registry team mailing list archive

[Martin Wildam <1@xxxxxxxxxxxxxxxxxx>]

On Mon, Oct 25, 2010 at 01:50, Faldegast <1@xxxxxxxxxxxxxxxxxx> wrote:
>> Nobody thinks of paying a yearly license for having tubes in the wall
>> - no - people pay the plumber when they need him/her for putting
>> additional tubes or doing repair work. I think, it should be the same
>> for software. The advantage of software is that it can be easier
>> duplicated and easier offered than the tubes.
> Actually they do. Its called insurance. You pay a monthly fee and if your tubes need to be repaired you are not hit with a overwhelming cost.

You are right - and I mixed up a few different models. My tube example
was comparing to a Windows license and not to a model where you get
the software and pay extra and optional for support - as Red Hat
offers for example.


> Perhaps this should be integrated with Ubuntu brainstorm, and other sites where such pledges get more exposure?
> There are also cofundos.org, and possibly other sites?

I would find it good to combine/link Launchpad with brainstorm and
integrate something like pledgebank or cofundos - Whether it is a bug
or a new feature request, I don't see so much difference in how it
should be handled. Even for bugs sometimes it turns out that a
complete different solution is better than a quick-and-dirty fix.


> I am talking about a commercial version with support, marketing and distribution. And more important, an OEM program.

Oh, I understand - I thought that such also already exists. ASFAIK
everyone can buy support from Canonical. I personally find those extra
commercial versions which maybe include additional patches or maybe
not is not very attractive. I like to have one thing and pay for the
appropriate support when I have a problem that could need the
implementation of a particular fix.


> Actually i think Microsoft expect Students to buy a computer with an OEM
> license and then get the student version, and actually pay more then
> they have to....

That was exactly what I meant. They try to create a dependency.


>> DonationCoder (http://www.donationcoder.com/) also have several
>> approaches for funding software development, from micro-donations up
>> to custom-made software request handling through forums. I think, to
>> build an "Ubuntu" for people seeking for programs to be written or
>> bugs to be fixed, combined with a simple way of donation/payment,
>> would be a big gain for Ubuntu. - Especially for companies with need
>> to optimize their overall performance (and the computer is often an
>> important tool for efficient administration in a company).
> Actually i think that is great. But most users don't take time for this. They would rather buy a box and expect the source of that box to fund such development.

Yes, agree - not the home users - but you were refering to companies
with commercial versions and support. The companies are important to
address. IMHO in the home user field Ubuntu cannot be stopped any more
in the long run - especially if Ubuntu continues to develop so well as
it did in the last one or two years.

Of course - as also has been mentioned - it is important to attract
developers to address Linux also. Most Windows developers I know are
completely ignorant of everything else but Windows although they get
beaten by their master (this year use winforms, the next year - oh no
- switch to wpf please). One reason might be that software development
is under pressure anyway already (so many software vendors - and
basically no money needed for a startup or single person to get
started). Addressing multiple environments increases complexity
signiicantly. And packaging then also must be done separately for each
platform.

And what options are for a platform agnostic way? Java, Python and
Mono - and several very little other players. I do not consider C(++)
here because it needs recompile on all supported platforms which then
must be maintained at developer side (compiling from source is not an
option for getting widely adopted). From those 3 only Java is very
well established and many try to ban it to server side only although
IMHO the option with the least required effort to make it work on all
platforms. Python would also be a good choice but the runtime not
widely spread so far (I don't know any Windows workstations at
customers where the Python runtime is installed). So for the
developers or software vendors it is not an easy thing to develop for
multiple platforms - maybe a reason why so many new softwares come as
web application and desktop applications still focus on the Windows
platform. What does Canonical recommend for companies planning a
migration to Ubuntu on the desktop within the next - let's say - 5
years? What should companies do if they need some software development
now?

OK, this maybe again not seems very relevant to Bug #1, but IMHO there
are many factors playing together. Home users buy there hardware often
at different shops where companies do and I guess Bug #1 would like to
address both groups.
-- 
Martin Wildam

http://www.google.com/profiles/mwildam

-- 
Microsoft has a majority market share
https://bugs.launchpad.net/bugs/1
You received this bug notification because you are a member of Registry
Administrators, which is the registrant for Debian.