[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> Certainly we could list enough objectives that we > could look at this as a > failure, but as I've pointed out, Sun has made huge > strides to open the > sources up and there has been participation from > folks that have the > knowledge and skills to do it. Juergen Keil a case in > point. Even before the > sources were out, Juergen didn't let that from > stopping him from devising > patches, some of which patched memory at run > time! I'm sure he's much > happier to have the sources today. Juergen Keil is a genius. I remember working with him on the ATAPI DMA issue back in the days of Solaris 9, and him giving me a commented portion of a reverse-engineered ATA driver. I used this commented text to locate and patch the ATA driver in mdb and after that, DMA worked. Simply an outstanding guy with very deep knowledge of the Solaris kernel. Too bad there aren't more people like him. I wouldn't lose any sleep over Juergen doing putbacks to the OpenSolaris source code; that guy knows what is going on! Isn't he the outside OpenSolaris contributor with most kernel fixes to date? This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> I think that is a negative way of looking at things. > It is all about improvements - so if there are less > Solaris kernel developers the idea should be to > create more of them and not be complacent about it. Developers at any cost will not help OpenSolaris. Software development is NOT like masonry -- "as many masons -- that much wall". That does not work for software development. Actually what we *really* need is the application developers to start developing on Solaris as their primary development platform instead of Linux. And for the most part, they need not concern themselves with the internals. Solaris is well enough designed, and it shouldn't be necessary. And for that, one doesn't need concern oneself with the engineering processes behind OpenSolaris. > (Last I checked there were 1500+ routinely active > Linux kernel developers - many of them self taught > and without a prior kernel development background - > they started fixing, their changes were taken in and > they were not bogged down by bureaucracy ) "Prior kernel development"? > I would consider that as a wasted opportunity and > potential. If that is your level of confidence then I > have to wonder why did Sun open it in first place > and named it "Open"Solaris? What you said > contradicts the definition of open in free software > parlance - specifically because you discount > collaboration and improvement and block freedom of > use - they way I want it, where I want it. OpenSource means that the source code is freely available. Nothing more, nothing less. And OpenSolaris code is freely available to anyone willing to download it or even just browse it at one's leisure. > Sorry but I already pointed out a key thing before - > people are not going to participate unless you give > them a right platform, unless they feel owning and > driving the changes. People are not going to work for > Sun (disguised as community) according to Sun's > rules, for Sun's purpose. I am going to participate no matter what! Solaris has been with me since I was a kid, many years ago, and has *never* let me down. A great operating system truly worthy of respect, in every technical sense. I don't need any kind of a GPL license "to sweeten the deal", nor will "red tape", as you call it, change that sentiment. I truly love Solaris for what it is, not what someone else expects it should be. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
Frank Hofmann <[EMAIL PROTECTED]> wrote: > > But more importantly this was never about accepting any and all changes > > - it was about making it better for people to propose changes and people > > to review it and then accept the quality ones. > > > The review/integration process on Linux can be as tedious as on Solaris, > and worse. Projects as: Linux is definitely worse here. On Solaris you get things in if you follow the rules. On Linux, you finish everything and then get a note from Linus that he does not like your project. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> Sun and may even explain the degradation of quality > once Sun tries to > ship it (re: JDS versus normal Gnome) but this is > hardly appropriate > for an Open Source project. I fail to understand why OpenSolaris would have to behave as some other open source projects. Why must we mimic the Linux development model? Why can't we be allowed to do development and participation in our own way? Why is the "Linux way" the only correct way, and why are all the other ways wrong no matter what? This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> No sir. They do not accept Solaris because firstly > they believe freedom is priceless and that a > for-profit company in drivers seat driving things the > deem fit, there cannot be freedom and no one likes to > work for free for somebody else's cause. Secondly > most use x86 and Solaris won't work there and if they > are to fix it they would have to go thru some good > amount of red tapism, instead of just fixing it and > integrating it in the main tree. Excuse me, but how is it OK that the Fedora project effectively works for RedHat *for free*, where RedHat gets to pick and choose what flows back into RedHat AS and ES, but it's *not OKay* when Sun wants to do the same? Or is the difference really that the Fedora community gets to mess up Fedora core really bad, which has happened several times so far, and the OpenSolaris community remains protective of OpenSolaris? It seems to me, that that's your beef really. You don't like the fact that there are processes established which effectively prevent messing up OpenSolaris everwhichway one pleases, and you also seem to have beef with the fact that most of the OpenSolaris community are Sun engineers which have quite a bit of influence on the process. Personally, I have no problem with that; when I decided to be part of the OpenSolaris community, I accepted that, and in fact, I'm grateful that that is the state of affairs; I don't want OpenSolaris screwed up just because someone wants to scratch their immediate itch, and compatibility be damned. No sir, I don't want that, no ifs, no buts, and no maybes. And just for the record, I am (unfortunately) in no way associated with Sun Microsystems. > Fix both of the above and do away with your > protectionist and wrong ways and you will see a surge > in usage and participation. I have more i86pc systems running Solaris 10 than I have SPARC systems running the aforementioned OS and revision. I believe that speaks for itself quite clearly apropos "Solaris not running on x86". Not only does Solaris 10 (which is quite behind Nevada) run well on *generic* x86 hardware. it *screams* how fast and how good it runs. I'd sure like to know which issue(s) you had with Solaris so that it doesn't run. So far I've either missed you mentioning any details, or you haven't given any. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
On Sat, 3 Feb 2007, S Destika wrote: Well there is only one Linux the kernel which Linus releases. All other changes are development branches and eventually all acceptable stuff gets merged in mainline. I don't think you understand how Linux development works at all. "eventually all acceptable stuff gets merged [ ...]"... Do you really believe what you write there ? If so, then you need to explain "acceptable", and explain in what sense "acceptable" on Linux is easier to achieve for big project foldbacks / commercially-motivated development than you claim it is on Solaris ... But more importantly this was never about accepting any and all changes - it was about making it better for people to propose changes and people to review it and then accept the quality ones. > The review/integration process on Linux can be as tedious as on Solaris, and worse. Projects as: - ReiserFS4 - Linux support for the ToUCam webcam family - MadWIFI / the Atheros driver - Xen - lkcd have been at the 'bad end' of "can it be integrated" "why can it not be integrated" "it needs to change to integrate" set of arguments, sometimes for years. There are bone-headed discussions about code quality, [ bad ] code reuse, licensing issues, GPL enforcement/enforcibility. Just as we have on OpenSolaris. In that sense, this sort-of-flamewar has just shown we've already caught up. As far as the 0.02c goes: Until I can freely exchange, both ways, code between the Solaris and Linux kernels, without having to worry or even know about licensing particulars, the "interoperability" issue between the two isn't solved. This isn't a one-way road. If I want to port Linux driver code to Solaris, I can only do so if the license on the Linux sources allows me to integrate them into a Solaris source assembly of whatever sort. If I want to port ZFS to Linux I can only do so if the Solaris source license allows me to integrate the result into a Linux source assembly of sorts. Now if you can explain in which way the GPLv3 allows for this sort of two-way free exchange of sourcecode, I'd be happy to hear and more than happy to give a +1 for such a change. But does it really ? If a license change / dual licensing to GPLv3 doesn't achieve a path for two-way exchange of code, Solaris has nothing to gain from it, and Linux has nothing to gain from it. It'd be a marketing spin. Or what ? FrankH. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
