Why haven't cash-strapped American schools embraced open source?
informationweek.com
informationweek.com
I had just started using linux that year, so I ran a class introducing students to linux. I taught students to install and maintain ubuntu systems, and suddenly all 30 computers were working again. Students maintained those 30 computers for the next year and a half. It was a bittersweet time when the district funded hardware again, and we had 30 new machines running Windows 7. I am fortunate to work in a district where I was supported in teaching students to dual-boot half of those machines.
It has been an overwhelmingly positive experience to bring linux into a school system. Many students are now running linux on their own machines. I still teach an intro to programming class that uses linux exclusively, and on any given day a small number of students are choosing linux for their everyday school work.
I also saw the politics of trying to make open source more mainstream in the district. When the district was deciding what machines and what software to purchase, we almost went with openoffice. I didn't take a side in the conversation, I just kept reiterating that both MSOffice and LibreOffice work as long as everyone in the district is using the same software. The real friction comes when people are trying to share formatted documents between systems. The district was just about to go with openoffice, but for no clear reason at the last minute the superintendent made a unilateral decision to go with MSOffice. I think it was easier to appease the few loud MSOffice supporters, knowing everyone else would move on, than go with something new and risk having made the "wrong" decision.
I've been getting more into programming, including trips to PyCon last year and this year. The more involved I get in the professional programming community, the more I see how much the open source world has to offer the education world. The obvious benefit is open software. There is a tremendous need for high-quality open software to make educational work more efficient. My favorite example is that most teachers still write their lesson plans in Word, which is like programmers working in an old version of Notepad. Programmers have had vim and emacs for 40 years, but teachers still don't have a dedicated curriculum planning tool.
I love to talk about this stuff, so feel free to get in touch through my profile.
http://www.groklaw.net/articlebasic.php?story=20071023002351...
Juicy quote below:
"""Our mission is to establish Microsoft's platforms as the de facto standards throughout the computer industry.... Working behind the scenes to orchestrate "independent" praise of our technology, and damnation of the enemy's, is a key evangelism function during the Slog. "Independent" analyst's report should be issued, praising your technology and damning the competitors (or ignoring them). "Independent" consultants should write columns and articles, give conference presentations and moderate stacked panels, all on our behalf (and setting them up as experts in the new technology, available for just $200/hour). "Independent" academic sources should be cultivated and quoted (and research money granted). "Independent" courseware providers should start profiting from their early involvement in our technology. Every possible source of leverage should be sought and turned to our advantage.
I have mentioned before the "stacked panel". Panel discussions naturally favor alliances of relatively weak partners - our usual opposition. For example, an "unbiased" panel on OLE vs. OpenDoc would contain representatives of the backers of OLE (Microsoft) and the backers of OpenDoc (Apple, IBM, Novell, WordPerfect, OMG, etc.). Thus we find ourselves outnumbered in almost every "naturally occurring" panel debate.
A stacked panel, on the other hand, is like a stacked deck: it is packed with people who, on the face of things, should be neutral, but who are in fact strong supporters of our technology. The key to stacking a panel is being able to choose the moderator. Most conference organizers allow the moderator to select the panel, so if you can pick the moderator, you win. Since you can't expect representatives of our competitors to speak on your behalf, you have to get the moderator to agree to having only "independent ISVs" on the panel. No one from Microsoft or any other formal backer of the competing technologies would be allowed – just ISVs who have to use this stuff in the "real world." Sounds marvelously independent doesn't it? In fact, it allows us to stack the panel with ISVs that back our cause. Thus, the "independent" panel ends up telling the audience that our technology beats the others hands down. Get the press to cover this panel, and you've got a major win on your hands.
Finding a moderator is key to setting up a stacked panel. The best sources of pliable moderators are:
-- Analysts: Analysts sell out - that's their business model. But they are very concerned that they never look like they are selling out, so that makes them very prickly to work with.
