[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [school-discuss] K-12 efforts



Hi Tim,

When I admin'ed & taught at a small school my tactics were different from a pure FL/OSS approach, in that the desktop was almost universally Windows.

Using your BCERAC program would have cost my school $675 -  $1,050/year. (300 students @ $300, 20 teachers  @ $375 or $750 depending on MS CAL level). Adding Photoshop would be another $400/yr (assuming I could find such a proration for PS for such a small site), bringing us to $5,375 - $7,250 over a 5 year period. For starters my software costs would've exceeded my initial workstation costs by a fair margin, let alone the impact to my budget. And that's just on the desktop alone - Windows server costs would've impacted my budget even further.

However my materials acquisition budget came to little more than $2,500/yr (although I did have extra funds in another pool that came from a special project grant.

On the server side, although Linux server admins don't just grow on trees, Karoshi Linux renders that point moot, b/c KL is very easy to admin. KL requires almost *NO* Linux admin skills, only typical power user skill levels are required to run a KL LAN. Best yet KL includes every imaginable service needed. I set it up, I ran it, it was a no-brainer (quite unlike my brain-racking Unix admin efforts as a DB/2 UDB admin....).

OK, that all said, ubiquity was as much a deciding factor as cost. Indeed Windows is the de facto ubiquitous OS *BUT* the Windows versions of GIMP, OpenOffice, Abiword, Gnumeric, etc. were readily made ubiquitous. They can be freely distributed to the students as portable freeware on USB flash drives &/or w/ complete installers on CD-ROM. Were I running the school now I'd have implemented more dual-boot & USB-boot Linuces, e.g. Puppy Linux w/ CinePaint, Rosegarden, Ardour, Cinelerra, etc.

So b/c of the budgetary constraints at small schools we are really back into an apples vs. oranges comparison. The advantages of closed source in large school districts stems from a pricepoint effect, which is fine & understandable, but one size doesn't fit all.

There's a opportunity cost question that arises however, from the monopoly of closed-source apps. Don't these extra commercial licenses functionally take budget away from additional hardware?

Admittedly in your BCERAC program the licenses are all-or-nothing (based on FTE & district size), granting district-wide freedom to use them. But there's a problem therein: Isn't it either more expensive b/c it's *NOT* functionally prorated by actual need/use?

If not, mightn't it lend to a bureaucratic license lock-in mentality that *ALL* workstations require the paid software *AND* need no other?

I ask b/c I've observed this problem at several school districts: NO FL/OSS apps are INSTALLED or ALLOWED on the school's computers! The only explanation I can come up with is the bureaucratic mindset that the coverage is there, why duplicate something costly with something free? But this again brings us back to the all-or-nothing vs. proration-by-need-only question, which brings us back to the cost problem - it's independent of size again.

And with that I've concluded there's an ethical problem, one that can be widely observed in the almost universal practice of keeping a closed-source monopoly in ever so many school districts. Basically the widespread practice at public school districts in the USA is to keep FL/OSS applications away from students - even though the free applications exist and there's a demand that they be used now, shockingly the closed source-only practice continues within the schools.

In terms of parity and equality, doesn't restricting access from FL/OSS apps (x-platform Gimp, OOo, Avidemux, etc.) restricts *which* workstations (and hence, which students) have access? Granted this isn't a problem when portable FL/OSS apps are made freely available but again, it appears that paying for software lends to this lock-out of FL/OSS apps on Windows & Mac WS in the schools.

And with that a no-FL/OSS policy enculcates dependency on closed source applications at home as well as at school, which encourages piracy of the same commercial software for home use (warez shopping at piratebay....). We don't need to be forcing our kids into becoming scofflaws, the music industry has done a grand job of that all by itself, why should the schools aid & abet the software industry's old backdoor tactic of creating software crack addicts?

So, on one hand we have had ubiquitous piracy of commercial software & a culture of piracy that has been tacitly (if not inadvertently) condoned by the schools. On the other hand we can have true ubiquity of utility computing at a decent level with far fewer costs & with far fewer fiduciary risks.

Where's the education establishment's awareness on this? I admit it's not an easy issue but if little ol' me can address it succinctly why is it so far below radar in the current dialog about technology in education?

Mysteries abound....

/Lee


From: Tim Dressel <tjdressel@xxxxxxxxx>
To: schoolforge-discuss@xxxxxxxxxxxxxxx
Sent: Tue, February 2, 2010 12:57:49 AM
Subject: Re: [school-discuss] K-12 efforts

OK, here are the details and I've left nothing out, nothing hidden.