"Josh Hurst" <[EMAIL PROTECTED]> wrote: > On 2/5/07, Joerg Schilling <[EMAIL PROTECTED]> wrote: > > "Josh Hurst" <[EMAIL PROTECTED]> wrote: > > > > > > Hmmm, how did JDS/gnome come up here??? Since it has, I assume > > > > you don't like the look & feel of JDS, and would prefer Gnome. This > > > > is a subjective view rather than a "degradation of quality" or > > > > inappropriate > > > > for a Open Source project. > > > > > > Few examples: > > > Linux+Gnome is much faster than Solaris+JDS on the same AMD hardware. > > > JDS+Solaris needs more memory than Linux+Gnome (same hardware). 512MB > > > are enough for Linux, Solaris starts swapping after a few hours of > > > usage and doesn't stop until swap is full. JDS doesn't grok Gnome > > > session files from Linux. > > > > So you are missinterpreting things The different behavoir is caused > > by Netscape and your Netscape usage > > The browser choice does not explain the problems with session files, > memory usage and other desktop performance problems. And I'm using > FireFox and not Netscape. Funny statement! You are using netscaspe code that if full of memory leaks. Whether and how fast it grows depends on what pages you visit and your problem is obviously caused by a memory leak from the netscape code. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
On 2/5/07, Joerg Schilling <[EMAIL PROTECTED]> wrote: "Josh Hurst" <[EMAIL PROTECTED]> wrote: > > Hmmm, how did JDS/gnome come up here??? Since it has, I assume > > you don't like the look & feel of JDS, and would prefer Gnome. This > > is a subjective view rather than a "degradation of quality" or inappropriate > > for a Open Source project. > > Few examples: > Linux+Gnome is much faster than Solaris+JDS on the same AMD hardware. > JDS+Solaris needs more memory than Linux+Gnome (same hardware). 512MB > are enough for Linux, Solaris starts swapping after a few hours of > usage and doesn't stop until swap is full. JDS doesn't grok Gnome > session files from Linux. So you are missinterpreting things The different behavoir is caused by Netscape and your Netscape usage The browser choice does not explain the problems with session files, memory usage and other desktop performance problems. And I'm using FireFox and not Netscape. Josh ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
"Josh Hurst" <[EMAIL PROTECTED]> wrote: > > Hmmm, how did JDS/gnome come up here??? Since it has, I assume > > you don't like the look & feel of JDS, and would prefer Gnome. This > > is a subjective view rather than a "degradation of quality" or inappropriate > > for a Open Source project. > > Few examples: > Linux+Gnome is much faster than Solaris+JDS on the same AMD hardware. > JDS+Solaris needs more memory than Linux+Gnome (same hardware). 512MB > are enough for Linux, Solaris starts swapping after a few hours of > usage and doesn't stop until swap is full. JDS doesn't grok Gnome > session files from Linux. So you are missinterpreting things The different behavoir is caused by Netscape and your Netscape usage Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
On 2/5/07, Doug Scott <[EMAIL PROTECTED]> wrote: > On 2/2/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> > wrote: > > > > >And how far have the star or ksh projects > progressed? The last one > > >appears to be in serious trouble now because Sun > has to complain about > > >every little detail and the star project makes > either zero progress or > > >no progress announcements. > > > > > > The only problem in the ksh93 project is people who > are not part of > > the project team and who are not participating in > the review and > > who understand bugger all of the processes we > created for OpenSolaris > > and which have worked reasonably well for Sun > internally, butting > > in with inflamatory remarks when there's even the > slightest hint > > of constructive criticism in messages from Sun > employees. > > The major problem with the ksh93 project is Sun > Microsystems who is > adding more and more mindless rules. Once one task > has been finished > Sun always comes up with two more items. Which kind > of cooperation is > this? I really think there are too many rules. They > may work within > Sun and may even explain the degradation of quality > once Sun tries to > ship it (re: JDS versus normal Gnome) but this is > hardly appropriate > for an Open Source project. Hmmm, how did JDS/gnome come up here??? Since it has, I assume you don't like the look & feel of JDS, and would prefer Gnome. This is a subjective view rather than a "degradation of quality" or inappropriate for a Open Source project. Few examples: Linux+Gnome is much faster than Solaris+JDS on the same AMD hardware. JDS+Solaris needs more memory than Linux+Gnome (same hardware). 512MB are enough for Linux, Solaris starts swapping after a few hours of usage and doesn't stop until swap is full. JDS doesn't grok Gnome session files from Linux. Sure, this is s subjective. Let's ignore these problems and add more colorful branding to JDS. Josh ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> On 2/2/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> > wrote: > > > > >And how far have the star or ksh projects > progressed? The last one > > >appears to be in serious trouble now because Sun > has to complain about > > >every little detail and the star project makes > either zero progress or > > >no progress announcements. > > > > > > The only problem in the ksh93 project is people who > are not part of > > the project team and who are not participating in > the review and > > who understand bugger all of the processes we > created for OpenSolaris > > and which have worked reasonably well for Sun > internally, butting > > in with inflamatory remarks when there's even the > slightest hint > > of constructive criticism in messages from Sun > employees. > > The major problem with the ksh93 project is Sun > Microsystems who is > adding more and more mindless rules. Once one task > has been finished > Sun always comes up with two more items. Which kind > of cooperation is > this? I really think there are too many rules. They > may work within > Sun and may even explain the degradation of quality > once Sun tries to > ship it (re: JDS versus normal Gnome) but this is > hardly appropriate > for an Open Source project. Hmmm, how did JDS/gnome come up here??? Since it has, I assume you don't like the look & feel of JDS, and would prefer Gnome. This is a subjective view rather than a "degradation of quality" or inappropriate for a Open Source project. It is mere branding, and you are only a few mouse clicks from changing to suit your tastes. Personally I quite like the new JDS look, and prefer it to the default Gnome look. This though is purely down to what I prefer. I certainly don't think it is something to get upset about By the way when you make quotes like "mindless rules", it really helps your point, if you produce some examples. Otherwise your words look rather irrelevant to the discussion. Doug This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
[...] pax ??? pax is an archiver. http://en.wikipedia.org/wiki/PaX http://www.opengroup.org/onlinepubs/009695399/utilities/pax.html based on the tar format with cpio legacy support. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
I've seen more domineering, tyrannical behavior in "voluntary" neighborhood improvement associations than in for-profit businesses. And I'm quite sure I could find in writing times that one of the Linux bigwigs has in effect said "not only no, but _HELL_ no!" to something. The good thing about making money is that at least it shows someone is willing to spend it for what you offer. The good thing about red tape is it keeps garbage out; once something gets in, it works. That doesn't have to mean that things _have_ to happen all that much slower, it just means you work in a private branch until you're done, resync, and then you better be prepared to show that what you've got is valuable to someone, doesn't break or degrade anything else (or that you've solved that too), and is maintainable. I'd _really_ like to see a count of x86/x64 hardware support new from Solaris 10 on; I bet it's a lot more than those that think x86 is being ignored would believe. Even better would be a bar graph by month, with explanations for some of the peaks (such as if a bunch of drivers were aquired or adopted from the same source pretty much all at once). I'd guess they'd show steady or even increasing progress. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
