /usr initramfs here?

"Upstream requirements" refers to the requirement that /usr, and indeed all partitions traditionally mounted as part of localmount, such as /var and /opt, are present or mounted before udev starts. Which if you were unaware of, you could have found out just by clicking and spending two minutes reading, instead of asking me to explain the background. "udev upstream has moved to requiring /usr mounted before it starts" seems pretty clear to me, followed by mention of /var and "helper scripts for devices."
Quote:
Those pressure forms can work in both ways (not necessarily both directions); while its intention might intend one way or the other, the effect of it can cause the opposite to happen or perhaps neither of both.

And that adds absolutely zero to the conversation: it's just another very general statement that no-one can really argue with, but is semantically null afaic.
Quote:
steveL wrote:
TomWij wrote:
Exactly. The bash scripts I usually write don't reach end-users, but circle around my home directory and use Bash; so, the only cases where it is necessary (eg. init scripts) I switch to POSIX compliance.

And what does that have to do with "the bad old days" of crappy initscripts written by amateur Linux "developers"?

No idea, you tell me. And what does that have to do with this systemd thread?

In case you've forgotten, you brought up the "old days" above, which started this whole sub-topic about sh, and then refused to let it die when I said I didn't see anything to argue about. So there, that's what it had to do with the thread: you raised it.

No offence, but I've had enough of going in circles. For some reason I find your style of discussion about generalities and conceptual philosophy (way too much along the lines of "it depends how you define", when the person in front of you just used it with a known meaning) instead of considering the points made and their relevance to the specific situation, quite tedious, and annoying, however much I find myself in agreement with you on the broader picture, in general, and support your aims. Because I know you better, I know you are not trolling. If I did not, I'd be forced to conclude you were.
So I'll bow out of this for now.

You should consider this quote (no I do not want a response: some things, such as the group dynamics I asked you to think about, you should take and ponder, not immediately give the first response that comes to mind. However smart you are, thinking takes time, even more so when it is outside your comfort-zone):
Kernighan & Plauger wrote:
Good Programming is not learned from generalities, but by seeing how significant programs can be made clean, easy to read, easy to maintain and modify, human-engineered, efficient and reliable, by the application of common sense and good programming practices.
Careful study and imitation of good programs leads to better writing.

And fgs stop calling systemd "minimal"; it is the antithesis of that. The counter is openrc, which really is minimal, and does as little as possible, and then gets out of the way. The more you insist on calling systemd "minimal", the more people will lose respect for your stance, imo. It's only minimal in terms of the capability it exposes to the admin, not in any other sense of the word.
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Mon Oct 28, 2013 2:13 am    Post subject: Reply with quote

steveL wrote:
I see; so are you claiming there are no config files in /usr/lib, then some other place, and then /etc, such that an admin must be aware of all them, if s/he wants to track down all relevant settings? Or that systemd has no service configuration, only what the service itself provides?
If so, you're just plain wrong, to the best of my knowledge, which is as full of holes as anyone else's.
And if not, then there is no difference between conf.d and the several directories that systemd uses, except that the former is a lot easier to work with.


There are indeed other places were default and example configs reside, but they are not the main place for configuration; as then they wouldn't be FHS compliant.

steveL wrote:
Nonsense: taking away functionality that is streamlined (if you're not a nub) and fits in with everything else the admin does, to give less capability and a completely different syntax to everything else the admin does, in a language notorious for being even easier to mess up in than sh, is simply a regression. No ifs nor buts about it: you may prefer the more limited project, but don't for an instant pretend it has not taken anything away, even if the "reason" for that is that "users simply can't be trusted", which as shown is in fact more a case of "amateur developers need proper supervision, code review and QA, and are likely to blame their shortcomings on anything and everyone but themselves."


Of course it takes things away, in order to be minimal and reach its design goal because services can cover the rest; reaching such goal is not a regression, it is a different paradigm.

steveL wrote:
What is supported by both? Using a shellscript as a service? Amazing, especially when you consider that every executable may be a script, thanks to the traditional Unix shebang.


Exactly, so it is no ground for comparison; glad we agree on that, as this thread is not about that anyway.

steveL wrote:
No no no: I pointed out how an admin now has to be aware of 3 possible locations at least for the configuration. That is more work, and the only reason for it is that red-hat doesn't have etc-update, at least according to bonsaikitten, who administers cross-distro for a living. Certainly nothing which applies on Gentoo requires us to use those 3 directories, apart from "conformance with upstream" which is just another way of saying "groupthink and fashion told us to." Distributions used to pride themselves on value-add: if you are just going to blindly follow upstream even to the detriment of your users, you might as well just run mirrors of upstream sources, and get out of the way.


In general, it has always been that way for as far as I am aware of the topic; with /etc under CONFIG_PROTECT, the configuration in /usr no longer necessarily needs to be installed there., yet it still serves as useful for defaults as well as examples.

steveL wrote:
Or are you claiming that awareness of those directories is "working against" systemd, when it is systemd which introduced them in the first place?


No, systemd follows the very same approach as we discuss above.

steveL wrote:
Quote:
steveL wrote:
Now, can we have our Gentoo back without being labelled haters?

What do you mean by "have our Gentoo back"? Which change do you aim to get undone?