We belong to a buying group called BCERAC (google is your friend). We actually pay dues to belong sort of like a Costco card. The membership cost is $1 per FTE student. For sake of easy math, we have 2400 students so to buy software with the BCERAC discounts we pay $2400 per year. Due date is at educational fiscal year end which is July 31st.

BCERAC goes out and negotiates with vendors under the guise of volume potential volume purchases going towards educational institutions, primarily K-12.

The deal specifically with Microsoft for a school district is based upon the FTE teacher count. So for easy math again, we have 190 teachers district wide. We have to supply the FTE count to BCERAC via the official Ministry of Education filings so its not a round number,,, it actually something silly like 187.12. The Microsoft offer comes in 4 packages.

The first package is $27.96/FTE and consists of any version of MS Office. It also includes both the online, and the offline learning essentials and the SCORM content for all the Office Products. I've loaded the SCORM content onto our Moodle server and we have an instant Office 2007 training platform.

The second package is the Windows Upgrade Package for $23.41. This licenses us to install any copy of Windows. From an "edition" standpoint, it is the Enterprise version or lesser (i.e. no "Ultimate" bet slightly better than "Professional"). We get downgrade rights as well to any desktop operating system back to Windows 3.0 theoretically. We are standardized on XP, but began migrating to Windows 7 this past fall.

The third package is $26.46. This is CAL's for pretty much all Microsoft's major server products, including pricey ones like Office Communications Server, System Center Config, Windows Right Management, etc. It also includes ForeFront Client Security Suite.

(I know I said there was 4 packages,,, there is actually a lower cost CAL package for $12.84, but you don't get the exotic CAL's and you don't get ForeFront. I needed a replacement for Symantec the the price difference along with the possibility of looking at some centralized VOIP made the more expensive CAL package more enticing)

So all in we are $77.82 per FTE teacher for software on the desktop. We have 190 teachers (less than that, but whatever) brings us to a yearly Microsoft bill of $14786.

Don't forget we also paid $2400 just to come to this party, but we buy a lot of other software via BCERAC, so lets say that the portion of the membership fee is, 25% of that, so $600. So our desktop cost is $14768 + $600 = $15368/year.

But we can install on an unlimited number of computers district wide. Right now we have just short of 1000 computers. So our per computer cost for software is $15, or $60 over its 4 year lifetime.

Now just like linux or any other platform out there, we are almost a completely managed desktop. We push images using a combination of RIS and FOG where it makes sense. Out actual desktop management time is actually very little. What's killing me right now is travel time between schools as sometimes the pipes into the schools are so packed I can't even get in remotely so I have to send a tech to site.

Now BCERAC also has a program with Photoshop as well. If all you are talking is photoshop, the cost is something like $7000. It's actually a bit less than that, but I haven't actually purchased it yet but plan to shortly. The license is not as generous as the Microsoft one as it is confined to the walls of a single school. We will shortly have just one high school which is where we will put this license. In that school today there are 190 computers, but by the time we purchase the Adobe license we will have close to 300 computers (we are amalgamating 2 smaller schools into one larger one). The adobe license permits you to install 500 copies within the boundaries of the walls of a single computer, plus it gives you all the educational materials, videos, etc, on how to not only use the tools but also how it can be used in the classroom with sample lesson plans. Finally it permits you to send home 50 copies to teachers. So we don't have 500 computers in that school but we will easily be able to give away 50 copies to teachers at home.

So the adobe license becomes $7000/350 = $20/computer. This license should realistically also have a 4 year lifecycle. It also comes with some nice deployment and update tools so its brainless to patch, update, maintain and deploy/redeploy.

Those are far cry's from the prices that are listed below. But they are realistic prices that are commonly available to educational institutions. Win on those merits and compare apples to apples. The software that you have selected for comparisons are very good. Gimp, although not as polished as its Adobe equivalent is just as powerful. We live in an area where there is next to zero linux on any desktop, so its a hard sell. We tried linux on some netbooks. It ran fast, had a great selection of software, and in my opinion worked very well. The cloning/imaging tools were a bit ugly in my opinion, but they did work in the end. My techs have very limited linux experience, but collectively my team has over 50 years of microsoft experience. Things that just came naturally to us with any new version of windows or office were always a struggle with linux. 

