Mini-Education Summit LinuxCon Boston 2010

I spent most of last week at LinuxCon helping Spot at the Fedora booth. However, the day before the main conference, Sebastian Dziallas organized a Education Mini-Summit to take place in conjunction with LinuxCon.
I gave a talk on the Inkscape class Red Hat has done plus some other related initiatives, including one we are planning for next fall with the Free Software Foundation. The slides are available here.
There were so many great talks. Here’s an overview of the ones I attended (my apologies for having no details on Caroline Meek’s talk on “Computers in US Schools: Realities and Challenges and how Open Source can help” – I had to leave before her talk for the FSF Women’s Caucus dinner in Cambridge.)

“Can Open Source Save The World…?” Bryant Patten, NCOSE

Bryant split his talk into three sections: the bad news, the good news, and the better news.

The Bad News

Here’s an overview of the points Bryant made in this section:

• Waiting for Superman, Two Million Minutes: A Global Examination, The Global Achievement Gap by Tony Wagner, What the internet is doing to our brains: the Shallows by Nicholas Carr are books and films that discuss some of the problems facing education today.
• There’s a perennial pull in education between content and skills. Bryant said we should be talking about content AND skills, and his slide had a full-screen ‘&& ! ||’. This kind of argument can cause issues in discussions about improving education, though.
• For the last 10 years or so, schools don’t want to teach tech – they just want to have it seamlessly integrated across subjects. Bryant disagrees with this position, though. He believes we need to teach technology too. One of the excuses he hears against it is that “kids know it better than we do, we don’t need to teach it!” Bryant counters, however, that kids know how to entertain themselves with the computer – their actual knowledge of technology is wide but very shallow.
• Within what fields are these students’ future jobs likely to be in? 18,000: number of jobs available for physicists in 2016. 91,000: number of jobs for chemists in 2016. 4,006,000: number of jobs for technologists in 2016
• Finally, the last bad piece of news is that there is still an attitude about free and open source software in schools: “It can’t be good it’s free.” “We can’t teach it – it’s not industry standard.” One other current misconception is that free & open source software is the same thing as web 2.0. No, Google Docs is not free software, even though it’s free as in beer. That it costs no money isn’t the most important point.
• I believe it was Algot, a K12 educator with 36 years of experience, in the audience who brought up the point that in k12, you’re not given the opportunity to retry if you fail. “We tell kids to scratch their itch as long as they get 70-75% or over on their test. If they don’t, they stop caring, or they care like crazy and turn themselves into dunces because they can’t afford failure again so they just turn off. The kids are left behind because they stop engaging.”

The Good News

• The free & open source software available today is the good news.
• Digital Equity – you can give it to kids, they can have it in both their homes if their parents are split, there is no monetary barrier to access.
• Teachers get more choice in what apps they choose to use.
• He mentioned theingots.org – a normative testing standard for technology.

The Better News

• NETS – the National Educational Technology Standard for students: the next generation. The guidelines have been loosened such that FLOSS is now eligible to compete – it no longer mandates Microsoft-specific technologies.
• project FOSS4ward – students and FOSS teams, submitting documentation, testing, creating tutorials, template and clipart creation for free software.
• Emerging maker culture in the US – Maker Bot, for example. A lot of opportunities for students to learn and create.

Ruth Suehle wrote up a great recap of Bryant’s talk on opensource.com: Bryant Patten on open source education (LinuxCon session recap).

“Open source improving education around the world” Ruth Suehle, Red Hat, Inc.

Ruth reviewed a number of projects and initiatives involving open educational materials and FLOSS:

• OER commons – makes it easy for teachers to license, share, and categorize educational resources so they can find them and use them.
• Common Core state standards initiative – a non-federal gov’t initiative to build open standards for schools
• Curriki, a wiki providing quality free educational content.
• IT@Schools – a project that is bringing FLOSS to 3 million Indian students.

I’ve kept my summary of Ruth’s talk short just because many of these projects are covered in great detail at OpenSource.com’s Education channel – so please pop on over there to read more if these pique your interest!