If you read a bit further you'd see. *sigh* In essence: stop labelling people haters, and indulging in groupthink, and get on with helping us, even if we don't do things how you'd prefer. Welcome to the real world of imperfect circumstance and "unreasonable" users who usually pay your wages. In this case, you don't get paid, so presumably you have an/other motive/s. It does not matter what that is: without users you aren't a distro: you're a clique.


Who are you talking to exactly? I have not labeled anyone here a hater. If you want help then you're free to start another thread and invite people there; but this is a systemd thread, so we shouldn't and can't expect people to blindly change the topic. The only help to be received here is that on the topic of this thread; so, in essence, help with using and/or understanding systemd and its changes in implementation. Please don't further distract away from the topic; you know very well from last time that my goal is to support users, thank you for your consideration...

steveL wrote:
I've been labelled that directly on IRC by a gentoo developer, who got an /ignore for his troubles, especially since he was way off-topic as well as wrong. I've seen the same label being used on mailing-lists, as if it were an astute summary of the situation, and not simply groupthink leading to flipping of the bozo bit, a well-known phenomenon I suggest you read up on. Note some of this was in discussions where we were simply trying to ascertain the upstream requirement for udev, and indeed where I suggested that since the problem of early init is exactly the same whether you are in an initramfs or not, both "sides" should collaborate. OFC no-one took that up, since they're not really interested in the best result for everyone, only in validating their own choices by foisting them on others.


Sad to hear that; as for the ML, that indeed got heated at some points. That kind of focus distracted people from attempts at collaboration; so yeah, eventually we do have sys-apps/systemd-sysv-utils which installs a /sbin/init symlink (thus blocking sys-apps/sysvinit) which is kind of the packaged approach as opposed to having a wrapper or something along those lines. It works out well enough and is quite handy if init=... is already used for something else (eg. bootchart).

steveL wrote:
Further, I must again point out that it is only systemd zealots who were even interested in taking sides in the first place, as if the whole world were against them, and they needed to deliver unto us a steaming pile of turd, for our own good (after all "we can push a new version later.") Even when we said "no thanks".


systemd, on Gentoo, is still used in a minority; so, with the feedback it receives it has put people in a more defensive position. There are people not posting something on the ML and want someone else to do it mainly because that person is affiliated with systemd development on Gentoo; I mean, the perception of "sides" on a "distribution that tries to support different choices" really shouldn't go this far...

steveL wrote:
That is not my main goal, thanks. However, if I am not interested in a piece of software, it should not be forced upon me, and certainly not in such an underhand, and frankly obnoxious manner when it comes to Gentoo "developers" flinging poo in their sandpit.


How exactly is the software forced upon you then? Apart from using GNOME 3.8 (though it provides USE flags instead of a hard dependency now) nothing pushes it through when you update the world set; unless, well, in that odd event that the virtual picks it due to a difference in USE flags on udev.

steveL wrote:
No I know it's not: that was my point. You will never get away from that initial setup, the most you will do is pretend it doesn't exist, by hiding it within initramfs, and labelling it "a really hard problem to be handled by scripts", while more experienced people laugh, or simply turn away in disgust.


I don't use an initramfs or separate /usr; so, I can't comment on this. Regardless, I wonder if the Gentoo Council has been made aware of such supportable setup; from what I understood it rather had to do with a large share of developers not willing to support it when packaging as well as support in the terms of attempting to defer from upstream can be pursued for prolonged times.

steveL wrote:
Quote:
How are upstream requirements limiting it? Or do you mean GNOME and/or separate /usr initramfs here?

"Upstream requirements" refers to the requirement that /usr, and indeed all partitions traditionally mounted as part of localmount, such as /var and /opt, are present or mounted before udev starts. Which if you were unaware of, you could have found out just by clicking and spending two minutes reading, instead of asking me to explain the background. "udev upstream has moved to requiring /usr mounted before it starts" seems pretty clear to me, followed by mention of /var and "helper scripts for devices."


Sorry, I meant to ask "which upstream requirements"; so, regardless, thank you for clarifying what that referred to. I have followed the Gentoo systemd discussion about this on the ML as well as the meeting on IRC; as part of that, I've read through those links from freedesktop documenting this in detail. So, yes, I understand and agree, however as far as I understood we couldn't provide long term support downstream.

steveL wrote:
Quote:
Those pressure forms can work in both ways (not necessarily both directions); while its intention might intend one way or the other, the effect of it can cause the opposite to happen or perhaps neither of both.

And that adds absolutely zero to the conversation: it's just another very general statement that no-one can really argue with, but is semantically null afaic.


It adds as much as the pressure forms itself do; so, I'd like not to get off-topic on that matter. It does have semantic value; but, I won't document that in the scope of this thread.

steveL wrote:
Quote:
steveL wrote:
And what does that have to do with "the bad old days" of crappy initscripts written by amateur Linux "developers"?

No idea, you tell me. And what does that have to do with this systemd thread?

In case you've forgotten, you brought up the "old days" above, which started this whole sub-topic about sh, and then refused to let it die when I said I didn't see anything to argue about. So there, that's what it had to do with the thread: you raised it.


I did however not bring up "the bad old days", "crappy" or "amateur".

steveL wrote:
No offence, but I've had enough of going in circles.


