Some thoughts
27 views
Skip to first unread message
Bill Smart
Jul 19, 2012, 8:22:00 PM7/19/12
to ROS Education SIG
Colleagues,
First, my apologies for promoting myself as coordinator, then dropping
off the map. I'm now firmly ensconced in my new office at Oregon
State, and fully back on-line again.
I wanted to propose some concrete first steps towards the goals that
we'd been talking about on the list. The first of these is,
recursively, to decide on a set of concrete goals for the SIG, both in
the short term and for the longer term. As a starting point, here's a
straw man for the goals. This is purely a starting point, and it
strongly reflects my wants and biases. Pitch in with your thoughts.
Short Term
- Provide an easy-to-install, out-of-the-box ROS experience for novice users
- Provide initial supporting materials to allow this to be used in a
classroom setting
- Prototype this in a very small number of university-level classes in
the fall, with the goal of bringing it to a proper beta release status
by the start of the spring session next year.
Medium Term
- Beta-test in a larger number of universities in the spring, and
gather feedback
- Integrate feedback, and aim for a 1.0 release by the fall 2013 session
- Recruit people to add open educational content (lecture material,
assignments, and *working* packages)
Long Term
- Integrate contributed material into a stable release
- Promote in CS education circles (SIGCSE, for example)
This leaves a couple of open questions, which I think we also need to
answer soon. Answers are mine, and are likely to be wrong and
offensive. Pitch in!
Q: What's a novice?
A: Advanced undergraduate or graduate student, at a traditional
university, able to program moderately-well in Python
Q: What's the infrastructure?
A: An LTS version of Ubuntu and a similarly-frozen version of ROS
(coordinated with the Enterprise ROS folks)
Q: What's the deployment mechanism?
A: Memory stick.
Q: Real robots, or just simulations?
A: Definitely simulation (both stage and gazebo) on the stick. I'd
like to support turtles, but that's another layer of complexity.
In order to move things along, I also propose that we bring the list
of goals to a vote on July 31st. That's only a week and a bit away,
but I think that it should be enough time to settle on things, and
will let us get started on parceling up the work and making actual
progress. If this seems too short for anyone, then speak up now.
I'm teaching a class to a bunch of engineers in the fall, and I'd be
delighted to test all of the stuff that we come up with in that
context. If anyone else is in a similar situation, then please speak
up. The more we can actually deploy our ideas, the more we can figure
out what the hidden problems are (like sysops objecting to memory
sticks hijacking their computers, for example).
cheers
-- Bill
First, my apologies for promoting myself as coordinator, then dropping
off the map. I'm now firmly ensconced in my new office at Oregon
State, and fully back on-line again.
I wanted to propose some concrete first steps towards the goals that
we'd been talking about on the list. The first of these is,
recursively, to decide on a set of concrete goals for the SIG, both in
the short term and for the longer term. As a starting point, here's a
straw man for the goals. This is purely a starting point, and it
strongly reflects my wants and biases. Pitch in with your thoughts.
Short Term
- Provide an easy-to-install, out-of-the-box ROS experience for novice users
- Provide initial supporting materials to allow this to be used in a
classroom setting
- Prototype this in a very small number of university-level classes in
the fall, with the goal of bringing it to a proper beta release status
by the start of the spring session next year.
Medium Term
- Beta-test in a larger number of universities in the spring, and
gather feedback
- Integrate feedback, and aim for a 1.0 release by the fall 2013 session
- Recruit people to add open educational content (lecture material,
assignments, and *working* packages)
Long Term
- Integrate contributed material into a stable release
- Promote in CS education circles (SIGCSE, for example)
This leaves a couple of open questions, which I think we also need to
answer soon. Answers are mine, and are likely to be wrong and
offensive. Pitch in!
Q: What's a novice?
A: Advanced undergraduate or graduate student, at a traditional
university, able to program moderately-well in Python
Q: What's the infrastructure?
A: An LTS version of Ubuntu and a similarly-frozen version of ROS
(coordinated with the Enterprise ROS folks)
Q: What's the deployment mechanism?
A: Memory stick.
Q: Real robots, or just simulations?
A: Definitely simulation (both stage and gazebo) on the stick. I'd
like to support turtles, but that's another layer of complexity.
In order to move things along, I also propose that we bring the list
of goals to a vote on July 31st. That's only a week and a bit away,
but I think that it should be enough time to settle on things, and
will let us get started on parceling up the work and making actual
progress. If this seems too short for anyone, then speak up now.
