On 02/18/2013 03:21 AM, Thibault Kruse wrote:
So
specific feedback leads to improved infrastructure.
Are you suggesting my earlier post to the mailing list over a month
and a half ago was not specific enough? Was my post wrapped in a
rage present? Absolutely. Should I have probably waited 24hrs before
hitting send. Maybe. But if this mailing list is only for
announcements/general discussion and the ONLY outlet for community
support is a site that does not accept your cry for help - you can
imagine how frustrating that is.
Speaking of frustrating, have you ever tried using ROS answers to
search for anything? Its a joke. I hate to promote anything
Microsoft, but this commercial is more appropriate for Askbot than
Bing's real competitors:
http://www.bing.com/videos/watch/video/bing-commercial-search-overload-cell-phone/qmk62tm0
The only thing that works logically is the tag system -- but even
that is inherently flawed as it requires users to tag their own
content. Some do that well and others might completely miss the
feature.
My specific comments about Groovy: First, how do you release a build
system that is ~18x slower than its predecessor? Did anyone go,
"hmm, build server is sure taking significantly longer..." or "wow,
its taking me 20x longer to do my job"? And if it was noticed how on
earth did someone ok it?
http://answers.ros.org/question/54688/why-is-groovys-rosmake-so-slow/
http://answers.ros.org/question/53925/rospack-became-agonizingly-slow/#53946
This has already been addressed and I believe there is even a
ticket, but that is appalling. That, by itself, would have kept me
from releasing catkin.
This is a bit of an edge case as I know it doesn't effect a lot of
ROS users, but right now, and I don't think I am reading between the
lines, the advice from the Orocos project is to stick with Fuerte if
you are an Orocos user:
http://lists.mech.kuleuven.be/pipermail/orocos-users/2013-February/006700.html
You'd have to consult them on what specific issues they are running
into. But the long build times plus lack of Orocos support is
keeping me from upgrading to it. Which is unfortunate because its
not all negative - I think the overhaul to Rviz is great. Much
better design/look. But then its the little things. Rviz now
supports Python! Great. But then there's no documentation at all on
the Python API. It looks, by examining the source, that it was just
wrapped by sip. On top of that, the python API goes missing for
interactive markers.
http://answers.ros.org/question/54278/what-happened-to-the-python-api-for-interactive-markers/
Answered and fixed, great response. But how does that get
overlooked?
Software as complex and as large as ROS will have some bumps and
bruises at a major release - but from the outside looking in, it
appears as though too much was changed in this release (as evidenced
by the above plus the longer than normal release cycle). Sweeping
changes are obviously needed to keep up with new use cases or
changing requirements. But more care needs to be taken by those in
the driver seat. With the size of the ROS community and its
popularity, more weight should be given to minimizing disruption
rather than favoring the hot new shiny stuff. Disruption breeds
discontent - discontent leads to searching for alternatives. And the
only way projects like ROS work is with a strong community
involvement. For once, one of these component-based approaches to
robotics has finally taken off (and history is full of failed
previous attempts). I don't want to see it happen again.