You can't proceed and reach a constructive discussion if you defer from the topic of the thread.

steveL wrote:
For some reason I find your style of discussion about generalities and conceptual philosophy (way too much along the lines of "it depends how you define", when the person in front of you just used it with a known meaning)


I'd rather know the meaning than make wrong guesses about it; so, when something does in fact depend on how something is defined, it really does. In this exact same case I could use "it depends on how you define 'known'"; but let me take a different approach and just expand the possibilities, if I were to assume that it means that it is known to the user typing that then in that case I can't really know what is going on, if I were to assume that it means that it is known to me then that could end up being a mistaken assumption of the writer. While this works for a single word that can have two meanings; on the scale of a discussion it quickly becomes hectic to enumerate possible meanings, therefore in order to not unnecessarily introduce branches that the person I am talking to isn't interested in the first place. I simply ask the person to define the right branch right away so the discussion can be continued.

steveL wrote:
instead of considering the points made and their relevance to the specific situation, quite tedious, and annoying, however much I find myself in agreement with you on the broader picture, in general, and support your aims. Because I know you better, I know you are not trolling. If I did not, I'd be forced to conclude you were. So I'll bow out of this for now.


That can indeed be perceived as tedious and annoying because as a side effect they are valuable indicators that measure if a person is willing to hold a constructive discussion towards a consensus or is rather basing oneself on opinions to defer from such a consensus. This has nothing to do with provocation, which from our private conversation I think you know why; rather has to do with improving discussion, providing support, not being judgmental about a particular piece of software or group of users and the list goes on...

steveL wrote:
You should consider this quote (no I do not want a response: some things, such as the group dynamics I asked you to think about, you should take and ponder, not immediately give the first response that comes to mind. However smart you are, thinking takes time, even more so when it is outside your comfort-zone):
Kernighan & Plauger wrote:
Good Programming is not learned from generalities, but by seeing how significant programs can be made clean, easy to read, easy to maintain and modify, human-engineered, efficient and reliable, by the application of common sense and good programming practices. Careful study and imitation of good programs leads to better writing.

And fgs stop calling systemd "minimal"; it is the antithesis of that. The counter is openrc, which really is minimal, and does as little as possible, and then gets out of the way. The more you insist on calling systemd "minimal", the more people will lose respect for your stance, imo. It's only minimal in terms of the capability it exposes to the admin, not in any other sense of the word.


Well, I don't agree with the last argument, I will however respond with something that will make you laugh; but I think it concludes this response well and you should think about: It indeed depends on how I define "minimal". :) (Hint: Assumptions)
Back to top
View user's profile Send private message
grey_dot
Tux's lil' helper
Tux's lil' helper


Joined: 15 Jul 2012
Posts: 142

PostPosted: Mon Oct 28, 2013 4:20 am    Post subject: Reply with quote

TomWij wrote:

I will however respond with something that will make you laugh; but I think it concludes this response well and you should think about: It indeed depends on how I define "minimal". :) (Hint: Assumptions)


That made me laugh indeed. You are one of that kind who believe GNOME 3 to be minimal and lightweight, aren't you?
Back to top
View user's profile Send private message
saellaven
l33t
l33t


Joined: 23 Jul 2006
Posts: 646

PostPosted: Mon Oct 28, 2013 5:10 am    Post subject: Reply with quote

First, let me point out, that, like SteveL, I've basically bowed out since it's a waste of time. There are the opinions of a handful of Gentoo devs and nothing anyone else says matters. They will do what they want while they spin, spin and spin some more in an Orwellian attempt to say that less is more, restriction is freedom and force is choice.

TomWij wrote:

steveL wrote:
Nonsense: taking away functionality that is streamlined (if you're not a nub) and fits in with everything else the admin does, to give less capability and a completely different syntax to everything else the admin does, in a language notorious for being even easier to mess up in than sh, is simply a regression. No ifs nor buts about it: you may prefer the more limited project, but don't for an instant pretend it has not taken anything away, even if the "reason" for that is that "users simply can't be trusted", which as shown is in fact more a case of "amateur developers need proper supervision, code review and QA, and are likely to blame their shortcomings on anything and everyone but themselves."


Of course it takes things away, in order to be minimal and reach its design goal because services can cover the rest; reaching such goal is not a regression, it is a different paradigm.


By taking things away, you are limiting choice and removing freedom. So much of what you've said makes so much more sense now that I see you actually use and favor systemd, and thus, are defending your choice by forcing everyone else to eat the same food.

It's like I said earlier, if the goal is to take things away in an effort to make Gentoo behave like Red Hat, why bother with Gentoo and not simply use RH? Why have use flags when it's so much easier to force a set configuration on people since it makes life easier for the devs?

And again, systemd is anything but minimal, it infects every part of the system and does everything from init to logging to managing devices to you name it. That's invasive and it is hostile to anything that might not fit its paradigm.


Quote:
systemd, on Gentoo, is still used in a minority; so, with the feedback it receives it has put people in a more defensive position. There are people not posting something on the ML and want someone else to do it mainly because that person is affiliated with systemd development on Gentoo; I mean, the perception of "sides" on a "distribution that tries to support different choices" really shouldn't go this far...


a minority which increasingly is forcing ITS choices on everyone else, restricting their choice... at best, they make our systems more fragile yet, again, in an Orwellian move, claim that doing so makes it more robust. And they claim ignorance every time the discussion comes up like there hasn't been a year long fight by a lot of users rejecting initramfs for a separate /usr, /usr merge (which is coming next to suck in /var, I guarantee), continue to ignore how the new network naming system fails every goal it was intended to do and makes things worse for users, etc.

It doesn't matter what we say, you and the other devs have your agenda and will implement it regardless and then continue to spin.