"Ignacio Marambio Catán" <[EMAIL PROTECTED]> wrote: > On 2/3/07, S Destika <[EMAIL PROTECTED]> wrote: > > Well there is only one Linux the kernel which Linus releases. All other > > changes are development branches and eventually all acceptable stuff gets > > merged in mainline. I don't think you understand how Linux development > > works at all. > > ohh, i think i do, let me give you one more example, check a project > called grsecurity, it is a set of kernel patches to implement PAX, > RBAC and other stuff, that is not in any of the developement branches pax ??? pax is an archiver. http://www.opengroup.org/onlinepubs/009695399/utilities/pax.html based on the tar format with cpio legacy support. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> Because as I have said hundred times or so - the > process is unnecessarily bureaucratic and dictated by > Sun based on their interests instead of community > inspired - not something I can work with. Also it is > too slow and additionally suffers from unfair > prioritization according to Sun's interests - it is > impossible to be of practical use - see all those > Solaris bugs open since 1999. I don't understand. If you work on a bug and request a sponsor, all the sponsor has to do is audit your code. So how does priorization come into play? If you are the one working on something, how fast it gets done depends on you. Or did I miss something? Have you actually tried? This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
On 2/3/07, S Destika <[EMAIL PROTECTED]> wrote: Well there is only one Linux the kernel which Linus releases. All other changes are development branches and eventually all acceptable stuff gets merged in mainline. I don't think you understand how Linux development works at all. ohh, i think i do, let me give you one more example, check a project called grsecurity, it is a set of kernel patches to implement PAX, RBAC and other stuff, that is not in any of the developement branches of the linux kernel and will likely never be there despite having been around for quite some time ( some years ). But more importantly this was never about accepting any and all changes - it was about making it better for people to propose changes and people to review it and then accept the quality ones. i think that is actually how the process works right now, and it will get better over time. Opensolaris is a lot of code and this is a fairly new community with many things to solve, please let it go one step at a time, slowly but safe and steadily. perhaps we could have a more sane conversation about licensing and other core things that need to be addressed by the community once the new ogb is elected, and that should happen fairly soon. There is also a draft of the constitution for those interested, and just like the gplv3 it's open for debate. nacho ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
On Feb 4, 2007, at 00:26, S Destika wrote: Why don't you simply open up an RFE or pick an already existing bug ID and request a sponsor? Because as I have said hundred times or so - the process is unnecessarily bureaucratic and dictated by Sun based on their interests instead of community inspired - not something I can work with. It's actually working much better than I had expected - I was a negative as you about the prospect of a FOSS project with no public VCS. When non-Sun committers are able to have direct VCS access across the whole of OpenSolaris I would be in favour of largely retaining it, with the only difference being that the committers able to act as sponsors would include non-Sun staff. I'd be pleased to hear your process suggestions, but I'm afraid a process of some sort is inevitable. All non-trivial FOSS communities have a process that ensures commits are only made by people the community trusts. S. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> Why don't you simply open up an RFE or pick an > already existing bug ID and request a sponsor? Because as I have said hundred times or so - the process is unnecessarily bureaucratic and dictated by Sun based on their interests instead of community inspired - not something I can work with. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
On Feb 3, 2007, at 19:26, Ignacio Marambio Catán wrote: In any case, evangelizm can solve the issue, CDDL is free by any standard, even the FSF thinks so, their only problem with it is that it is just not GPL compatible. That might change with GPLv3, there is some focus in license compatibility these days. Based on the advice I have most recently received, I am not expecting the GPLv3 to be compatible with any Mozilla-like (what I call "Category B"[1]) license. S. [1] http://www.sun.com/software/opensource/whitepapers/ Sun_Microsystems_OpenSource_Licensing.pdf ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> But more importantly this was never about accepting > any and all changes - it was about making it better > for people to propose changes and people to review it > and then accept the quality ones. That's exactly how the process works now. Why don't you simply open up an RFE or pick an already existing bug ID and request a sponsor? This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
Well there is only one Linux the kernel which Linus releases. All other changes are development branches and eventually all acceptable stuff gets merged in mainline. I don't think you understand how Linux development works at all. But more importantly this was never about accepting any and all changes - it was about making it better for people to propose changes and people to review it and then accept the quality ones. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
S Destika <[EMAIL PROTECTED]> wrote: > Somebody should just freaking replace Linux with latest Solaris version for > sites like Wikipedia, kernel.org (running at 400+ load as of today) and see > how it stands up - I doubt they'll even get past the hardware incompatibility > issues. If the Engineering cannot be put to use, why even boast about it? > Stop bashing Linux before Solaris can even parallel it in hardware > compatibility - An OS which cannot run on hardware I want it to run on is of > ZERO use to me. Looks like you did never really try to run OpenSolaris on a typical web server hardware. There is no problem to run OpenSolaris on it and it definitely outperforms Linux. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
I think that maybe x86 *is* the main area where we need help from device driver writers who have done the compatibility heavy lifting for Linux already. Is there a licensing problem in getting their work onto Open Solaris for x64/x86? Would the GPLv3 even solve the problem? Outside of this camp, I think we are looking at a green field of folks who are not necessarily coming from Linux to Solaris, but maybe getting involved with kernel development for the first time by getting their feet wet with Solaris. -- mark Alan Coopersmith wrote: S Destika wrote: Secondly most use x86 and Solaris won't work there I'll admit there are some areas that need improvement, but Solaris certainly works on x86, and more Solaris users are now on x86 than SPARC, so it's getting a lot of attention to fix the deficiencies. ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
No sir. They do not accept Solaris because firstly they believe freedom is priceless and that a for-profit company in drivers seat driving things the deem fit, there cannot be freedom and no one likes to work for free for somebody else's cause. marketing, just marketing, changing the license won't solve the problem for the simple fact that Solaris is sun's product and most of the changes to the ON will still come from sun. In any case, evangelizm can solve the issue, CDDL is free by any standard, even the FSF thinks so, their only problem with it is that it is just not GPL compatible. That might change with GPLv3, there is some focus in license compatibility these days. Secondly most use x86 and Solaris won't work there and if they are to fix it they would have to go thru some good amount of red tapism, instead of just fixing it and integrating it in the main tree. "find an application that solves the problem in hand, then pick the os that runs it the best and finally pick the right hardware for the os" that should be the number one rule for any sysadmin picking hardware. Hobbyists might have some problems with hardware and solaris, I agree there, but it is not nearly as bad as you paint it. i had 0 problems getting solaris to work in my desktop and except for the wireless card everything in the laptop i'm using works, this is a Dell 640m. Red tape is a necesary evil to get high quality software and you'll find that most here agree. In fact it is that high quality that drives people to solaris, without that it would just be another linux, a product that just works fine most of the time Fix both of the above and do away with your protectionist and wrong ways and you will see a surge in usage and participation. nacho ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