-- Consultants: These guys are your best bets as moderators. Get a well-known consultant on your side early, but don't let him publish anything blatantly pro-Microsoft. Then, get him to propose himself to the conference organizers as a moderator, whenever a panel opportunity comes up. Since he's well- known, but apparently independent, he'll be accepted – one less thing for the constantly-overworked conference organizer to worry about, right?"""It's all open source. I took the job because I believe that open source solutions for government and non-profits are the Correct Answer.
Now, a few years later, I'm disillusioned.
The work I do, one cannot push harder. Like an octopus, the harder you push, the harder the problem fights back. Rather, you have to come at everything sideways. Covertly lay out a tarp to the side of the problem. Sneak up on the problem. Whack it on the head. Hope that it falls onto the tarp. Once it's unconscious, drag the tarp in a generally useful direction. Quickly. Before the problem wakes back up. Before it starts to wander off in some random direction and resumes eating everything in sight.
The IT needs for higher ed are so varied and complicated, there is no one size fits all. I don't see how any code I've written could possibly be used as-is any where else. It's barely useful across all our colleges and departments.
My prior real gig was basically a lot of ETL for medical records for regional health organizations. The higher ed stuff I do now is far more complicated, overall. Which really surprised me.
The community governance model for collaborative development of open source software is very, very, very inefficient. I came from doing startups. Working higher ed, I can't believe how long stuff takes. And what does get accomplished is lowest common denominator stuff, stone soup style. Imagine doing pair programming XP style with the stupidest people you've ever worked with. Now have those people do "software architecture".
I don't know so much about K-12 and community colleges.
But methinks continuous improvement, cultural change for higher ed IT is hopeless.
First, it's wicked hard problem. Scheduling, resourcing, registration, financing, payroll, etc. You'd never guess how much work goes into determining if a class is open (for enrollment) or closed. Now imagine that deliberation for absolutely every function within an organization as heterogenous and weird as a university.
Second, every single mailstop within the organization insists on their own unique super double plus awesome way of doing things. Not better or worse, just different.
Third, there is no existential crisis which can be used to ram change down every one's throats. A president or provost or whoever really is just a fund raising figure head. They're not an executive like a CEO. They're more like the secretary general of the UN.
The only way I can imagine higher ed IT can be "solved" is for GATT style agreement for how stuff should be done to be forged. But I don't want to see that. I love universities for their complexity, weirdness, variety, unruliness. To become an organization with straight forward IT needs, it'd become just a diploma mill. And that'd suck.
The tech budget is funded by separate tech levies, which can only be spent on computers or tech related staff.
This leaves the district in the strange place of being able to afford iPads all around for a couple grades in middle school, but not being able to afford a librarian for the primary school. There can be a media specialist, but only if the media is not books. They can have smart whiteboards in each classroom. But not a librarian.
No way in hell do I want to support the desktop users of my school district.
(for reference, I've taught an intro cs/programming class there as a volunteer (using open source software, yay!), and have some idea of what their systems are and what kind of environment they're up against.)
First of all there was some up front payment the school/district made to their contracted IT provider. Wehn purchasing computers, no matter what type of computer you wanted, it would cost at least $1400/unit. This was because it came as a package deal with "support". Typically this meant the choice of an e/iMac or iBook or some cheap Dell box with a cheap 15" CRT. As a result the much better equipped, teacher preferred, Macs were typically what were purchased for classrooms and the bargain basement Dells made it into admin offices. Additionally a server for At Ease for Workgroups[1] was occasionally purchased for some other ridiculous number I don't remember.
At this particular school, even if you could switch the entire campus to FOSS software, it wouldn't really affect costs because the school signed a contract to lock the price in at a certain rate. The contracting company would have just pocketed the savings.
Now, the worst part was the the price premium for "support" bought you the privlidge of having the the buck passed to Apple for Mac support (basically some sort of Applecare for Education) or shipping in a new Dell that would sit in a supply closet until they sent someone out... if they ever did. Hence why I was volunteering.