Being Present – a Beginners Guide to FLOSS Outreach in Education Karlie Robinson, Webpath Technologies

Karlie has been very involved in both the Fedora project and the Rochester Institute of Technology (RIT) and its usage of OLPCs in computer science courses. She outlined how she began her involvement – it started when her company, On-Disk.com was asked by Greg DeKonigsberg to produce Fedora 10 SD cards for use with the OLPC laptops. Afterwards she attended what she classified as a random meeting on RIT’s campus, where she connected some students and professors with the OLPC project and which culminated in Professor Jacobs using OLPCs in a computer science class in which students developed math software for 4th grade students using OLPC as a platform. Karlie talked about how she is not a technical contributor, rather she served as a ‘router’ to connect people and projects together to help make things happen. She emphasized how impactful it is to simply show up, to be available, make connections, and share your knowledge.
Karlie also talked about some other initiatives happening at RIT involving open source. Currently they are working on improving video chat on OLPCs in the hopes of providing enough fidelity for sign language video chat, with the help of funding from the National Institute for the Deaf. The project is leading to better video drivers and software for the OLPC. Typically for college computer science students, their buddy working on a project with them or in the same class is right down the hall, and they can chat in-person and help each other out and work on things. In FLOSS, the help you need is in an IRC room, not a dorm room down the hall, and working within that social space takes a bit of adjustment. For example, Karlie gave the example of a student struggling to find an answer to a question he had on the software he was working on. He told Karlie he couldn’t find the answer. “Who did you ask? ” asked Karlie. “Oh, it was late” the student replied. Karlie responded, “It’s never late!Ppeople all over the world come online all the time who can help you.” She’s had to explain to the students that time doesn’t exist in the FLOSS world and if a chat room is silent, it’s time to hit the mailing list – just don’t give up. She relating teaching community interaction to teaching a foreign language. These students know python and C and java, but they don’t know who to ask for help. This is in a world where proprietary software developers are trained to not ask anybody else – loose lips sink ships. And what’s considered cheating and plaigarizing vs. collaboration? FLOSS challenges some of these computer science instructional notions. Karlie suggested that with FLOSS, if you write something from scratch you should get an F. Karlie also mentioned how the FLOSS world has made computer science instruction easier, and she related an example of a professor from Seneca College who told her “It’s much easier to grade a class based on git and wiki commits. You can see who did the work and who didn’t, and how well they managed their time.”
Finally, Karlie said that open source isn’t code. It’s a method, it’s a culture.

The State of Open Data in Education Colin Zwiebel and Andrew Pethan, Olin College

Colin and Andy started off by introducing themselves. They are both currently Olin College students. Colin started with Linux at age 11 – he found his way to FLOSS. Andy built a computer when was younger. He recently took a year off from Olin to start a software business in education. He didn’t know a lot about business, though, and he learned a lot about it and education. Colin worked at a non-profit this summer in New York City that focuses on public transportation and open source technology. Andy worked at IBM this summer on Lotus.
Colin and Andy then painted their idea of data utopia, where everyone’s app makes their data usable by other apps and easy to share. They’d like to see much smarter software, and more learning based on data findings. There are some practical concerns – What can you do with that data? Who do I allow to use that data? What is the license of that data? But they think the community around building standards and interoperation is important.
Colin gave the example of how New York City recently released a public domain transportation dataset in an open format. What are people doing with this boring timetable data, he asked? He then gave some examples of applications that make novel usage of that data. One is called “Exit Strategy NYC”, which, if you tell it where you are trying to go, will tell you not only which way you need to head out of the station to best get where you’re going, but even which train car to get into for the most efficient trip.
They both went on to talk about how poor the data on education is today, and what kinds of things might be possible if schools had some standard open format and tools with which to report up that data. Today schools rely on proprietary software that locks up their data and makes it hard for them to share data as well as migrate to better tools. Andy and Colin mentioned some initiatives that might help achieve this goal:

• National Education Data Model (NEDM) – just releasedthis year, it’s a standardized way / db schema to describe everything, from students / blood type to bus routes to neighborhood demographics. They couldn’t find any reference implementations of it. It’s XML-based so you can transform it and submit it for federal reporting you need to do, good incentive for schools to use it.
• Schools Interoperability Framework – It’s been around for about 10 years and hasn’t been implemented in any product so far. There’s a project (Open ZIS) to develop a zone integration server to work with SIF, where you should be able to hook something else in as a teacher, to do interesting analysis, help you track grades differently, or other innovative things. A single teacher could just access the students in their class, and outside of the school the data could be anonymized.
• Common Cartridge – a common way for publishers to create something that will plug into a CMS. It’s gained buy-in from Blackboard, Pearson, McGraw-Hill, and is supported by apps such as Moodle.
• School of One, NYC Dept of Education – a pilot program of a data-driven education model, one kid per computer. It involved structured content to switch instruction based on the best individual learning model – for example, teaching a math concept for a visual learner vs. a math concept for aural learner, etc. They tried to correlate performance with the order in which it was learned, the style in which it was presented in. Aggregate data can’t provide that kind of insight.

