Most Useful OS For High-School Science Education?

Clayperion writes "I teach at a high school program for gifted students which emphasizes math, science, and technology. Currently we have two computer labs for the students: A new programming lab (all Dell PCs running XP, MS Visual C++, Eclipse, and SolidWorks for programming and CAD) and an old general-purpose lab (all Macs running OS X 10.3, with software ranging from some legacy OS 9 science applications to MathCad). Most of our students eventually pursue graduate degrees in science and engineering, and we would like them to have experience with the tools they will find out in industry. As we look to replace the old machines, there has been a push to switch to PCs with XP so that there is only a single platform to support. There are over 5000 machines on the district's network and the IT department is very small (fewer than 10 people), so the fewer hardware and software differences between the machines, the better. Without opening a flame war as to which one is 'better,' I'd like to know what those of you in the science and engineering fields actually use more in your labs (hardware, OS, software), so that we can decide which platform to support. It will most likely have to be either XP or OS 10.6, with very restricted permissions to students and teachers, as that is the comfort level of IT and administration, but I'll push for whatever would benefit the students the most."

Comments

Windows XP?

[ta bu shi da yu]

I'm not sure I'm following the logic... Windows XP is getting close to EOL. Why wouldn't you use Windows 7? Certainly it and Windows Server 2008 has more features to make admin'ing easier.

Re:Windows XP?

[Red_Chaos1]

XP may be close to EOL, but it has massive support behind it still. There are numerous applications that extend its features, make it very easy to customize, etc. Not to mention most of the bugs and such that are left are well documented and easy to fix or work around.

Re:Windows XP?

[Kaboom13]

XP is dead. If you aren't stuck to a legacy system (as this guy isn't) you would be a complete fool to stick with XP. It would be a mistake you will constantly regret. Most of the things you would want to extend XP's features are built into 7/Server 2003. Remote administration, Patching, Application Control, Network Image Deployment, locking down the desktop like deep freeze does, all can be accomplished with built in (and supported) features. Security is also better (requiring drivers to be signed, built in support for full disk encryption, Memory address randomization, better default settings, better implementation of SFC, etc), and the systems are a lot more usable running as a non-admin without lots of extra scripting work. You also get better ip v6 support, and improved network performance in general. Just the fewer headaches in patching alone makes it worthwhile (even with a WSUS server, I find myself frequently manually updating XP machines, I've never once had to do it on a 7 machine).

The 7/Server 2008 networks we have deployed require substantially less maintenance then the XP networks. Support for XP is being phased out on new hardware, as it is you have to stick to certain long-term support models to get support for XP from the big OEM's (there's a difference between "heres some drivers, good luck" and officially supported). 7 is a mature OS, if it makes you feel better think of it as Vista service pack 3. Furthermore if you don't have the cash to shell out for VLC licenses, expect trouble when Microsoft drops downgrade rights on OEM licenses. Setting up a brand spanking new network with Windows XP is like making a brand new web app from scratch, and designing it in Visual Basic to only work in IE 6. You can do it, and the technology is tried and true, but you will be creating more work for an inferior result that will bite you in the ass in a short time frame. The only reason for not deploying 7 on new hardware where you are not constrained by legacy code is you want to stay in your comfort zone, and are scared to learn new things. If that's the case, you need to GTFO IT, it's the wrong field for you, and you are doing your clients/employers a disservice. Being skeptical of new technology is fine, but being irrationally afraid of it is stupid. As far as Engineering/Science goes, any commercial software package that can't run at all under 7 is probably on it's way out anyways. Whats bleeding edge today will be a generation behind by the time the students get into the real world.

All that said, I think XP/7 is the WRONG way to go. If you want a Windows environment, your best bet will be to buy some thin clients, network boot them with something like ThinStation, and have them RDP to a farm of nice beefy 2008 R2 Terminal Servers. Thin clients are the only thing I've seen hold up to a school environment. Unlike a corporate environment where you can expect the employees to only cause damage out of ignorance, high school students will be actively malicious, and will destroy/break/steal things just to do it. If you lose a thin client, the teacher can yank it out, pull a spare from the closet, and send the old one to be diagnosed/redeployed in your spare time. Because they are stateless, if one is stolen you are out a couple hundred bucks and not any information. It will be easier to setup a consistent environment, and you can shop around to different hardware vendors if needed while maintaining a consistent experience for the students. It will be easier to create flexible lesson plans, install software, and you can often really cut down on licensing costs. Thin client tech has come a long way, and if you spec your servers properly, and have a decent network, you can't tell the difference. I took a class in Solidworks (a ram hungry and CPU hungry 3d CAD program that makes your average office workstation dog slow) that was taught in a lab using thin clients and terminal servers, and it ran better on them then my personal laptop, despite having 20 other users on the same serve

Science or Engineering, huh?

