Saturday, January 31, 2009

The room gets closer still, the district server and class server advance, too

Peggy and John came from the district and dropped off the rest of the 25' power strips and cord protectors. Thanks guys (guy/gal)!

After volunteering for 8 hours last week, Fred again found time to volunteer this Thursday after school. We worked till 6 (and I spent another hour finished up a couple things on Friday AM) and got close to finishing the other side of the room! I just need to route 6 more Ethernet cables and plug in a few power cords, keyboards and mice :-) Next week?

Fred looking like a student, kind of

On Saturday Jack came to my house to continue the work of transferring the various Plone sites (requiring the re-installation/transferring of 3 different Plone instances) to the new district web server. This new server will have the capacity to quickly create content such as Vee's, discussions, upload images from many members at once. We/he of course had a challenge with the old Plone versions as he had to separately build the appropriate versions of Zope and Python (2.7.8 and 2.3.5, respectively). Then, he moved the content from the U server to the District server (255 megs in 1.5 minutes!). And... it didn't work. We got closer. Anyone know what this means: "ImportError: cannot import name base_hasattr"? A step at a time...

I did an experiment with the solutions proposed by LaserJock and others. I simply did a search for "firefox" on the thin client server, 'chgrp firefox name.of.file' to change the group from root to 'firefox'. Then, I changed the permissions on the file, 'chmod 750 name.of.file', making it so that the owner (root) can do anything, the group (firefox) can read/execute it, and (quite importantly in this situation) others can not read/write/execute it. Then, I tried to open firefox and... it didn't work! That was conditionally good. Then I added myself to the group, 'firefox'. Then I logged out then back in and... I could now access it! This is a simple solution, pretty much, at least in a temporary way. This way I can get the room up and running. Now I need to repeat this process with the different apps and many users... And work on getting a user friendly version of this for other teachers... I wonder if this solution could integrate with the edubuntu-menus?

Sunday, January 25, 2009

Ah plone... sometimes it doesn't work

I'm sure it has a good reason. Or at least maybe it has a good reason. I spent probably 12 hours this weekend striving to move my main site on the old server in my adviser's office at the U. of Minnesota to the new district server. It shouldn't have been too hard, exporting the 93 MB "Zope export file, .zexp" and then transfer it first at my computer, then over to the district server and from the air importing its into the new Zope instance that Jack and I were able to get working.

After several hours of trying to make that work including time spent on #plone irc I found out that's there is a bug in a program called, "wicked" which prevents it from being successfully imported. Arrrg.

After doing several things I tried a very different approach involving copying over the singular, large, database file from the U server to my house and then over to the district server. This too, ought to have worked. But it didn't. Jack tried to help me and so did a couple people from the plone community via the irc. all to no avail.

Well, did they say that Edison had to make 1000 failures before he could make one that worked? Hopefully next weekend I'll be able to move the site!

Friday, January 23, 2009

Really, half of room is done

Friday is the end of semester so, we teachers had most of the day dedicated to grading work and reporting the grades. So, I took advantage of his time to work on the server in the district office which will serve out the Plone sites and the CmapServer. I think I have most of the configuration sheet for the CmapServer done, well, it will be when I hear back from Rodrigo at the IHMC.

Also, I had invited Fred to a Friday workday. He got there at 9:00 and left at 5:30. However, there was a potluck lunch for the entire staff so at least he had something good to eat! As you can see the following photos he cleaned up my earlier work -- we had some good ideas, and finished another pair of tables.

Monday, January 19, 2009

More connections, more community

I needed help with the server (sound new?) so I jumped onto the #edubuntu irc last Tuesday (1-13) looking for help on using the Edubuntu-menus--an alternative to Sabayon I hoped. So good deal, Jordan Mantha AKA LaserJock, the lead person behind edubuntu-menus was on line and responded to my request for help.

Looks like edubuntu-menus might not be the solution (has 'right-click loophole') for what I need. I had a long and helpful chat with him that night. It's amazing: there are so many people 'out there' who are willing to (and currently) volunteer their precious time to help out education. As a matter of fact, part of my agenda for this project is to provide a medium through which these very smart and energetic people can make an important impact in education. Think Jack, Eddie, Christen, James, Fred, Brandon, Conner, Chris, Brian and Don.

LaserJock shared his ideas resulting from our chat that eve. His comments sparked a response from quite a few people -- at last view there were 31 comments on that blog entry. The internet provides the tools for communities of people to connect and create something much bigger than the sum of what they could, on their own, create. And, that's what I want to do within my classrooms of scientists, as well as connecting the scientist-students to the immense community outside the classroom.

