Re: [arch-general] [arch-dev-public] [signoff] vc/* -> tty* transition
Dan McGee wrote:
On Sat, Jul 18, 2009 at 8:02 AM, Thomas Bächler<thomas@archlinux.org> wrote:
Allan McRae schrieb:
From experience... not necessarily. I got into X without doing that although I had no tty's. But exactly how do we deal with this? Post a new item before the move?
Fix it: apply a cool sed line that seds vc/$NUMBER to tty$NUMBER on inittab. This will not destroy anything, but potentially fix it.
Those of us not using X on certain machines would get burned here, so its definitely worth attempting to fix...
-Dan
Although you are correct, I do not think automatically fixing the problem is a good solution here. First, I am not a big fan of the idea of package modifying a core file. Secondly what will happen when a user reinstalls the initscripts? If my memory holds me correctly the sed command will fail which will cause the post_install() to fail, because the file already has the correct modifications. This will also happen with users who switch between the official initscripts and unofficial initscripts (faster boot, bootsplash, etc...). I would suggest including a post_install() message and posting a news item a couple days before the move, instead. This way users can find out about this change via the news or via pacman's output. Finally this might be a good "reality" check for users who just install packages and don't pay attention to pacman's output or the news. ~pyther
Matthew schrieb:
Although you are correct, I do not think automatically fixing the problem is a good solution here.
First, I am not a big fan of the idea of package modifying a core file. Secondly what will happen when a user reinstalls the initscripts? If my memory holds me correctly the sed command will fail which will cause the post_install() to fail, because the file already has the correct modifications.
No, all that will happen is ... nothing. And there is a difference between post_upgrade and post_install! I don't want to spend the next 3 weeks responding to forum posts, mailing list threads and irc rants about the same little thing that could have been fixed safely and easily! Do you?
On Sat 18 Jul 2009 22:11 +0200, Thomas Bächler wrote:
Matthew schrieb:
Although you are correct, I do not think automatically fixing the problem is a good solution here.
First, I am not a big fan of the idea of package modifying a core file. Secondly what will happen when a user reinstalls the initscripts? If my memory holds me correctly the sed command will fail which will cause the post_install() to fail, because the file already has the correct modifications.
No, all that will happen is ... nothing. And there is a difference between post_upgrade and post_install!
I don't want to spend the next 3 weeks responding to forum posts, mailing list threads and irc rants about the same little thing that could have been fixed safely and easily! Do you?
Hah. It seems that Arch is turning into one of them user friendly distros where things are automatically configured and all eh? Really I think the proper thing would be to put out a notice that it will be changed, give users a chance to adjust, then change it. Don't bother answering those who are unattentive. Let them suffer or choose another distro.
Loui Chang wrote:
On Sat 18 Jul 2009 22:11 +0200, Thomas Bächler wrote:
Matthew schrieb:
Although you are correct, I do not think automatically fixing the problem is a good solution here.
First, I am not a big fan of the idea of package modifying a core file. Secondly what will happen when a user reinstalls the initscripts? If my memory holds me correctly the sed command will fail which will cause the post_install() to fail, because the file already has the correct modifications.
No, all that will happen is ... nothing. And there is a difference between post_upgrade and post_install!
I don't want to spend the next 3 weeks responding to forum posts, mailing list threads and irc rants about the same little thing that could have been fixed safely and easily! Do you?
Hah. It seems that Arch is turning into one of them user friendly distros where things are automatically configured and all eh?
Really I think the proper thing would be to put out a notice that it will be changed, give users a chance to adjust, then change it.
Don't bother answering those who are unattentive. Let them suffer or choose another distro.
