[opensuse-project] openSUSE Strategy Discussion: Status quo strategy proposal
Today we continue with public discussions about strategy proposals, this time with the "Status quo" strategy proposal: ---8<------8<------8<------8<------8<------8<--- == Statement == We deliver a well-balanced GNU/Linux platform for modern computers (workstation, laptop, netbook, server) that equally appeals to end users, power users, developers and server/network infrastructure administrators. It shall protrude professionally and let the user be productive. == Background == This strategy tries to quantify what we tried to do in the past — as it was not written down earlier. So, this is what most users expect from openSUSE today, but does not give a vision for change looking forward. In the context of other distributions, we differentiate ourselves from Ubuntu targeting the newbie and further differentiate from Fedora being experimental bleeding edge — instead we pick "the middle ground". == Key ideas == * Creation of a general purpose distribution that ** anyone can use without too much effort ** is known for good quality (stable and usable but neither outdated nor bleeding edge) ** has good and sane defaults so the user can do what s/he wants to do ** has programs that work out of the box ** focused on modern hardware and their use cases (workstations, laptops, netbooks and servers) ** is targeted towards end users, but is reasonably equally usable for other workloads * Critical analysis of hyped items before inclusion == Activities == === We need to be excellent in the following === * Do as we always did! That is, ** good compromise between actuality and stability ** agreeable release cycle of 8 months ** support for the three most recent releases * Supporting our target customers ** End users: *** Delivering multiple desktops, focusing on both GNOME and KDE *** Focus on providing tools for being productive and creative (IDEs, editors, authoring tools, graphics manipulation, office productivity, etc.) ** Developers: *** Development environments for especially C, C++, Perl, Python, Java, Ruby: IDEs, tools and support libraries ** Power users and system administrators: *** Providing admin tools that are powerful yet (reasonably) easy *** Agreeable command line experience *** Virtualization technique, e.g. KVM, Xen *** Standard networking services * Continue the naturally growth of openSUSE:Factory by incorporating contributors' submissions. === We will try to do the following effectively === * Innovate and keep up with latest upstream developments. * Include a more minimalistic desktop environment. * Provide a low entry barrier for potential contributors. With the openSUSE Build Service, it is easier to make contributions than any other Linux distribution to date. * Offer easy creation of specialized install media (appliances) through SUSE Studio. * Good presentation and marketing, in particular communicating our existing strengths and unique features (i.e. competitive advantages). * The usual niceties: speed, less bloat, possibility of minimality. === As project, we will not focus on the following === (fill in if exists) -- Andreas Jaeger, Program Manager openSUSE, aj@{novell.com,opensuse.org} Twitter: jaegerandi | Identica: jaegerandi SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
Tirsdag den 10. august 2010 09:32:56 skrev Andreas Jaeger:
Today we continue with public discussions about strategy proposals, this time with the "Status quo" strategy proposal:
There's a lot to like about it - if you don't think so, what are you doing here? ;-) But I have a couple of gripes with it. It's not easy to sum up with one or two sentences or a couple of keywords to answer the question "Why openSUSE?" which I consider a requirement for the strategy - like e.g. Debian (stable, old, server), Fedora (innovation, bleeding edge, open source), Ubuntu (newbie, bad python code, more newbie). It doesn't clearly identify an audience to (mainly) focus on. -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
* Martin Schlander
Tirsdag den 10. august 2010 09:32:56 skrev Andreas Jaeger:
Today we continue with public discussions about strategy proposals, this time with the "Status quo" strategy proposal:
There's a lot to like about it - if you don't think so, what are you doing here? ;-)
I do, too. It lists the reasons why I use it and maintain it on other peoples' boxes.
But I have a couple of gripes with it.
It's not easy to sum up with one or two sentences or a couple of keywords to answer the question "Why openSUSE?" which I consider a requirement for the strategy - like e.g. Debian (stable, old, server), Fedora (innovation, bleeding edge, open source), Ubuntu (newbie, bad python code, more newbie).
It doesn't clearly identify an audience to (mainly) focus on.
Actually I think it does mention openSUSE's competitve advantages if you read it carefully: * a reasonable compromise between stability and up-to-dateness (distinguishing it from Fedora, Ubuntu, and Debian) * several desktop environments of high quality (unlike Ubuntu, Fedora) * a broad range of packaged software for desktop, server usage, development etc. (unlike e.g. Mandriva) * configuration management through YaST * easy collaboration and low hurdles for contributing through OBS IMO an unique combination which should naturally appeal to a broad range of consumers, including non-technical users, sysadmins, developers, or generally anyone focused on being productive. Usage on production servers seems less attractive due to the recently shortened support cycles. IMHO one of openSUSE's major weaknesses is its marketing failing to communicate its competitve advantages adequately. -- Guido Berhoerster -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Tuesday 2010-08-10 13:39, Martin Schlander wrote:
Tirsdag den 10. august 2010 09:32:56 skrev Andreas Jaeger:
Today we continue with public discussions about strategy proposals, this time with the "Status quo" strategy proposal:
There's a lot to like about it - if you don't think so, what are you doing here? ;-)
But I have a couple of gripes with it.