Quote:

How exactly is the software forced upon you then? Apart from using GNOME 3.8 (though it provides USE flags instead of a hard dependency now) nothing pushes it through when you update the world set; unless, well, in that odd event that the virtual picks it due to a difference in USE flags on udev.


how about the /usr initramfs decision made to placate the systemd devs (which now includes udev since they're the same package)? First try to hide behind the 16k gentoo packages saying that it's too hard to maintain all of them for a separate /usr (because all 16k packages are boot relevant) but didn't you personally commit a kernel patch to make it easier for people wanting to switch to systemd? So again, it's ok to complicate things for other people to make your favored package easier, but if we ask to keep decades old functionality, well, suddenly 16k patches have to be maintained.

Quote:

steveL wrote:
No I know it's not: that was my point. You will never get away from that initial setup, the most you will do is pretend it doesn't exist, by hiding it within initramfs, and labelling it "a really hard problem to be handled by scripts", while more experienced people laugh, or simply turn away in disgust.


I don't use an initramfs or separate /usr; so, I can't comment on this. Regardless, I wonder if the Gentoo Council has been made aware of such supportable setup; from what I understood it rather had to do with a large share of developers not willing to support it when packaging as well as support in the terms of attempting to defer from upstream can be pursued for prolonged times.


Again, doesn't fit your personal paradigm, so it doesn't matter... and we've had multiple discussions here but, well, either the Council doesn't listen or doesn't care. The devs have certainly shown that they will stick their heads in the sand in an effort to defend their personal choices while pushing them on to everyone else.

Quote:
That can indeed be perceived as tedious and annoying because as a side effect they are valuable indicators that measure if a person is willing to hold a constructive discussion towards a consensus or is rather basing oneself on opinions to defer from such a consensus. This has nothing to do with provocation, which from our private conversation I think you know why; rather has to do with improving discussion, providing support, not being judgmental about a particular piece of software or group of users and the list goes on...


The past year of these discussions have shown that they are pointless. You devs will do what you want regardless of the feedback you get and you'll spin, spin and spin some more instead of just cutting to the chase.

You like systemd, you want systemd to be easier on Gentoo and that means that you're going to make changes to their systems to make your life easier and you don't care how it effects their system. There is something wrong with them or their systems for not submitting to your model, but you won't simply admit that it is pure arrogance that you and your fellow systemd supporters favor, it's that everyone else is wrong and when we prove you guys are wrong, like with network renaming not being persistent, predictable or more robust, it's a matter of continuing to pretend we've said nothing.

This discussion is pointless. You aren't going to change your mind and you've gone full blown Orwell in your cognitive dissonance whereby less choice is more and something that has grown to the point where it replaces half the system is minimal.

The more these choices are forced on us, the more you DESTROY Gentoo... but keep telling us that you aren't, that you're listening to us, etc and keep making the same push regardless. The minute my Gentoo system no longer boots without an initramfs, I'm going back to my own LFS type system that Gentoo replaced since it automated things for me. The new Gentoo is destined to tell me how I can setup my system without having to fight the one true way. It's less of a headache to have to go back to keeping track of packages and compile them by hand for multiple systems than to keep fixing Gentoo when the devs decide to arbitrarily break it for their new fanboy package. And if the Linux ecosystem gets to the point where it's impossible to not swallow RH's load, I'll leave for BSD.

As for me debating, there's no debate happening... just spin and I'm sick of it. It doesn't matter what we say, so why bother saying it? Why bother just going in circles where we say X is broken, you ignore it and try to tell us Y is better, we point out the flaws in Y and say X doesn't have them, then you insist that Y is better and we're doing it wrong, like we never actually said anything.

If you guys actually want to have a discussion, I'll be back. If you want to just continue to spin and justify breaking things, pretending you didn't in an effort to boost your ego and give yourself a smug chuckle about how clever you are for redefining words, then why even bother posting? I give up. Congratulations... you didn't win the argument, it just means it's pointless to continue talking to you.
_________________
Ryzen 3700X, Asus Prime X570-Pro, 64 GB DDR4 3200, GeForce GTX 1660 Super
openrc-0.17, ~vanilla-sources, ~nvidia-drivers, ~gcc
Back to top
View user's profile Send private message
schorsch_76
Guru
Guru


Joined: 19 Jun 2012
Posts: 450

PostPosted: Mon Oct 28, 2013 8:26 am    Post subject: Reply with quote

Gentoo is all about choice. udev/eudev/systemd/openrc and others should coexists. The problem is that systemd is so upstream invasive, that the projection of the shown behaviour to the future shows no good for the choices of the user. This is not gentoos fault or the gentoo developers fault. It's this software. Its bad, that udev came from the same author and he wants to integrate this into his new baby. Let the first born die and use its good parts.
If systemd gets too invasive, that a lot of packages can not work without it, it has the gnu/linux ecosystem under control, which would be bad for choices.

What can we as users or developers do, to defend the freedom of choices?
1. Discussion with the devs like on this thread

2. abandon linux in total and switch to windows?
3. switch to BSD?

2. is not really an option.
3. BSD desktop side is gnome/kde/xfce like on linux. Is gnome dead on BSD? I think so. A few peolpe ehich have less choice, because of systemd. Am i wrong?

The only possible solution for users is to talk to the devs like on this thread. People should not be called haters, when they try to discuss with the devs in an attemt to defend their freedom.