S Destika wrote: Secondly most use x86 and Solaris won't work there I'll admit there are some areas that need improvement, but Solaris certainly works on x86, and more Solaris users are now on x86 than SPARC, so it's getting a lot of attention to fix the deficiencies. -- -Alan Coopersmith- [EMAIL PROTECTED] Sun Microsystems, Inc. - X Window System Engineering ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> > Those guys aren't going to accept Solaris. They're > fundamentalists who don't use something based on > technical merit, but based on ideological merit. And > to me, that's the wrong reason to use an OS. > > The only way those people might ever be *compelled* > to accept Solaris is if Solaris delivers everything > Linux has, but in a Solaris way -- a clearly better > way -- and markets that, point for point. It's the > software availability and the end users and customers > that will make or break Solaris -- not a few million > Linux geeks stuck in their ideological dogma. No sir. They do not accept Solaris because firstly they believe freedom is priceless and that a for-profit company in drivers seat driving things the deem fit, there cannot be freedom and no one likes to work for free for somebody else's cause. Secondly most use x86 and Solaris won't work there and if they are to fix it they would have to go thru some good amount of red tapism, instead of just fixing it and integrating it in the main tree. Fix both of the above and do away with your protectionist and wrong ways and you will see a surge in usage and participation. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> What I am about to say is fairly brutal, so if you're > already upset, > don't read further. I'm certainly not upset; this is a disucssion, and I appreciate contrarian views so long as they're stated in a non-ad hominem manner. > I am an admin on wikipedia, was very active before my > son was born, > and have some 3000+ edits logged there. It's written > in PHP, with > apache and mySQL, and linux (Suse, a few Debian, a > few others). ... > Yet, it's "Good Enough" to generate 4 billion > pageviews per day. I get your point. The numbers speak for themselves, and Linux has results to back it up. However, what you neglected or are perhaps unaware of is that IT/CS field is unlike any other field out there. The problem with IT/CS is that it is inherently comple; extremely complex. Think of it in terms of literacy: there are *very few* people on this planet, out of 6.6 billion population, that are IT / CS literate. It's a strange side effect that stems from the fact that the field is so vast, that it is almost impossible to master it all. To have the insights and to understand what is truly a good technology and what is not, one would have to be involved with it practially since diapers, and also be extremely interested in it, to the point of eating, breathing and sleeping it. And there are indeed very few people on this planet that are like that. My point: Windows is the most widespread platform and has the highest numbers, but what does it prove? It proves that most people *don't care* what runs on the computer -- they want the *damn thing* to just *do stuff* that they want to do. In other words, they expect this extremely complex tool to be as simple to use as a washing machine and they don't know and even don't care that there might be highly advanced washing machines for professionals out there. Actually most "professionals" don't care how much their "washing machine" is capable of. I should know, I have to watch this horror every day of my life, and yes, before you rush out to write back, I'm extremely frustrated by it, because I care. In other words, people don't know any better, and they will use *whatever* as long as it is advertized enough, in one way or another. At least as far as a general consumer and as far as your average IT guy (which are *far away* from pros and engineers based on my experience). Like so many things, it's all about awareness. > For people who have a problem with that, look here: > http://www.google.com/finance?client=ig&q=GOOG > and see that GOOG market capitalization is $147 > Billion. > SUNW's is $23 Billion. > For comparison, IBM's market cap it $149 Billion., > Microsoft's is > $295 Billion, but that only means IBM + GOOG > MSFT > (think about that > for a moment). Don't fall for the Wall Street numbers; the share price sways from one day to another as the wind turns; Google, at the end of the day, has only one useful product, and that's a search engine. Other than that, the "huge" Google has *NOTHING*, and I repeat *NOTHING* *tangible*. And to someone in a 3rd world country that has nothing to eat, is sick, and doesn't even have access to clean water to drink, which is still most of the aforementioned 6.6 billion population, your GOOG is *worthless*. Do you understand that? WORTHLESS. You can't eat it, you can't drink it. The share price is meaningless, as is the market capitalization, if any of those people would even know what market capitalization is. I think you get my point. > What you fail to realize is that Linus does not write > paychecks. > People do it because they "believe". And if you think > the Solaris > codebase is forbidding, you ought to take a look at > Linux and the GNU > userland. It's daunting to most people. Yet they do > it. They go in > and roll up their sleeves and stay up until the wee > hours of the > morning and drink coffee (Hi Dennis) and they Get It > Done. People believe because they are ignorant. In fact, thay are no more enlightened than the serfs were in the middle ages and the church was stuffing religion down their throat, and they believed it, because they didn't know any better. Same scenario is playing out here, just the props and the stage have been altered. And I for one have no fear of yelling "the emperor is naked!!!" > Get off your high horse. Your technology is too > complex, too > slow-moving, too difficult to get running, to > difficult to patch, > change, and too difficult to write applications for. Have you actually looked into writing applications for Solaris? It's actually way easier than for Linux. Let me tell you a real story: when Nvidia started on writing drivers for Solaris, they were suprised when they learned that *one version* of their drivers will work *on all* versions of Solaris, and that they didn't have to develop separate versions for every Solaris release. Linux can only dream of something like that, in fact, the very people you believe in hav
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> According to Alexa, > http://www.alexa.com/data/details/traffic_details?q=&u > rl=http://www.wikipedia.org > The agglomerated wikipedia sites are generating 4 > billion page views > per day and are ranked 12th in the world for all > traffic. > > Now, PHP is admittedly one of the most insecure, > badly designed > languages out there. It certainly does not compare to > Lisp, Perl, > Python, or even Ruby in style and cohesion. It does > not compare with > Java in enterprisey-modularization. It does not > compare well to most > other languages for that matter. > > Yet, it's "Good Enough" to generate 4 billion > pageviews per day. > Somebody should just freaking replace Linux with latest Solaris version for sites like Wikipedia, kernel.org (running at 400+ load as of today) and see how it stands up - I doubt they'll even get past the hardware incompatibility issues. If the Engineering cannot be put to use, why even boast about it? Stop bashing Linux before Solaris can even parallel it in hardware compatibility - An OS which cannot run on hardware I want it to run on is of ZERO use to me. Sun isn't fixing that and doesn't expect community to do it and claims everything is fine with this open project which have no takers due to it's inability to work with most common hardware. > You might think that people will want the Solaris > because it's > advanced, rock solid, and of "great quality". You're > wrong. For the > marketplace, Linux does that well enough. That's not > a > differentiator. Exactly. If Solaris has to gain territories apart from where it stands today, greater strides need to be made in order for people to contribute and for other projects to share an enhance the source. That will not happen with bashing Linux (now admittedly it's the stupid community members who do that, not Sun) and Sun being the control freak they are. Sure they can control their internal engineering, not the community. > > What you fail to realize is that Linus does not write > paychecks. > People do it because they "believe". And if you think > the Solaris > codebase is forbidding, you