It's not easy to sum up with one or two sentences or a couple of keywords to answer the question "Why openSUSE?"
http://news.opensuse.org/2009/04/23/people-of-opensuse-jan-engelhardt/ could add a pointer to that question.
which I consider a requirement for the strategy - like e.g. Debian (stable, old, server), Fedora (innovation, bleeding edge, open source), Ubuntu (newbie, bad python code, more newbie).
You forgot to add: Ubuntu (quite-forking GNOME and its implications, see more on LWN)
It doesn't clearly identify an audience to (mainly) focus on.
q = !fedora_stuntmen && !ubuntu_newb; As it happens, this also seems to evaluate for Debian users (*handwave*). -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
Le 10/08/2010 14:28, Guido Berhoerster a écrit :
IMHO one of openSUSE's major weaknesses is its marketing failing to communicate its competitve advantages adequately.
probably, but don't forget, openSUSE (with it's new name and status) is fairly new, newer than the competitors, except ubuntu, but ubuntu spent a lot of money in marketing, money we don't have. getting a large user base is a long range work. That's why I think we have to keep what we did, but discuss on the evolution for the next years (not stricly the same as changing strategy) some ways we could: * state clearly that we wont support hardware configs more than 5 (or 7) years old (of course teams can work on other versions) - say PIV, 256Mo ram, 20Gb HDD * state we will support kde and gnome long term, and the other desktops only is teams provide support (a la lxde) - of course terminal/ncurse yast is essential; * state we will attentively follow the evolution of cloud and mobile computing * state that our infrastructure (obs, studio) allows teams to build they own appliance and that we will host them * state that we won't be incentive for complete newbies installing openSUSE, some computer skills being usefull. * quote more often "Linux". We are openSUSE GNU/Linux dstribution. Almost any computer user knows at least the name "Linux", and this is who we are. -- http://www.dodin.net http://pizzanetti.fr -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Tue, 10 Aug 2010, Andreas Jaeger wrote:
** support for the three most recent releases
Three? I count two plus an extra two months for the oldest when a new one comes out (that is, two for 75% of the time, three for the other 25%).
=== As project, we will not focus on the following ===
(fill in if exists)
Hmm, if there is nothing that is not in focus, it's not actually a
strategy (by definition). :-)
Gerald
--
Dr. Gerald Pfeifer
On Wednesday 2010-08-11 01:23, Gerald Pfeifer wrote:
On Tue, 10 Aug 2010, Andreas Jaeger wrote:
** support for the three most recent releases
Three? I count two plus an extra two months for the oldest when a new one comes out (that is, two for 75% of the time, three for the other 25%).
Yeah, I noticed the counting mistake. The period of 18 months drops one release halfway through time, which is very different from what some other distros do (namely, obsoleting an old version at, or shortly thereafter, the blink of a new release; but not phasing one out in the midst of time-nowhere.)
=== As project, we will not focus on the following ===
(fill in if exists)
Hmm, if there is nothing that is not in focus, it's not actually a strategy (by definition). :-)
Well, I thought it would be obvious that everything that is not explicitly listed as being in focus (such as KDE/end users) would automatically fall under no-focus. I really had hoped to spare The Obvious: === Will not focus === - beyond KDE/GNOME and end users - antique and baroque hardware - mobile environment - equipping zypper with a cow "zypper moo" command that returns a hedgehog(?). -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Wednesday 11 August 2010 01:49:54 Jan Engelhardt wrote:
On Wednesday 2010-08-11 01:23, Gerald Pfeifer wrote:
On Tue, 10 Aug 2010, Andreas Jaeger wrote:
** support for the three most recent releases
Three? I count two plus an extra two months for the oldest when a new one comes out (that is, two for 75% of the time, three for the other 25%).
Yeah, I noticed the counting mistake. The period of 18 months drops one release halfway through time, which is very different from what some other distros do (namely, obsoleting an old version at, or shortly thereafter, the blink of a new release; but not phasing one out in the midst of time-nowhere.)
18 months means 2 months overlap, so that's IMO phasing out, Andreas -- Andreas Jaeger, Program Manager openSUSE, aj@{novell.com,opensuse.org} Twitter: jaegerandi | Identica: jaegerandi SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
On Wednesday 2010-08-11 11:45, Andreas Jaeger wrote:
On Wednesday 11 August 2010 01:49:54 Jan Engelhardt wrote:
On Wednesday 2010-08-11 01:23, Gerald Pfeifer wrote:
On Tue, 10 Aug 2010, Andreas Jaeger wrote:
** support for the three most recent releases
Three? I count two plus an extra two months for the oldest when a new one comes out (that is, two for 75% of the time, three for the other 25%).
Yeah, I noticed the counting mistake. The period of 18 months drops one release halfway through time, which is very different from what some other distros do (namely, obsoleting an old version at, or shortly thereafter, the blink of a new release; but not phasing one out in the midst of time-nowhere.)