Loui stated it very well. I must admit that I forgot about post_upgrade, however I still think it is a poor idea to let a package modify a critical system file. In this instance, modifying the file is not that big of a deal, but by modifying the file a precedents gets sets. Where does the line get drawn, then? What do we start automatically modifying next? Over the past year or so, we have seen a great deal of new users, which IMHO do not fully appreciate arch for what arch truly is. Does that mean we should make arch easier? I think by automating this processes it would be the start of arch becoming a ubuntu, fedora, suse type of distribution. One of the things that made arch great was that you knew exactly what was going on with your system and how it worked. By automating this task, you lose a bit of that. A few years ago, this idea would have been shot down in an instant. As for the users who don't like it, simply ignore them. If you want to be "helpful" post a link to the news entry and leave it at that. Those who want to help such users can do so in the Newbie conner of the forum. ~pyther
On Sat, Jul 18, 2009 at 4:37 PM, Matthew<pyther@pyther.net> wrote:
Loui Chang wrote:
On Sat 18 Jul 2009 22:11 +0200, Thomas Bächler wrote:
Matthew schrieb:
Although you are correct, I do not think automatically fixing the problem is a good solution here.
First, I am not a big fan of the idea of package modifying a core file. Secondly what will happen when a user reinstalls the initscripts? If my memory holds me correctly the sed command will fail which will cause the post_install() to fail, because the file already has the correct modifications.
No, all that will happen is ... nothing. And there is a difference between post_upgrade and post_install!
I don't want to spend the next 3 weeks responding to forum posts, mailing list threads and irc rants about the same little thing that could have been fixed safely and easily! Do you?
Hah. It seems that Arch is turning into one of them user friendly distros where things are automatically configured and all eh?
Really I think the proper thing would be to put out a notice that it will be changed, give users a chance to adjust, then change it.
Don't bother answering those who are unattentive. Let them suffer or choose another distro.
Loui stated it very well.
I also agree with Loui. We've always tried to avoid these automatic sed type things. I think a message and a news item should be enough
Aaron Griffin schrieb:
Loui stated it very well.
I also agree with Loui. We've always tried to avoid these automatic sed type things. I think a message and a news item should be enough
It seems wrong to me to let so many people perform the same step by hand when we could have done it automatically and completely safe. We have to think about what's simpler here: Have a short and safe sed-line in post_upgrade, or have a shitload of users spend their time booting with live CDs and editing files because they weren't careful. This thing is a huge breaker.
2009/7/19 Thomas Bächler <thomas@archlinux.org>:
We have to think about what's simpler here: Have a short and safe sed-line in post_upgrade, or have a shitload of users spend their time booting with live CDs and editing files /and opening bugs and shouting in the forums and crying on the mailing lists/ because they weren't careful. This thing is a huge breaker.
There, fixed it for you. I concur :)
Thomas Bächler wrote:
Aaron Griffin schrieb:
Loui stated it very well.
I also agree with Loui. We've always tried to avoid these automatic sed type things. I think a message and a news item should be enough
It seems wrong to me to let so many people perform the same step by hand when we could have done it automatically and completely safe.
We have to think about what's simpler here: Have a short and safe sed-line in post_upgrade, or have a shitload of users spend their time booting with live CDs and editing files because they weren't careful. This thing is a huge breaker.
What if the post_upgrade() message gives the users the sed command to run? Who cares about the users? Arch has been a distro that is made the way the developers want it, not the users. The users just reap the benefits of all the developers hard work. It seems as if more devs care about the users, especially the new users. We get lucky if you (the devs) listen to us. So what if we loose 50% of the user base? Does that really matter? Back in the day, the devs got their way. They listened to user input, but more times than not they did what they wanted. They didn't give a squat about the users. IMHO this was one of the qualities that made arch great! And know it is disappearing. It is sad to see arch turning into just another distro. The many things that made arch great are dissolving. :-( ~pyther
Matthew schrieb:
What if the post_upgrade() message gives the users the sed command to run?
Seems kind of pointless.
Who cares about the users? Arch has been a distro that is made the way the developers want it, not the users. The users just reap the benefits of all the developers hard work. It seems as if more devs care about the users, especially the new users. We get lucky if you (the devs) listen to us. So what if we loose 50% of the user base? Does that really matter?
What the developers want, at least me, is not spend the next two weeks being bitched at because philosophy forbids us to change a file automatically.
