A poll to gage the end user's satisfaction with the way kde is being developed. As of the inception of this poll KDE4 is 2.5 years into development.

very good idea, this survey! I think that the linking together is the most difficult part! There should be more coordination, more communication with users and especially more coordinating of the realisation of the overall KDE idea. The parts must fit together, that's the reason to use a platform like KDE!

Regards, Martin

It would work if and only if the sample was indeed representative of the majority of the users, and that's difficult even with "offline" polls (that's why you need more or less complex statistics).
As polls online can be faked very easily, no matter what tricks you put online, their usefulness is very limited, from a statistical point of view, and thus they would not be representative of the community.

even the forum for sure is not represantative! there are thousands of users out there who never will enter a forum and not even will write bug reports.

That's why an online poll's results will never be solid enough to be considered properly.

einar wrote:That's why an online poll's results will never be solid enough to be considered properly.

Well, then einar, what IS representative? How does one derive at any kind of consensus? What is the standard?

That is the point: A survey like this is not represantative but not trying is even worse. Einar's answer shows that nonone within the KDE team can be sure to find out what the users really would like to have if they do not start to look for different ways. The communication between development and development coordination on the one hand and the users on the other hand must be improved. To find a better way is a task of the development coordination (does this not exist? as one of the "administrators" here in the forum said?)! Any proposal?

Regards, Martin

mshelby wrote:Well, then einar, what IS representative? How does one derive at any kind of consensus? What is the standard?

You'd need to study the landscape and do some statistical checking first: there are ways to determine optimal samples and sample sizes.

You've never done any real live statistics, did you?

Considering that einar is a scientist, that's a fatuous remark.

Every major and minor release of KDE 4 has contained some compromises from the developers. Things that did not fit their original dream and plan, but that users insisted that they want. I know you're going to ask for examples, so here are some -

Making it possible for the whole desktop to be a folderview so that it would feel more like KDE 3.

Allowing activities to be linked to virtual desktops so that people could have different wallpapers on their desktops - and just why so many people got steamed up about a wallpaper I'll never understand. Content concerns me far more than decoration.

Stopping the automatic 'peeking' into subfolders in a folderview - something that I found extremely useful and resent having taken away so that I now need mouse-clicks to get what I want.

These are just the more obvious ones. There are dozens of smaller ones. Of course you could always give some of your valuable writing time to helping achieve the things that you value.

And just one last remark - you'll notice that by giving in to pressure from one group I lost something I value. This will be true every time a compromise is made. One group wins something, another group loses, and the losing group matters as much as the winning one.

Considering that einar is a scientist, that's a fatuous remark.

I wrote "real life". The last time I have seen an objective statistic was back at university (and thats been quite a while). Oh, I am a scientist, too. I'm in the medical business .. you would not beleve how statistics and proofs are made there

Stopping the automatic 'peeking' into subfolders in a folderview - something that I found extremely useful and resent having taken away so that I now need mouse-clicks to get what I want.

Well, is it now consistent with your choosen focus policy? If you have "click to focus" it makes sense.

samhain wrote:xD You've never done any real live statistics, did you?

Oh yes, I did, sadly. It doesn't take back the validity of my statement anyway. You want a representative sample size, and an Internet poll is certainly not that.

You're right. But what better do you have? Maybe the poll should be on www.kde.org, first page, to get more users?

It won't work. It can be gamed too easily. I think it's evident.

So no user feedback is better?