Is there an "gentoo social contract" like the one for debian? [1] Is there a need to establish one? Is there an DSFG [1] for gentoo?

Edit:
relevant quote
Quote:

4. Our priorities are our users and free software

We will be guided by the needs of our users and the free software community. We will place their interests first in our priorities. We will support the needs of our users for operation in many different kinds of computing environments. We will not object to non-free works that are intended to be used on Debian systems, or attempt to charge a fee to people who create or use such works. We will allow others to create distributions containing both the Debian system and other works, without any fee from us. In furtherance of these goals, we will provide an integrated system of high-quality materials with no legal restrictions that would prevent such uses of the system.



Even if there is one, the problem is that redhat has not such a social contract.... its the "too big upstream" problem.

Bye
schorsch

[1] http://www.debian.org/social_contract
Back to top
View user's profile Send private message
grey_dot
Tux's lil' helper
Tux's lil' helper


Joined: 15 Jul 2012
Posts: 142

PostPosted: Mon Oct 28, 2013 9:07 am    Post subject: Reply with quote

schorsch_76 wrote:

3. BSD desktop side is gnome/kde/xfce like on linux. Is gnome dead on BSD? I think so. A few peolpe ehich have less choice, because of systemd. Am i wrong?


Actually OpenBSD dudes made quite an afford to make GNOME 3 run on that OS. I didnt test it since I dont use gnome, but they say it works quite well.
Back to top
View user's profile Send private message
mv
Watchman
Watchman


Joined: 20 Apr 2005
Posts: 6747

PostPosted: Mon Oct 28, 2013 9:09 am    Post subject: Reply with quote

steveL wrote:
configuration data in /usr/lib of all places

Actually, this is one of the very few things which I like about systemd: You have a clear distinction of what is "upstream" and what you manually configured. It is somehow giving "unionfs" abilities to user: /etc is treated like the possibility of overwriting whatever you want by whatever you want (including the possibility to delete files by linking them to /dev/null).
However, this is convenient only in the beginning: Later you are not informed about changes in upstream configuration files, and so your configuration may fail in a subtle manner. Well, everything has pros and cons...
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Mon Oct 28, 2013 10:23 am    Post subject: Reply with quote

grey_dot wrote:
TomWij wrote:

I will however respond with something that will make you laugh; but I think it concludes this response well and you should think about: It indeed depends on how I define "minimal". :) (Hint: Assumptions)


That made me laugh indeed. You are one of that kind who believe GNOME 3 to be minimal and lightweight, aren't you?


You can use it that way if you prefer; I've started with the gnome-light meta package, and as time progressed I removed any meta package and use a more stripped down version with just the basics needed.
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Mon Oct 28, 2013 11:01 am    Post subject: Reply with quote

saellaven wrote:
By taking things away, you are limiting choice and removing freedom.


Not exactly, because these things are still there; they are just not part of systemd, so, there is no limit in choice and freedom in that perspective. Minimalism on its own can be a feature...

saellaven wrote:
So much of what you've said makes so much more sense now that I see you actually use and favor systemd, and thus, are defending your choice by forcing everyone else to eat the same food.


There is no need for me to defend it, this is not an OpenRC VS systemd thread; I don't see any indication of "forcing" around here, what makes you believe so?

saellaven wrote:
It's like I said earlier, if the goal is to take things away


Where does Gentoo state that that is the goal? Where do they do these things on a large scale?

saellaven wrote:
in an effort to make Gentoo behave like Red Hat, why bother with Gentoo and not simply use RH? Why have use flags when it's so much easier to force a set configuration on people since it makes life easier for the devs?


Because we can support it to a certain extent? But as there are limits in resources, there are limits in what can be supported. Nothing comes from nothing...

saellaven wrote:
And again, systemd is anything but minimal, it infects every part of the system and does everything from init to logging to managing devices to you name it.


That it does so by default doesn't mean it have to; as shown before, it takes a one liner to disable most of that.

saellaven wrote:
That's invasive and it is hostile to anything that might not fit its paradigm.


Providing good defaults isn't invasive; it's to ensure you start out with something that works, having a logger available when you are trying to switch to systemd is handy just like device management support is.

If you don't like it then you can disable or replace those default features.

saellaven wrote:
a minority which increasingly is forcing ITS choices on everyone else, restricting their choice... at best, they make our systems more fragile yet, again, in an Orwellian move, claim that doing so makes it more robust. And they claim ignorance every time the discussion comes up like there hasn't been a year long fight by a lot of users rejecting initramfs for a separate /usr, /usr merge (which is coming next to suck in /var, I guarantee), continue to ignore how the new network naming system fails every goal it was intended to do and makes things worse for users, etc.


I don't see any indication of that minority forcing choices.

saellaven wrote:
It doesn't matter what we say, you and the other devs have your agenda and will implement it regardless and then continue to spin.


Or perhaps we don't?

saellaven wrote:
how about the /usr initramfs decision made to placate the systemd devs (which now includes udev since they're the same package)?


That systemd containing udev (as uptream does) has nothing to do with the initramfs decision as far as I am aware.

saellaven wrote:
First try to hide behind the 16k gentoo packages saying that it's too hard to maintain all of them for a separate /usr (because all 16k packages are boot relevant)


There's much more than just the boot to do on Gentoo; and when you need to take all in account, we're limited by the tools, resources and manpower we have.

saellaven wrote:
but didn't you personally commit a kernel patch to make it easier for people wanting to switch to systemd?