What kinds of innovative things could come from schools sharing their data? Colin gave one example, “What if a teacher tweeted each topic she was about to cover during a typical school day, and that was correlated with student attendance. You could see based on attendance and teacher tweets what topics a student missed when they were gone, and try to correlate the absences with to questions missed on an exam.”
Schools are trying to buy packages and services from companies, but Colin and Andy think that’s the wrong model. If the data is provided, then more free applications could surface to allow schools to do more with the data they already collect
Ruth Suehle wrote up a fantastic article on Colin and Andy’s talk on opensource.com: The importance of open data in education.

The Open Source Way: Leveraging Communities Sebastian Dziallas, Fedora Project & Sugar Labs

Sebastian first got involved in open source two years ago as a high school student in a computer club. He was looking for a way to improve his high school’s computer system. He started looking at Linux and poked around, and was very lucky things worked out how they did. He did an internship at 11th grade at a computer magazine, and a person he met there was a former leader of Fedora Engineering Steering Committee. This person challenged Sebastian to get involved in open source. Sebastian started approaching open source by looking at educational initiatives, as he was a high school student. The wiki pages he found on the topic were outdated, so he started poking around on the related mailing lists and also did not see much stuff going on. He decided that he wanted to create an education spin of a Linux distro and chose Fedora. From the start though, people were discussing technical issues such as if it should be CD- or DVD-sized. Sebastian really didn’t care, he just wanted to get started and have a spin at all. Greg DeKonigsberg emailed him off-list and they started talking…. that helped Sebastian a lot, and eventually Greg pinged him in IRC and asked him if he wanted an XO. He really liked OLPC but didn’t know how to get involved because they were so far away – he is just in HS. Greg asked him if he wanted an XO, and Sebastian said of course, and this really jump started his involvement.
One of Sebastian’s first projects was developing the Fedora 10 SDcards for OLPC. Through this project, he got to know Walter Bender and Mel Chua and found his way through OLPC and Sugarlabs. He ended up creating Sugar on a Stick – a Sugar desktop environment (as seen on the OLPCs) that can be booted off a USB key so you don’t need OLPC hardware to use Sugar. There is now a 1st grade teacher in Boston who is using Sugar on a Stick in her classroom. Although Sebastian was in Germany, he was called into their classroom three times. Her students have provided bug reports and have learned about how open source projects work. One bug report they made was on the maze activity, which resulted in a fix. The team working on Sugar on a Stick now includes Sebastian along with Peter Robinson and Mel Chua.
Sugar on a Stick, as a project, is an example of both how things can go right and how they can go wrong. First Sebastian showed us the ‘Blueberry’ release’s release cycle. It involved the orchestration of three main upstream components:

1. Fedora / core OS
2. Sugar core / Sugar Labs
3. OLPC activities / third-party developers

Combining these three upstreams to build Sugar on a Stick worked well for the first release, but it didn’t work so well for the second release. As Sebastian was about to jump on a plane for FUDcon Toronto last December, he still had to do composes for final image as there were still bugs occurring. The team had a schedule, but they couldn’t stick to their freeze. The problem was that they had the idea to include some nice ebooks, but at the last minute it turned out that there were licensing issues with the books – they had an ND clause in their Creative Commons license. The ND clause is against Fedora licensing guidelines, so they had to recompose the spin at last minute to remove the content.
After learning from the issues in the Blueberry release cycle, the Sugar on a Stick team changed up how release engineering happened for the latest Mirabelle release. Sugar on a Stick is now an official Fedora spin. Practically this means the Sugar on a Stick folks don’t have to take care of builds anymore – from becoming a Fedora spin, they get automatic nightly builds. All they have to do now is to make sure the OLPC activities and the Sugar core are packaged for Fedora, and they get pretty much everything else automatically.
Sebastian then talked a little bit about how he’s worked on enabling people to contribute. The new Sugar on a Stick webpage links to a contributors’ portal to help show how folks can help. The Sugar on a Stick team continues to try to make it easier for folks to get involved in their community. One vital thing you need to do to build community is to communicate. If you don’t talk about what you are doing, people won’t know what you’re doing, so they won’t know how to help. Sebastian mentioned a group development theory by Bruce Tuckman – it involves four phases:

• forming
• storming
• norming
• performing

Sebastian closed off his talk by suggesting when you’re trying to develop a new project and community, it’s important to look for related projects. It’s okay to fork, but try not to waste resources.

