Silicon Valley and the Edtech Revolution
imaginek12.com
imaginek12.com
I know of multiple school districts that are absolutely plagued by overzealous IT admins, including the director, who lock down teacher and student computers to the point where they're unable to install or use the software they need to learn or login to all these cool educational services that use Google or Facebook login. Watching educational YouTube videos? That's not even open for discussion.
They're totally blocked from learning. Funny thing is: the kids all know how to get around these blocks, but the teachers don't dare do it for fear of losing their job.
This on top of the fact that these schools' LANs fail on a regular basis, including when students are taking computer-based standardized tests.
So teachers spend hours each week filing bug reports and calling IT to get these sites unblocked, software installed, or LAN/wifi back up. Most people here wouldn't believe what is going on in many school districts because of overzealous and incompetent IT admins. It's a structural problem in at least 3 states I'm familiar with (all of them low-performing states on standard tests).
Maybe build a "screened" portal to educational sites (including ones with YouTube videos) and software? But it would have to be accompanied by some serious lobbying towards the IT folks to whitelist it.
We've now setup OpenStack on a server, and are in position to give every student in the school their own Linux VPS next year. They'll ssh in using putty, have full admin rights to their instance, be able to apt-get whatever they like and all of a sudden the IT policies of the school no longer apply.
I'm suddenly really looking forward to teaching Computing next year.
You're very fortunate. I'm guessing you're probably in a high-performing, relatively well-to-do school district (please correct me if I'm wrong).
>and are in position to give every student in the school their own Linux VPS next year.
The IT depts at the schools I'm familiar with (through friends, I'm not a teacher myself) wouldn't even know what OpenStack is. They're 100% Microsoft shops. And the idea of giving a student root access to his/her own virtual instance would simultaneously blow the IT staffs' mind, terrify, and enrage them.
I've suggested numerous setups to get around all of the excuses tech people at the district give. Would you mind sharing personal contact info for either yourself or your IT staff so that I can go to administration with a "this works in practice, here, go see" argument?
We're starting the whole OpenStack/Linux experiment in September, so I can't give you concrete evidence that it works yet, although I can't see any issues at present.
You're 100% correct. I work in a UK grammar school, so we have a predominantly middle-class catchment area, and students are selected on academic ability.
TBH I didn't know what OpenStack was a few months ago. I went to my tech, after issues with running Python on the network, and explained that I wanted the kids to have root access to 'something'. I wanted them to play around with a CLI and be able to explore different tools.
We explored RaspberryPis with KVMs, dual booting and paying for VPSs from a third party. My Tech did his research and came up with the OpenStack solution.
My point was that these things are possible, but I agree that to many IT admins the concept is terrifying. The sad thing is that offering students such an experience is well within the budgets and technical capabilities of all schools in the UK.
I think this is a bigger problem than most people realize. Expensive proprietary software has no place in education. It is antithetical to the students' interests in learning, and it inherently prevents them from tinkering. Techies with children entering the public education system really need to push for a wholesale transition to open source. If it's good enough for the office, it's good enough for schools.
>'I know of multiple school districts that are absolutely plagued by overzealous IT admins, including the director, who lock down teacher and student computers to the point where they're unable to install or use the software they need to learn...'
Locking down desktops is a pretty basic part of successfully managing more than a handful of them, particularly when they're distributed across an entire county. Obviously, there should be a better process for identifying and distributing (in a controlled manner) the software that will benefit these classes, but handing out administrative rights isn't likely a realistic option.
>'Watching educational YouTube videos? That's not even open for discussion.'
The reasons for this should be obvious and you allude to it just below.
As soon as one of the kids catches a glimpse of some cleavage in a recommended video or YT pre-rolls an ad featuring a girl in a bikini the system is looking at a lawsuit, the teacher - unemployment.
>'Most people here wouldn't believe what is going on in many school districts because of overzealous and incompetent IT admins.'
It's a bit much to pile this on the admins. In my experience, those jobs don't pay enough to attract/keep the level of talent they need nor are they given the autonomy to effect change on their own.
I think it's right to call it a structural problem, one with a whole slew of factors. To name a few...
* Vendors abusing E-Rate funds and the like to oversell districts on things they can't afford to maintain or properly utilize.
* The wacky funding schedules of school systems. They tend to get lump sums for 'modernization' and the like, but are restricted from spending it on services with recurring costs or technical hands to run them.
* Individual schools run like fiefdoms which work against centralizing for efficiency. A notable example of this from my own experience is project which could have served an entire county of 100+ schools for well under <$1M being dashed in favor of a grossly inferior, $50k per school solution.
* Super parents pushing their own agendas and self-interest. It's not at all uncommon for someone who coincidentally works for a particular vendor to decide that the vendor they work for is the only viable solution for the district and raise hell until they get their way.