Now our teachers just plain balked at linux on the desktop. For sure there were a couple of technology leaders out there like there are anywhere, but I can count on one hand how many there were. We ended up putting Windows 7 on the netbooks.

And I am proud to say that our Windows desktop is rock solid. We know how to setup an image, how to protect our networks, and how to prevent service calls. So the argument that there are fewer support issues with operating a linux or open source desktop don't fly when you know what your doing.

If you are talking server software (like a Windows Operating System license) they are also on this scale of cheap. A copy of Windows Server 2008 R2 standard runs about $139. For Datacenter it runs around $1000, but remember that datacenter is licensed by CPU but it gives you unlimited virtualization rights. So we have a 2 node virtual cluster running 12 production windows servers. We had to license all 4 processors (2 sockets per server) but we can deploy as many virtual Microsoft Servers as we want (because in a cluster you never know where virtual servers might be reallocated you have to licence all possible processors that might end up with a microsoft server on them,,, you can restrict where things migrate to but then you also restrict your flexibility in getting the maximum performance out of your hardware). We are also running 6 linux servers in that environment and have the overhead for at least another 10 virtual servers if we need them depending on the kind of workload of course. So our copies of a server license have cost us to date $333 per virtual server. Each additional virtual server drops that price. But because they are all virtual we have full failover, hot migration, disaster recovery, etc etc. You could not get that with two windows physical servers running standard, so we are ahead of the game.

I'm not sure what a Red Hat Enterprise license costs now, but I was under the impression it was around $500. It would be interesting to see what real world educational costs are for Red Hat because I doubt they are that expensive in the real world, and I'm willing to bet that the virtulization capacity of linux is probably very good.

The last thing I'll throw into the ring is tech costs. First, its very hard to find experienced linux techs in my area. You could do this on a contract remotely, but if your desktop techs can also support the backend and the desktop isn't taking much of their time, then your ahead of the game. I've got 2 full time techs, plus another two techs who are 0.2 tech and 0.8 teacher. The full time techs are underpaid in my opinion, but its a union shop. But I could find in my little town at least 2, maybe 4 more techs to replace my techs at the drop of a hat with a couple of phone calls. The term "a dime a dozen" comes to mind,,, but its disrespectful of their talents. Today I am convinced that a shop that has clue can run a closed source, or hybrid environment at the same or less cost to a purely open source shop. I see this changing though when I look at projects like FOG.

Where I see possibility for a real game changer is what google is doing. If you can convince people that the operating system is actually the web browser for everything, that's the real revolution. If you can address things like hardware graphics acceleration to make it pretty, and abstract the end user experience from the tool sitting in front of them, I think the open source argument should get dropped just as fast as the Microsoft argument. Both become irrelevant overnight.

Interesting times.





On Mon, Feb 1, 2010 at 8:28 AM, lee rodgers <sregdoreel@xxxxxxxxx> wrote:
Hi Tim

If you don't mind, yes that would be very helpful.

Understand those prices are cited from elsewhere and although are putative are probably more representative of the costs borne by small schools that don't enjoy volume pricing.

/lee

From: Tim Dressel <tjdressel@xxxxxxxxx>
To: schoolforge-discuss@xxxxxxxxxxxxxxx
Sent: Fri, January 29, 2010 12:33:44 PM
Subject: Re: [school-discuss] K-12 efforts

Would you like numbers that reflect real purchases in Canada? They are significantly lower than what you list there.

On Fri, Jan 29, 2010 at 8:49 AM, lee rodgers <sregdoreel@xxxxxxxxx> wrote:

More going on in the K12 space... searching NECC 2009 workshops & presentations yielded some individual efforts:

http://center.uoregon.edu/ISTE/NECC2009/program/search.php

Classroom 2.0

.: Open Technology Resources :.

Open Source Software Solutions for K-12

Open Options | What are educators using?
Open Options | Why do educators choose it?

Open Options | On the backend

etc

oss-school.pdf (application/pdf Object)

buchanan_krasnoff_libraries.pdf (application/pdf Object)



TaskCommercial ApplicationAprox. Cost per SeatOpen SourceCost
Operating SystemMicrosoft Windows$150 Linux K12LTSP$0
OfficeMS Office$161 Open Office, etc.$0
GraphicsPhotoshop$274 GIMP / CinePaint$0
GeometryGeometer's Sketchpad$30 Dr. Geo$0
Server seatsMS Windows Server seat license$40 Linux SAMBA / LDAP domain$0