Using OpenHatch to find student projects and mentors Asheesh Laroia, OpenHatch

Asheesh got through his talk admirably well considered he faced some pretty rough hardware failure and ended up having to give his talk using a borrowed laptop. 🙂 That being said, he introduced the group to OpenHatch, a project he started with a couple of friends and now runs on his own. The website is meant to help match up FLOSS projects with potential contributors – and the assumption of the site is that the potential contributors searching for projects to work on are ‘self-starters’ who want to help out. OpenHatch itself is open source and is written in Django. Asheesh works on the site half-time and has some start-up funding.
Asheesh introduced some aspects of the site that would be of particular interest to an educator looking for projects their students could get involved with:

• OpenHatch has an inventory of bugs & tasks to work on, and you can filter this list to show ‘bite-sized’ projects. He compared this category of bug to GNOME Love bugs, which are typically marked as bugs that are good for a newcomer to work on. The bug browsing interface of OpenHatch is powerful and will let you search across all bugs that are ‘bite-sized’ or view only the ‘bite-sized’ bugs of a particular project of interest.
• OpenHatch has a mentorship system. There are some caveats to it, however. There is no delineation between skills and projects in mentoring system, so you can mentor in both a particular project (e.g., GNOME Shell) and/or a particular skill (e.g., python).
• There is a location-based system where you can search for projects that have people near you. This could be of particular use to educators who might be interested in having a developer come and visit the classroom for some in-person mentorship. You can also go to a location on the map and check out what projects your neighbors are working on. One caveat here is that if a developer decides not to put in where they live, they are placed on an inaccessible island in the Atlantic. 🙂 The map is a CPU-hog though!

We had a discussion about how contributions to FLOSS projects don’t just consist of code, and during the session I actually submitted the Fedora Design team’s ticket system to OpenHatch, and Asheesh said he would use our bug tags as design-related tags in OpenHatch. (Examples of our design-related tags include html/css, icon request, hackergotchi request, logo, interaction design, usability.)
Asheesh pointed out that in general, when you need help, or want something – say so! He said, “So Sebastian told us about how we should communicate, communicate, communicate. I like to say, “communicate – it can’t make things any worse!”Asheesh’s overall position is that the FLOSS community as a whole needs to do more outreach.
Here’s Asheesh’s summary for teachers on using OpenHatch:

• You can search projects for students to work on by language.
• You can also search for people nearby, maybe have them come in and talk to your students.
• If you are worried about submitting to a project and seeing no movement on your proposals, or a mentor dropping off the face of the earth –
  find a local person…. call them on the phone, meet them in person.
• Create a project page, say you want to help, and we’ll figure out what you want to work on together.

Some ideas for features suggested by the audience:

• Jeff suggested integration with trip-it to coordinate which folks were going to which conference so you’d know who you could meet at a conference.
• Colin suggested adding a feature to add ‘in-person’ as a mentorship possibility.
• Colin also suggested a way to broadcast my hackathons on OpenHatch. Asheesh noted that only way to do that now is to go to people’s people pages and contact them via email.

Asheesh’s grand world-domination scheme is to get professors all over the world to sign on to teach open source classes to find mentors to help structure the class for a semester – and then have students at colleges across the world working on a project during the same semester.
What’s next for OpenHatch? They have a summer of code stduent who is working on training missions, which you can preview and test out at http://openhatch.org/missions – these are interactive tutorials on using free software to teach you. For example, you’re on a mission, you’re an agent for mr good to gain the trust of mr. bad, and you learn how to use git along the way.

