Thoughts on the Iteration Sprint, discussions and criticism

[Sebastian Kügler]

Hey,

On Tuesday, June 05, 2012 13:31:44 Dario Freddi wrote:
> This mail was originally born as a reply to the survey thread, but it
> eventually drifted into something bigger which probably deserved his
> own thread.
> 
> First of all I'd like to speak both with a Plasma hat (which I think I
> always wore, at least somehow) and with an external hat, as it'll be
> probably clear in the next lines. At the same time, I want to
> underline these are my personal thoughts about all of this. The
> purpose of this mail is to try to answer once at all some of the
> recurring arguments against the planned sprint.
> 
> 
> Starting from the very beginning, I honestly can't understand the
> arguments about "we already have a vision". Assume this is true - this
> vision was created (and eventually evolved) over the last, I think
> now, 3 years. Maybe a bit more maybe a bit less, whatever. In the
> highly changing time we live, that's an *enormous* span of time. The
> question this sprint tries to answer is: do we need to create a
> new/change the existing/keep the current vision for the workspace?

I think at least it doesn't hurt to verify if our assumptions are still true, 
or if things happened that might change the assumptions.

At the very least (and I think we've made some good steps into that direction) 
we should try to communicate that vision better among ourselves, so we get 
more of that feeling that we're on the same page. I think a sprint like next 
week's is a wonderful opportunity to work into that direction.

> What should be our goal over the next months, especially for the
> desktop? If I were an author of the original vision, I would be HAPPY
> to see this happen and I would strive to take part in it. The argument
> about names doesn't really stand - this is as clear as it gets, it is
> a sprint for finding out who we are now and who we want to be. Does
> this have a clear answer now? I don't think so.
> 
> Also, I'd like to point out that Plasma != Active. Active IS Plasma,
> but Plasma IS NOT active. Plasma is (or should be) much more than
> that, should focus on other platforms more than tablets. 

Plasma Active is not about tablets, it's about a spectrum of devices which 
includes -- but certainly does not *ex*clude the desktop. What I meant by 
writing Plasma == Active is that in the majority of cases, it does not makes 
sense to look at them separately. There's a catch, though: The desktop comes 
with pretty strict requirements in the form of legacy behaviour which we have 
to keep in place to not cause a user revolt. That makes it a difficult ground 
to experiment on. In the Active UIs, we have much more freedom, can collect 
experiences there and have them, after they're verified and matured, also 
bleed over into the desktop. That's exactly what we already see happening, but 
which can be more emphasized.

In the end the desktop is really just another formfactor.

> At the same
> time, as much as consistency should be valued, it's arguably hard to
> see a potential common vision for Active and the Desktop Shell. One
> could have a common vision about Plasma as in *the framework*, but the
> shells NEED to have different goals, targets and visions. If there is
> a naming mismatch at all, it's here. What is Plasma then? A shell? A
> framework? Something else? If Plasma == Active really stood, why the
> need of a different name in the first place?

In order to get things flying, to be able to concentrate on creating a 
critical mass. I think we've reached that critical mass now, that's why we are 
merging the IRC channels and the mailinglists into one, it's all Plasma.

> The answer, in my opinion, is quite easy: Plasma is something *to
> build upon*. And as much as there are people working actively on
> active (pun unintended), I guess you should be happy if there's people
> who still care about the desktop

Again, do not make the mistake to see them as different projects. There are 
differences, but they're highly technical and I think we've found pretty 
decent ways to not have the need to think in separate projects. The target 
device should be an implementation detail.

(Usage scenarios are another thing which differs, but also here, it does not 
warrant thinking in terms of different projects.)

> and want to strive to make it a
> better place to be. I would also like to note that there was not all
> this ongoing bikeshed and worries when active was started/announced.

I do not think making up an artificial barrier between Active and the Desktop 
helps. Emphasize what they have in common and where they should be different 
instead, but do not deal with them as if they're separate things.

