--Boundary-02=_aymJ/GYVrfXHZci Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On Monday 28 July 2003 18:11, Ralf Nolden wrote: > The issue I see here is only that you have those three scenarios: > > - people running HEAD -> no problem for them to switch to Qt 3.2 Define "people running HEAD". I for one try to work as long as possible=20 with the latest stable branch of KDE except for certain small parts=20 (kio_smtp, KMail, other PIM + dependencies) which are from HEAD, so I=20 can work on them. I'm not at all interested in Hindi (other than from a=20 nice to have POV) support b/c I can't read Hindi anyway. Instead, I=20 want to stay as long as possible (ideally until the feature freeze is=20 in effect for the next version, 3.2 in this case) on a stable KDE=20 environment. I expect other mostly-apps-developers to share this=20 desire. > - people running KDE 3.1.x -> how does Qt 3.2 affect that ? I still > didn't get a complete answer on exactly *WHAT* does not work > correctly when upgrading to Qt 3.2 on a KDE 3.1 machine. That is what > should be fixed and therefore #ifdef's in KDE_3_1_BRANCH for > backports of fixes that go into HEAD to work with Qt 3.2. That makes > users happy while we can use 3.2 with HEAD automatically. Also people > can install Qt 3.2, KDE 3.1 and HEAD, all on the same machine. I am completely with you here. However, kde 3.1 issues with Qt 3.2=20 cannot be fixed by #ifdef'ing. It must be possible to compile a kde=20 3.1.x against Qt 3.1.0 and let it run on all 3.1.x and 3.2.x that are=20 out there when kde 3.1.x is released. A kde 3.1.4 should _not_ require=20 Qt 3.2, but require 3.1.0 and suggest 3.2.x and 3.1.y for maximal x,y. > - people running KDE 3.1 and HEAD -> see above Definition, please ;-) HEAD libs with 3.1 kdebase? 3.1 base and libs with HEAD kdepim? HEAD koffice with qt-copy and KDE 3.1 desktop? > So from my perspective it is *better* to switch to Qt 3.2 in HEAD and > fix any known issues with Qt 3.2 in KDE_3_1_BRANCH ASAP. Of course, it's better to switch. but we've already done that when Qt3.2=20 betas were committed to qt-copy :-o. What should _not_ be done is to=20 _require_ 3.2. In RFC speak: you MUST use 3.1.x and SHOULD use 3.2. Marc =2D-=20 There are lots of hackers in the world -- kids, mostly -- who like to play at politics and dress their own antics in the trappings of terrorism. [...] It's base and despicable, and it causes real damage, but it's cyberhooliganism, not cyberterrorism. -- Bruce Schneier, Crypto-Gram Jun 2003 --Boundary-02=_aymJ/GYVrfXHZci Content-Type: application/pgp-signature Content-Description: signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQA/Jmya3oWD+L2/6DgRAn8cAKCIGqGGrRqlQOXjPgzS22yoCwYETgCeNZlP NKbcnJ7M9BA4mFcCTnbVIUg= =3Bnn -----END PGP SIGNATURE----- --Boundary-02=_aymJ/GYVrfXHZci--