Back in the day, the devs got their way. They listened to user input, but more times than not they did what they wanted. They didn't give a squat about the users. IMHO this was one of the qualities that made arch great! And know it is disappearing.
In fact, it is not disappearing. I'll just listen to what you said, then ignore it and get it my way. I beat you with your own logic, how nice is that?
It is sad to see arch turning into just another distro. The many things that made arch great are dissolving. :-(
Oh my god, Arch is becoming one of those distros that will actually reboot after an upgrade. Anyway, I will go with whatever the rest says. If it's really necessary to be narrow-minded about our own philosophy, then I won't stand in anyone's way. You see, my own computer still works - as always.
On Sun 19 Jul 2009 01:39 +0200, Thomas Bächler wrote:
Matthew schrieb:
What if the post_upgrade() message gives the users the sed command to run?
Seems kind of pointless.
Who cares about the users? Arch has been a distro that is made the way the developers want it, not the users. The users just reap the benefits of all the developers hard work. It seems as if more devs care about the users, especially the new users. We get lucky if you (the devs) listen to us. So what if we loose 50% of the user base? Does that really matter?
What the developers want, at least me, is not spend the next two weeks being bitched at because philosophy forbids us to change a file automatically.
Back in the day, the devs got their way. They listened to user input, but more times than not they did what they wanted. They didn't give a squat about the users. IMHO this was one of the qualities that made arch great! And know it is disappearing.
In fact, it is not disappearing. I'll just listen to what you said, then ignore it and get it my way. I beat you with your own logic, how nice is that?
Haha. Yeah I just don't want packages to be messing with my configs behind my back. Post a message with a sed command, or a .pacnew file, or something. Don't do it without letting me have that control. That's rude.
On Sat, Jul 18, 2009 at 20:09, Loui Chang<louipc.ist@gmail.com> wrote:
Haha. Yeah I just don't want packages to be messing with my configs behind my back. Post a message with a sed command, or a .pacnew file, or something. Don't do it without letting me have that control.
That's rude.
Agreed. I'm very much against automatically modifying any config files.
First off, I don't like modifying config files. But, given I did this update and still managed to screw my system up when testing it with a reboot... So it is a question of which I hate more; post install messages or automatically fixing the file. A post install message means that I tell all complaining users that they should have read their pacman output. But hang on, they are already told that there is a .pacnew file for what is a very important config file. So I can say that anyway. So here is my prototype install file... post_install() { if [ -f "/etc/inittab.pacnew" ]; then echo "You are being very stupid if you did not take notice of that warning about a .pacnew file" fi } A simple sed of the config file means much, much less complaining users. And given the number of complaints I got about libjpeg7 (wheres the thanks now gtk and kde are working?), I am very, very tempted just to do the sed. Allan
On Sat, Jul 18, 2009 at 22:01, Allan McRae<allan@archlinux.org> wrote:
post_install() { if [ -f "/etc/inittab.pacnew" ]; then echo "You are being very stupid if you did not take notice of that warning about a .pacnew file" fi }
+1 to this solution from me.
Daenyth Blank wrote:
On Sat, Jul 18, 2009 at 22:01, Allan McRae<allan@archlinux.org> wrote:
post_install() { if [ -f "/etc/inittab.pacnew" ]; then echo "You are being very stupid if you did not take notice of that warning about a .pacnew file" fi }
+1 to this solution from me.
I guess you missed my sarcasm. It is difficult to convey across email. I see absolutely no point in repeating warnings. If they did not read the first, why would they read the second? Allan
First off, I don't like modifying config files. But, given I did this update and still managed to screw my system up when testing it with a reboot...
So it is a question of which I hate more; post install messages or automatically fixing the file. A post install message means that I tell all complaining users that they should have read their pacman output. But hang on, they are already told that there is a .pacnew file for what is a very important config file. So I can say that anyway.
...
A simple sed of the config file means much, much less complaining users. And given the number of complaints I got about libjpeg7 (wheres the thanks now gtk and kde are working?), I am very, very tempted just to do the sed.