The school was pretty lucky as computer purchases tended to be an uncommon event as this school was near Universal Studios and Warner Brothers and both would regularly donate a number of their two year old "outdated" machines. A lot of what I ended up doing was replacing aging LC IIs[2] and Performa 575s[3] with the new computers.
[1]http://en.wikipedia.org/wiki/At_Ease [2]http://en.wikipedia.org/wiki/Macintosh_LC [3]http://en.wikipedia.org/wiki/Macintosh_LC_500_series
It may be hard to understand here but tech in education takes a back seat in most standard schools. There are not a lot of (if any) resources in most schools available for tech support, upgrades, hardware, etc.
So what happened was that 10-20+ years back schools slowly integrated tech and computers - Windows was really the only sensible choice for most institutions in that it allowed for everyone (non-techies) to run software that met the needs of students, teachers, admins, etc. PCs and Windows was the sensible and natural choice back then and everyone jumped on.
The problem now is that the cost to migrate hardware, software, licenses, systems, and knowledge to better platforms is significant, daunting and costly even if the new stuff is free. Its analogous to building a city around one center point for years and then realizing that you want to move the whole city 5 miles down the road with everything basically continuing to function (oh yeah and on a shoestring budget).
On the bright side the education system is filled with many who are tech savvy. They know the tech sucks... its pretty obvious when you have to use IE6 for some online lesson plan and then you go home and watch Netflix on your iPhone. So there is pressure there and things are moving but it will be slow and painful and likely whenever you send your kids to school you will gasp in horror at the tech museum they call the computer lab.
(On a side note if you are entrepreneur in edtech - the most important thing you can understand is what the tech landscape looks like to students and teachers and administrators. Very different for all 3 groups and very detached from the tech saturated environment of Silicon Valley)
There's now a team of student developers at the school district who help keep the project going. I believe this makes it first the school district to create open source software in California.
[1] App (now also on Google Play): https://itunes.apple.com/us/app/ifusd/id454673943?mt=8
[2] GitHub: https://github.com/FremontUnified/
Then they have students use computers for the most simplistic things. Typing up essays and getting to piece together a video in windows movie maker is the absolute extent of what the above average Canadian high school can do. They don't understand, they being the old guard of teaching, the things a modern computer can do because they think in analogue methodologies, not digital.
Only within the last few years have we replaced overheads and transparencies with 'smart boards' and projectors and even then, few teachers have been found to use them in any manner other than an even more LIMITED version of the traditional overhead. Very recently the younger teachers have built small wordpress sites or shared google drives for notes because of ease and personal comfort which is a giant step forward but moot for all it does to solve the issue of the unsavvy teachers.
If a school were to wise up and switch to open source, the amount of headache from staff is simply impossible to imagine. The students can understand the rapid pace of technology and the very new or extremely impassioned teachers can cope with it where the older teachers throw their hands into the air and bring out the transparencies. A billion dollars to the person to come up with a way of easily transitioning them to modern method.
Sorry for the rant, but education hits close to home for me.
| Only within the last few years have we replaced
| overheads and transparencies with 'smart boards'
| and projectors and even then, few teachers have
| been found to use them in any manner other than
| an even more LIMITED version of the traditional
| overhead.
Point being some people will use the tools presented to them (even very basic tools) in the most bone-headed way possible. It's just a fact of life.
Many people who don't use tech a lot (many teachers!) aren't really familiar with open source. They learn the big suites, because that's what they learned in school. It's a cycle.
On the optimistic side: MOOCs and the like may change things. Crafty, budget-conscious teachers who have a tech bent may change things (raspberry pi, anyone?). Clever startups may change things. I'm with Everlag: A billion dollars to the person to come up with something that helps them transition.
Complaints like "We don't like it" just split proponents and opponents of the software into opposing factions and aren't productive.
The front page should not have a page number.
Please place a header on each page. On the even pages, the header should be on the left, on the odd pages, the header should be on the right. On the bottom-center of each page, place the page number, except for the front page, as specified above.
The above is probably a bit much, but it is, as far as I know, absolutely impossible to do in Libre Office, or it is not, it is extremely difficult to do, but this sort of formatting is absolutely trivial in MS.
