<br><br><div class="gmail_quote">On Thu, Sep 23, 2010 at 12:22 PM, Polyvertex <span dir="ltr"><<a href="mailto:polyvertex@gmail.com">polyvertex@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
Understood. That's only a part of my point.<br>
<br>
My main concern is that users don't always vote and do feedbacks plus<br>
the fact they currently don't know where to go if they want to. This<br>
is a fact.<br>
This problem cannot be seen with popular and visible applications but<br>
it becomes a blocking one with small applications, hence my initial<br>
post : the extras repository contains a too old version of my<br>
application since too long, and the right version is stuck into<br>
extras-testing. This is also a fact.<br>
<br></blockquote><div><br>The trouble is that no matter how popular apps like FAP or appdownloader
get, IMO there will always be a very dominant number of users who will
use the system-provided app manager. I agree that the scoring scheme for
Extras which we DO have on <a href="http://maemo.org">maemo.org</a> is grossly underutilized -
something that IMO stems from the previous point (giving feedback is
difficult and people don't install things from the web anyway). I feel
the best we can do for *fremantle* is continue streamlining
extras-testing (extras-assistant and FAP are of course very welcome to
make this experience better for at least a portion of users) and try to
get these points as clearly as possible through to the Harmattan program
so that this stuff in the ends finds it's way to next-gen app managers. <br>
<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
When I expose those two facts here, the answers I get is that things<br>
has been drastically improved since Chinook/Diablo epoch and one is<br>
even happy about the nature of my complain. Great.<br></blockquote><div><br>I'm not happy that there are things that are broken or inefficient - I just think it's a good thing that we ARE moving in a good direction of reducing delays without outright sacrificing the relatively safe-harbor nature of Extras. Not perfect, and needs continuous kaizen-style improvement, but moving ! (I *really* need to make some graphs...) Please forgive me if I discard something with something seen as a been-there-done-that, it's not dismissive of your enthusiasm or popping ideas in - it's just that some things have been debated over quite extensively and I don't see that anything has changed fundamentally that would make a difference. <br>
<br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
You see, in my example, I am perfectly aware that the testing version<br>
of my application will never reach this 10 karmas score and that it<br>
will be unlocked only because time has passed. This is a problem.<br></blockquote><div><br>Yes, that's why we will try to increase tester input volume with specialized tools. On a side note - it will be unlocked because you have at least 3 positive tester votes, the additional quarantine time is there only to avoid the tester votes 'overpowering' non-tester votes. And yes, I agree that we might need different rules for packages that are updates for things already in extras (as that is already listed in extras-testing procedure improvements).<br>
<br>@Naresh: Hey, I'm a dev with packages in extras-* too ! :)<br>
<br>
Best regards,<br>
Attila <br></div></div><br>