Allan Could someone please enlighten me why you and Thomas want to please the users that complain? I simply do not understand. You said yourself that you don't like modifying config files, so don't. To hell with the users
Allan McRae wrote: that don't like it. There are a lot of people that appreciate all the work that went into to the libjpeg and readline rebuilds.You don't hear from the users that appreciate all of your hard work. All you guys see/hear are the users that complain. Many times we take it for granted that things just works! And we are at fault for not speaking up and saying, "Hey thanks for the hard work!" I haven't yet heard any users on the Mailing List who are in favor of sedding /etc/inittab. Many of the ML followers probably use testing and reports bugs when ever they encounter anything. Going out on a limb, I would say nearly all users that read the ML want to be involved with arch in one way or the other. I, personally, run testing and I am always amazed at how smoothly things run. I was on IRC earlier today and somebody posted a youtube video on how to install arch linux: http://www.youtube.com/watch?v=BIVcF5t1kZw Now just look at some of the comments on the video. These users are probably on the extreme, but these are some of our users. We also have many users that have just a little bit more knowledge, but they are still left clueless. I'd guess these are the users that complain. Point is users, that complain, probably lack the skills needed to run arch. If you think that automating file modifications is good and in the best interest of the users that you are targeting then go for it. To hell with me and the others. After all, the distro is your work. If that is the path chosen, it is sad to see something so great dissolve away. In that case thank you for the ride and good luck in the future. ~pyther
On Sat, Jul 18, 2009 at 23:03, Matthew<pyther@pyther.net> wrote:
Could someone please enlighten me why you and Thomas want to please the users that complain? I simply do not understand. You said yourself that you don't like modifying config files, so don't. To hell with the users that don't like it. There are a lot of people that appreciate all the work that went into to the libjpeg and readline rebuilds.You don't hear from the users that appreciate all of your hard work. All you guys see/hear are the users that complain. Many times we take it for granted that things just works! And we are at fault for not speaking up and saying, "Hey thanks for the hard work!" Too true. totally agree with all of the above
On Sat, 2009-07-18 at 23:03 -0400, Matthew wrote:
First off, I don't like modifying config files. But, given I did this update and still managed to screw my system up when testing it with a reboot...
So it is a question of which I hate more; post install messages or automatically fixing the file. A post install message means that I tell all complaining users that they should have read their pacman output. But hang on, they are already told that there is a .pacnew file for what is a very important config file. So I can say that anyway.
...
A simple sed of the config file means much, much less complaining users. And given the number of complaints I got about libjpeg7 (wheres the thanks now gtk and kde are working?), I am very, very tempted just to do the sed.
Allan Could someone please enlighten me why you and Thomas want to please the users that complain? I simply do not understand. You said yourself that you don't like modifying config files, so don't. To hell with the users
Allan McRae wrote: that don't like it. There are a lot of people that appreciate all the work that went into to the libjpeg and readline rebuilds.You don't hear from the users that appreciate all of your hard work. All you guys see/hear are the users that complain. Many times we take it for granted that things just works! And we are at fault for not speaking up and saying, "Hey thanks for the hard work!"
I haven't yet heard any users on the Mailing List who are in favor of sedding /etc/inittab. Many of the ML followers probably use testing and reports bugs when ever they encounter anything. Going out on a limb, I would say nearly all users that read the ML want to be involved with arch in one way or the other. I, personally, run testing and I am always amazed at how smoothly things run.
I was on IRC earlier today and somebody posted a youtube video on how to install arch linux: http://www.youtube.com/watch?v=BIVcF5t1kZw Now just look at some of the comments on the video. These users are probably on the extreme, but these are some of our users. We also have many users that have just a little bit more knowledge, but they are still left clueless. I'd guess these are the users that complain.
Point is users, that complain, probably lack the skills needed to run arch.
If you think that automating file modifications is good and in the best interest of the users that you are targeting then go for it. To hell with me and the others. After all, the distro is your work. If that is the path chosen, it is sad to see something so great dissolve away. In that case thank you for the ride and good luck in the future.