I'm teaching a class to a bunch of engineers in the fall, and I'd be
delighted to test all of the stuff that we come up with in that
context. If anyone else is in a similar situation, then please speak
up. The more we can actually deploy our ideas, the more we can figure
out what the hidden problems are (like sysops objecting to memory
sticks hijacking their computers, for example).
cheers
-- Bill
Bill Morris
Jul 19, 2012, 9:31:54 PM7/19/12
to ros-sig-...@googlegroups.com
On Thu, 2012-07-19 at 17:22 -0700, Bill Smart wrote:
> Q: What's the infrastructure?
> A: An LTS version of Ubuntu and a similarly-frozen version of ROS
> (coordinated with the Enterprise ROS folks)
Progress is being made on this front. IHE is initially focusing on
> Q: What's the infrastructure?
> A: An LTS version of Ubuntu and a similarly-frozen version of ROS
> (coordinated with the Enterprise ROS folks)
building ISOs for a TurtleBot install target, followed by a generic ROS
Workstation target. We have forked some of the Debian packaging work of
Austin Hendrix and Michael Brindle at WG, and hope to merge things in
the future.
I'm not sure things are quite ready for testing yet, but if any one
wants to try a _beta_ of a new TurtleBot ISO running ROS Fuerte and
Ubuntu 12.04, it is available here.
https://www.iheartengineering.com/files/turtlebot-fuerte-ubuntu-12.04-amd64.iso
Warning, you will probably need to reinstall from scratch once the final
release is made.
If you want to try rolling your own, our code is can be found below.
However keep in mind that it is currently being refactored so package
names will probably change.
UCK build scripts
https://github.com/TurtleBot-Mfg/turtlebot-iso
Here is our fork of turtlebot-debs
https://github.com/TurtleBot-Mfg/turtlebot-debs
Our goal is to provide a new ISO release by mid August.
Thanks,
Bill
--
Bill Morris <bi...@iheartengineering.com>
I Heart Engineering
http://www.iheartengineering.com
<3
Joe Tojek
Jul 19, 2012, 9:39:23 PM7/19/12
to ros-sig-...@googlegroups.com
hi Bill - Great to have you back. I am excited to get back to this and I am putting together a machine for the ROS > URDF > Gazebo workflow that can support real-time screen capture in HD. So that might come in handy for video demos. I also looked into places to build
FOSS content like wiki books, but I imagine we will build it on ros/wiki to start? I am open to a goal vote anytime. Thanks!
Bill Smart
Jul 20, 2012, 12:50:27 PM7/20/12
to ros-sig-...@googlegroups.com
>> Q: What's the infrastructure?
>> A: An LTS version of Ubuntu and a similarly-frozen version of ROS
>> (coordinated with the Enterprise ROS folks)
>
> Progress is being made on this front. IHE is initially focusing on
> building ISOs for a TurtleBot install target, followed by a generic ROS
> Workstation target. We have forked some of the Debian packaging work of
> Austin Hendrix and Michael Brindle at WG, and hope to merge things in
> the future.
Bill,
>> A: An LTS version of Ubuntu and a similarly-frozen version of ROS
>> (coordinated with the Enterprise ROS folks)
>
> Progress is being made on this front. IHE is initially focusing on
> building ISOs for a TurtleBot install target, followed by a generic ROS
> Workstation target. We have forked some of the Debian packaging work of
> Austin Hendrix and Michael Brindle at WG, and hope to merge things in
> the future.
Good stuff. IMO, 12.04 and Fuerte is as good a combination as any,
and I'm more than happy not to have anything to do with actually
producing the iso. Anyone else have any thoughts or comments on this?
If not, then I'd propose that IHE takes the lead on this (or,
continues to take the lead). Bill, if there's no dissention, the can
I get you to put a bit on the wiki (Proposed Infrastructure, or
something like that) that briefly describes what you're up to. Once
this is up and people have a chance to look it over, we can add it to
the vote.
cheers
-- Bill
> I'm not sure things are quite ready for testing yet, but if any one
> wants to try a _beta_ of a new TurtleBot ISO running ROS Fuerte and
> Ubuntu 12.04, it is available here.
> https://www.iheartengineering.com/files/turtlebot-fuerte-ubuntu-12.04-amd64.iso
> Warning, you will probably need to reinstall from scratch once the final
> release is made.
>
> If you want to try rolling your own, our code is can be found below.
> However keep in mind that it is currently being refactored so package
> names will probably change.