18 months means 2 months overlap, so that's IMO phasing out,
Indeed. Graphing helps, thus I updated http://en.wikipedia.org/wiki/Template:Timeline_openSUSE with approximations of 11.4 and future. Mh. Our timeline looks so fedoraish now, with the 3-overlap time being so overly short. -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Tue, 2010-08-10 at 09:32 +0200, Andreas Jaeger wrote:
Today we continue with public discussions about strategy proposals, this time with the "Status quo" strategy proposal:
---8<------8<------8<------8<------8<------8<---
== Statement ==
We deliver a well-balanced GNU/Linux platform for modern computers (workstation, laptop, netbook, server) that equally appeals to end users, power users, developers and server/network infrastructure administrators. It shall protrude professionally and let the user be productive.
== Background ==
This strategy tries to quantify what we tried to do in the past — as it was not written down earlier.
So, this is what most users expect from openSUSE today, but does not give a vision for change looking forward.
In the context of other distributions, we differentiate ourselves from Ubuntu targeting the newbie and further differentiate from Fedora being experimental bleeding edge — instead we pick "the middle ground".
== Key ideas ==
* Creation of a general purpose distribution that ** anyone can use without too much effort ** is known for good quality (stable and usable but neither outdated nor bleeding edge) ** has good and sane defaults so the user can do what s/he wants to do ** has programs that work out of the box ** focused on modern hardware and their use cases (workstations, laptops, netbooks and servers) ** is targeted towards end users, but is reasonably equally usable for other workloads * Critical analysis of hyped items before inclusion
== Activities ==
=== We need to be excellent in the following ===
* Do as we always did! That is, ** good compromise between actuality and stability ** agreeable release cycle of 8 months ** support for the three most recent releases * Supporting our target customers ** End users: *** Delivering multiple desktops, focusing on both GNOME and KDE *** Focus on providing tools for being productive and creative (IDEs, editors, authoring tools, graphics manipulation, office productivity, etc.) ** Developers: *** Development environments for especially C, C++, Perl, Python, Java, Ruby: IDEs, tools and support libraries ** Power users and system administrators: *** Providing admin tools that are powerful yet (reasonably) easy *** Agreeable command line experience *** Virtualization technique, e.g. KVM, Xen *** Standard networking services * Continue the naturally growth of openSUSE:Factory by incorporating contributors' submissions.
=== We will try to do the following effectively ===
* Innovate and keep up with latest upstream developments. * Include a more minimalistic desktop environment. * Provide a low entry barrier for potential contributors. With the openSUSE Build Service, it is easier to make contributions than any other Linux distribution to date. * Offer easy creation of specialized install media (appliances) through SUSE Studio. * Good presentation and marketing, in particular communicating our existing strengths and unique features (i.e. competitive advantages). * The usual niceties: speed, less bloat, possibility of minimality.
=== As project, we will not focus on the following ===
(fill in if exists)
The lack of "what we won't do" is a fundamental problem with this proposal. By not defining what it is that does not fit within our priorities, we effectively lose any strategic purpose to the proposal. And we end up not moving forward. All of the above is good at highlighting some of the strengths that exist within our Project, but it also highlights a "everything under the sun" approach. What's important to note here is that no proposal is going to exclude anything that someone wants to do. A strategy can be very broad or very narrow, depending on what all parties agree with and can in fact provide the resources for. So, in the above listed items, its basically concluded that we reach a middle ground. Middle ground implies that we make certain sacrifices here or there. That's fine. That's actually what a strategy statement is supposed to do. But this proposal does not define *what* the middle ground sacrifices are. This Status Quo proposal has some merits and supports some personal beliefs I have about the Project's benefits. But, until it defines clearly what openSUSE is all about to the world, I cannot support this proposal in its currently written form. If you go back to the first paragraph of the proposal (the Statement), its simply a claim that just about every other distro makes. If a new person to Linux is presented with three different distro CD's and has to pick one to install, what is our argument why openSUSE should be picked? Or more importantly, if a new Linux user is approached by three different openSUSE advocates separately, would the message each person gives to the new user the same? Would it be different? Or would it be different but have some common thread embedded within that message? I think we all agree that there isn't a consistent message, and that sometimes when we talk to each other about what we think openSUSE means, we think... "What is that person talking about?? That's not what openSUSE is about." And thus, a strategy statement not only gives some focus and direction to what we are doing on the technical/community side, but also unifies us in how we communicate to the world what openSUSE is about. This proposal, in its current written form doesn't achieve that. But it could, if it was re-written somehow. Bryen -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Wednesday 2010-08-11 17:19, Bryen M. Yunashko wrote:
=== As project, we will not focus on the following ===
(fill in if exists)
The lack of "what we won't do" is a fundamental problem with this proposal.