* Inability to retain qualified help. IT people who want to get paid don't work in education. Experienced IT people who want to work in education anyway know to avoid primary education and aim at colleges and the numerous benefits like free-rides for their own children.
I agree with you here. What I wrote there was careless -- I meant to lay most of the blame squarely at the feet of the overzealous and incompetent district IT directors and their staff. I wrote "admins" when I was thinking of the people in charge. The mostly incompetent admins certainly don't help but they aren't at the root of the problem. You're correct that it's hard to attract competent help at the salaries public school districts pay.
And based on my friends' experiences, I agree with your list of structural problems. The 'super parents' doesn't apply as much since they're working in poor districts where many families are fragmented and uninvolved. But otherwise, I'd say you're spot on.
I do strongly disagree with your view on locking down computers to such an extent, particularly your view on handing out admin rights. When a teacher is teaching a computer science class that requires installing new libraries and executable every other week, they should be given admin rights or else the computer pre-installed with all needed programs (that never happens in my friends' schools). They can't even follow a nationally standardized curriculum due to their IT depts' obstructionism, not only because they can't get the right software installed, but also because their kids can't access the various Google-related projects (i.e, Google docs) since all Google logins are blacklisted.
It's absolutely ridiculous and it infuriates me. These are motivated teachers who want to teach students who desperately need to learn, and they're being actively blocked by their own district policies. It's maddening for me to hear about these things.
Ideally, they'd be given root accounts on Linux machines as in MarcScott's enlightened district, but I think my friends would be happy to just get Python installed on their kids' computers.
I see. That's just broken.
I'm thinking of desktops, the sort that need some amount of persistence and security.
What you're describing is a lab - which should be built in such a way that it's as wide open as the class requires, but can be restored to a clean base quickly and easily.
Unfortunately, this sort of thing - automation is generally beyond the folks charged with running those systems.
Years ago, I encountered one well-meaning, but woefully unequipped guy who was re-imaging 20+ machines every few weeks with a single CD. I got him rolling with a networked solution at least.
This is one place 'we' can help though - by documenting or ideally packaging up the tools for building proper environments.
Unfortunately, these schools I'm talking about don't have computer labs. They're lucky if the AC works and the roof doesn't leak. They have a half-dozen of Windows machines in some classrooms, and now (thanks to several generous donors) one friend has a half dozen Android tablets. One has a "mobile" lab consisting of 10-15 laptops on a cart shared throughout the school, but they are as locked down or more as the classroom computers.
>This is one place 'we' can help though - by documenting or ideally packaging up the tools for building proper environments.
If you are someone who still has influence with district-level IT staff, where you could help the most is by making them aware that what may seem like abstract policies can have major impact on a child's education, often in ways they would never imagine if they're not teaching.
Sadly, I've seen this sort of thing repeatedly. It's the ancient practice of carting around film projectors and the like applied to an age when there are better options.
>'If you are someone who still has influence with district-level IT staff, where you could help the most is by making them aware that what may seem like abstract policies can have major impact on a child's education, often in ways they would never imagine if they're not teaching.'
Unfortunately, at this point all I can do is offer solutions when they ask for help (Politics precludes us from being proactive about this) - it's still up to the district to take it or not.
It's a really tough problem as there are countless players who feel entitled to opinions on and profits from these things.
One of many dreams I have is to become CTO of a notable school system and set an example. We'll see how it goes.
In my school we have an MIS that seems to have evolved from an Access Database created in the 90s, that manages all student data, but has no API. We also have a VLE that was designed to look pretty, but offers almost no functionality and has a UI that is tortuous to use and actually lengthens the time it takes to mark work and record results.
Every teacher I work with has a different way of operating. Some write assignments on the board and take in hard copies. Other's use the official VLE. Some use email or shared network drives, and others Edmodo or whatever platform is flavour of the month.
Work is marked and fed back to students, and then inevitable placed in an Excel spreadsheet, exported as a CSV or transcribed into a paper mark book, only then to be then double entered into our MIS at a later date.
The system is a mess, and every new Edtech tool I see just adds to the confusion - another URL to remember, another username, password, interface. I wonder how many Edtech startups have founders that were actual teachers, or bring teachers on board at an early stage. Judging by the tools on offer, I would guess that it's not many.
There are a couple of exceptions. EdX, I think, is an amazing platform, but even though it is OpenSource, it is well beyond the capabilities of most schools to set up an in-house platform. Codecademy seem to genuinely want to engage with teachers (in the UK at least) but where are the Codecademy startups for Geography or French or Science.