ought to take a look at > Linux and the GNU > userland. It's daunting to most people. Yet they do > it. They go in > and roll up their sleeves and stay up until the wee > hours of the > morning and drink coffee (Hi Dennis) and they Get It > Done. Truer words never been spoken. Sun needs to understand this, deeply. (They ignore this crucial point repeatedly.) > Get off your high horse. Your technology is too > complex, too > slow-moving, too difficult to get running, to > difficult to patch, > change, and too difficult to write applications for. > > There is a reason it's free to download and use in > production: it > costs as much to get it running and keep it running > as it provides in > benefit to the user. So in financial terms, it's > value added is zero. > I wonder why this is so hard for Sun to understand and easy to ignore. > The > burtal fact remains: your UNIX customers don't pay > your bills. Dump > them. Go with the millions of hackers out there who > can take Solaris > to the next level. Sun should pay you a big fat check for pointing out the crucial thing that is like a key to all locks, permit to fly without barriers, without resistance. Recipe for freedom - real. I would also add that Sun should reflect on their current view - "Solaris is too advanced, our Engineering processes are also very advanced and we do not think there are smarter people out there and so we will continue to direct and develop Solaris and we don't need no stinkin community "hackers" to spoil our code and processes". That is just too disgustingly insane, too childish, too protectionist, too wrong to be of any real benefit and as a added bonus it kills all sorts of wonderful possibilities. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
"Richard L. Hamilton" <[EMAIL PROTECTED]> wrote: > So unless you can point out _specific_ needs, wants, etc. that can't be > met either now or with actions already underway, I just don't see what > your point is. No particular license is IMO going to make that much of > a difference in a positive way, and dual-licensing would just result in the > giant sucking sound of code leaving and once modified a bit, not coming back. > I have no problem with Solaris, Linux, and the *BSDs feeding each other > ideas, but I think they'd mostly all be better off not using each others code > all that much anyway (with some major exceptions that are largely possible > except that the Linux folks just don't seem to want to go there; like porting > zfs as a loadable module for Linux). The important point is that the Linux folks just don't seem to want to use code from OpenSolaris and you cannot change this by dual licensing. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> You know, there are people over there who say the > same thing about > us. I don't agree with them either. I'd say you're right, there probably are; but, people like me care as much for them as they do for me. No problem there. The thing is this: they said Sun is closed and proprietary -- you've given them most contributions of any company; they said Solaris is "Slowlaris" and that "Linux rulez" -- Solaris now runs as fast or faster than Linux; then they said Solaris is closed source and proprietary, and that open sourcing Solaris is a lie -- OpenSolaris exists for more than a year now; they complained and moaned how the tools are not gratis -- Sun has released the whole middleware stack and the compilers for free-as-in-beer. You -- we all -- consistently proved them wrong, point for point -- but what good did it bring? Those people are stuck -- no matter what we as a community and no matter what Sun did, they kept making up one excuse after another. They're still not using Solaris, still prefer GCC to Sun Studio, still complain how Solaris isn't GPL. So let me ask you all this: if Solaris is GPLed, what will the next excuse be? Those guys aren't going to accept Solaris. They're fundamentalists who don't use something based on technical merit, but based on ideological merit. And to me, that's the wrong reason to use an OS. The only way those people might ever be *compelled* to accept Solaris is if Solaris delivers everything Linux has, but in a Solaris way -- a clearly better way -- and markets that, point for point. It's the software availability and the end users and customers that will make or break Solaris -- not a few million Linux geeks stuck in their ideological dogma. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
John Plocher <[EMAIL PROTECTED]> wrote: > I think his point was that, even if there were 1000 non-Sun developers > contributing to OpenSolaris, the number of application developers, > students and users "participating" in the OpenSolaris community > would still dwarf them. > > As I said earlier, there is a pyramid here; the same one you will find > in those Linux kernel numbers: lots of small bugfixers, many module > creators, some systems owners and few heavy lifters. This is doubtlessly true. We need to inform people about OpenSolaris and about it's advanced debug features. Then we will attract application developersto at least port to OpenSolaris. We may (will most likely) attract bug fixers and probably a few driver developers. We will most likely in the close future only stay with the people who do already work on the Solaris kernel, but this is no real problem. Jörg -- EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED](uni) [EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> > If Sun would just get out of the way as you > suggest, > > and let the external > > folks do what they wanted, OpenSolaris would be a > > real mess. > Wow - that is so wrong. You would not want to apply > the same analogy in say a child's case, forget adults > for a moment. Cause then it will never learn, never > venture into depths and never take risks. You > basically are saying that everyone else is a fool. > That could not be right. You limit the possibilities > by asserting some totally wrong assumption and being > a control freak. When a child learns, it needs guidance by parents to help it interpret and understand experiences. There is more than one way to learn, but unfortunately some ways are worse than others. Right now, even as we are discussing this, "children" are learning and being taught to interpret their experiences in many new and different contexts. This is a process that might take a while, but eventually the attentive children will learn and will become just as capable adults as their "parents". Others who were "less attentive" will give up and fall out. Such is the way of life. One can see it every day in schools and families; good students are good because they don't give up and because they realize that they have to *seek to understand*, and those who do not never reach that level of understanding. To apply this to the OpenSolaris case, if you let children learn without guidance, chances are high they'll create a mess in the learning process -- that is why guidance is both needed and efficient. If you let people do ad-hoc "necessary" changes to OpenSolaris, it would turn into a hacked-patched-up *mess* really fast. Without the experience to properly interpret the context and without the insights, one might believe certain changes to be "necessary", when they are actually completely unnecessary, or when there is a better way. What you are suggesting is exactly what the GNU/GPL/Linux community did. While they have a pretty "clicky-bunty" operating system, the core of it is poorly engineered and can not scale. The context to understand here is "scale", and what that really means. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
[...] > > If Sun would just get out of the way as you > suggest, > > and let the external > > folks do what they wanted, OpenSolaris would be a > > real mess. > Wow - that is so wrong. You would not want to apply > the same analogy in say a child's case, forget adults > for a moment. Cause then it will never learn, never > venture into depths and never take risks. You > basically are saying that everyone else is a fool. > That could not be right. You limit the possibilities > by asserting some totally wrong assumption and being > a control freak. You don't let a five year old play in the street. But you don't keep a 14 year old locked up in the house all the time, either. This community is still maturing, and probably not enough so to handle everything even if all the mechanisms were in place _now_. It _is_ certainly doing more than it used to, and plenty of things are moving faster than your descriptions suggest. > > > > > We're moving to a model where changes will go into > > OpenSolaris first, then Sun > > will pull those changes into their own Solaris > > distribution. This will still > > take time to hash out and make work, a huge step > that > > has never been > > attempted previously. > > That is all fine-n-dandy if community controls what > goes into OpenSolaris. Well, even if that happens, I know I wouldn't casually change the processes, and I'm reasonably sure I can think of at least a few others that don't work for Sun that wouldn't either, because they _understand_ why things are done a certain way. Maybe someone from Sun that's been around a long time needs to do a series of blogs on how their process evolved, and just why it is the way it is, and the constant balance between quality and speed of integration (if they haven't already). Even if you think the community could do better, I think you need to start out understanding _why_ things are the way they are. Hint: I'm pretty sure it has very little to do with being a corporation. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