It is an assessment of what openSUSE used to do. Much of that happened behind closed doors just years ago so excuse me if I could not type up everything. For example, assume "not focusing on power users anymore" would be on the list. I would dismiss such a claim, because I, as a power user, feel having been satisfied over the years, as such, openSUSE can't possibly have not focused on power users. Sure there are areas openSUSE does not focus on, but I could not quantify them easily. And that's why it says (fill in) rather than (there's nothing here). And, as I probably mentioned elsewhere already, consider everything that is not explicitly listed as focus and strength to be implicitly listed under non-focus, but it seems, that, too, has been overread. Since I had originally come up with it, it obviously is a little biased, and some people already joined efforts to merge, streamline, unbias, whatever. If you haven't yet, it maybe time to do so. I suck at politics, and that's probably good so. There is this wiki page right now, and in a developer world you simply send a patch - or in this case, a fragment of an antiproposal for a single point - and we will see.
This Status Quo proposal has some merits and supports some personal beliefs I have about the Project's benefits. But, until it defines clearly what openSUSE is all about to the world, I cannot support this proposal in its currently written form.
If you go back to the first paragraph of the proposal (the Statement), its simply a claim that just about every other distro makes. If a new person to Linux is presented with three different distro CD's and has to pick one to install, what is our argument why openSUSE should be picked?
My simple stance[1] is my personal pasttime experience: "Because I run it, because I develop with it, and outlook for free support from me is dim for others. Because Ubuntu drives me nuts and shows that you might have not carefully enough considered all options." That alone was sufficient to turn my surrounding world to openSUSE. [1] http://news.opensuse.org/2009/04/23/people-of-opensuse-jan-engelhardt/
Or more importantly, if a new Linux user is approached by three different openSUSE advocates separately, would the message each person gives to the new user the same? Would it be different? Or would it be different but have some common thread embedded within that message?
I am sure two of the advocates would try to /influence/ this user with some persuasing techniques ("give it a try", "it's free"). My stance is clearly on Rule Of Acquisition #190 ("Hear all, trust nothing.") So I tell them what openSUSE has and where its strengths and weaknesses are. Then the user has to choose where he thinks he best fits in. Yes, it's an elitist view which seems to be completely disjunct from this proposal, but I am fine with going with this proposal (or the poweruser one) simply because it allows me to have my share of "expressional freedom" (lets me do what I want).
I think we all agree that there isn't a consistent message, and that sometimes when we talk to each other about what we think openSUSE means, we think... "What is that person talking about?? That's not what openSUSE is about."
Is it about green fields and backgrounds?
This proposal, in its current written form doesn't achieve that. But it could, if it was re-written somehow.
Patches welcome. ;) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Wed, 2010-08-11 at 10:19 -0500, Bryen M. Yunashko wrote:
On Tue, 2010-08-10 at 09:32 +0200, Andreas Jaeger wrote:
Today we continue with public discussions about strategy proposals, this time with the "Status quo" strategy proposal:
---8<------8<------8<------8<------8<------8<---
== Statement ==
We deliver a well-balanced GNU/Linux platform for modern computers (workstation, laptop, netbook, server) that equally appeals to end users, power users, developers and server/network infrastructure administrators. It shall protrude professionally and let the user be productive.
== Background ==
This strategy tries to quantify what we tried to do in the past — as it was not written down earlier.
So, this is what most users expect from openSUSE today, but does not give a vision for change looking forward.
In the context of other distributions, we differentiate ourselves from Ubuntu targeting the newbie and further differentiate from Fedora being experimental bleeding edge — instead we pick "the middle ground".
== Key ideas ==
* Creation of a general purpose distribution that ** anyone can use without too much effort ** is known for good quality (stable and usable but neither outdated nor bleeding edge) ** has good and sane defaults so the user can do what s/he wants to do ** has programs that work out of the box ** focused on modern hardware and their use cases (workstations, laptops, netbooks and servers) ** is targeted towards end users, but is reasonably equally usable for other workloads * Critical analysis of hyped items before inclusion
== Activities ==
=== We need to be excellent in the following ===
* Do as we always did! That is, ** good compromise between actuality and stability ** agreeable release cycle of 8 months ** support for the three most recent releases * Supporting our target customers ** End users: *** Delivering multiple desktops, focusing on both GNOME and KDE *** Focus on providing tools for being productive and creative (IDEs, editors, authoring tools, graphics manipulation, office productivity, etc.) ** Developers: *** Development environments for especially C, C++, Perl, Python, Java, Ruby: IDEs, tools and support libraries ** Power users and system administrators: *** Providing admin tools that are powerful yet (reasonably) easy *** Agreeable command line experience *** Virtualization technique, e.g. KVM, Xen *** Standard networking services * Continue the naturally growth of openSUSE:Factory by incorporating contributors' submissions.
