Hello, know someone why was annihilated '/etc/sysconfig/i18n' file? It seems as there is '/etc/locale.conf' instead, but which was reason for this?
TIA, Franta
On Mon, 25 Feb 2013 15:05:36 +0100 Frantisek Hanzlik wrote:
Hello, know someone why was annihilated '/etc/sysconfig/i18n' file? It seems as there is '/etc/locale.conf' instead, but which was reason for this?
Redhat can sell more support if they constantly change things for no apparent reason?
Frantisek Hanzlik wrote:
Hello, know someone why was annihilated '/etc/sysconfig/i18n' file? It seems as there is '/etc/locale.conf' instead, but which was reason for this?
See http://docs.fedoraproject.org/en-US/Fedora/18/html/Release_Notes/sect-Releas...
Starting with 2.3.2. Some /etc/sysconfig files have been deprecated
And, test list thread during f18 development on the topic, http://lists.fedoraproject.org/pipermail/test/2012-October/111230.html
Rex Dieter wrote:
Frantisek Hanzlik wrote:
Hello, know someone why was annihilated '/etc/sysconfig/i18n' file? It seems as there is '/etc/locale.conf' instead, but which was reason for this?
See http://docs.fedoraproject.org/en-US/Fedora/18/html/Release_Notes/sect-Releas...
Starting with 2.3.2. Some /etc/sysconfig files have been deprecated
And, test list thread during f18 development on the topic, http://lists.fedoraproject.org/pipermail/test/2012-October/111230.html
Rex thanks for reference. But reasons for this changes isn't there. With 'hostname' maybe this is for easier hostnamectl manipulation, as in new file is only this one value, nothing else - as was in '/etc/sysconfig/network'.
But reasons for change '/etc/sysconfig/i18n' I cannot see. Maybe some developers think that Fedora users have unlimited amount of time to learn all their flip-flops?
Am 25.02.2013 21:08, schrieb Frantisek Hanzlik:
But reasons for change '/etc/sysconfig/i18n' I cannot see. Maybe some developers think that Fedora users have unlimited amount of time to learn all their flip-flops?
that is the point
there are WAY TOO MANY changes which are useless and only for making over decades usebale documentations crap
well, i have the knowledge and energy to follow all this bullshit, but the ordinary user and newcomer which finds everywhere in the internet useless doucmentations may have not and go back to a operating system wehre developers fun is not fixing things which are not broken
P.S: yes, i know about the direction to remove distribution-specific configurations - but WHAT THE HELL why do we not kill all the different distributions at all when the differences are bad?
Hi
On Mon, Feb 25, 2013 at 3:08 PM, Frantisek Hanzlik
But reasons for change '/etc/sysconfig/i18n' I cannot see.
The goal of the change is to unify the configuration files and format between distributions using systemd.
http://0pointer.de/blog/projects/the-new-configuration-files.html
standardization has the usual benefits for both users and developers.
Rahul
Am 25.02.2013 21:46, schrieb Rahul Sundaram:
Hi
On Mon, Feb 25, 2013 at 3:08 PM, Frantisek Hanzlik
But reasons for change '/etc/sysconfig/i18n' I cannot see.
The goal of the change is to unify the configuration files and format between distributions using systemd.
http://0pointer.de/blog/projects/the-new-configuration-files.html
standardization has the usual benefits for both users and developers
in theory
in real life only for developers and not for Fedora/RHEL users which are on board since many years - they are only finding a lot of references and documentations/howtows ihe web which are no longer true
Hi
On Mon, Feb 25, 2013 at 3:55 PM, Joe Zeff joe@zeff.us wrote:
On 02/25/2013 12:46 PM, Rahul Sundaram wrote:
standardization has the usual benefits for both users and developers.
Standards are a wonderful thing; there are so many of them to choose from.
I am not sure why the stock quote is consider a useful argument here. Distributions are already adopting this.