[...] > Again some of those folks are complaining about much > of the same things that I did. > (see elsewhere on the forum). I think there wasn't a > need for the source to be made open if the idea was > just people developing apps and creating distros on > top of what Sun provides. That you believe no one > else can contribute in a significant way to > OpenSolaris kernel then I have to tell you that you > are wrong. There are plenty of talented developers > outside of Sun. Significant innovations happened > outside of Sun and will continue to happen. And > Solaris as a mainstream operating platform is nowhere > near done - lot still needs to be done and I don't > see that happening without active community > involvement. Where exactly has anyone ever said that significant kernel work can only happen within Sun? I think the most that's been said is that in any community, those who do such work are a small minority. In fact, people were writing device drivers for Solaris long before OpenSolaris. Even a few filesystems, although the interface was never really standardized or public. And whether or not they were with the author's consent adopted by Sun (which I think had happened sometimes), they still had some influence (like just pointing out that it wasn't that big a deal to get it done). There's really _nothing_ to prevent you from grabbing your own copy of the source, and making huge giant massive pervasive changes to the whole darn thing if you like. There's nothing to prevent you from forming your own community around it elsewhere. Just don't expect all those who have come to value stability and reliability over the whatever benefits the Linux model of development offers to beat a path to your door. > > If you want it to go faster, then participate. > > Stephen Lau posted a > > ood way to help a couple of days ago. Put your > money > > where your mouth > > is. Sorry to be blunt, but really.. > > > Sorry but I already pointed out a key thing before - > people are not going to participate unless you give > them a right platform, unless they feel owning and > driving the changes. People are not going to work for > Sun (disguised as community) according to Sun's > rules, for Sun's purpose. > > So please open up everything and make it GPLv3 (I > believe GPL2/3 will continue to be the most favorite > licenses in open source community for variety of > reasons) and you will have people taking the source > and putting it out in open and doing their things. > That will bring in innovation, progress and > significant changes. Most Sun people think Sun's is > the only one, true way - that is wrong. That's where > the conflict really is. That's where the community > dissatisfaction really is. Your last two paragraphs don't follow; GPLvX != owning and driving changes, which you could do now if you wanted to, at least with your own tree. > Sun can choose what, if anything, to take back > whatever they like from the community. Sun can > continue to enforce their processes, directions, > quality etc. internally on their own tree. That way > Sun won't have to be burdened with doing everything. > (Sun has proved that it cannot make changes fast > enough while not hurting their business interests). > People can step up at a far higher level and do > amazing things efficiently if they feel like they are > owning and driving it. Again, CDDL vs GPL has nothing to do with that, except in the minds of those who love to believe in a One True Way and think that GPL is somehow more nearly that One True Way than anything else. You argue for the benefits of lots of people doing their own thing, yet you want one community to conform to another. That sounds like a contradiction to me. > Don't stop just with _you_ can do it - Think why > people are not doing it. Think what could be done in > order to get people to do it. Otherwise it just > becomes very narrow and very limited and isn't much > of an achievement at all. > > But I take it that Sun isn't interested in this kind > of thing at all apart from using community as a > vehicle to get to Sun's goals. That ain't gonna work. I think there's plenty for them to work harder at: opening more code, or at least assisting in it (since some of it they _cannot_ open without expensive purchases of rights or expensive rewrites), getting an SCM that encourages participation out there, streamlining, and so on. And I've said that more than once before. As far as I know, those things are happening, they just aren't _done_ yet. But I think that both license and touchy-feely visions of community are largely irrelevant to getting those things done. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> >The size of an OSS community is irrelevant if the > community > >cannot provide quality code to resolve issues. I > include > >both bugs *and* RFEs in "issues". > > That is your view, sir. You completely, totally > disregard the "community" here. You are not the > community. Community is a distinct, independent > entity with their own needs, wants and definitions of > right and wrong. They are not here to work toward > "your" goal of "quality code" and "resolved issues". > To work towards creating conducive atmosphere for the > community to thrive is absolutely the first aim of > any open project that wants to be successful. > > Your comment is a very good example of not > understanding the basics of open source projects. > This is what most people imply when they say {When it > comes to open source} "Sun just doesn't get it". Well, that's what you imply, anyway. I probably want very different things, and while it's entirely possible I'm alone in that, it's perhaps not likely. I've seen organizations grow in size faster than the core grew in maturity; it's usually a disaster for the organization, for what it's allegedly trying to accomplish (I personally suspect that any organization larger than about a dozen people, i.e. all that can each personally hold all the others accountable, ends up being interested in continuing to exist first, and only with constant effort manages to keep focused on accomplishing their nominal goals), and harmful to the participants. I want a better Solaris. Yes, I want the rest opened up too. And more outside participation at all levels. But with most of the expertise and investment from one place, I'd only expect most of the work to come from that same place, and accordingly, most of the agenda as well. That doesn't preclude anyone from participating and thereby increasing the outside investment. And I'd be as glad as the next person when some sort of externally accessible SCM is fully functional, along with some outside committers, a bit more streamlining of process (but not to the detriment of quality!), and so on. And I like the focus on quality over rapid introduction of new code, although I'm certainly not opposed to the increase in desktop/laptop support (of which quite a bit has already taken place in the last couple of years!). So unless you can point out _specific_ needs, wants, etc. that can't be met either now or with actions already underway, I just don't see what your point is. No particular license is IMO going to make that much of a difference in a positive way, and dual-licensing would just result in the giant sucking sound of code leaving and once modified a bit, not coming back. I have no problem with Solaris, Linux, and the *BSDs feeding each other ideas, but I think they'd mostly all be better off not using each others code all that much anyway (with some major exceptions that are largely possible except that the Linux folks just don't seem to want to go there; like porting zfs as a loadable module for Linux). I for one don't automatically suspect the motives of corporations (and for example see no more reason to regard Sun with suspicion that RedHat); or rather, I respect their motivations, as long as they have a real clue what _enlightened_ self-interest is; and I see no reason thus far to doubt that in the case of Sun and OpenSolaris. As far as I can see, they've been doing the things they said they'd do, give or take some schedule slippage. That's all you can expect from anyone, really. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
