Slashdot Log In
How Do I Start a University Transition To Open Source?
Posted by
Soulskill
on Sun Feb 08, 2009 01:01 PM
from the baby-steps-version-control dept.
from the baby-steps-version-control dept.
exmoron writes "I work at a small university (5,500 students) and am in a position to potentially influence future software purchasing decisions. I use a number of FOSS solutions at home (OpenOffice.org, Zotero, GIMP, VirtualBox). My university, on the other hand, is a Microsoft and proprietary software groupie (Vista boxes running MS Office 2007, Exchange email server, Endnote, Photoshop, Blackboard, etc.). I'd like to make an argument that going open source would save the university money and think through a gradual transition process to open source software (starting small, with something like replacing Endnote with Zotero, then MS Office with OpenOffice.org, and so on). Unfortunately, I can't find very good information online on site licenses for proprietary software. How much does a site-license for Endnote cost? What about a site license for MS Office for 2,000 computers? In short, what's the skinny on moving to open source? How much money could a university like mine save? Additionally, what other benefits are there to moving to open source that I could try to sell the university on? And what are the drawbacks (other than people whining about change)?"
Related Stories
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
money is not the way (Score:5, Insightful)
You won't be able to win this with the money argument. Microsoft will swarm all over you, giving free stuff away. They have a fund just to give away free licenses to anyone who's even thinking about trying open source.
No, since you're a university, the way to approach this is to let the undergrads explore. Sell it as a learning experience. Why is OSS so popular nowadays? Maybe the University itself, as a place of learning, should offer this? Don't limit it to just OSS, bring up OSX as well, to be fair. Let the students explore.
Now, how to get everything work well together? Why, we depend on open standards of course! The entire Internet is built on open standards, RFCs and so on. All the software must be open interfaces (exchange has imap, for example, and AD has ldap). Keep doing this. Get in touch with the contracting office, and ask them to consider putting language in for their RFPs and RFIs to include "must work with appropriate open standards".
Slowly, but surely, things will get better.
Re:money is not the way (Score:5, Insightful)
One more thing - recognize the shortcomings of OSS too. Not everything that's OSS is perfect. There are shitty OSS things too. For example, openoffice sucks, compared to MS Office. Be open about things.
Also look at external offerings. Why run your own mail server, when you can do google apps - I think it's free for non-profits and .edus. Gmail, and instantly, you just saved a bunch of money, and a bunch of work. Now those people can be put to working on other higher priority stuff.
Parent
Re:money is not the way (Score:5, Insightful)
Always remember that a bit of job training is in there too. Your artists *need* extensive Photoshop experience. Same with Maya, 3ds max, protools, etc. And asking non-techies to switch from MS office is like convincing 70-year-olds to drive on the other side of the street.
E-mail is a perfect place to start the transition, especially if nobody uses meeting requests. But go one piece at a time, and realize that people in academia are frequently motivated by things other than money.
Parent
Re:money is not the way (Score:5, Insightful)
Exactly. Trying a global migration to OSS, or anything else, is doomed to failure. I saw a similar thing in a crazy "lets get rid of Linux" effort at a big bank: doomed to failure because a few groups really wanted Linux as the compute farm OS. One size does not fit all.
The best thing to do is find bottlenecks that are tying the users to a specific OS - IE only webpages, mail servers, print services, weird apps, etc. Spend your effort prying these loose. Fight pointless mandates (you must use XYZ software to do random task ABC.) Get support in place for other OSes: if your helpdesk thinks in terms of MS software only, you are screwed - get them used to MacOS, Linux, etc. Then let the users do what they want: they'll be happier, and you'll see a lot more software diversity, which will in turn encourage more infrastructure openness.
Parent
Re:money is not the way (Score:5, Insightful)
This is huge and vastly underestimated. Your goal should not be to transition to open source--that's just as bad as an all closed-source ecosystem. Your goal should be to transition to infrastructural openness so people can use what they want. If they want to use Office, great--just make sure that their documents save in ODF so everybody can access them, etc. etc.