Rahul
Am 25.02.2013 22:09, schrieb Rahul Sundaram:
On Mon, Feb 25, 2013 at 3:55 PM, Joe Zeff <joe@zeff.us mailto:joe@zeff.us> wrote:
On 02/25/2013 12:46 PM, Rahul Sundaram wrote: standardization has the usual benefits for both users and developers. Standards are a wonderful thing; there are so many of them to choose from.
I am not sure why the stock quote is consider a useful argument here. Distributions are already adopting this.
and why "they" are not adopting /etc/sysconfig?
Hi
On Mon, Feb 25, 2013 at 4:12 PM, Reindl Harald h.reindl@thelounge.netwrote:
Am 25.02.2013 22:09, schrieb Rahul Sundaram:
I am not sure why the stock quote is consider a useful argument here.
Distributions are already adopting this.
and why "they" are not adopting /etc/sysconfig?
You will have to ask them for specific reasons. As a general trend, different distributions have over a period of time come up with various configuration files and formats to do the same thing which leads to pointless differences and makes it a fragile development platform for any third party developer. Adoption to a neutral location is easier than asking them to buy into /etc/sysconfig on the whole because of the legacy involved, for both technical (non extensible file formats, distro specific quirks) and political reasons (why should we adopt a Red Hat or SUSE decision etc). For distributions using systemd, a change like this is easier to adopt now because they have already bought into the benefits of systemd and compatibility has been maintained during the transition period.
Rahul
Am 25.02.2013 22:21, schrieb Rahul Sundaram:
Hi
On Mon, Feb 25, 2013 at 4:12 PM, Reindl Harald <h.reindl@thelounge.net mailto:h.reindl@thelounge.net> wrote:
Am 25.02.2013 22 <tel:25.02.2013%2022>:09, schrieb Rahul Sundaram: > > I am not sure why the stock quote is consider a useful argument here. Distributions are already adopting this. and why "they" are not adopting /etc/sysconfig?
and political reasons (why should we adopt a Red Hat or SUSE decision etc). For distributions using systemd
who needs political reasons in software-development?
things are fine and working or not yes, so easy the world could be
it would be easy to drive with sytemd the Radhat way and let other distributions decide if they use sysVinit forever or follow this way
Reindl Harald wrote:
Am 25.02.2013 21:08, schrieb Frantisek Hanzlik:
But reasons for change '/etc/sysconfig/i18n' I cannot see. Maybe some developers think that Fedora users have unlimited amount of time to learn all their flip-flops?
that is the point
there are WAY TOO MANY changes which are useless and only for making over decades usebale documentations crap
well, i have the knowledge and energy to follow all this bullshit, but the ordinary user and newcomer which finds everywhere in the internet useless doucmentations may have not and go back to a operating system wehre developers fun is not fixing things which are not broken
P.S: yes, i know about the direction to remove distribution-specific configurations - but WHAT THE HELL why do we not kill all the different distributions at all when the differences are bad?
big confirmation. Now is nearly faster download appropriate SW sources (usualy with lot of patches) and study it, than finding some on net.
H
On Mon, Feb 25, 2013 at 4:38 PM, Reindl Harald h.reindl@thelounge.netwrote:
who needs political reasons in software-development?
things are fine and working or not yes, so easy the world could be
it would be easy to drive with sytemd the Radhat way and let other distributions decide if they use sysVinit forever or follow this way
You are self-contradicting. Driving systemd "the Red Hat way" would be a political decision. systemd developers specifically don't want to do that at all. They want to make technically good decisions and sometimes that means looking at what other distributions are doing and adopting the good parts. Sometimes, that means coming up with new configuration file formats or locations.
