Re: [kde-linux] Users guide to KDE?

On Sunday 06 July 2008, Jim Philips wrote:
On Sunday 06 July 2008 09:55:09 Kevin Krammer wrote:

I work in the world of commercial software and it's always tricky business
to start talking future features with your customers. They have unrealistic
expectations or they will start pushing you to do things with the code that
don't make any sense from an architectural point of view. Or they will
hurry you into a feature before it's ready. The same caveats would seem to
apply to the open source world.

Indeed, the proprietary software industry has, in its self inflicted downwards
spiral which sacrifices the very basics of engineering for time to market,
trapped itself and consequently everybody else interacting or competing with
them.

However, Free Software developers, especially the ones with respective
education, really value software development as an engineering discpline and
will probably under no circumstances "bend over" to unrealstic expectations
because the usually have to when doing proprietary software.

Which results in behavior which is totally new and unexpected to the
miseducated customer side. Will be interesting to see if this strong
resistance against the wrong model can free others from this trap as well.

I don't know that you would have to push it to that extreme, but as the
user base for KDE expands, the nature of communications is bound to change.

Right, it already has changed a lot.
Originally all communication was very geek oriented, e.g. by using
mailinglists and usenew groups, but it has been taken to more widespread
public by putting information onto the web, e.g. article series on
dot.kde.org. blogs.

I have been a happy user of KDE since 2.0. I can truthfully say that the
change from 3.0 to 4.0 has been far more jarring than the change from 2.0
to 3.0. If my memory isn't lying to me, the alpha versions of 3.0 (I rushed
out to compile them) were more stable than the 4.0 release version. I can
live with that and it isn't going to turn me off to KDE. But if you have a
release version that is so far from feature complete, you have to expect
some backlash. This isn't, however, a reason to cut off communications with
your user base.

The KDE2 to KDE3 transition isn't actually quite comparable. I was basically a
porting to the new Qt and improving of the already existing implementations,
no major changes in architecture or frameworks.
This is probably one of the reasons why the difference of KDE4 is so
substantial, because the previous incarnation has been around for two major
versions.

Of course, ideally any new incarnation should at least be capable of all the
things the previous one could do, but given limited resources not all
compatability measures could be made available at this time and had to be
postponed.
Of course it was expected that this less than ideal approach would trigger
concern or irritation, but it was quite unexpected that people would totally
be unaware of the principle of improvement and would assume that the
currently available feature set would all that would ever be.

Given that there were quite some substantial improvements during the KDE2 and
KDE3 series, I was shocked to learn that some users obviously think that KDE4
will never go beyond the scope of KDE 4.0

I'm all for creating blogs that act as channels for targeted
audiences...and for not opening them up to people who only rant about the
difficulty of using A or B.

The problem with this is that is defies the reason why developers blog at all.
The reason is to allow those who are not directly involved an insight in what
is going on.

It will be dark day in the Free Culture world when we can no longer believe in
people's abilities to decide for themselves what kind of information they
want to get but rather have someone deciding it for them.

Most users will not appreciate the openness of a new architecture until it
has concrete benefits for them. If you target your audience carefully, you
won't get useless complaints about a new architecture that hasn't yet
produced tangible benefits for the end user.

True, but the main problem I can see is how does one control that media
outlets like news site only report about things suitable for their target
audience.
Naively I would have expected that to be their responsibility since they have
probably a better understanding of their audience than anyone else.

I am also not sure if they would like to be told what to report and what not
to, especially since they would have to check back whenever creating
something they want to publish.

Cheers,
Kevin

--
Kevin Krammer, KDE developer, xdg-utils developer
KDE user support, developer mentoring

Attachment: signature.asc
Description: This is a digitally signed message part.

___________________________________________________
This message is from the kde-linux mailing list.
Account management: https://mail.kde.org/mailman/listinfo/kde-linux.
Archives: http://lists.kde.org/.
More info: http://www.kde.org/faq.html.

Relevant Pages

  • Re: FC4 or FC5
    ... cost it is equivalent in Windows. ... won't happen for the individual private developer. ... happens one more license was needed for the Matrox DigiSuite SDK. ... If somebody sees an income model for the feature ...
    (Fedora)
  • Re: gEBI wrapper (Code Worth Recommending Project)
    ... The developer is screwed if the docs say "$returns DOM ... You apparently don't understand feature detection at all. ... I can write robust Web applications that degrade gracefully in Z- ... feature testing altogether, which is hardly a realistic option. ...
    (comp.lang.javascript)
  • Re: [RFC] Unify KVM kernel-space and user-space code into a single project
    ... slows down to a crawl - while the complexity of the change might be very ... If you ever tried to implement a combo GCC+glibc+kernel feature you'll know ... developer exhaustion ... contributions skyrocketed and basically all contributions were top-notch. ...
    (Linux-Kernel)
  • MSITOSIT: [2] ERROR processed: has no associated directory
    ... Developer Comment:: has no associated directory, verify the Feature table is correct ... HRESULT: -2147467259 ...
    (microsoft.public.vsnet.setup)
  • Re: browsers & standards
    ... >> One may not be able to do object/feature detection. ... >> FAQ should indicate a test value that can be used to prove toFixed to ... > feature behaves according to standards or expectations. ...
    (comp.lang.javascript)