Efficient management of student user computer privledges

In his book, "The Courage to Teach", Parker Palmer talks about the complexities of being the teacher, that is leader, of the classroom. In his book he tries to, "think the world together." He uses the concept of, Paradox to do this. I find it useful to consider Palmer's six paradoxes when designing classroom sessions:
  1. The space should be bounded and open.
  2. The space should be hospitable and "charged."
  3. The space should invite the voice of the individual and the voice of the group.
  4. The space should honor the "little" stories of the students and the "big" stories of the disciplines and tradition.
  5. The space should support solitude and surround it with the resources of community.
  6. The space should welcome both silence and speech
The first paradox, bounded and open, applies to different dimensions including student management. If one has no rules, the environment is so open that little exploration can occur as it is an emotionally unsafe environment. However, if it is too restrictive, likewise no exploration will occur. When I'm designing the student management system for the growing communities of scientists classroom I try to balance the needs of openness and boundedness. In this spirit, consider the 4 level permissions approach. At the bottom of that page it shows ideas for accomplishing this provided by Jordan Mantha and response to his blog post.

Saturday, January 17, 2009

New and improved cable management in the classroom

Note the 2 power strips on the bottom side of the un-painted 2x6 in lower left. The 2x6 is used to connect pairs of tables.

I called John at the district a week ago Friday. I was unable to reach him but did leave a message to call me. I had e-mailed him right before winter break a couple weeks prior about cables and cord protectors that I needed but hadn't heard back from him. He returned my call around 7:30 Monday morning. After about five back and forth e-mails and 48 hours, 2 people from the district delivered several boxes of power strips, patch cables, and cord protectors, all part of the plan to meet the fire marshals concerns. A few parts are on back order but enough things came to outfit my room of course. So, I spend a couple of hours this last week and fixed up a pair of tables. The following images trace the power route, from power strips to wall outlet:

I attached the power strips on the underside of the table-connector. We are putting it underneath to avoid tempting science-minded students to 'mess around with it' i.e. turn on and off, stick wires in it, etc. No, I didn't start teaching yesterday :-)The cords from the power strips join the CAT 5e patch cables and are held in place with strap metal.
Instead of using strap metal at the last attachment point on the table, we're using plastic tubing w/screws and washers--this won't cut into cord insulators.
Here you can see the cord protector bridge the gap. These are large cord protectors: both powerstrip cables and 4 ethernet cables fit through it! I've taped it down slightly. The plastic tube is serving as hardware to route the cables up the side of the cabinet.A split 1.5 inch pvc tube protects the cable as it goes over the counter top. (We've better ideas for this last step--check back in a few days). Soon the patch cable will plug directly into the drop in the wall and the switch will be in the drop room! We're not quite there, yet.

Saturday, January 10, 2009

A bit dry, but project proceeds

The physical setup
Fred Olson came over to Roosevelt after school on a Friday and we worked for a few hours. He worked on making a few additional Pentium 3's thin-client ready--removing the hard drive, disconnecting the CD and Floppy drives. No more disconnecting the main fan--it looks like I might have been wrong thinking that the thin clients operate well w/out them. Time will tell.

Additionally, there were several thin clients which wouldn't boot, and after a while we figured out that the problem was that they were set to use 220 volt systems and not 110 Volt! Upon flicking the small red switch they worked great.

Fred also made an improvement in how we affix the TC boxes under the tables. W/out immobilizing them they migrate quickly and wires get all over the place and there becomes no good place for the mouse/keyboards. Anyway, he screwed down the back lip of the box and put down a few strategically placed screws and, simple, quick, affixed!

The Plone Front
Jack Ungerleider Has put in two appearances over winter break in efforts to load plone onto the virtual machines (VMs) on our server at the district office. Of course this is all done remotely and there isn't a GUI to fall back on (I'm the one who wants it, not the pro's). Also, district IT people only allow VPN connections (then ssh or vnc), so Jack can only work on this at Roosevelt, my house where I have VPN or... through me via e-mailed instructions!

He's twice dealt with insufficient size limitations on the VMs on the server and with Brian Dolan-Goecke's help over the phone the 2 of them were able to set up disk quotas on all 5 VMs. Next, Jack tried to run the "buildout" file he had written--it basically contains a couple of hundred lines of instructions telling the computer to down load this and put it here then download that and put it there etc until the plone instance is all set up. Well, he ran that script and it should have been somewhat straightforward but alas... it hicupped. He has been researching stuff on the web, getting me info and I've been VPN/ssh-ing into the server and maybe it's up and running, but because of firewall issues I can't yet access it. I wasn't able to go to the site and of course have not been able to migrate the present site. The current site looks quite boring as anything worth looking at can only be seen if one is logged into the site. In conclusion, a few silent steps were made in this component of the GCoS project.

Lots of details on the server
Joyfully, I've been studying the book that Conner gave me, Linux Pocket Guide (by O'Reilly), much more understandable than I'd thought it would be. I've used knowledge that was pulled together by reading that book to do many little things like successfully install CmapTools and add java plugins on the thin client, to understanding better how to navigate using Terminal and how to use -options on a command, and where to guess that things might be located on the Linux system. Also, I now open and edit files using "vi", an editor, within Terminal. Really, I've done quite a few other things too, like trying to figure out how to edit the edubuntu_menus config files that Conner installed. This last effort is really the last major hurdle in the thin client server setup that is stopping a full scale classroom trial of the CESC. Soon, I hope!

I just found laserjock's blog which lead me to joining Edubuntu User and Developer mailing lists--better late than never as my dad used to say :-). This community I need to join!