Rahul
Am 25.02.2013 22:47, schrieb Rahul Sundaram:
On Mon, Feb 25, 2013 at 4:38 PM, Reindl Harald <h.reindl@thelounge.net mailto:h.reindl@thelounge.net> wrote:
who needs political reasons in software-development? things are fine and working or not yes, so easy the world could be it would be easy to drive with sytemd the Radhat way and let other distributions decide if they use sysVinit forever or follow this way
You are self-contradicting. Driving systemd "the Red Hat way" would be a political decision. systemd developers specifically don't want to do that at all. They want to make technically good decisions and sometimes that means looking at what other distributions are doing and adopting the good parts. Sometimes, that means coming up with new configuration file formats or locations
in other words:
we are changing things for EVERYBODY so that no one can complain it was changed only for him - technical nonsense
sometimes it feels more developers have contracts with book wirters to give a godd reason to buy the next version of a book because you can throw away all existing ones
there where books written in 1998 which where perfectly until 2006/2007 and mostly 2009, they contained informations which where valid for decades and now things are chnages each month - and this is the right direction?
it is NOT! this has NOTHING to do with the unix paradigms
Hi
On Mon, Feb 25, 2013 at 4:52 PM, Reindl Harald h.reindl@thelounge.netwrote:
in other words:
we are changing things for EVERYBODY so that no one can complain it was changed only for him - technical nonsense
Not really what I said at all. On the contrary, it all depends on the configuration file and we have to consider it on a case by case basis. For instance, systemd now uses /etc/hostname which was originally a Debian convention. In other cases, a neutral location and format was created. example: /etc/os-release
http://0pointer.de/blog/projects/os-release
Rahul
Am 25.02.2013 23:00, schrieb Rahul Sundaram:
On Mon, Feb 25, 2013 at 4:52 PM, Reindl Harald <h.reindl@thelounge.net mailto:h.reindl@thelounge.net> wrote:
in other words: we are changing things for EVERYBODY so that no one can complain it was changed only for him - technical nonsense
Not really what I said at all. On the contrary, it all depends on the configuration file and we have to consider it on a case by case basis. For instance, systemd now uses /etc/hostname which was originally a Debian convention. In other cases, a neutral location and format was created. example: /etc/os-release
which does ALL not change the fact that if you have a valid Redhat knowledge / certification you can throw it away and the current development does not show a sign that this will stop in the next years
technical nonsense AKA change for the sake of the change
why do we not kill all the linux distributions and go ONE GLORY way? becaue there is NO one glory way!
Hi
On Mon, Feb 25, 2013 at 5:09 PM, Reindl Harald wrote:
which does ALL not change the fact that if you have a valid Redhat knowledge / certification you can throw it away and the current development does not show a sign that this will stop in the next years
You are right in the sense that Fedora might not be really suitable for someone who has conservative requirements. On the other hand, Red Hat certification is typically valid for only a limited period regardless of other changes anyway. IT industry knowledge tends to need updates now and then and the pace of development in open source tends to be even higher. Since Fedora is on the leading edge, you will get those changes rapidly. This is just the fundamental nature of the distribution.
Rahul
Am 25.02.2013 23:26, schrieb Rahul Sundaram:
Hi
On Mon, Feb 25, 2013 at 5:09 PM, Reindl Harald wrote:
which does ALL not change the fact that if you have a valid Redhat knowledge / certification you can throw it away and the current development does not show a sign that this will stop in the next years
You are right in the sense that Fedora might not be really suitable for someone who has conservative requirements. On the other hand, Red Hat certification is typically valid for only a limited period regardless of other changes anyway. IT industry knowledge tends to need updates now and then and the pace of development in open source tends to be even higher. Since Fedora is on the leading edge, you will get those changes rapidly. This is just the fundamental nature of the distribution.
fine, changes for technical reasons are fine changes for the sake of the change are nonsense maybe Fedora should realize this
and YES in times of Fedora Core 6 / Fedora Core 7 as long as the distribution was mostly controlled by redhat there where way less useless changes for the sake of the change
Hi
On Mon, Feb 25, 2013 at 5:31 PM, Reindl Harald wrote:
and YES in times of Fedora Core 6 / Fedora Core 7 as long as the distribution was mostly controlled by redhat there where way less useless changes for the sake of the change
You seem to be implying that the community outside of Red Hat is pushing through needless changes. This is not a responsible claim and not backed by any real evidence since most of the changes that you seem to against are fairly major changes backed by Red Hat developers substantially and nothing to do with core/extras merge. Fedora as a project has seen tremendous growth in volunteer contributors after the merge and I would say users are better off because of it.