That same patch includes OpenRC; and if you take a closer look, you'll see which one is the default.

saellaven wrote:
So again, it's ok to complicate things for other people to make your favored package easier, but if we ask to keep decades old functionality, well, suddenly 16k patches have to be maintained.


Did things get more complex? Can we maintain those patches? Does the majority of developers want to maintain those patches? Or are they even able to with the other work there is to be done?

saellaven wrote:
Again, doesn't fit your personal paradigm, so it doesn't matter... and we've had multiple discussions here but, well, either the Council doesn't listen or doesn't care. The devs have certainly shown that they will stick their heads in the sand in an effort to defend their personal choices while pushing them on to everyone else.

The past year of these discussions have shown that they are pointless. You devs will do what you want regardless of the feedback you get and you'll spin, spin and spin some more instead of just cutting to the chase.


You should point out those discussions to the council; if they don't know about their existence, then they can't act based on it.

saellaven wrote:
You like systemd, you want systemd to be easier on Gentoo and that means that you're going to make changes to their systems to make your life easier and you don't care how it effects their system.


No, I don't; I don't have anything to do with systemd maintenance or development.

saellaven wrote:
There is something wrong with them or their systems for not submitting to your model, but you won't simply admit that it is pure arrogance that you and your fellow systemd supporters favor, it's that everyone else is wrong and when we prove you guys are wrong, like with network renaming not being persistent, predictable or more robust, it's a matter of continuing to pretend we've said nothing.


We take all users into perspective, not just the small group that objects to it; so, you may very well be right about USB, that doesn't mean you are right about the whole picture, it rather means you have a strong opinion about it that we can take into account. But if most of the Gentoo users are not verbal about it; then, why should we break it for everyone for the sake of some people that have a strong opinion?

saellaven wrote:
This discussion is pointless. You aren't going to change your mind and you've gone full blown Orwell in your cognitive dissonance whereby less choice is more and something that has grown to the point where it replaces half the system is minimal.


Does changing my mind about it matter? I do not maintain systemd or have anything to say about it on Gentoo; so, a change in my train of thoughts is not going to change anything about systemd on Gentoo.

saellaven wrote:
The more these choices are forced on us, the more you DESTROY Gentoo... but keep telling us that you aren't, that you're listening to us, etc and keep making the same push regardless.


Because you keep picturing me as a developer on the systemd herd; which I am not, as you can see on https://www.gentoo.org/proj/en/metastructure/herds/herds.xml

saellaven wrote:
The minute my Gentoo system no longer boots without an initramfs, I'm going back to my own LFS type system that Gentoo replaced since it automated things for me. The new Gentoo is destined to tell me how I can setup my system without having to fight the one true way. It's less of a headache to have to go back to keeping track of packages and compile them by hand for multiple systems than to keep fixing Gentoo when the devs decide to arbitrarily break it for their new fanboy package. And if the Linux ecosystem gets to the point where it's impossible to not swallow RH's load, I'll leave for BSD.


Gentoo, despite wanting to reach out to as much users as possible, can't fit everyone; they have to cut particular choices because of the limited resources, yet they do our best not to...

saellaven wrote:
As for me debating, there's no debate happening... just spin and I'm sick of it. It doesn't matter what we say, so why bother saying it? Why bother just going in circles where we say X is broken, you ignore it and try to tell us Y is better, we point out the flaws in Y and say X doesn't have them, then you insist that Y is better and we're doing it wrong, like we never actually said anything.

If you guys actually want to have a discussion, I'll be back. If you want to just continue to spin and justify breaking things, pretending you didn't in an effort to boost your ego and give yourself a smug chuckle about how clever you are for redefining words, then why even bother posting? I give up. Congratulations... you didn't win the argument, it just means it's pointless to continue talking to you.


Because you are trying to convince the wrong person; there is a fair bit of separation between when you state "you" or "your guys", I'm part of the former but not of the latter. I'm just a systemd user, nothing more...
Back to top
View user's profile Send private message
saellaven
l33t
l33t


Joined: 23 Jul 2006
Posts: 646

PostPosted: Mon Oct 28, 2013 11:07 am    Post subject: Reply with quote

One final note... was digging through the council meetings.

http://www.gentoo.org/proj/en/council/meeting-logs/20130813.txt

Go ahead and read it, it backs up everything I said about /usr being a firewall to stop the rest of the Poettering/systemd mess, the /usr merge probably coming later, they use Poettering's ramblings as justification and then follow up with "if we want to support gnome and systemd..."

No spin... the Council is shoving the "choice" down our throats and flat out state that they don't care if they intentionally break existing systems relatively immediately (30 days notice) and may very well set out to intentionally break util-linux to force the change.

Yay for Gentoo's infamous "choice."

So, if we're just going to clone Red Hat, again, I ask why anyone bothers with Gentoo to begin with?

In my 7 years of using Gentoo, I've never been as pissed off as I am right now. I reiterate - first package that breaks my system, I'm done. Enjoy your sandbox.
_________________
Ryzen 3700X, Asus Prime X570-Pro, 64 GB DDR4 3200, GeForce GTX 1660 Super
openrc-0.17, ~vanilla-sources, ~nvidia-drivers, ~gcc
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Mon Oct 28, 2013 11:07 am    Post subject: Reply with quote

schorsch_76 wrote:
The only possible solution for users is to talk to the devs like on this thread. People should not be called haters, when they try to discuss with the devs in an attemt to defend their freedom.