If the above sounds to complex, then try creating a working TOC in Libre Office.
The grammar and spell-checking capabilities are not very good in LO.
I've also had many doc -> PDF conversion problems in LO. There are many other smallish issues.
LO is about as much of a replacement for MS Suite as Gimp is a viable replacement for Photoshop for productivity and real-world work. I use the FOSS stuff at home, but my overall productivity is way lower at home as well. I've ran into many things I simply cannot do. It really is a cost-to-productivity ratio here. MS Suite really isn't that expensive if you compare the speed and accuracy you can accomplish common tasks.
Please place a header on each page. On the even pages, the header should be on the left, on the odd pages, the header should be on the right. On the bottom-center of each page, place the page number, except for the front page, as specified above.
The above is probably a bit much, but it is, as far as I know, absolutely impossible to do in Libre Office, or it is not, it is extremely difficult to do, but this sort of formatting is absolutely trivial in MS.
If the above sounds to complex, then try creating a working TOC in Libre Office.
... on the other hand, those are all trivial thing to do in LaTex. Maybe the problem is people using Libre Office for things which Libre Office is not designed to do?
Why isn't Libre Office designed to create a report which can be printed and bound? Seems like a basic word processing task.
The first group is the Libre Office -- and Office 2013 -- audience, and they generally don't need to do anything sophisticated with headers, footers, and page numbers. The second group uses LaTex or Adobe InDesign or similarly more "sophisticated" software, which of course provides all the requires functionality.
In any case, I have virtually no Libre Office experience, and I was able to generate dizzystar's test case in about 2 minutes. My apologies for leading us down this dead-end discussion. (The only confusing part was that you have to manually turn on headers & footers before you can click on them to edit.)
When it comes down to it, LO is probably not missing many features, but they haven't received the UI research/testing effort to make them accessible. The general look & feel is also 'old fashioned'.
Chances are, they wouldn't even try it. And nobody had what it takes to tell 'em "Well, give it a try, it should suit your needs just fine. But feel free to set up your own tools if you don't like it."
I'd love to contribute!
It is a god damned miracle Microsoft Office even works at all. The design is a monolithic do everything fully integrated suite. The combinatorial explosion of interacting features that MSOffice has going on is enough to make a mathematician blush (and the testing staff cry). A super large scale SDLC type one size fits all solution, not based on a generic powerful concept, but a gigantic feature blob that deals with problems as they come.
It's a god damned miracle, and LibreOffice is what happens when you don't get a miracle. It's a feature blob, it's buggy (last time on my system it used 100% CPU, and every time I adjusted a field in a dialog box it didn't redraw the background), it has surprising behavior (play with tables some time, now try sizing them). Hey, I'm no fan of Microsoft Office, either.
I'd say I like LaTeX, but that's a generic powerful concept with absolutely no concessions made for John Q Hacker. Let alone John Q Pedestrian.
Personally, I don't know where to turn.
I'm in University and I still find myself explaining to "cyberinfrastructure" developers that their "novel idea" is really just cron + shell + GUI, of which the first two have been around since the 70's (and their "novel" GUI sucks). Combine that with the legion of Matlab programmers cranked out by engineering departments (Python? you mean this is free?!), and I think schools primarily succeed at creating more consumers of technology, rather than the producers they purport.
Matlab is unique in its expressiveness for numerical and matrix computation, and value semantics.
Value semantics matter a lot because they result in functional programming without even having to think about it, while copy-on-write means that you can pass big objects into pure functions and they never get copied if you don't modify them.
e.g. in Matlab, the following function is pure (and never makes a copy of big_matrix)
function y = my_pure_function(x, big_matrix)
x = x + big_matrix(i, j);
y = x ^ 2;
endMatlab and Python are perfectly interchangeable for the overwhelming majority of undergrads and graduate students. Claiming that they are not interchangeable based on the difference between pass-by-reference and copy-on-write is just silly.