Rahul
Am 26.02.2013 00:04, schrieb Rahul Sundaram:
On Mon, Feb 25, 2013 at 5:31 PM, Reindl Harald wrote:
and YES in times of Fedora Core 6 / Fedora Core 7 as long as the distribution was mostly controlled by redhat there where way less useless changes for the sake of the change
You seem to be implying that the community outside of Red Hat is pushing through needless changes. This is not a responsible claim and not backed by any real evidence since most of the changes that you seem to against are fairly major changes backed by Red Hat developers substantially and nothing to do with core/extras merge. Fedora as a project has seen tremendous growth in volunteer contributors after the merge and I would say users are better off because of it.
however
i am on board since Fedora Core 3 until Fedora 14 the disrtibution grow better and better
since F15 finally the distribution grows also much better BUT with much more negative impact and careless all over the subsystems, whatever goes wrong, there should be more care again by consider changes/impact _____________
and no, i am not that conservative
i was pretty sure one of the first admins out there which had running Apache 2.0/Apache 2.2/MySQL 5.0/MySQL 5.1 as also PHP5/5.1/5.2/5.3 in production as many other server software over the last 10 years (dbmail3 as example where i spent many nights to help upstream get it stable)
On Mon, 25 Feb 2013 22:27:44 -0500 Rahul Sundaram wrote:
This is something we can all agree with. The problem is that finding the balance in practice turns out to be quite hard to do.
Not really, it is quite easy to do: if it is an entirely gratuitous and cosmetic change, then don't do it. The hard part is restraining the developers who think they are contributing useful work when all they are doing is making life hell for all the people trying to get real work done and have better things to do than dig through every blog and mailing list in the universe to try and discover what pointless change is being made next.
On Mon, Feb 25, 2013 at 09:16:02AM -0500, Tom Horsley wrote:
Redhat can sell more support if they constantly change things for no apparent reason?
No. That is definitely not it.
Hi
On Tue, Feb 26, 2013 at 6:16 AM, Tom Horsley wrote:
On Mon, 25 Feb 2013 22:27:44 -0500 Rahul Sundaram wrote:
This is something we can all agree with. The problem is that finding the balance in practice turns out to be quite hard to do.
Not really, it is quite easy to do: if it is an entirely gratuitous and cosmetic change, then don't do it. http://ask.fedoraproject.org
People don't make changes for the sake of it. They are putting effort, often voluntarily because they really do believe it is helping users or customers
Rahul
On 02/26/13 19:16, Tom Horsley wrote:
On Mon, 25 Feb 2013 22:27:44 -0500 Rahul Sundaram wrote:
This is something we can all agree with. The problem is that finding the balance in practice turns out to be quite hard to do.
I've been avoiding this thread. But I do have to ask.
Not really, it is quite easy to do: if it is an entirely gratuitous and cosmetic change, then don't do it.
And who/what do you envision as the arbitrator of changes? Who/what decides that a change is simply "gratuitous/cosmetic" or not? I think you may be saying that you feel, at times, changes are made purely for the sake of change by some. Is this opinion on your part or do you have documented evidence?
The hard part is restraining the developers who think they are contributing useful work when all they are doing is making life hell for all the people trying to get real work done and have better things to do than dig through every blog and mailing list in the universe to try and discover what pointless change is being made next.
Again, it sounds as if you think a layer of "management" needs to be placed on developers. Who else to "restrain" them from making your life miserable?
This list has become tiresome as of late. People being critical without offering solutions *and* involving themselves in the process. Bitching and moaning on this list isn't being "involved".
Are their shortcomings to Fedora in particular and any Linux distro in general? You betcha. Can we help each other get around the shortcomings and seek to improve what is handed to us? Yes. Without tearing down or insulting those who have given their time without seeking fame or fortune? I think we can.