In my own teaching I find myself increasingly shunning new Edtech tools, and now barely bother clicking on the links I receive in my inbox every day. Until Edtech startups really engage with teachers, rather than just treating them as beta testers, I really don't see anything in the future other than increasing fragmentation and frustration.
(i'm not affiliated, but have been impressed by the founder. They are bootstrapped)
It seems like if you build a school around these tools, it could work a lot better than bolting on to existing schools. In that sense, charter schools may be good for experimenting with technology.
The other observation - teachers are very busy. Each technology initiative takes time and attention. Reminds me of this Onion article. http://www.theonion.com/articles/magical-office-worker-able-...
I am part of and see daily the frustration that teachers have being bombarded with tools clearly untouched by teachers and where teachers have never been asked for their opinion and have certainly never been engaged or approached in the early development stage.
In response to this massive problem, I created a platform where EdTech developers can actually collaborate and share their products with our community of willing teachers. They get real time feedback and expert eyes on their products in the early stages. In your words, we hope that there will no longer be an excuse to not 'engage with teachers.' You can check out our site:
We have 500 teachers onboard and they are taking on paid jobs for the summer to Tinker with new tools and take part in meaningful interviews and user testing studies. Their input leads to iterations of Edtech products that meet real classroom needs and solve real problems. It is a lot of fun and rewarding to both sides of the community, both teachers and EdTech entrepreneurs.
"...both sides of the community, both teachers and EdTech entrepreneurs"
Aren't students part of this community? Are EdTech companies trying to solve problems for teachers, or for students?
"I was surprised how insightful my six year olds were." TinkerEd teacher.
"My students loved that they got to share their ideas with a developer, they thought it was really cool." TinkerEd teacher.
Many schools make centralized decisions: you either sell the school board or the IT department or both. The people stuck with the decision aren't the people you need to convince for a conversion.
By the time teaching is happening, you are one level removed. By the time learning is happening (or not), you are two levels removed.
If I were being adversarial and trying to design a system to achieve suboptimal results, this is the sales and feedback channel I would set up.
This may be one of the most surprising innovations we've made at AltSchool; it's completely non-technical but running your own school lets you iterate on the metric that matters: learning.
At every edtech or digital learning meetup/conference my developer colleagues and I went to, it was all other developers, non-profits, and edtech startup people, but almost never any teachers. And yet, all of us were dying to connect with classroom instructors! We understood of course that teachers were super busy, but edtech is a two-way street. Teachers have to meet developers at least part of the way if they want edtech products to improve, right?
Anyway, I'm really thrilled that you're finding those teachers who are willing to give feedback and beta test and putting them in touch with developers. I wish this had been around a few years ago.
I have done lots of research in the field and found the 3 top 'no no's' and why the disconnect can still exist:
EdTech developers (not all) make the mistake of talking and forgetting to listen once they finally get in touch with a teacher.
Teachers are tired of being told "we have the BEST tool for ...." Let them make that judgement.
Giving teachers your sales pitch instead of a 2 way collaboration process.
I hope that with TinkerEd we will continue to narrow the gap. As you say it is a 2 way process and it is exciting to see some great advances from all parties.
1) Don't know how to get it
2) Only get it from only a handful of teachers that don't represent a wide range of teacher interests
To be fair, I've seen the same problem in other industries too. In edtech, this problem is particularly noticeable because there's a new wave of edtech startups right now, driven by advances in web technology, decreased startup costs, and increased investor interest. This new wave has catapulted dozens upon dozens of new offerings, so it's very easy to get lost in the sea of choices.
If it makes you feel any better, since markets tend to be cyclical, this expansion will eventually collapse and consolidate into fewer choices, I'd imagine.
But back to edtech startups and teachers. It's hard for many founders to know how to find and engage with teachers meaningfully. Those that don't do it are doomed, IMO. For those that want to find out how, a handful of resources are rising up to fill this need, such as:
+ http://edtechhandbook.com/ - A collection of informational articles.
+ https://www.edsurge.com/summits - A series of conferences to bring teachers & startups together.
+ http://tinkered.co - A new teacher-created service to help startups employ teachers.
If anyone else knows of others, please share!
One way to get teacher engagement and feedback is to run your own schools, with your own teachers. Works like a charm.
Disclaimer: I work there.
It's an infomercial, full of Silicon Valley patting itself on its own back, by a former YC partner. Recursive much?
Sometimes you guys have to stop chugging from the SV Kool-Aid fire hose. Especially when it comes to educational technology.
It seems that a lot of companies in the space (or at least the ones I hear about) are geared towards this one topic, and only a notable few are tackling other areas of academia or infrastructure issues around education. On surface it makes sense - tech companies geared towards education are probably going to focus on technology as the subject. It's also relatively cheap to source material, employees have first-hand experience in the field, and there's high demand with straightforward business models available.