~pyther
1st off I am just one of those "ignorant users". Yes I do complain but I am only complaining so others don't have the same problem I did. I am only trying to help. If I find a problem in a PKGBUILD or when building a package I file a bug report. I think packages should build with little or no problems. If devs or anyone at Arch thinks it is bogus you can simply close it. If I post something to help with a problem and it is wrong or can be done better, then some one corrects me then everyone has an opportunity to learn, that is how me/us "ignorant users" learn so have a little patient with us. I don't see anything dissolving away. It is just you have a problem with a update and need to solve it. I am for solving this problem on its own merits and not "only by the rules" then doing the proper thing. Any way you have been warned, that one of your config files may have been edited.. so respond appropriately. Oh see it works both ways.
Allan McRae schrieb:
First off, I don't like modifying config files. But, given I did this update and still managed to screw my system up when testing it with a reboot...
So, the average advanced user won't even notice the problem, even you didn't (and you did get a .pacnew and a warning, didn't you?). It would take someone like me to notice it on his own - and let's admit it, there's not too many people like me. Let's view this from another angle: It's not just the noobs and the unattentive users that will fall for this, it's also over half of the advanced and experienced users. That said, we do modify configuration files all the time. We run grpck on a shadow update so users can still log in, some gtk update generate files in /etc so it still finds its plugins and more. We just don't do it ourselves, but hide behind some program provided to us and tell ourselves "It's okay, upstream wanted it this way". And guess what, nobody even notices.
So it is a question of which I hate more; post install messages or automatically fixing the file. A post install message means that I tell all complaining users that they should have read their pacman output. But hang on, they are already told that there is a .pacnew file for what is a very important config file.
Yes, and there has been a warning for inittab.pacnew several times in the past few months, always with some completely irrelevant added comment or added default lines. So, we give the user a pacnew with irrelevant things until he knows he can ignore it and THEN we break his system, how nice is that? A short and simple message explaining what about this .pacnew is rather important might be in order. It could be as short as "Please read http://www.archlinux.org/news/1234 before you reboot." Or a bit more pragmatic, like "Your system is cannot reboot now. Please thank the Arch developers".
Allan McRae wrote:
And given the number of complaints I got about libjpeg7 (wheres the thanks now gtk and kde are working?)
Thank you! :-) DR
I have to agree with pyther. You, devs, have been doing all you can to warn the users. There is the arch-announce mailing list, there are messages from pacman when it installs something that might break others, there is the forum, there are announcements on Arch's home page... Damn! there are even RSS feeds! Even though, many users will complain, just like when X got updated and HAL needed to be running. IMHO, you should say "We're sorry, but we did try to warn you". I mean, what else could you do? Call every user and tell them that there'll be a big update? You're doing a great job, don't let the few people from the Arch community that don't know how to read upset you.
Loui Chang wrote:
Hah. It seems that Arch is turning into one of them user friendly distros where things are automatically configured and all eh Feature request for pacman-4.0: Please add a Clippy like assistant :P __ / \ _______________________ | | / \ @ @ | Its looks like you are| || || | trying to update your | || || <--| system, are you sure? | |\_/| \_______________________/ \___/
-- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D
Gerardo Exequiel Pozzi wrote:
__ / \ _______________________ | | / \ @ @ | It looks like you are | || || | trying to update your | || || <--| system, are you sure? | |\_/| \_______________________/ \___/
The ASCII-art is by far the best post in this thread. And I am for installing a .pacnew on meeting a changed inittab and a big message. And I'm sort of against packages doing out-of-bailiwick business for me. clemens
participants (12)
-
Aaron Griffin
-
Allan McRae
-
André Ramaciotti
-
Baho Utot
-
bardo
-
clemens fischer
-
Daenyth Blank
-
David Rosenstrauch
-
Gerardo Exequiel Pozzi
-
Loui Chang
-
Matthew
-
Thomas Bächler