[manjaro-dev] Static (non-rolling) option

[Guillaume Benoit]

My thought:
When updating you system there is always a risk a breakage, even if it's 
a "small" or "big" update. The only way to minimize it, it's to do as 
much testing as possible.
So for me a "big" update is more dangerous for two reasons:
- it's more difficult to prevent breakage with a "big" update than a 
"small" one.
- if the system breaks it's hard to find what are the packages which 
caused that breakage.

Also dealing with security fixes is very big job, especially with Arch 
packages which are not made for this.

Le 08/11/2013 12:48, Rob McCathie a écrit :
> Unless I were to try to keep up with comprehensive backporting of
> security patches/fixes, which certainly to start with I wouldn't be (I'd
> just clearly state it's not recommended for internet facing servers),
> the workload should actually be pretty light most of the year, except
> surrounding release time.
>
> That said, it does concern me a bit, i'd really want a second person on
> board. Originally I contemplated a non-rolling Arch project with a
> friend who is a capable Linux user/admin & developer (housemate at the
> time, years back now) and the plan was to both run it. I've been meaning
> to get in touch with him and see if he's still interested, but my hopes
> aren't that high if it's to be based on Manjaro, because he's a bit of
> an Arch purist.
>
> Regarding my attention on Manjaro proper, don't take the hypotheticals
> and the questions i've been asking as meaning i'm definitely going to
> start this project, it's still a very big maybe at this point. I'd
> *like* to have something that both I can run and that i'm comfortable
> giving to friends/family of varying computing literacy, but I certainly
> don't *need* it. I can just drop openSUSE on their systems and I'll just
> run it in a VM or on a secondary system.
>
> Also a couple of weeks ago I started putting together another Manjaro
> respin that is designed to be very comfortable for novice users, with
> basic needs, coming from Windows. Though almost completely
> unconfigurable (user can't even choose which apps they install) i'm
> hoping it will be, partially as a result, unbreakable. It's using a
> minimal installation and JWM, which these days has an ultra-conservative
> release cycle.
> Although it wouldn't cover the full spectrum of users that I think the
> static/non-rolling release would, it should cover a lot of them... I
> might just get back to that.
>
> Lastly btw, I have been meaning to test the new RC ISO's and try to help
> more with Manjaro... it's mainly just i've had a HDD space issue going
> on recently, lol... but i've just bought multiple new drives yesterday :D
>
> Regards,
> Rob.
>
>
>
>
> On Fri, Nov 8, 2013 at 9:32 PM, Carl Duff <cdrw2400 at gmail.com
> <mailto:cdrw2400 at gmail.com>> wrote:
>
>     This has been an interesting read.
>     All I would say is: make sure that you would not be heaping too much
>     on yourself, Rob. As it would be a spin-off rather than just another
>     flavour, obviously the amount of work required would be substantial.
>
>      From a selfish point of view, it would also be a shame to have your
>     attention and expertise directed away from the main system...
>
>     Carl
>
>     On 8 November 2013 08:46, Rob McCathie <korrode at gmail.com
>     <mailto:korrode at gmail.com>> wrote:
>
>
>
>
>         On Thu, Nov 7, 2013 at 5:28 PM, Philip Müller <philm at manjaro.org
>         <mailto:philm at manjaro.org>> wrote:
>
>             This is possible and we already have some ideas regarding
>             this topic. Actually Roland is working on such a thing with
>             a different prospective. This will mean a different concept
>             as we have it now. Also lots of testing. When we have some
>             we can inform you. Might also want to look at frugalware,
>             which basically does this concept you're talking about.
>
>
>
>         Any chance you can drop some more hints on what you've got
>         planned? How it's different?
>
>
>
>             Rolling releases have some downs but we try to fix them. We
>             already slowed it down a little but most of the people need
>             or like the current speed.
>             I think adding a longer testing period in testing might help
>             to get it much more stable. Maybe you point me to the edges
>             on our current concept and what you miss right now.
>
>
>         Mainly it's just the simple fact that there is a risk of
>         breakage with every update (updates to system operation
>         components anyway). Maybe if there was about a thousand more
>         people running Manjaro testing branch so we really started
>         getting some serious hardware coverage, and also wider coverage
>         of all the packages in the Arch repos, rolling might be ok for me.
>
>         I want something I can run myself and roll out to friends and
>         perhaps even business clients. Right now Manjaro is not for
>         newbies. I know you're all probably thinking "That's exactly who
>         it's for!", but i'm talking about REAL novice computer users.
>         Right now most all of Manjaro's users are at least
>         *enthusiastic* enough about computing (even if they don't have
>         much knowledge) to have heard of Linux and decided to try to
>         install it. I want an Arch based distro I can roll out to people
>         who aren't pondering what features the next release of their DE
>         will have, or if the driver for their hardware in the Linux
>         kernel is going to improve, because they don't know what any of
>         those things are.
>
>
>         Regarding a longer period in testing; it certainly has my vote.
>
>         I run testing but will not perform pacman -Syu unless I know I
>         have enough time then and there to deal with any degree of
>         breakage, if i'm having a busy week it may be some days before i
>         update, so yeah - a longer period would be good.
>
>         Regards,
>         Rob.
>
>         _______________________________________________
>         manjaro-dev mailing list
>         manjaro-dev at manjaro.org <mailto:manjaro-dev at manjaro.org>
>         http://lists.manjaro.org/cgi-bin/mailman/listinfo/manjaro-dev
>
>
>
>
>
> _______________________________________________
> manjaro-dev mailing list
> manjaro-dev at manjaro.org
> http://lists.manjaro.org/cgi-bin/mailman/listinfo/manjaro-dev
>