Yes and no; I wouldn't use my participance as an example, because I can not lead to a possible solution (unless the intention is to convince me to join the systemd herd and change things, or something else; but then I am simply not interested, so, there is no need to convince me). But yes, more discussion with the actual developers might lead to improvements; while I don't know where they hang out, you could try the gentoo-systemd ML.

schorsch_76 wrote:
Is there an "gentoo social contract" like the one for debian? [1] Is there a need to establish one? Is there an DSFG [1] for gentoo?


Yes and no; on the forums, in the first instance the forum rules count.

Outside the forums, or when things really get out of hand; when it is not just a misunderstanding but rather intentional, there is the Gentoo CoC.
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Mon Oct 28, 2013 11:22 am    Post subject: Reply with quote

saellaven wrote:
One final note... was digging through the council meetings.

http://www.gentoo.org/proj/en/council/meeting-logs/20130813.txt

Go ahead and read it, it backs up everything I said about /usr being a firewall to stop the rest of the Poettering/systemd mess, the /usr merge probably coming later, they use Poettering's ramblings as justification and then follow up with "if we want to support gnome and systemd..."


Both systemd as well as the /usr merge link are just used as examples; they are in no way the only reasoning to it (as per prior discussions leading up to this meeting), please also note that WilliamH is a part of the Gentoo systemd project so it is not using it to point out what you deem to be mess or ramblings.

saellaven wrote:
No spin... the Council is shoving the "choice" down our throats and flat out state that they don't care if they intentionally break existing systems relatively immediately (30 days notice) and may very well set out to intentionally break util-linux to force the change.

Yay for Gentoo's infamous "choice."


We can't care about what we can't support. Quoting from the summary of the meeting, "make supporting a proper barrier between root and /usr filesystems increasingly difficult".

saellaven wrote:
So, if we're just going to clone Red Hat, again, I ask why anyone bothers with Gentoo to begin with?

In my 7 years of using Gentoo, I've never been as pissed off as I am right now. I reiterate - first package that breaks my system, I'm done. Enjoy your sandbox.


Because I'm glad it provides choice; if I have reason to switch to OpenRC tomorrow, I will be happy that I can. If I need a separate /usr, I know Gentoo has a hard time to support it; so, I won't mind running an initramfs for that. And if I would mind, I might even try to produce some form of kernel patch to not have to use an initramfs; it really depends on how much I care to step away from Gentoo itself, in terms of support I'd rather choose to take the supported paths though. As long as it works, I consider Gentoo to be the right thing for me; if not, I'll be as verbal about it as you guys do, or maybe I just quit and go back to running Windows...
Back to top
View user's profile Send private message
_______0
Guru
Guru


Joined: 15 Oct 2012
Posts: 521

PostPosted: Mon Oct 28, 2013 12:28 pm    Post subject: Reply with quote

saellaven wrote:

No spin... the Council is shoving the "choice" down our throats and flat out state that they don't care if they intentionally break existing systems relatively immediately (30 days notice) and may very well set out to intentionally break util-linux to force the change.


Damn!! Is this fer realz or a joke?

If it's for real, I guess time to learn LFS ...

oh well ...

this the asshls responsible of destroying gentoo?

Code:
Michał Górny    mgorny    Lead
Robert Piasek    dagger    Member
Mike Gilbert    floppym    Member
Fabio Erculiani    lxnay    Member
Pacho Ramos    pacho    Member
William Hubbs    williamh    Member


but how can this be? What about gentoo's effort for eudev? only 30 days countdown??
Back to top
View user's profile Send private message
saellaven
l33t
l33t


Joined: 23 Jul 2006
Posts: 646

PostPosted: Mon Oct 28, 2013 12:47 pm    Post subject: Reply with quote

_______0 wrote:
saellaven wrote:

No spin... the Council is shoving the "choice" down our throats and flat out state that they don't care if they intentionally break existing systems relatively immediately (30 days notice) and may very well set out to intentionally break util-linux to force the change.


Damn!! Is this fer realz or a joke?


Don't take my word for it, go through and read the logs... you can find the rest here

Quote:

but how can this be? What about gentoo's effort for eudev? only 30 days countdown??


I haven't heard much on eudev lately. But here on the forums, it's been the policy of one of the gentoo devs to constantly smear the project, otherwise, it's been pretty quiet.

If I do decide to stick around Gentoo, I'm probably going to test out SteveL's project to avoid initramfs. Wonder if the Council knew about that... they seemed to have done little due diligence before taking their vote, other than once again citing Poettering's discredited FUD.

Apparently, every Gentoo user is supposed to keep up on the Council's agenda and contact them ahead of time so we can argue our positions, but there is no duty for them to actually inform themselves of what the community's concerns may be before taking a vote. It's the very same dev arrogance that I complained about a few pages back, which in turn has kept me from becoming a dev myself. Funny part is, I don't know why they want to be Gentoo devs if they just want to run the RH default setup.
_________________
Ryzen 3700X, Asus Prime X570-Pro, 64 GB DDR4 3200, GeForce GTX 1660 Super
openrc-0.17, ~vanilla-sources, ~nvidia-drivers, ~gcc
Back to top
View user's profile Send private message
saellaven
l33t
l33t


Joined: 23 Jul 2006
Posts: 646

PostPosted: Mon Oct 28, 2013 12:49 pm    Post subject: Reply with quote

PS - that 30 day countdown officially ends November 1... so it's more like the end of this week.
_________________
Ryzen 3700X, Asus Prime X570-Pro, 64 GB DDR4 3200, GeForce GTX 1660 Super
openrc-0.17, ~vanilla-sources, ~nvidia-drivers, ~gcc
Back to top
View user's profile Send private message
mv
Watchman
Watchman


Joined: 20 Apr 2005
Posts: 6747

PostPosted: Mon Oct 28, 2013 1:11 pm    Post subject: Reply with quote

Dr.Willy wrote:
But if the first line of that script says #!/bin/sh it better be POSIX.