You're ignoring the typical use case for an engineering student: general math & plotting. In Python, that is easily accomplished with NumPy, SciPy, and Matplotlib. These are all free.
Kudos on knowing how these languages work and understanding some low-level design specifications, but negative kudos for not understanding the education system and the immense fiscal burden it places on students for books & software. Sure, many schools provide a free license... here is a little tidbit of information for you: ain't nothing free.
Actually I gave one more reason.
It really isn't. http://wiki.scipy.org/NumPy_for_Matlab_Users
I read that link already, in fact used it as a reference when I did convert from Matlab to python. I converted back when I realized that in spite of the apparent syntactic similarities, matlab is still very different to scipy.
In matlab, everything is a matrix. The only "gotchas" when it comes to types, is that
f = X(X > 0)
f = X(double(X > 0))
Python arrays are not scipy matrices or arrays. And a scipy 1 x n array is not a matrix.
All these little things add up to more time spent thinking about programming and less about mathematics.
Don't think of me as some biased matlab fanatic.
I am a grad student, and I found the appeal of python as strong as those who advocate it as a substitute for matlab. The difference is that when it came down to doing my actual work, I just couldn't do it as easily in python. I am the student whose interests you claim to be advocating, and I'm explaining why I, and so many others, make the choice of matlab.
As for your car analogy, I think we should do a better job of teaching new drivers how cars work, how they perform under different operating conditions, the physics and the mechanics involved, etc. Again it's anecdotal, but I saw my child's driving improve quite a bit after taking one of those car control clinics, in which a solid understanding of the physics and construction of cars played an important part.
I know that if I found myself suddenly a school technology administrator person, I would be very inclined to take the easy way out and just install windows on all the lab computers. I would do this if only so that I did not have to fight the stupid mentality of "You have to teach these 5th graders how to use obscure features of a version of MS Word that will be a decade and a half out of date by the time they graduate"
Seriously, every "vocational" computer skills class I ever had was absolutely worthless. They should shitcan all of those classes and teach something that is actually worldview expanding... but that will never happen. The people in charge of making those decisions will all inevitably think that those courses are important because they are teaching the tools that they are currently using.
Sounds like a startup opportunity to me!
> redhat
They'd probably be decent at it...if they made a conscious high-level decision to devote resources to focus on the education market, and followed through.
However, the volunteers run into massive roadblocks with school IT departments. Want to install a new browser or (a god forbid) an IDE on lab computers? Expect at least a semester of hand wringing.
This isn't the fault of the IT staff, it's a result of massive underfunding of public education. Everyone does the best they can but school employees are stretched so thin already that it's very hard to drive change to computing resources that are a prerequisite for computing education for those students who don't have access to their own DEV environment.
Long story short: school systems that are strapped for cash don't have the capital resources to bring an alternative architecture up to parity as a teaching tool. They don't even have the resources to support volunteer teachers.
Such a person would also be counterproductive using the criteria of capitalism, and I think the rest follows from that.
Of these, perhaps #1 is the most important, and to some extent impacts #2 and #3. Linux has a 1% usage share [1]. Schools would be doing students a disservice by teaching them such a marginal desktop technology, when the opportunity cost of teaching them a mainstream operating system is so high. To the masses, Windows skills are more valuable than Linux skills. I'd love to squeeze my own favorite esoteric topics into the curriculum, but there are only so many things you can teach in a day or a year, and you have to choose based on maximum benefit.
[1] http://en.wikipedia.org/wiki/Usage_share_of_operating_system...
I handle a small network, and deploy Windows and Office mainly because most people know it, easing the training burden on me.
This isn't a software company here. If it were, obviously, using LibreOffice would be the way to go.
Eventually I believe the economics will change, but we are not there yet. There may be special cases where talented individuals have donated their time to set up and maintain open source software in schools, but this is no more indicative of the true economics than if Microsoft had donated an entire computer lab for free. Rather than evangelize open source for all use cases, I think that we should render unto Microsoft what is Microsoft's.