Parent
Re:money is not the way (Score:5, Interesting)
If you assume that the needs of the users are being met, all open-source has *many* advantagess that closed-source doesn't (auditability, ease of maintenance, transparency, lack of vendor lock-in.)
Because I don't assume that the needs of the users are met. I do graphics work. I've used both the open and closed ecosystems' products. The open ecosystem's products, to be pretty frank, suck.
If they were as good, great, use it. But they are not, and realistically speaking, almost certainly never will be. Ignoring that in favor of ideology is stupid.
Parent
Re:open or closed ecosystems (Score:5, Insightful)
I tried CinePaint back when they still called it Film Gimp. It was a lot better than the Gimp in its core features, but it inherited its predecessor's shit UI and shit workflow. I haven't heard anything to suggest that that's changed, so I haven't spent any time with it. If I were going to be using nothing but Film Gimp I might consider it, but I actually use my tools in a novel (to OSS, anyway): in concert. I expect my tools to benefit my workflow, not hinder it by switching up everything and anything when I jump from (say) Photoshop to Illustrator.
Inkscape - meh. Yes, it's the SVG reference implementation for all intents and purposes--good for it, I don't care about its features if it sucks at presenting them. It's a clunky tool with a poor UI and--surprise surprise!--little to speak of in terms of horizontal integration. It'd be fine if I could do everything in Inkscape without any other tools, but that's a very rare occurrence.
Blender - eew. Internally it's not bad. The feature set is nice and it's a solid program. But...again...shit UI, shit workflow. No horizontal integration to speak of. I mean, hell. For example: I can modify a texture in Photoshop and see its effects propagate right to my textured 3D model in Maya. It's easy there. Such integration needs to be the standard with open source apps if they want to be taken seriously and it simply is not.
The capabilities of these programs are fine (CinePaint is head-and-shoulders better than the GIMP, which is probably praising with faint damns), it's just that their workflows all suck enormous amounts of donkey cock and I don't see their workflows improving anytime soon. Don't get me wrong: it's not that I can't do what I want with the open source tools, it's that doing it sucks with the open source tools. It takes longer and is more of a hassle. And, for burgeoning professionals in a university environment, having them not use industry-standard tools is mindfuckingly stupid. People always trot out CinePaint as "oh, look, people are using this IN INDUSTRY!"--great, go CinePaint, may you someday have all the success there is. But a hell of a lot more professionals are using ProTools, Premiere, After Effects, and other proprietary software packages, so it's pretty stupid not to teach what's actually used.
Parent
Re:money is not the way (Score:5, Informative)
Parent
Re:money is not the way (Score:5, Funny)
I think this is the problem with the Ask Slashdot Question.
This is how I read it: "Dear Slashdot. I don't use Word, Excel, Photoshop or any other proprietary software in my day to day work. But on the occasion that I need to crop a photo or graph a single column of data OSS seems to be just as good as the software I never use. How do I switch everyone else out despite their obviously ignorant desire to be held captive by giant evil corporations?"
The answer might be: "Dear Slashdot Questioner. Before you try and pry the closed source software out of your user's hands... perhaps you should remember that you're just an IT help monkey and might not be the best person to evaluate the merits of Open Office Calc vs Excel for the business department. You might not be able to judge the various merits of Photoshop and Gimp in the graphic design department. Or you might be completely clueless as to the relative value of Blender vs 3dsmax and Maya. Please leave tool selection to the various department heads. And we'll leave your decision between a Cisco router and a Netgear router to you.
Cordially,
The Educational Faculty.
P.S. We know where you live.
"
Parent
Re:money is not the way (Score:5, Interesting)
Microsoft will swarm all over you, giving free stuff away. They have a fund just to give away free licenses to anyone who's even thinking about trying open source.
If one of his major goals is to save money (and not be an OS zealot for example, changing to OS just because etc) then doing something that causes MS to open the charity chest be an alternate, possibly acceptable alternative?