I certainly noticed all these things when I launched Bento (http://bentobox.io/), and I still think about it today. I do try and maintain a healthy skepticism though: is the edtech being referenced in this article still largely in the narrow band of technology, or are we seeing it expand?
It's rare to find an effective staged learning system with excellent content. I haven't seen it much outside of academic platforms.
We are tooling up the industry, but not revolutionizing it. We are the same old industry, with shiny toys, and online classes that supplement schooling, but only replace it for a very small minority of very clever children.
The investors don;' signal a revolution. They signal a market opportunity, which is not the same thing. When parents start pulling their kids out of the public schools, because the online options surpass the local school district at a functional level for lower income working families, THEN we have a revolution on our hands.
This is presented as a novel idea, but governments have been subsidizing education for a long time because of this long held value (see: student loans).
"You know a revolution is happening in Silicon Valley when the money shows up."
Correlation does not equal causation? Money does not a revolution make.
Not only that, but the UK has had the Open University since 1969 which provides mass long-distance learning, with grants and loans being easy to find for those taking less than 120 credits a year. They've been using computers and the Internet since they were first accessible to people (using TV and radio for broadcasting lectures before that), and have been releasing some free course material since 2006. The OU also provide a degree track, giving out degrees that are (theoretically) on par with brick-and-mortar unis.
Of course, MOOCs are interesting in that the course material tends to be free, but they're also far behind the Open Uni in most other aspects.
When has it ever? Unless we're talking about Cold War sponsorships, which I don't think would count for these purposes.
The concern lies here. The blog post references a record $509 million in ed tech financing in 2014. However, in the last 10 years I believe we have only seen two companies go public (2U and Chegg). Both trade at modest multiples to invested capital and neither is focused on the K-12 space. And those are the success stories. Other than those two, I can't think of a nine figure acquisition other Wireless Generation.
My concern is that, unless we start seeing large exits soon, we'll have a lot of investors fleeing the space similar to what happened to the "e-learning" companies in the wake of the dot com crash.
I sincerely hope that I'm wrong though. I'm a big fan of ImagineK12 and many of the companies referenced in the blog post. I'd love to see a number of IPOs and big exits which will drive more capital towards innovation in this sector.
The Khan Academy and Coursera do not automate education, they simply are the modern version of correspondence courses.
I think many institutions are set up as human proof-of-work verification machines, when they should be set up as human brain debuggers. Debuggers show you counter-examples to your mental model. You set the goals and drive the process.
Are people getting hired based on a resume of a MOOC education? Do you know anyone who really learned a subject by taking a MOOC? If so, that's fantastic.
1. Employers completely ignore them.
2. The vast majority of courses are not very good - poor preparation, bad lecturing, not well thought out course design. A small number of courses (mostly those based on existing courses at top universities) are very good. I think that this actually reflects the university experience in general.
As it stands, I think the people that (1) understand the topic well, and (2) are getting jobs are also people that take that understanding and have something to show for it. It's not enough to say that you learned code through Codecademy, you do need to build something of your own to show in order to be employable. I think that's the case even with traditional education: it's not enough to simply have a CS degree, but something to show for it as well.
At a conference I was at, one of the speakers claimed that a disproportionate number of the successful students in the group they were studying not only had some college, but had actually finished college and had a degree (though not always in the same area). It'd be nice to see some more solid stats on that. Their interpretation was that where MOOCs are most successful, so far, is as "continuing education" for people who already finished conventional higher education. Those people can use MOOCs to keep up to speed in their field, or broaden by taking a few courses in other fields that they didn't get a chance to study at school.
I think if there is a problem with MOOCs, it is purely one of expectation. Of course there isn't some magical "if you MOOC it they will learn" effect. Of course people aren't getting hired because of putting a few classes on their resume. But both of these things are true of college education as well. You can't force people to learn, you can only give them the opportunity. Nobody hires on the basis of individual college courses either. It will be interesting to see if the "degree program" equivalents that the MOOCs are trying will be successful, but it's too early to tell.
But ignoring all that, there are now hundreds of opportunities for motivated individuals to learn about an enormous breadth of topics completely for free, with lots of good lecture videos, lecture notes, exercises, projects, and peer discussion. This really can't be anything but a really great thing.
I've done quite a few MOOC courses. The best courses demonstrate what I believe the future of higher education should be. They're better than anything I've at an accredited university. Of course, it bears noting that I'm dealing with courses at a public university versus MOOCs by the likes of MIT and Stanford.
>'Do you know anyone who really learned a subject by taking a MOOC?'
No, but I see no reason it couldn't happen. If the best courses I've taken were stacked together and extended to half a dozen semesters of escalating content I think they'd produce formidable results.