Hello All, Please excuse me if I'm late to the party and a little out of context. I would like to see a section that specifically identifies the elements of the core operating system (kernel,udev,package managment) that must be rock solid. And honestly, that is THE reason I subscribed to this group. I originally choose to use SuSE back in 2000 because it was rock solid, well integrated, and worked with most hardware. These are the key reasons I kept using SuSE, privately and professionally, even though I never used KDE and gnome has often been a weak spot for SuSE. I think the core qualities of OpenSUSE have slipped since the 10.3 release. I would REALLY like to see a release that re-focuses on those core qualities (once identified) above all else. IMO, until some of the lingering items from the last several releases are clearly identified and solved, until the core qualities that made SuSE a great product to begin with are rediscovered, until a succinct and specific set of features that define the OpenSUSE OS are identified and perfected, goals such as "has programs that work out of the box" won't get me or the folks coding the next release overly enthused. I don't know who participates in this group. How about a definition of OpenSUSE that looks like it was written by engineers and conveys a defacto set of requirements with a force that marketing phrases can't begin to conjure? Is OpenSUSE suffering some kind of corporate "identity crisis"? Why would OpenSUSE want to define itself in terms resembling "somewhere in-between Ubuntu and Red Hat" Cheers! Charles Wight -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
jdd wrote:
some ways we could:
* state clearly that we wont support hardware configs more than 5 (or 7) years old (of course teams can work on other versions) - say PIV, 256Mo ram, 20Gb HDD
With such a statement, you have quite accurately ruled out a lot of server installations. -- Per Jessen, Zürich (17.2°C) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
Le 18/08/2010 10:51, Per Jessen a écrit :
jdd wrote:
some ways we could:
* state clearly that we wont support hardware configs more than 5 (or 7) years old (of course teams can work on other versions) - say PIV, 256Mo ram, 20Gb HDD
With such a statement, you have quite accurately ruled out a lot of server installations.
do you expect more than 7 years old servers to install the very last distro? I have some of them, always running the same distro for 5 years now :-). it's already very difficult to install 11.3 on 256Mb ram computers, I don't expect to be able to do so with 11.4 but in fact the years number don't really matters, what matters is to say when we leave a hardware out. Supporting very old hardawre is not that easy (and the kernel already let alone some, including SCSI stuff - and always did) jdd -- Jean-Daniel Dodin -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
Jean-Daniel Dodin wrote:
Le 18/08/2010 10:51, Per Jessen a écrit :
jdd wrote:
some ways we could:
* state clearly that we wont support hardware configs more than 5 (or 7) years old (of course teams can work on other versions) - say PIV, 256Mo ram, 20Gb HDD
With such a statement, you have quite accurately ruled out a lot of server installations.
do you expect more than 7 years old servers to install the very last distro?
Sure I do, and why not? Server hardware does not change as often as desktop ditto. I ran most of the test-versions of 11.3 (and 10.x for that matter) on a Compaq test-server from around 2000. When I reported an issue in the PCI hotplug code, it was also attended to right away.
it's already very difficult to install 11.3 on 256Mb ram computers, I don't expect to be able to do so with 11.4
Amount of RAM is not directly related to the age of a server (or desktop). My three year old workstation has 4Gb of memory, but my ten year old server has 8Gb.
but in fact the years number don't really matters, what matters is to say when we leave a hardware out. Supporting very old hardawre is not that easy (and the kernel already let alone some, including SCSI stuff - and always did)
I think the really important thing is to accurately define what we mean when we say "support". For instance, does "not supported" equal "will not work"? As we know, no it doesn't, nor does it mean that a bug reported on old hardware will not be looked at. IMHO, it's perfectly acceptable to adjust the minimum requirements, but that is not the same as explicitly excluding "support" for hardware from before 200x. -- Per Jessen, Zürich (17.9°C) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Wednesday 2010-08-18 11:02, Jean-Daniel Dodin wrote:
* state clearly that we wont support hardware configs more than 5 (or 7) years old (of course teams can work on other versions) - say PIV, 256Mo ram, 20Gb HDD
With such a statement, you have quite accurately ruled out a lot of server installations.
do you expect more than 7 years old servers to install the very last distro?
Absolutely. Let me remind you, 7 years, that's 2003ish, that's the MHz wars, that's AMD Athlon XP-ish. And it runs 11.2. So yes, I damn expect it. Yes it was originally shipped with 256MB (and 512 for hardcore gamers) back then, but upgraded to 1G since I am heavily working with linux and git, where cache is good. Needless to say even with 256MB it runs. Unless you try to run kde and candybars, that is enough for ye basic home-server-in-a-cellar.
it's already very difficult to install 11.3 on 256Mb ram computers, I don't expect to be able to do so with 11.4
but in fact the years number don't really matters, what matters is to say when we leave a hardware out. Supporting very old hardawre is not that easy (and the kernel already let alone some, including SCSI stuff - and always did)
Huh? -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
Le 18/08/2010 11:27, Per Jessen a écrit :
I think the really important thing is to accurately define what we mean when we say "support". For instance, does "not supported" equal "will not work"? As we know, no it doesn't, nor does it mean that a bug reported on old hardware will not be looked at. IMHO, it's perfectly acceptable to adjust the minimum requirements, but that is not the same as explicitly excluding "support" for hardware from before 200x.
