Hi Stefan, first of all thank you so much... my responses are in line ...
On Nov 22, 2007 3:30 PM, Stefan Teleman <stefan.teleman at gmail.com> wrote: > Hi. > > My 0.0.2: > > KDE4 is the future (obviously). So, i think we should concentrate on KDE4. > Back in May, when we had our big KDE Solaris IRC meeting (#kde-solaris), > we decided to focus on KDE4. > Great ! > I would be much more interested in focusing on Nevada and/or OpenSolaris, > for no other reasons than: > > - they deliver more components (many of which are KDE dependencies) > - it's easier to integrate new components in Nevada/OpenSolaris than in > Solaris > - the general idea about the relationship between Solaris and > Nevada/OpenSolaris > seems to be: New Things(TM) appear in Nevada/OpenSolaris first, then they may > (or may not) appear in Solaris. > > So, i agree with you in principle: Nevada/OpenSolaris is our focus. > I think so too. > On the one hand: many of KDE[3|4] dependencies will probably not make it into > Solaris: OpenLDAP, MIT Kerberos, Cyrus-SASL2, recent (younger than a year) > Net-SNMP, and others. I believe it's very important to have the > ability to include > these "Big Things" in a Nevada/OpenSolaris distro, and for KDE to use these > components as system-installed ones, as opposed to KDE delivering privately > installed ones (if you want to see what happens when you start delivering all > these components as depenencies, look at Blastwave's package dependency mess). > Yes i saw it, it's strange. > But: many are still using Solaris. If we focus on Nevada/OpenSolaris > exclusively, > we run the risk of leaving Solaris out, simply because we assume > certain components > are already in Nevada, but they are not in Solaris. Or, we are telling > this userbase > "update to Nevada, or you can't build/run KDE". > > My proposal: let's set a "sunset" Solaris release. In other words, let's say: > > - we support Solaris 10 Update 4 and up, and Nevada/OpenSolaris > - we will support Solaris 10 Update 5 (and continue to support > Nevada/OpenSolaris) > - at some point in the future, after the release of Solaris 10 Update > 5, depending on > what components are delivered by S10U5, we'll have to re-evaluate. if S10U5 > delivers most of KDE's dependencies, then we no longer have to do anything > special for Solaris Updates, and we can just say: either Solaris 10 Update 5, > or > current Nevada/OpenSoalris will work. If this does not happen, we'll > have to make a > decision: drop support for "official" Solaris Updates (Update 6, > Update 7, etc), or > continue supporting them. Continue supporting "official" Solaris Releases > means: > continue to deliver dependency components, big or small, simply because they > are > not available in Solaris. > > Ultimately it's a matter of "who is the indended audience". > Yes, that's true. > If the intended audience is made of hackers/developers who do not mind > updating > their operating system every 3 months, and don't care about things > being broken once > in a while, then we don't have to worry about Official Solaris > Releases. But i'm not sure > this is a wise approach. > I changed my point of view. I completely agree with you. I'm happy, because I'm driving on the same way as you now. Thank you. > Now about KDE3: i am planning a KDE3 (3.5.8) binary release. Nice > > --Stefan > -- Lukas Oboril When dealing with people, let us remember we are not dealing with creatures of logic. We are dealing with creatures of emotions, creatures bristling with prejudices and motivated by pride and vanity. Dale Carnegie