> The question one should pose himself, in my opinion, should be: how
> this could be harmful or beneficial to KDE? If we take apart our egos
> and sentimentalism and try to see things in a rational way, there is
> an obvious benefit in revaluing what (I personally believe - wouldn't
> have worked on those if I didn't think so) were good ideas, but could
> probably be laid down in a better way now, maybe with a spin of new
> things. We learned from our mistakes and successes, and now we want to
> see how we move forward. How could such a thing be harmful to KDE?
> 
> And to be 100% honest, I'm quite let down that almost every person
> involved in Plasma (as in the framework/shell, not the workspace) at
> the time of writing won't be there out of their own decision.

Not correct, Marco will attend, I will attend, Martin will attend, Aaron will 
be pulled in through skype/telepathy/whatever.

> Alienating just because of a different name or intended purpose is a
> dull argument - we communicated early, timely and *on plasma-devel*
> most of all. 

This I probably communicated wrongly, so let me try again: The reason why I 
didn't register immediately is that it felt like a group of new people thought 
we were doing things wrong and started their own initiative. It just occurred 
to me yesterday that this was the reason I didn't initially buy in, but just 
thought "oh, I hope they won't try to change direction and screw up entirely" 
and "I don't think they understand the challenges we're facing right now of 
Qt5, Frameworks5 and libplasma2". To me, it would probably have been much more 
apparent from the beginning what this sprint is about if people came in and 
went "Let's do a Tokamak in which we focus on ABC, quite some people are 
interested in working on these topics".
Don't interpret this any differently than that is my own fault understanding 
the meaning.
I now understand that much better, so let's not put it in the way of progress.

> If we compare this to the Active sprint, which was kind
> of a fail from this perspective (even though there were a number of
> factors not imputable to organizers which I reckon), I think we did
> way better. It's hard to argue that people who are not coming are
> feeling left out instead of leaving themselves out for their own
> decision.

Could you be more specific about what has failed and how it should have been 
handled in the first place?

An explanatory note: We started Active because we needed a space to move 
forward in without others telling us what to do and what not to do. We needed 
that experimental freedom to create a critical mass (which we have achieved). 
This meant that we'd work with the people who actively showed interest to 
contribute to the result, quite purposefully, we didn't rally everybody to 
take part since this would have stood in the way of getting this critical 
mass. The result was that the people who have started contributing come from a 
much wider background: We attracted new people from outside of KDE (even UX 
people, which we are traditionally very short of!), but also from inside of it 
(Calligra, for example, Marble, etc). The team may not have exploded, but it 
grew steadily and beyond what we had hoped for.

> All in all and in fair honesty: I don't really like where this and
> other threads are going. The feeling I am getting is that we are
> trying to sacrifice a potential innovation just for the sake of
> sticking to a goal which is not clearly defined and hasn't been
> revised or maybe even shared by the people working/willing to work on
> the project *now*. When Active was started everyone tried to be
> positive about it and trusted people working on it. We are not
> newbies. We know and love KDE and we are trying to do this just to
> benefit KDE. A bit of trust in us would not hurt, just like we did
> with Plasma, we did with Active, and will do with the next big thing
> out there.

To be quite honest, after a long chat with afiestas last night, I think people 
should care less about artificial barriers and think more in terms of goals. 
We are one big team, working on technologies that belong together, we share 
the same goal overall. 

In short: let's not be such pussies but just sit down and do the necessary 
work together.

> Ending this long mail: if you want to ask me for my proposed plans for
> this "new vision", they're currently the following: improve what we
> have, make it better, and get it up to speed with what we learned from
> these years, and what we can do now which wasn't there in 2007. If
> this is how you feel as well, then this discussion is pointless, and
> we likely still have a couple free beds in the house. We're waiting
> for you.

This is how I feel as well, and yes, I think the discussion is mostly 
pointless. :)

Hugs,
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9