KDE4 release discussion, Was: KIO::NetAccess static methods question

[Kurt Pfeifle]

Aaron J. Seigo wrote:
> On Thursday 25 October 2007, Aaron J. Seigo wrote:
>> because not releasing doesn't make an open source software better. it just
> 
> it occurred to me while reading this that maybe there's a difference in 
> expectations here.
> 
> personaly, i do not expect kde 4.0 to be a release that comes preinstalled on 
> computer systems or a release that one would put out into production 
> deployments.
> 
> this is not because of a failure of KDE 4.0 or anything like that. 4.0 should 
> simply not be that kind of release at all. period. it should be something 
> that lets us get that working draft out into the hands of people who are 
> closest to us, but not us. emphasis on "working" but also on "draft".

I understand that. But to make the rest of the world understand too (that
rest that cares about KDE at all), it seems to me that we've got to invent
a new name for that release. A name that reflects what you said above.

Because what you just said, re-defines a "x dot zero" release. It re-
defines what "x dot zero" (especially for cases where "x > 1") means to
most people.

Most people (while still be a bit cautious because of the "zero" in the
name, and expecting more bugs and less polish) will *not* see it as a
"draft". "drafts" are what "Alpha" releases are perceived as. "Betas"
start to become usable and "RC" means: "we don't know of any more bugs
ourselves other than the 5 we list in the release notes -- not *you*
*users* please help us find the ones we didn't see".

So what kind of additionally naming tag could we add to the "KDE4 4.0"
release that transports (or helps to transport) your redefined meaning
of that release number?

"KDE4 0.8 platform technology preview release for Beta Testing End Users"?
(not meant to be a serious name suggestion -- just throwing a few name
tags into a single mix as food for thoughts...)


> the only way it could be a "ready for the enterprise" release is if 4.0 was 
> very modest in its goals (it isn't) or if we held it back until it was "done" 
> (which really wouldn't do us much good, either, though).
> 
> if this were a proprietary product we'd have the mandate to release this 
> version to our pilot projects and 4.1 to the world with 4.2 being the 
> first "service pack" or "hotfix" or "patchlevel" or whatever terminology was 
> decided to be used.
> 
> but we're an open source project; such project live by (or die by not) 
> releasing early and often. the pace is more important than anything.
> 
> we seem to have gotten so full of ourselves and so scared of releasing 
> something that isn't "ready for the enterprise" that we seem to have 
> forgotten that process.
> 
> remember the days when free software projects would release at 0.01 and slowly 
> reach 0.9 and then one day after years of use by tons of people would 
> announce a 1.0? didn't k3b do that recently, in fact? granted things have 
> changed since then, but the pendulum has swung too far in the other 
> direction, imho.
> 
> somewhere along the way a lot of us seem to have forgotten the mechanics of 
> how these things work. the linux kernel project "gets it": their dot-oh 
> releases are nearly always insanely bad and get pushed out almost purely 
> because they need to start the process of public consumption. and you may 
> have noticed, that the linux kernel project is slightly more "ready for the 
> enterprise" than we ever have been. yet we seem to think they can get away 
> with it and we can't.
> 
> if you are concerned about what the public perception is, first step is to 
> stop being worried yourself. people take their cues from you. second, let the 
> marketing and communications team do their job. that may also include asking 
> them for input and/or direction when doing public presentations or tech show 
> booths if you're uncertain how to present it. but public perception is 
> managable, and shouldn't come into play when trying to figure out when to 
> release Four Dot Oh.
> 
> that is a technical decision, and one that needs to be made in light of this 
> being an open source project.
> 
> now, if my expectations were that 4.0 must be like Leopard or Vista, a 
> finished completed product that Dell should be bundling on the latest laptops 
> and pushing out next quarter, i'd be freaked out too. is that where some of 
> these concerns are coming from? if so, please lean back in your chair a bit, 
> breath, and think about how the *open source software* cycle works in 
> contrast to the proprietary one.


-- 
Kurt Pfeifle
System & Network Printing Consultant ---- Linux/Unix/Windows/Samba/CUPS
Infotec Deutschland GmbH  .....................  Hedelfinger Strasse 58
A RICOH Company  ...........................  D-70327 Stuttgart/Germany