>
> UCK build scripts
> https://github.com/TurtleBot-Mfg/turtlebot-iso
>
> Here is our fork of turtlebot-debs
> https://github.com/TurtleBot-Mfg/turtlebot-debs
>
> Our goal is to provide a new ISO release by mid August.
>
> Thanks,
> Bill
>
> --
> Bill Morris <bi...@iheartengineering.com>
> I Heart Engineering
> http://www.iheartengineering.com
> <3
>
> You received this message because you are subscribed to the Google Groups "ROS Education Special Interest Group" group.
> To post to this group, send email to ros-sig-...@googlegroups.com.
> To unsubscribe from this group, send email to ros-sig-educat...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
Bill Smart
Jul 20, 2012, 12:52:14 PM7/20/12
to ros-sig-...@googlegroups.com
One question just occurred to me. August is getting close to the
Groovy release date. Should we target Fuerte (which we can test now),
or Groovy (which will be newer) for the fixed version? Some of this
might be determined by the change lists for Groovy (which I can't
recall right now).
-- Bill
Groovy release date. Should we target Fuerte (which we can test now),
or Groovy (which will be newer) for the fixed version? Some of this
might be determined by the change lists for Groovy (which I can't
recall right now).
-- Bill
Bill Smart
Jul 20, 2012, 12:54:06 PM7/20/12
to ros-sig-...@googlegroups.com
Joe,
I'd imagine that the wiki is the right place to start, at least in the
short term, since it's easy and already in place.
-- Bill
> https://groups.google.com/d/msg/ros-sig-education/-/hxutE0imBq8J.
I'd imagine that the wiki is the right place to start, at least in the
short term, since it's easy and already in place.
-- Bill
> --
> You received this message because you are subscribed to the Google Groups
> "ROS Education Special Interest Group" group.
> To post to this group, send email to ros-sig-...@googlegroups.com.
> To unsubscribe from this group, send email to
> ros-sig-educat...@googlegroups.com.
> To view this discussion on the web visit
> You received this message because you are subscribed to the Google Groups
> "ROS Education Special Interest Group" group.
> To post to this group, send email to ros-sig-...@googlegroups.com.
> To unsubscribe from this group, send email to
> ros-sig-educat...@googlegroups.com.
> https://groups.google.com/d/msg/ros-sig-education/-/hxutE0imBq8J.
Bill Morris
Jul 20, 2012, 1:12:52 PM7/20/12
to ros-sig-...@googlegroups.com
I can probably get something written up by Monday as to where we are now
and where we are currently planning on going.
The switch from Fuerte to Groovy should be relatively painless. The
biggest issues we are facing now is dealing with building system Debian
packages for 12.04 and the thirty minute compile/test development cycle
for the ISO installer.
Thanks,
Bill
and where we are currently planning on going.
The switch from Fuerte to Groovy should be relatively painless. The
biggest issues we are facing now is dealing with building system Debian
packages for 12.04 and the thirty minute compile/test development cycle
for the ISO installer.
Thanks,
Bill
Yujin Stonier
Jul 23, 2012, 1:37:34 AM7/23/12
to ros-sig-...@googlegroups.com
On 21 July 2012 01:52, Bill Smart <sma...@engr.orst.edu> wrote:
One question just occurred to me. August is getting close to the
Groovy release date. Should we target Fuerte (which we can test now),
or Groovy (which will be newer) for the fixed version? Some of this
might be determined by the change lists for Groovy (which I can't
recall right now).
We're currently working on turtlebot2 software with fuerte. I don't anticipate there'd be too much trouble slipping into groovy despite the build environment upheaval, so we should be fine either way.
Phone : +82-10-5400-3296 (010-5400-3296)
Home: http://snorriheim.dnsdojo.com/
Yujin Robot: http://www.yujinrobot.com/
Bill Smart
Jul 23, 2012, 7:18:19 PM7/23/12
to ros-sig-...@googlegroups.com
Given this, and Bill Morris' earlier answer, it makes sense to me to
aim for 12.04 and Groovy for the first target. This will make sure
things are as up-to-date as possible on the ROS side, although it does
mean some last-minute fiddling to make sure everything works with the
new release.
-- Bill
aim for 12.04 and Groovy for the first target. This will make sure
things are as up-to-date as possible on the ROS side, although it does
mean some last-minute fiddling to make sure everything works with the
new release.
-- Bill
Reply all
Reply to author
Forward
0 new messages