Call up MS's volume / edu license group and ask for quotes, saying you're comparing TCO with MS and looking at switching. Not only will you get your quotes, but the Free Gifts Fairy at MS will call you and offer all sorts of nice things to drop the idea of FOSS. Even if you're not seriously considering FOSS, that's a nice way to say, cut the bill for next year's software upgrades in half or better isn't it?
I mean, if MS is going to try to bribe you, may as well take advantage of it if you can, as a serious option.
Parent
Re:money is not the way (Score:5, Interesting)
Parent
*NOT* free for the *Students* (Score:5, Insightful)
You won't be able to win this with the money argument. Microsoft will swarm all over you, giving free stuff away. They have a fund just to give away free licenses to anyone who's even thinking about trying open source.
They give free license (or outrageously cheap site license) for universities. *BUT* not everyone is getting the softwares for free !!!!
No, since you're a university, the way to approach this is to let the undergrads explore.
Also undergrads *DON'T* benefit from all softwares. Most often the students ends-up torrenting their office suites off pirate bay.
Does the University really wants to indirectly encourage software piracy ?!?
Usually, the licensing agreements with stuff like MSDNAA are :
- University gets a dead cheep site license as part of MSDNAA.
- Professors, teaching staff, etc... *DO GET* the right to obtain all these softwares *also for home*.
but
- Students *DO NOT* obtain license for MS-Office for home/personal laptops. They officially have *TO PAY* to get the same software that everyone else is getting for free and that everyone has declared necessary. (Usually, the students actually end up pirating it).
MS-Office is the critical point here.
Microsoft think that, as long as they have seeded the nest (the university) and the important influencial figure (the people giving the lessons), MS-Office will get automatically adopted as the de-facto standart and every body will start using it.
Student will probably get pirated copies anyway, so there's no point in trying to give them free licenses. At least they are getting used to it, they get brainwashed into the notion that there's nothing else worth beside MSO (like all other sheeple), and probably 5-6 years down the line when they finish studying and enter the professional world, they will ask at their workplace to use whatever is the then MSO du jour.
The strategy to bring open source into the university should work on two points :
- not only going open source can save licensing money in the long run.
- open source is also a way for the *students* to get the necessary software for home.
Currently OpenOffice.org is functionally equivalent to MSO. (And is indeed used as a replacement in several public administrations here around in Europe)
At least, even if the university refuses to switch open source, the *students* might be interested getting it for home because it's free, it's compatible with MSO to open university's documents, is functionally equivalent, and even is currently EASIER to migrate to from older MSO 2003 than migrating to MSO 2007, as the OOo's classical interface is closer, unlike 2007's criticized ribbons.
So even if the university refuse to change its stance you have a way to encourage a significant part of the university's population to switch to open source.
Now you can try to use these arguments with the university :
- if they go with MSO, not only do they have to pay (a small) site license, but they are using a solution that WON'T be accessible to the students outside the computer labs (and everyone has seen how currently there are lots of laptops everywhere. Modern students tend to use much more a laptop they carry everywhere, rather than going to the uni computer labs).
- if they go with OOo, the licensing is cheaper (free). They also will be offering a solution that absolutely everyone can use : teacher, staff, university computer labs, students at home, on their laptops... all this regardless of the system : Windows, Mac OS (very popular on student laptops in some richer region) or the customized Linux wich are the latest craze in the netbook segment.
*AND* as an icing on the top of the cake, the current version of OOo will require much less retraining as it looks much more like classical MSO than latest MSO. This is a really ironic argument given the fact that usually Microsoft have always been fast to point retraining as "hidden short-term costs" against ope
Parent
Here is at good way to start (Score:5, Interesting)
Go for the two easy wins first.
Cut your costs on licensing. Get ALL of the decison makers together and get them to put out a 100% unified front. Announce a total conversion to open source for the 2011-2012 year so as to be plausible. Then wait for your Microsoft rep to show up and offer the incentives. Take them.
Now you are a hero to everyone in the university who is in on the con you just pulled. This will be useful to you as you slowly do the real conversion.
