I would like Debian to stop shipping XScreenSaver
Posted: 2016-04-06 16:47
No they shouldn't and yes it does.spacex wrote:Debian should upgrade Stable with the versions from testing. It doesn't make Stable any less stable.
It's been backported "for others" the first day that it was reported. Piece of cake using the OBS.spacex wrote:Nah, but Debian should get a newer version of "xscreensaver" and "xscreensaver-data" into stable, so that people don't get the outdated warnings anymore, and the annoying message about our distribution not doing us justice. Besides, the version in Testing are better, as in better screensavers. Obviously I have fixed it for myself, picked them directly from testing as all the dependencies are covered in Stable. That's the easy way to do it, but not the recommended way.
But it works for me, and nah, I have no intention of backporting it for others, because that comes with obligations. Debian should upgrade Stable with the versions from testing. It doesn't make Stable any less stable.
Well, that's how it is today, but allow me to disagree with this policy. In my mind, Stable should be kept reasonably stable, but not frozen and dead. In my mind, Stable should get a new definition, as in "not to many bugs", not as in "never going to change". We'll probably disagree about that, and that's fine. But in my mind, Debian should become more current. Not as current as some of the competitors, but more current than today.dilberts_left_nut wrote:No they shouldn't and yes it does.spacex wrote:Debian should upgrade Stable with the versions from testing. It doesn't make Stable any less stable.
The 'stable' (does not change) policies have been there for a long time and have proven their worth in providing a dependable release.
I see no reason to make exceptions just because upstream wants it - that is one of the important points of free software, you must be free to modify it to suit your use case.
There are certain things that change faster and need special handling (browsers, AV etc.) but this isn't one of them - security fixes are backported to the released version as with everythng else, and the nag code is an artificial obsolescence introduced for ridiculous (and easily handled in other ways) reasons.
^^This.montagdude wrote:Is it really that much trouble for the developer to simply redirect bug reports about old versions to the maintainers of the various distributions? His method of bug handling seems to be his personal email, so maybe that's the problem if it's causing that much annoyance for him.
I don't know who Peter Nowee is but I think he has a point here. The GPL is the GPL. No 'moral' or other residual rights.With regard to future versions of Debian, again I think you should be
more clear in your license terms. Don't pretend to publish it as free
software, but then use pretty "please", time bombs and cursing to get
users to not use their freedoms.
-- Peter Nowee
Allow me to disagree.spacex wrote:Free software or not, Debian shouldn't go against the developers wishes.
Sure, you may disagree.mor wrote:Allow me to disagree.spacex wrote:Free software or not, Debian shouldn't go against the developers wishes.
First of all, you can't say "free-software or not" and do as if it is something one can put aside in this matter. It is in fact all that matters.
The whole point of free software is to leave freedom to the user, it can't be freedom if it comes with caveats.
Bye
I don't agree. Common decency and moral always apply. It's the developer who do Debian a favour. Not the other way around.keithpeter wrote: I don't know who Peter Nowee is but I think he has a point here. The GPL is the GPL. No 'moral' or other residual rights.
No, because that is how distributions work, by curating, integrating and distrubuting upstream software - otherwise every end user gets to grab code from each project and do all their own compiling integration etc.spacex wrote:If you change the code and redistribute something, you should also change the name and sign it with your email or other contact info. No matter what the license is. Because it's cleaner, and because it's the right thing to do.
Stop for a second.spacex wrote:I don't mind end users changing the code for themselves. Obviously, that is a freedom the user has. But, if you change the code and redistribute it to others, then you should fork it and change the name instead. No matter how small the change may be, the app is no longer authentic,
That is true, it shouldn't be that Debian users go directly to him with problems on the obsolete version, but as dln already explained there are parts of the mechanisms that are out of Debian's control.spacex wrote:and the original developer can in no way be expected to support or be responsible for an app where someone else has changed the code.