S Destika wrote: >>Do you have a specific device that is really >>preventing you from using >>Solaris/OpenSolaris? >> >> > >About 8 different x86/64 boxes - and I am not alone by any means. Only place >where it works reasonably is VMware. Has Sun any interest in fixing x86 >hardware support? When? If there is genuine interest in fixing this I can file >bug reports. You lost one community member - Sun won't fix x86 h/w support and >community is not even expected to fix it. > > > > x86 hardware support has been steadily improving over the last couple of years and the rate of improvement has accelerated in the past year. I have only failed to install on one x64 system (and I have done a lot of installs) and that was an HP DL380. So the answer is x86 hardware support is being very actively fixed. Ian ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> > Yes, because Sun employees are the majority of > engineers that understand and > can work with the code. A good majority of the > folks > outside of Sun have not > even pulled or looked at the sources. You don't > expect the engineers who have > been working on this code for man years to just > allow > any willy-nilly person > to have some willy-nilly patch accepted and > putback, > do you? > > > Do you have a patch/fix that you would like to > submit? > I have the ability to submit patches in a significant > way but I am not gonna bother doing it until basics > are fixed - I do not believe it's a fair and > intuitive process that OpenSolaris has currently. It > is not open enough, it has intolerable red tapism > and it can only work in the interests of Sun. > Someone independent is needed to control the affairs > and the process should be designed with inputs from > community rather than dictated by Sun. > Lets see to create a patch your going to need a bug id that tracks the patch. No problem file a bug and get a contributors agreement to supply the patch to. Hey I have a great idea what if there was a 'suggested fix' in the bug database to put the patch in. Wait there is a suggested fix field. But this is not part of the opensolaris community. But then somebody internal to Sun see a flaw with the patch and adds a comment to the comment field to fix it. But that is not part of the opensolaris community either. But the Sun engineers can work with you through the resolution of the bug and patch. Or are patches applied willy-nilly without a way to track and communicate what is going on just not viewable by the community? Did you want the community to be part of this? ---Bob This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> NOTE: send an email to Derek Cicero to have your > email changed. > Stevel agreed to help me with that - he has my new email and I am waiting on him to update it. > No, if you understood the process better, Sun made > some huge progress. > Not having SCM, not having a proper commit/review process, not having even insignificant outside contributions, after more than a year, that does not fit in my definition of progress especially if you consider a "open" project's goals. Sorry. > > In regards to these, it will take time to not just > put in place, but to help > it grow. > You are missing the point though - community participation can speed things up a lot and we are missing participation and we continue to ignore the root causes of lack of participation. > Yes, because Sun employees are the majority of > engineers that understand and > can work with the code. A good majority of the folks > outside of Sun have not > even pulled or looked at the sources. You don't > expect the engineers who have > been working on this code for man years to just allow > any willy-nilly person > to have some willy-nilly patch accepted and putback, > do you? > No I do not expect low quality high volume changes, AT ALL. I don't know why you exclude high quality high volume changes - perhaps because you believe only Sun engineers can do that? No one will agree with you there. It's not like all smart people work for Sun. There are lot many smart people who did many amazing things, all outside of Sun. > Do you have a patch/fix that you would like to > submit? I have the ability to submit patches in a significant way but I am not gonna bother doing it until basics are fixed - I do not believe it's a fair and intuitive process that OpenSolaris has currently. It is not open enough, it has intolerable red tapism and it can only work in the interests of Sun. Someone independent is needed to control the affairs and the process should be designed with inputs from community rather than dictated by Sun. > Do you have a specific device that is really > preventing you from using > Solaris/OpenSolaris? About 8 different x86/64 boxes - and I am not alone by any means. Only place where it works reasonably is VMware. Has Sun any interest in fixing x86 hardware support? When? If there is genuine interest in fixing this I can file bug reports. You lost one community member - Sun won't fix x86 h/w support and community is not even expected to fix it. > Most of them shouldn't as they have done little or no > work on it. If you > worked on this code for 15 years, you would feel that > you are a part of it. > This is how the Sun engineers feel, many of them have > worked on the code for > a good number of years. It is their work that has > made our community > appreciate the system, enough to be here. That is beside the point though. You already *have* employees doing their work - this thread is about active community participation. > > > 3) Contributing changes remains hard > > I expect them to be, and I think the current process > does work. It's no easier > to do a putback on the inside of Sun, this is where > the illusion is. It is > hard to do a putback, because there are engineers > that are passionate about > the product and strive to keep the quality. > It is also a process which has proved to be big time slow. It is time to recognize that better alternatives exist - two trees, one development controlled by some one independent and driven purely by community interests and one Sun's own tree. Let community set their standards, do what they care about and let Sun cherry pick what they need and what passes their quality bars. Benefit everyone, unlike years without progress and only Sun drives and benefits. > If Sun would just get out of the way as you suggest, > and let the external > folks do what they wanted, OpenSolaris would be a > real mess. Wow - that is so wrong. You would not want to apply the same analogy in say a child's case, forget adults for a moment. Cause then it will never learn, never venture into depths and never take risks. You basically are saying that everyone else is a fool. That could not be right. You limit the possibilities by asserting some totally wrong assumption and being a control freak. > > We're moving to a model where changes will go into > OpenSolaris first, then Sun > will pull those changes into their own Solaris > distribution. This will still > take time to hash out and make work, a huge step that > has never been > attempted previously. That is all fine-n-dandy if community controls what goes into OpenSolaris. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
S Destika wrote: I will never expect a huge number of people developing OpenSolaris itself. I would consider that as a wasted opportunity and potential. I think his point was that, even if there were 1000 non-Sun developers contributing to OpenSolaris, the number of application developers, students and users "participating" in the OpenSolaris community would still dwarf them. As I said earlier, there is a pyramid here; the same one you will find in those Linux kernel numbers: lots of small bugfixers, many module creators, some systems owners and few heavy lifters. You will get no disagreement from me (or I would suspect, *anyone* else) that the current OpenSolaris/ON development process (intent + deployed externally usable tools) makes it hard for the people at the top of the pyramid, extremely difficult for those in the middle, and absolutely impossible for the majority at the bottom. And that this is a high priority problem that needs to be addressed ASAP. [As an aside, I would suggest that those of us who have been dominating this discussion need to shut up and listen to other voices for a while] -shutting up for now, John ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> If you are defining "participation" as "putting back > code", yes, you are > correct. But really, how many kernel developers are > out there in the > world? I'm not expecting a huge number here. I think that is a negative way of looking at things. It is all about improvements - so if there are less Solaris kernel developers the idea should be to create more of them and not be complacent about it. (Last I checked there were 1500+ routinely active Linux kernel developers - many of them self taught and without a prior