Evaluation issues Outside Resources Question on Experience with e-Labs Upgrades Status Swift/OGRE New Hire LIGO report Status of SWIFT/LIGO Data News Dale's Status Use of Mashups IT Survey

Present: Mihael, Eric, Liz
Absent: Mike W., Randy, Kris, Joao, Dan K., Bob, Tom L., Tom. J

A parallel telecon on evaluation took place while we talked about technical issues. Marge, Bob, the two Toms, and Dale participated. They should have their own set of notes.

Evaluation issues

Outside Resources

Tom L. brought the issue of using outside resources at the last meeting, and Eric has been thinking about it more. What role do/did outside resources and/or outside data play in a teacher's use of the [LIGO|cosmics|CMS] e-Lab?

An example of an outside resource would be the Quake Catcher Network.

Questions for Jean: Do we want to avoid outside influences? If we allow them, then ask –

• Did you use outside data?
• Did you use outside internet resources or did you just use the e-Lab barebones?
• Did this make a contribution to your investigations (distract, helped, no effect)

There are two alternate ways to work with outside data –

• Copy it from other place and add data sources to our e-Labs.
• The user interface allows you to get data from some other data source (we don’t’ have to copy it over, get it directly from the source). (NOAA)

Examples of data that might be collected by students and added to the e-Lab: Students would have classroom weather stations in classrooms and upload data from their own weather station.

Question on Experience with e-Labs

Eric suggested we should ask "Have you done an e-Lab for another content area?".

Upgrades

System folks are moving along at their own pace. Only www12 has been touched so far, and that needs further testing. Eric’s feedback has been somewhat helpful. They are going to do everything except the production machine. Eric is hoping to use the packaged forms of Apache web server and PHP. (Note added later: this seems less likely -EAM)

Status Swift/OGRE

Transfer of CMS Testbeam data. – nothing on this.

New Hire

Tony has accepted the job; (Note later in the week: Tony rescinded his acceptance to take a permenant job, so we have to restart the process.)

LIGO report

Bluestone production is to be moved to ANL cluster. Eric found that www13 is slower than tekoa. We might overload the production machine. Eric also found a race condition on www13 -- a script was not setting its lock file soon enough. Fixed in future release. Will users run analysis locally (on www18)? Mihael thinks we should try it out first. He does not think it will become unresponsive. If it is the case, then we should find strategies to deal with it. We could run it on a different machine. But worry about that when it's a problem. (Eric is designing Bluestone so that it can be run on the same machine or on a different machine in the cluster, just in case.)

Status of SWIFT/LIGO

Last fall Mihael and Eric ran an analysis on Teraport from the command line. Since then, their accounts got wiped out on Teraport, so Eric has to do this again to set things up. His notes should make this go faster, and lead to better documentation. In the meantime, Eric worked out how to have Bluestone "login" to the e-Lab on behalf of the student, so save (upload) a plot. Having that working will also help in getting Bluestone to run jobs through SWIFT.

Data

Eric thinks we are getting bad data from DMT; He switched the source of DMT frames from /dmt/New_Seis_Blrms to the "archive" /archive/frames/dmt/New_Seis_Blrms, at Greg Mendell's suggestion (and because this is how it will work on the Caltech machine) on 8 Sept. At about the same time we started getting wacky data, so Eric thought he'd done something wrong. Greg Mendell wrote back to say that all of DMT was screwed up about that time, and will be fixed. So it was just a coincidence. Eric has to wait for Hanford to fix this. We may lose any interval of bad data which we may have to delete it. It may get backfilled with the correct data so we could keep the data. Eric logged this problem in the Boiler Room, which is not open to public. Should we add a News item? Eric can talk to Bob about what to announce.

Eric is working on improvements to the script which collects the data. Part of this is preparing for the new machine at Caltech. He's also created a page in the CI Wiki (see LIGO Relay Server) so that Greg or someone else can maintain the machine at Caltech even if Eric is no longer around. Greg will follow Eric's notes to set up the machine, and give Eric feedback on improvments. For starters it's really just documenting what tekoa does now. Once the machine at Caltech is up and running then www13 will be switched to get data from there, with tekoa acting as a backup. Eric probably won't have a login on the Caltech machine, unless there are problems, so he's using tekoa to get things in shape before the Caltech machine is set up. He has put nothing in RSS News. Should he?

News

Dale liked seeing Bob's notice in the news box on the LIGO e-Lab. What should we have in the future? If it is a mashup, it may not matter where the news is stored.

Dale's Status

Dale may be tied up with planning the building of a new Outreach Center at Hanford which NSF is going to support.

Use of Mashups

Mashups may allow schools to access information from wiki without the school’s needing to be able to access the wiki directly. The i2u2 server would do the mashup and the information would look like it was coming from the i2u2 machine and not the wiki. Schools may have a policy to block sites that run a wiki (unfounded as that might be). Having information on the wiki allows changing of content by us or people we want to help, rather than having to edit jsp pages and put them in SVN.

IT Survey

Eric proposed we do a survey about school’s policies about wikipedia, firewalls and site blocking, JavaScript and Java or desktop installation, etc..

Example of questions:

• Are wikis in general an issue at schools?
• Do students have separate emails?
• Is JavaScript allowed?

Eric wondered what we asked before when we asked schools. Whenever we do a workshop, we could ask them these questions of their schools. It would give us a better picture.

Note: this proposal for an IT Survey is completely separate from the evaluation process! There was some (understandable) confusion about that when these minutes were first posted.

-- Main.LizQuigg - 16 Sep 2008
-- Main.EricMyers - 16 Sep 2008