The strategy discussion was started assuming we don't have the ressources to do all and every thing My feeling is than we have to leave alone some old hardware. I know there are many on the wild (I just use a at least 5 years old computer), but these are usually very difficult to use with recent distro and, by the way, don't need them. Why Change distro on old hardware? That's said, it's true on desktop/server (that is all but laptop) it's not easy to give a date, given many component can be changed anytime. So it's Ok to speak about power. For example, it's pretty difficult to run modern distro on 486... How can we do? Hardware is only supported if somebody in the community willing to fix have one to test :-( jdd -- http://www.dodin.net http://pizzanetti.fr -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Wednesday 2010-08-18 13:10, jdd wrote:
I think the really important thing is to accurately define what we mean when we say "support". For instance, does "not supported" equal "will not work"? As we know, no it doesn't, nor does it mean that a bug reported on old hardware will not be looked at. IMHO, it's perfectly acceptable to adjust the minimum requirements, but that is not the same as explicitly excluding "support" for hardware from before 200x.
The strategy discussion was started assuming we don't have the ressources to do all and every thing
Compatibility for old hardware comes at a very low price. The strategy cuts somewhere otherwise we get into circular discussions again. ("Focus does not rule out the existence of other areas of interest.") -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
jdd wrote:
Le 18/08/2010 11:27, Per Jessen a écrit :
I think the really important thing is to accurately define what we mean when we say "support". For instance, does "not supported" equal "will not work"? As we know, no it doesn't, nor does it mean that a bug reported on old hardware will not be looked at. IMHO, it's perfectly acceptable to adjust the minimum requirements, but that is not the same as explicitly excluding "support" for hardware from before 200x.
The strategy discussion was started assuming we don't have the ressources to do all and every thing
Which is certainly true.
My feeling is than we have to leave alone some old hardware. I know there are many on the wild (I just use a at least 5 years old computer), but these are usually very difficult to use with recent distro and, by the way, don't need them.
My experience is quite the opposite. So far I have only had minor issues in running openSUSE 10 and 11 on older HP Proliants (up to 10 years).
Why Change distro on old hardware?
It doesn't have to be a change, it could be a new install. The other day I bought 10 second-hand HP Proliants all from around 2004-2005.
That's said, it's true on desktop/server (that is all but laptop) it's not easy to give a date, given many component can be changed anytime.
So it's Ok to speak about power. For example, it's pretty difficult to run modern distro on 486...
Really only because openSUSE is built for i586 - otherwise there is no problem. -- Per Jessen, Zürich (18.9°C) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Wednesday 2010-08-18 13:57, Per Jessen wrote:
That's said, it's true on desktop/server (that is all but laptop) it's not easy to give a date, given many component can be changed anytime.
So it's Ok to speak about power. For example, it's pretty difficult to run modern distro on 486...
Really only because openSUSE is built for i586 - otherwise there is no problem.
And, as I've probably mentioned a number of times elsewhere already, these packages also work on i386 - provided that you do have a native kernel-default.i386 and glibc.i386. Everything else can be i586. -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Wednesday 11 August 2010 19:00:45 Charles Wight wrote:
On Wed, 2010-08-11 at 10:19 -0500, Bryen M. Yunashko wrote:
On Tue, 2010-08-10 at 09:32 +0200, Andreas Jaeger wrote:
Today we continue with public discussions about strategy proposals, this time with the "Status quo" strategy proposal:
---8<------8<------8<------8<------8<------8<---
== Statement ==
We deliver a well-balanced GNU/Linux platform for modern computers (workstation, laptop, netbook, server) that equally appeals to end users, power users, developers and server/network infrastructure administrators. It shall protrude professionally and let the user be productive.
== Background ==
This strategy tries to quantify what we tried to do in the past — as it was not written down earlier.
So, this is what most users expect from openSUSE today, but does not give a vision for change looking forward.
In the context of other distributions, we differentiate ourselves from Ubuntu targeting the newbie and further differentiate from Fedora being experimental bleeding edge — instead we pick "the middle ground".
== Key ideas ==
* Creation of a general purpose distribution that ** anyone can use without too much effort ** is known for good quality (stable and usable but neither outdated nor bleeding edge) ** has good and sane defaults so the user can do what s/he wants to do ** has programs that work out of the box ** focused on modern hardware and their use cases (workstations, laptops, netbooks and servers) ** is targeted towards end users, but is reasonably equally usable for other workloads * Critical analysis of hyped items before inclusion
== Activities ==
=== We need to be excellent in the following ===
* Do as we always did! That is, ** good compromise between actuality and stability ** agreeable release cycle of 8 months ** support for the three most recent releases * Supporting our target customers ** End users: *** Delivering multiple desktops, focusing on both GNOME and KDE *** Focus on providing tools for being productive and creative (IDEs, editors, authoring tools, graphics manipulation, office productivity, etc.) ** Developers: *** Development environments for especially C, C++, Perl, Python, Java, Ruby: IDEs, tools and support libraries ** Power users and system administrators: *** Providing admin tools that are powerful yet (reasonably) easy *** Agreeable command line experience *** Virtualization technique, e.g. KVM, Xen *** Standard networking services * Continue the naturally growth of openSUSE:Factory by incorporating contributors' submissions.
Hello All,
Please excuse me if I'm late to the party and a little out of context.
I would like to see a section that specifically identifies the elements of the core operating system (kernel,udev,package managment) that must be rock solid. And honestly, that is THE reason I subscribed to this group.