The other easy win is to cut the costs to your students. Office and Blackboard.Mandate ODF for any document that crosses the barrier between the school and the students. This relieves them of the requirement to obtain Office and YOU the cost of buying that big site license out of the student fees that is the real reason the students get those low low prices in the bookstore.
You of course continue to offer Office Student at the regular student rates for those who want it because your Microsoft rep is sniffing around. You also be sure to have OpenOffice.org 3.1 DVDs hanging at the register for $5. Be fuzzy about just where those came from, but heck in this economy it sure does save the students money. It's just too popular to pull off the counter.
Blackboard is a never ending cause of cross platform pain (at least it was a couple of years ago) so ditch it. It not being a Microsoft product you can probably get away with it while running the con above. You tell them that will be your token (picked because it IS no visible) conversion to be able to 'claim victory' on your previous grandious project.
After this step students should be able to use whatever the heck they want. Many will probably be using netbooks in this down economy, thus they can buy the really cheap Linux ones. The college bookstore can be encouraged to stock with this in mind. Linux and open source would then be in a position to bubble up.
Surely (Score:5, Insightful)
it's called a cost analysis (Score:4, Informative)
i did one recently to justify the purchase of a new backup system. i got the purchase orders and added how much it all cost over the last 3 years for support, maintenance, offsite tape storage, etc. then compared to a new LTO-4 and estimated a few years out. put everything in a nice easy to read PPT to show how buying a new tape library will save a lot of money going forward.
Same here. get all the costs associated with whatever you run. You might need to ask your boss of finance department. estimate the costs of transition and running the new solution and compare the two.
MS licensing is a nightmare and there are a bunch of programs depending on how much users you have and which program you buy into. ask your finance people to pull copies of the purchase orders.
I work in a 95% MS shop. Reason MS rules is 90% of all MS software is stupid little scripts to make things easier. like the box to create a new user in AD. With Open Source you need to customize a lot of it and it may cost money for the consultants, extra support, etc. I help manage 30 or so SQL servers and in the last 2 years our support costs were around $1000 for a few support cases. In all cases MS released a hotfix after we opened a case. No need for custom coding.
we do have a lot of internally depeloved apps and it's like Quake point releases with them. constant updates and fixes.
Site licenses (Score:5, Informative)
It doesn't surprise me that you can't find good information about this. Even if you found valid pricing for a medium-sized business, I doubt that universities have the same pricing. Universities themselves also negotiate directly with Microsoft (at least the larger ones do), leading to differences in pricing and terms. Unis also often negotiate to obtain student pricing on products like Office. For example:
University of Wisconsin Office 2007 Enterprise [wisc.edu]: $72
University of Michigan Office 2007 Enterprise [umich.edu]: $47
The real question is, if you're "in a position to potentially influence future software purchasing decisions", how do you not have access to the current expenditures on software licensing? What you really need are current expenditures and knowledge about when the current contract expires.
Pay attention to what your users need (Score:5, Insightful)
You can do it by paying attention to what your users need, not just what you want. OpenOffice.org may be an acceptable substitute for MS Office apps in your organization. Or, you may hobble the faculty because they're required to submit Word documents for various publications, using Word templates. It's bad enough having to suffer through this in Word, but having to manage this with another layer of indirection sounds utterly intolerable. That situation sucks, but you aren't going to change it by unilateral decree.
Likewise, using the GIMP vs. Photoshop may be great for some of your users. But if they're using features daily in Photoshop that aren't supported in GIMP, soon they'll be GIMP'ing up dartboards with your face on it.
Simply put, users care about applications that meet their needs and organizations should too. If you are truly in a position to influence these decisions, then your responsibility is to understand and meet those needs, not serve your own ideology. Working contrary to users' needs is a terrible way to promote the OSS software cause; you'll make more enemies for OSS than friends.
Microsoft Licensing (Score:5, Informative)
With Vista (and "above" - 2k8, win7), Microsoft changed the way they do site licensing. Instead of having one key for every computer, every client does a DNS lookup for a Key Management Software Server (KMS server), which then simply activates the client computer. It does not keep a record of how many activations you have used, only the last 50.