This is OT, but I learnt the hard way that even for POSIX the first line of the script should never say "#!/bin/sh": Better make it "#!/usr/bin/env sh" as POSIX standard says only this implies a POSIX shell...
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Mon Oct 28, 2013 7:47 pm    Post subject: Reply with quote

mv wrote:
Dr.Willy wrote:
But if the first line of that script says #!/bin/sh it better be POSIX.

This is OT, but I learnt the hard way that even for POSIX the first line of the script should never say "#!/bin/sh": Better make it "#!/usr/bin/env sh" as POSIX standard says only this implies a POSIX shell...

Oh god no: set it at install-time using getconf to find the correct PATH: iirc it's only Solaris that does stuff like /usr/xpg4/bin/sh (unless you're on Windows ofc, in which case the shebang does not work in any case.) So by default we have #!/bin/sh and then check with getconf before we install. It's not hard at all.

I'm staying out of the rest of the argument: I'm sick to death of going round in circles, with detailed point-by-point breakdown of idiocy. Those with enough experience know what I'm saying, the rest can either use their brains, or leave me alone.

FWIW we're not giving up on Gentoo any time soon, and I will keep the udev without initramfs page updated for as long as we are on Gentoo (since we need it, and it's my backup notes;).
Ultimately the developers can do WTF they want: they can't stop us patching code or script, and masking packages. I'm just not very happy about being treated so badly for trying to cooperate with them. Seriously fscked-up project-management afaic.

Oh wait, they don't know what that means, they're all "hobbyists" and "unpaid volunteers" so apparently they cannot learn anything until they quit and get a real job, and we should be grateful for any crap they come out with, even if we've spent the last 2 years telling them it's crap and showing them how to do it better.</sarcasm>

@saellaven I totally agree with what you're saying. Don't bother arguing the toss though, some of the devs are real beginners, but very clever nonetheless. Unfortunately, like so many young smart males, they think that means they should respond as quickly as possibly, so as not to lose the "wonderful" thoughts spilling from their brains. It takes years, or good teaching, to appreciate slow thinking, and the education system is as fscked-up as the rest of society. CS ain't what it used to be, I can tell you that much.

"minimal" ha.
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Mon Oct 28, 2013 8:28 pm    Post subject: Reply with quote

_______0 wrote:
saellaven wrote:

No spin... the Council is shoving the "choice" down our throats and flat out state that they don't care if they intentionally break existing systems relatively immediately (30 days notice) and may very well set out to intentionally break util-linux to force the change.


Damn!! Is this fer realz or a joke?

If it's for real, I guess time to learn LFS ...

oh well ...


Why would they?

_______0 wrote:
this the asshls responsible of destroying gentoo?

Code:
Michał Górny    mgorny    Lead
Robert Piasek    dagger    Member
Mike Gilbert    floppym    Member
Fabio Erculiani    lxnay    Member
Pacho Ramos    pacho    Member
William Hubbs    williamh    Member


but how can this be?


How did they?
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Mon Oct 28, 2013 8:32 pm    Post subject: Reply with quote

saellaven wrote:
Wonder if the Council knew about that... they seemed to have done little due diligence before taking their vote, other than once again citing Poettering's discredited FUD.

Apparently, every Gentoo user is supposed to keep up on the Council's agenda and contact them ahead of time so we can argue our positions, but there is no duty for them to actually inform themselves of what the community's concerns may be before taking a vote.


There isn't much to find on the community till after the motion passes, as before that point most of the community doesn't know about the motion; so, in order to be aware of an upcoming motion and to get your concern heard, you can respond to the public community inquiry where motions are gathered and discussed that happens at least monthly on the gentoo-project mailing list.
Back to top
View user's profile Send private message
ulenrich
Veteran
Veteran


Joined: 10 Oct 2010
Posts: 1480

PostPosted: Mon Oct 28, 2013 8:41 pm    Post subject: Reply with quote

TomWij wrote:
_______0 wrote:
this the asshls responsible of destroying gentoo?
Code:
Michał Górny    mgorny    Lead
Robert Piasek    dagger    Member
Mike Gilbert    floppym    Member
Fabio Erculiani    lxnay    Member
Pacho Ramos    pacho    Member
William Hubbs    williamh    Member

but how can this be?

How did they?
@TomWij
It is the most important rule of Redhat-Poettering conspiration theory:
Code:
Gentoo is about choice. Supporting Systemd destroys Gentoo.
You cannot argue about that. If you tried the last pages of this thread, it is blown in the wind.

The only method would be to create an anti-conspiration-conspiration-theory: Google has paid some guys for red-herring purpose. Exactly what is told about Redhat is the case with Googles Android platform.
Back to top
View user's profile Send private message
Display posts from previous:   
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Gentoo Chat All times are GMT
Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next
Page 6 of 7

 
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