Open1to1.org (Linux and FOSS in the classroom) David Trask, Open1to1.org

David is a school teacher from Maine. He has taught every grade level; he’s taught special ed, high school, social studies, freshmen US history, coached football…. now he is teaching computers and he is the IT guy at a large school. In the beginning, he was working at the school until 11 pm trying to get things working; it was a windows shop and he was more of a fireman. He started position in 1999, and by 2001, the school was totally Linux. Now that Maine has a Macbook laptop program, it’s a Linux & Apple school system. MLTI – the Maine Learning Technology Initiative – went with an Apple-based solution, requiring ibooks and macbooks be used in the schools. People have learned a lot through the project; even though it’s an Apple solution it’s still an open solution: their primary word processor is OpenOffice.org, Firefox is the standard browser, and Gimp is the standard for imaging.
He started out with a Windows 2000 server, and was a Windows guy. With the Windows 2000 resource kit, he had to enter in 300 users one-by-one because it wouldn’t import the users, and the system crashed completely and totally. It was a ‘reformat and start over’ situation. David had been using free Cisco floppy-based Linux router distro as his first foray into Linux. He started asking people on that project’s mailing list if it would be possibleto replace win2k server with a Linux-based domain controller; the list-goers pointed him at esmith with is now smeserver: you click a checkbox and it’s a domain controller. It’s still around today. After this experience, David went to his principal and told him: “We can do Linux, our users won’t know about it…” So they sat down and drew out the pros and cons of going back to the Windows solution. The Windows cons ended up being a very long list, while the Linux pros we really long. So the principal ended up supporting David in trying to do Linux. The very first year of his school’s Linux usage, people noticed he wasn’t on the intercom anymore telling people they had to log off NT4 to reboot the server weekly (due to a famous memory leakin NT4.)
David’s staff and students are now used to things changing every year – they don’t know what to expect first day of school. A good illustration of this is when he finally pulled every Windows machine out of every classroom. All the machines were Linux except that they had a single Windows machine in every room for the teacher. David spent a lot of time creating a special Windows image just for the teachers, but the kids were doing just fine with Linux. He decided to pxe boot the teacher’s Dell optiplexes to be Linux thin clients that August. There was a kindergarten teacher, a 33-year veteran, who had the machine on right after the refresh to Linux, who was using OpenOffice.org…. The principal had been worried about the teachers not acclimating to Linux, and here this teacher was up and running all on her own. The teacher knew what to do, although she was upset about about the lack of U.S. holiday clipart.
Then the funding model for deploying laptops changed. The Maine government had originally paid for them, and nothing came out of each school’s budget directly. In their high school program, the funding formula is to you spend $242 per secondary student on technology. If you’re a poor community, most of money that comes from state subsidy. If you’re a wealthy community, you have to pay the $242 out-of-pocket. Sometimes the money was used to pay for things like the technology coordinators’ salaries. The state came up with a new rule – they would take the cost of the laptops out of that money. If you’re a poor school, that’s a good deal – they don’t lose any money since it’s all state-subsidized anyway. If you’re a wealthy school, though, you lose a lot of money. If you were using it to fund your tech coordinator, suddenly you’ve got an extra expense. School budgets get started in November and typically pass in January in Maine. The state made this decision in March, so schools were scrambling. It resulted in the wealthy school communities becoming the ‘have-nots.’
This situation is what led to the Open 1 to 1 project. Netbooks are becoming popular and very inexpensive to do 99.9% needed to do in education. Openoffice.org, web surfing, etc. Open 1 to 1 is not just for netbooks, but they are the platform of choice because of their low cost. Macbooks had been a yearly fee, $249 x 4 years. These netbooks are $289 once – no recurring fees. So how does David get Open 1 to 1 on 1,100 machines? He needed a total solution, not just a netbook image, but a way to get it on the machine in the first place. He got it down to two clicks by using and modifying an existing USB creator.
Something a lot of people seem to forget is that kids are not concerned about the operating system on devices they purchase at Best Buy – they don’t know and they don’t care. What they do care about is if they can take pictures on it and if they can listen to music. It takes kids two seconds to get on the internet…. before you know it they are learning it.
This project is across all grade levels…. Bryant has 24 dell 2100s used by a 1st grade teacher with the Open 1 to 1 image. He went with open 1 to 1 because it is so easy to customize. You can simply drag a new application like scratch on to it, save it, and it creates a new image… so you can basically add on to or modify the stock image for your own purposes very easily.
David ended his talk with a live demonstration of Open 1 to 1. You can learn more about this project at Open1to1.org.

Lunch

We had some lively discussions at lunch just down the street from the conference site.

Our Swag

A great idea, but unfortunately we weren’t actually provided with whiteboards in the room. 🙂

Conclusion

I hope this overview of the LinuxCon Education Mini-Summit has been useful for you! If you’d like to get involved and continue the conversation, you can visit the Teaching Open Source project. Many of the folks who presented are on the Teaching Open Source Mailing List. There is also a Teaching Open Source blog planet and an irc channel at #teachingopensource on irc.freenode.net.

Update: And if you'd like to catch up on the sessions during the main LinuxCon Conference, Alison Chaiken (whom I had the pleasure of meeting face-to-face at the conference) has put together a great set of notes on the sessions she attended, licensed CC-BY-SA!