Likewise, you just call them up, tell them how many computers you have, and they give you a price. A few minutes and many thousands of dollars later, you have a key to plug in to KMS. Magically, every Vista+ box that you have on site is licensed and activated. This can include student computers if you wish. The activations 6 months, after which time they *must* talk to the KMS server again.
http://www.microsoft.com/licensing/resources/vol/default.mspx [microsoft.com]
Now look. I run centos/debian/openbsd/gentoo/xp/vista/server 2008. I really hate (operating system) licensing. I hate the simple concept. But KMS is really the way to go. It takes right next to no system resources. In the KMS docs, they say that most 100k+ client customers are perfectly content with 2 KMS servers (with the same key). Next to zero system load.
Second, Office.
http://office.microsoft.com/en-us/suites/HA101080191033.aspx [microsoft.com]
There is also their Software Assurance program.
http://www.microsoft.com/licensing/sa/default.mspx [microsoft.com]
Software Assurance has one big downside, and one big upside. The downside is that it is a yearly fee. It is more or less a subscription. The upside is that you are entitled to free upgrades of "the product" as long as you keep paying. This means that if you purchased SA on Office 2003 a year before 2007 was released, your 2003 license can be automatically upconverted to 2007 free of charge. The same applies to... all of their products. XP --> Vista --> Win7, SQL 2000 --> 2003 --> 2008, Visual Studio, the works. It is not a required upconversion either - you choose if and when you upgrade.
As a result, buying your weight in gold worth of Software Assurance also gives you 24/7 software support. It more or less gives you everything. Tech support, upgrades, technical resources... it is essentially the equal to a Red Hat Enterprise Linux subscription in terms of the support you get, the products that you get, and the upgrades.
Really, your best bet to understanding MS licensing is to contact one of their reps. Gather everything that you can find before hand, and give them a call. Grill them endlessly. Ask questions, and don't let them leave until you know everything you needed.
What is the benefit of open source/free software? EVERYTHING ABOVE IS ENTIRELY IRRELEVANT.
Give up now. Pick your spots intelligently. (Score:5, Insightful)
I am being 100% honest here. I too work at a univeristy, a bit larger but same deal. You are shooting yourself in the foot big time, but well intentioned.
There are far too many individual needs in this setting to do what you propose. Instead identify and choose a few specific spots where open source actually makes sense and offers a huge advantage (there are a couple) and make it happen. Start small and be smart about it. If it goes smoothly and shows real savings and improvements you may have earned the chance to do the same in another area.
Openoffice sucks. Period. Large-scale monitoring and maintenance can also suck. Sometimes Mac OSX is even the best choice. You have to take off the rose-colored glasses and think critically about everyone's real needs not just your pie-in-the-sky dream.
Don't shortchange the whiners... (Score:5, Insightful)
In most organizations, it takes only a small group of whiners to transition the whole of an IT focus to something else. Trust me, I've been through this battle.
Make changes where it *makes sense*. Microsoft Office currently is best of breed, no offense meant to OpenOffice but seriously... it's not even in the same realm. Windows on the desktop obviously goes side by side with this.
Where you can make arguments are on the backend where users don't really have a say. Say you want to launch some web servers -- go *Nix and Apache instead of MS and IIS. Want a database cluster? Go *nix and MySQL. These are changes that *can* happen.
I have seen far too often that 'techies' get involved and just because the technology is more superior (in some way) they totally discount the business benefit from having it set up that way. What is your roadmap for the future of IT? What paths are you looking to cross? Say the CIO wants to invest some money into Sharepoint, or wants to use WIM (standard image format) for deployments, or wants to lock down users better (AD Policies). These things are *windows specific*. You can make the argument, but if you can't look at it from a business perspective, then you are already on the path to failing at your argument.
Usually the cost of changing everything, retraining users, and getting them to be AS PRODUCTIVE as they were before is far more expensive than to keep technology the same and use branches into other things to accomplish business tasks.