I originally choose to use SuSE back in 2000 because it was rock solid, well integrated, and worked with most hardware. These are the key reasons I kept using SuSE, privately and professionally, even though I never used KDE and gnome has often been a weak spot for SuSE. I think the core qualities of OpenSUSE have slipped since the 10.3 release. I would REALLY like to see a release that re-focuses on those core qualities (once identified) above all else. IMO, until some of the lingering items from the last several releases are clearly identified and solved, until the core qualities that made SuSE a great product to begin with are rediscovered, until a succinct and specific set of features that define the OpenSUSE OS are identified and perfected, goals such as "has programs that work out of the box" won't get me or the folks coding the next release overly enthused.
I don't know who participates in this group. How about a definition of OpenSUSE that looks like it was written by engineers and conveys a defacto set of requirements with a force that marketing phrases can't begin to conjure?
Is OpenSUSE suffering some kind of corporate "identity crisis"?
Good question, and imho, yes, to some extend it does. Some people feel it should focus more on newbies; others prefer a focus on powerusers. Some say - we should shorten the release cycle. Others want long(er) term support. Some want more up-to-date packages. Others ask for more stability. Somehow we have to define an answer to those questions, define where we stand and what we want. Seeing how popular some options seem to be, imho this proposal, the poweruser one and the focus on developer all seem to point in the right direction.
Why would OpenSUSE want to define itself in terms resembling "somewhere in-between Ubuntu and Red Hat"
Agreed, we need our own story, not a 'we're inbetween that and that distro'. But looking at the users they target can at least make clear where we are different. Fedora focusses strongly on the 'I want the latest, shineyest things' people. Ubuntu goes for anyone who knows little about computers and doesn't want to put in efforts. openSUSE imho seems to focus currently more on what you could call a productive poweruser - somebody who is capable and interested in finding out how something works. But at the same time doesn't want to WASTE time - things should work out of the box, but offer flexibility and configuratiblity where needed. Such a person is often developer or system admin, but could also just be a dude or girl who reads computer magazines and is interested in technology in general. Or somebody who works in an office and sits behind a computer all day. Neither Fedora nor Ubuntu cater to those people - either by being too unstable or by dumbing things down to the point where the poweruser is limited in what he/she can do. We can enable these people, and imho that is the path we as community are on - and we should continue to go there. Opinions very much welcome :D
Cheers!
Charles Wight
Le 18/08/2010 17:56, Jos Poortvliet a écrit :
Such a person is often developer or system admin, but could also just be a dude or girl who reads computer magazines and is interested in technology in general. Or somebody who works in an office and sits behind a computer all day. Neither Fedora nor Ubuntu cater to those people - either by being too unstable or by dumbing things down to the point where the poweruser is limited in what he/she can do.
pretty well seen. can you summarize this in max three words :-))) I think what we discuss (and it's good) it to discover what we already do :-)). Making things with word makes feel better and safer :-). But I think most Linux users do this, including Ubuntu and Fedora, the others uses Windows... jdd -- http://www.dodin.net http://pizzanetti.fr -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
Jos Poortvliet wrote:
On Wednesday 11 August 2010 19:00:45 Charles Wight wrote:
[snip] Is OpenSUSE suffering some kind of corporate "identity crisis"?
Good question, and imho, yes, to some extend it does.
I think we do to a great extent. At first we had SuSE Linux made in Germany (very clear, crisp, well-defined product identity), then SUSE Linux by Novell, then openSUSE (not very open), then openSUSE (open and apparently community driven), now openSUSE (open and partially/actually community driven). -- Per Jessen, Zürich (18.9°C) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 [Sent later] On 2010-08-18 18:40, Per Jessen wrote:
Jos Poortvliet wrote:
On Wednesday 11 August 2010 19:00:45 Charles Wight wrote:
[snip] Is OpenSUSE suffering some kind of corporate "identity crisis"?
Good question, and imho, yes, to some extend it does.
I think we do to a great extent. At first we had SuSE Linux made in Germany (very clear, crisp, well-defined product identity), then SUSE Linux by Novell, then openSUSE (not very open), then openSUSE (open and apparently community driven), now openSUSE (open and partially/actually community driven).
What about people living in countries like, for instance, Cuba? They can't even report a bug because they can not log-into Bugzilla, requiring registering at Novell, which refuses because of the laws as USA, I guess... I can understand why Novell has to do that, but do we (openSUSE) need to do that as well? We are commenting on such a case just now, in the Spanish list. A Cuban wants to become an openSUSE ambassador, but can't. We have to admire the difficulties of all sorts he has for installing openSUSE (with our help, folks of the Spanish list). This denial of registration is one more. - -- Cheers / Saludos, Carlos E. R. (from 11.2 x86_64 "Emerald" GM (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.12 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org/ iF4EAREIAAYFAkyIPRUACgkQja8UbcUWM1z/OQD/S1J/a21h4YMage2ttVidgps8 do9gnycSiSusxbn23V4A/0oFujW++Cd1/3RyLjC9SuGMhfTB5zwa7jAVBWJkXpZ2 =iPge -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Thursday 09 September 2010 03:49:09 Carlos E. R. wrote:
[Sent later]
On 2010-08-18 18:40, Per Jessen wrote:
Jos Poortvliet wrote:
On Wednesday 11 August 2010 19:00:45 Charles Wight wrote:
[snip] Is OpenSUSE suffering some kind of corporate "identity crisis"?
