On 10/12/13 10:36 AM, Dave Koelmeyer wrote:
On 12/10/13 07:06 PM, Nikola M. wrote:
I also don't want *anytime* to be left with an update of Hipster that
just break all everyday apps. (Plus all old apps binary compatibility)
It must be considered with the rest of the apps and people that are
using it.
People are willing to use Hipster in everyday installations if it
does not break horribly without announcement.
I would have thought it was pretty clear at this stage (also given
previous recent posts) that Hipster is bleeding edge,
developer-focused and breakages are to be expected. Given this I'm not
sure why you or anyone else would be using it in "everyday
installations" instead of the a8 release.
Like Sriram suggested, point of having Hipster is that someone will use
it, install and update from it
and test changes in that way while reporting bugs, so they could be
fixed prior putting updates to /dev release.
If it can Not do everyday tasks (working latest Adobe flash, opening
PDFs, using many other everyday applications (Firefox Thunderbird,
OpenOffice and so on)
no one would consider installing it, using it for a subset of desktop
tasks, that are gradually widen to server use with
Opensolaris/Openindiana management techniques.
It is whole point of the Openindiana and Hipster effort to have
something usable and to have users using it and reporting further real
problems toward component and packages changes.
If we don't have basic (desktop) functionality working undisturbed and
not broken for those everyday tasks, we would have unusable platform
from every user supposed to use it regularly,
that expand knowledge about the platform and report more important bugs
(and feature requests!) then UI and everyday applications.
We are mostly not people that are payed to do testing (and use Hipster),
to make it main job.
We are mostly casual users that are willing sacrifice benefits of other
open or closed source platforms in favor of being included in the
project that has clear future and growing rate
And things like breaking Gnome desktop (clearly untested before release)
and borking adobe flash support in 151a8 and Hipster, shows alarming
lack of understanding of this user-desktop user tester , user-server
administrator , developer-contributor and developer-distribution
maintainer relationship.
Where would be next OI /dev release be tested, but in Hipster?
If user-tester have all features he used to, gradually broken and
non-working and user-contributor make changes that affect wide range of
packages, _binaries and sub-projects, user-distribution maintainer needs
to stop that without planning, proposing and selecting people for
projects that deal with side-affect of user-developer changes and that
actually care for user-server admin, as the final consumer of the product.
This is my intention to discuss obvious problems that Openindiana have
with release management
and quality control and I see no justification to ignore user requests
for clear open testing process that is suppose to produce quality and
improvements, not drawbacks as a result.
Every user should have clear information how project actually functions
and who does what,
without need to read tons of IRC discussions, mailing lists and be there
for 5 years,
without having a f* idea what are steering paths and models of choosing
future of a distro.
_______________________________________________
oi-dev mailing list
[email protected]
http://openindiana.org/mailman/listinfo/oi-dev