And don't say you're an educational facility... you're a business first, and any good business is in the *business* of making money or showing results. That's what you call an organizational unit :)
Good luck to you, but make sure you have your ducks in a row before you go making arguments of vast change, because if you don't know what the future holds or what the goals are, you will just look like an idiot.
From a faculty member... (Score:5, Insightful)
I am an assistant professor. If you came to my office and told me to use anything, I'd kick your IT-fiddle-monkey-ass to the door.
Here's something I really want university IT guys to get through their thick skulls:
You work for us. Not the other way around.
If I want to use a Windows machine, you need to figure out how to let me. If I want to use a Mac (which I do), you need to make sure I can get to my servers. If I want to use Linux (which I hope to be doing one day--when the software I need to do my research is available on the platform), I expect your support there, too.
In the specific case of what you're proposing--moving to OSS for all everyday tasks, I have to be totally clear and honest here: You are wholly unqualified to make that call. It's not your job; it's not your responsibility; it's none of your damned business. You don't even know what I do; how could you know what I need?
Finally, let me say this: My first jobs in academia were in IT support, and I, too, got drunk on the power. I, too, was young and full of myself, and I, too, ran around telling people what they should do, instead of listening to what it is that they needed to do, and helping them do it. Now that I'm on the other side (and older and less full of myself), I see why I pissed people off so much in those days. I sucked at my job.
If you try to meddle in your customers' business, you suck at your job, too.
Re:Remind me not to send my kid there. (Score:5, Interesting)
A university is supposed to educate a child as to the world of software
Really? Maybe you are thinking of trade schools. A university is supposed to provide a well-rounded education. Indoctrinating into the world of Microsoft might be helpful in getting a white-collar-grunt job, but it is not in any way vital to a liberal arts education.
And anyway, a large percentage of universities use *nix and/or Macs. Are they all failing in their educational mission as well?
Parent
Re:Remind me not to send my kid there. (Score:5, Insightful)
Yes! Exactly! Your argument is so wonderfully persuasive. You've totally discouraged me from open source software! I will only send my children to universities that support convicted monopolists and their patent/copyright law abusing corporate pals. Also, I 100% agree that free as in speech OR beer software shouldn't exist to insure the enrichment of these companies.
It is of UTMOST importance that I spend two years of wages on an education designed to give my kids painstakingly detailed, precise instruction on where to point-click in MS-Word to make pretty charts! And to help cover the licensing costs, I will GLADLY support and requests to raise tuition. After all, it would be down right un-American to not work my ass off to help cloth and feed a bunch of rich assholes!
Parent
Re:Remind me not to send my kid there. (Score:5, Insightful)
After all, it would be down right un-American to not work my ass off to help cloth and feed a bunch of rich assholes!
So, while we are at it then, do you want to do something about universities owning huge patent portfolios paid for by your tax dollars, while at the same time raising tuition faster than even the price of gasoline?
Parent
Re:Declaration of independence (Score:5, Funny)
Bes
http://osor.eu/case_studies/declaration-of-independence-the-limux-project-in-munich [osor.eu]
Nah, too political and the political process takes too long.
No, what you have to do is the following:
1. Shoot all Windows admins. I know, it sounds brutal but trust me, it'll be better for everyone on the long run. It's no more than what they deserve after all. I mean, they freely chose to support the Evil Empire.
2. Send all the brainwashed Windows users to the appropriate re-education camp to have them deprogrammed. Now, some might say that this is no better than what Microsoft has done all these years, and I'd agree, but sometimes you have to fight fire with fire.
3. Send in the LPTs (Linux Proselytization Teams) to spread the Word amongst all those who didn't get the message the first time around. After all, there are always some for whom the deprogramming process doesn't work perfectly, or who managed to escape the initial roundup. It's necessary to root them out so they can be given proper guidance. Really, it's for their own good.
This may be hard for some of you stomach, I understand, but just think how free we'll all feel when Microsoft is gone forever.
Parent