kernel development background - they started fixing, their changes were taken in and they were not bogged down by bureaucracy ) > > But we define participation as, well, participating. > Over 40 > niversities are using OpenSolaris in their > curriculum. Students are > learning on OpenSolaris, using it, experimenting with > it, developing on > it. Are they participating? Absolutely. > Which of the above was not possible when there was no OpenSolaris but Sun was allowing downloads of Solaris for free? I don't think that's the point here - "Source" in "OpenSource" is the key. We continue to ignore that. > I will never expect a huge number of people > developing OpenSolaris > itself. I would consider that as a wasted opportunity and potential. If that is your level of confidence then I have to wonder why did Sun open it in first place and named it "Open"Solaris? What you said contradicts the definition of open in free software parlance - specifically because you discount collaboration and improvement and block freedom of use - they way I want it, where I want it. Sun could have just made the binary download available and people who need it may have used it. Why the source? > Where I see the potential volume is with > people developing apps > for OpenSolaris, and people using distros based upon > OpenSolaris. All > of those people are participating. > Again some of those folks are complaining about much of the same things that I did. (see elsewhere on the forum). I think there wasn't a need for the source to be made open if the idea was just people developing apps and creating distros on top of what Sun provides. That you believe no one else can contribute in a significant way to OpenSolaris kernel then I have to tell you that you are wrong. There are plenty of talented developers outside of Sun. Significant innovations happened outside of Sun and will continue to happen. And Solaris as a mainstream operating platform is nowhere near done - lot still needs to be done and I don't see that happening without active community involvement. > If you want it to go faster, then participate. > Stephen Lau posted a > ood way to help a couple of days ago. Put your money > where your mouth > is. Sorry to be blunt, but really.. > Sorry but I already pointed out a key thing before - people are not going to participate unless you give them a right platform, unless they feel owning and driving the changes. People are not going to work for Sun (disguised as community) according to Sun's rules, for Sun's purpose. So please open up everything and make it GPLv3 (I believe GPL2/3 will continue to be the most favorite licenses in open source community for variety of reasons) and you will have people taking the source and putting it out in open and doing their things. That will bring in innovation, progress and significant changes. Most Sun people think Sun's is the only one, true way - that is wrong. That's where the conflict really is. That's where the community dissatisfaction really is. Sun can choose what, if anything, to take back whatever they like from the community. Sun can continue to enforce their processes, directions, quality etc. internally on their own tree. That way Sun won't have to be burdened with doing everything. (Sun has proved that it cannot make changes fast enough while not hurting their business interests). People can step up at a far higher level and do amazing things efficiently if they feel like they are owning and driving it. Don't stop just with _you_ can do it - Think why people are not doing it. Think what could be done in order to get people to do it. Otherwise it just becomes very narrow and very limited and isn't much of an achievement at all. But I take it that Sun isn't interested in this kind of thing at all apart from using community as a vehicle to get to Sun's goals. That ain't gonna work. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
> disregard my previous post - I didn't realise that > you were having > issues changing it. email me your new email address > and i'll update > your account. sdest at startvclub dot com - Please update (People will now have to find new reasons to flame me - kidding :). But I would have been happy to see jive fixed to allow people do this on their own. Any way... Thank you. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
Stephen Lau wrote: .. and the first step in creating a conducive atmosphere for the community is having an email address that freaking works. Why even have the much-hated Jive forums if you're going to disown people who use them instead of the mailing lists? -- -Alan Coopersmith- [EMAIL PROTECTED] Sun Microsystems, Inc. - X Window System Engineering ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
>>The size of an OSS community is irrelevant if the community >>cannot provide quality code to resolve issues. I include >>both bugs *and* RFEs in "issues". > >That is your view, sir. You completely, totally disregard the "community" >here. You are not the co mmunity. Community is a distinct, independent entity with their own needs, wants and definitions of right and wrong. They are not here to work toward "your" goal of "quality code" and "resolved issu es". To work towards creating conducive atmosphere for the community to thrive is absolutely the fi rst aim of any open project that wants to be successful. > >Your comment is a very good example of not understanding the basics of open >source projects. Your statement excludes the possibility of a community wanting such properties of a development tree; as member of the community, quality all the time is a worthwhile goal; if I want to tinker with code, I don't see a reason to bestow the breakage that causes onto others. Casper ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
.. and the first step in creating a conducive atmosphere for the community is having an email address that freaking works. snarkily yours, steve On Thu, Feb 01, 2007 at 06:45:57PM -0800, S Destika wrote: > >The size of an OSS community is irrelevant if the community > >cannot provide quality code to resolve issues. I include > >both bugs *and* RFEs in "issues". > > That is your view, sir. You completely, totally disregard the "community" > here. You are not the community. Community is a distinct, independent entity > with their own needs, wants and definitions of right and wrong. They are not > here to work toward "your" goal of "quality code" and "resolved issues". To > work towards creating conducive atmosphere for the community to thrive is > absolutely the first aim of any open project that wants to be successful. > > Your comment is a very good example of not understanding the basics of open > source projects. > > > This message posted from opensolaris.org > ___ > opensolaris-discuss mailing list > opensolaris-discuss@opensolaris.org -- stephen lau // [EMAIL PROTECTED] | 650.786.0845 | http://whacked.net opensolaris // solaris kernel development ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
Re: [osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
S Destika wrote: >>The size of an OSS community is irrelevant if the community >>cannot provide quality code to resolve issues. I include >>both bugs *and* RFEs in "issues". >> >> > >That is your view, sir. You completely, totally disregard the "community" >here. You are not the community. Community is a distinct, independent entity >with their own needs, wants and definitions of right and wrong. They are not >here to work toward "your" goal of "quality code" and "resolved issues". To >work towards creating conducive atmosphere for the community to thrive is >absolutely the first aim of any open project that wants to be successful. > >Your comment is a very good example of not understanding the basics of open >source projects. > > > So you would rather have a hundred hackers rather than a few dedicated developers? Most open source projects are based around a small code of contributors. Open Solaris probably has more than most, they just happen to work for Sun. Ian ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
[osol-discuss] Re: Re: Community participation (was GPLv3 ravings)
>The size of an OSS community is irrelevant if the community >cannot provide quality code to resolve issues. I include >both bugs *and* RFEs in "issues". That is your view, sir. You completely, totally disregard the "community" here. You are not the community. Community is a distinct, independent entity with their own needs, wants and definitions of right and wrong. They are not here to work toward "your" goal of "quality code" and "resolved issues". To work towards creating conducive atmosphere for the community to thrive is absolutely the first aim of any open project that wants to be successful. Your comment is a very good example of not understanding the basics of open source projects. This message posted from opensolaris.org ___ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org