Good question, and imho, yes, to some extend it does.
I think we do to a great extent. At first we had SuSE Linux made in Germany (very clear, crisp, well-defined product identity), then SUSE Linux by Novell, then openSUSE (not very open), then openSUSE (open and apparently community driven), now openSUSE (open and partially/actually community driven).
What about people living in countries like, for instance, Cuba? They can't even report a bug because they can not log-into Bugzilla, requiring registering at Novell, which refuses because of the laws as USA, I guess...
I can understand why Novell has to do that, but do we (openSUSE) need to do that as well?
We are commenting on such a case just now, in the Spanish list. A Cuban wants to become an openSUSE ambassador, but can't. We have to admire the difficulties of all sorts he has for installing openSUSE (with our help, folks of the Spanish list). This denial of registration is one more.
Interesting issue. Yes, as long as the relationship between openSUSE and Novell is as it currently is, this is hard to solve. Another thing the foundation can help with!
Le 13/09/2010 21:23, Jos Poortvliet a écrit :
Interesting issue. Yes, as long as the relationship between openSUSE and Novell is as it currently is, this is hard to solve. Another thing the foundation can help with!
make the foundation live in man island, as canonical do :-)) jdd -- http://www.dodin.net http://pizzanetti.fr -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
Jos Poortvliet wrote:
On Thursday 09 September 2010 03:49:09 Carlos E. R. wrote:
[Sent later]
Jos Poortvliet wrote:
On Wednesday 11 August 2010 19:00:45 Charles Wight wrote:
[snip] Is OpenSUSE suffering some kind of corporate "identity crisis"? Good question, and imho, yes, to some extend it does. I think we do to a great extent. At first we had SuSE Linux made in Germany (very clear, crisp, well-defined product identity), then SUSE Linux by Novell, then openSUSE (not very open), then openSUSE (open and apparently community driven), now openSUSE (open and partially/actually community driven). What about people living in countries like, for instance, Cuba? They can't even report a bug because
On 2010-08-18 18:40, Per Jessen wrote: they can not log-into Bugzilla, requiring registering at Novell, which refuses because of the laws as USA, I guess...
I can understand why Novell has to do that, but do we (openSUSE) need to do that as well?
We are commenting on such a case just now, in the Spanish list. A Cuban wants to become an openSUSE ambassador, but can't. We have to admire the difficulties of all sorts he has for installing openSUSE (with our help, folks of the Spanish list). This denial of registration is one more.
Interesting issue. Yes, as long as the relationship between openSUSE and Novell is as it currently is, this is hard to solve. Another thing the foundation can help with!
personally, i have never understood the need for the open community to tie itself to Novell's iChain.. sure it _is_ convenient for the Novell employees and SLE_ owners who happen to already need Novell credentials to not have to be bothered with a completely separate system of access management (located in a place not encumbered with USA laws/politics).. i do not know: does Red Hat also market an access management application and tie the Fedora Community to it.. or, could the community not administer access totally independent of Novell? just questions, DenverD -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Monday 2010-09-13 22:34, jdd wrote:
Le 13/09/2010 21:23, Jos Poortvliet a écrit :
Interesting issue. Yes, as long as the relationship between openSUSE and Novell is as it currently is, this is hard to solve. Another thing the foundation can help with!
make the foundation live in man island, as canonical do :-))
www.opensuse.cx! Oh wait, that was Christmas Island... -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
On Tuesday 14 September 2010 09:30:08 DenverD wrote:
[...] personally, i have never understood the need for the open community to tie itself to Novell's iChain..
sure it _is_ convenient for the Novell employees and SLE_ owners who happen to already need Novell credentials to not have to be bothered with a completely separate system of access management (located in a place not encumbered with USA laws/politics)..
i do not know: does Red Hat also market an access management application and tie the Fedora Community to it..
or, could the community not administer access totally independent of Novell?
When we started the project, we needed iChain for bugzilla and it was natural to use it else where then as well. I think that the new connect.opensuse.org can help us to develop our own authorization infrastructure. Right now this is not the highest priority for those working on it but if some people want to help with that... But before we start developing, I suggest to write up what we want and how to do it - and answer questions like: Shall we go compeletly to openID? How will we access bugzilla? etc Andreas -- Andreas Jaeger, Program Manager openSUSE, aj@{novell.com,opensuse.org} Twitter: jaegerandi | Identica: jaegerandi SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126 -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org
participants (13)
-
Andreas Jaeger -
Bryen M. Yunashko -
Carlos E. R. -
Charles Wight -
DenverD -
Gerald Pfeifer -
Guido Berhoerster -
Jan Engelhardt -
jdd -
Jean-Daniel Dodin
-
Jos Poortvliet -
Martin Schlander -
Per Jessen