[Itninja]

You know those are meaningless unless we know what kind of science or engineering right? Civil engineering? Network engineering? Traffic engineering? Geneticist? PhD Researcher? Hell, Sexology??? What of donuts?! WHAT!?

Re:Science or Engineering, huh?

[value_added]

Maybe it's me, but 5,000 Dell computers all running XP suggests Microsoft Certified Systems Engineering.

Re:Science or Engineering, huh?

[spun]

A mechanical engineer, an electrical engineer, and a civil engineer are discussing God. They all agree He must be an engineer, but what kind? The mechanical engineer says, look at the human body, its skeleton, joints and musculature, mechanical genius! God must be a mechanical engineer. The electrical engineer says, Nonsense! Look at the brain, the nerves, God is an electrical engineer. The civil engineer says, "Nope. God is a civil engineer, who else would put the sewer outflow in the middle of the entertainment district?

But we all know donuts belong in the realm of theoretical physics. I quote the great Stephen Hawking, who said, "Your theory of a donut shaped universe is intriguing Homer, I may have to steal it."

Re:Science or Engineering, huh?

[Anonymous Coward]

Maybe it's me, but 5,000 Dell computers all running XP suggests Microsoft Certified Systems Engineering.

Why do I always hear circus music when anyone mentions them?
DOOT doot doodle oodle OOT doot doot doot, DOOT doot doodle oodle OOT doot doot doot, DOOT doodle oot doot, DOOT doodle oot doot, doodle oodle oodle oodle doodle oodle oot doot.
You know, the music they play when the clown car comes out, and the clowns start getting out, and you're all like, "Can there BE any more clowns?" But there are, there are always more clowns, and they just keep piling out of that car while the music plays.

Re:Science or Engineering, huh?

[Hognoxious]

My coffee mug has a handle that's only attached at the top

I hope you aren't a mechanical engineer. Oh, the stress concentration!

Re:Science or Engineering, huh?

[1s44c]

Maybe it's me, but 5,000 Dell computers all running XP suggests Microsoft Certified Systems Engineering.

Just what the world needs, another 5000 node botnet.

Linux in our labs

[King InuYasha]

Most of our labs in college use a mix of Fedora and Ubuntu Linux, with some Solaris speckled around.

I'd probably go for Fedora, since a lot of students will likely be working on some Fedora derivative, and it is easier (in my opinion) than Ubuntu to administer. However, it's really up to you.

I've also heard that many of the co-op companies our college partners with use some form of Linux. Though, for obvious reasons, a few design oriented companies use Mac OS X, though that may change in the future.

Windows is a rarity, from what I've seen and heard.

Technically real world use.... OSX

[dbarrycoyle]

I work at NASA and have many university colleagues I work with as well. A recent IP survey I had IT do at GSFC in MD showed a Mac OSX installation base of about 30%. This is similar at my freind's universities... at least in the physics and engineering depts. We recently moved our 20 or so PC's over to Mac a few years ago and have been very happy. I was able to show I saved the government approximately $60K-$90K a year in gained productivity and reduced IT support, salary, etc.... So, while Windows is used mostly now by the Best Buy consumer level base, which is 80% of the "market", the professional technical use of OSX is much higher. I suggest having a mix of new machines if possible and taking your own data. Track how often the machines are used, under repair, software costs, and how the students take to them and make your own conclusions. Good luck.

Habababdub

[Cylix]

So science, religion and porn have three things in common with your network. Neither of them are really going to play a huge role in the decision of the topology or specifics regarding your hosts.

What is important to consider are what are your requirements for the specific applications that apply to your curriculum today and in the near term. These things dictate what is necessary to support your environment. If you don't know what you should be using I would consult a similar audience rather then the general populace. In practice, I've generally found most educational institutes are staffed with at least some individuals who do thrive in the industry. (Hint, industry experience is a good thing).

In any event, this is a very long winded ask slashdot, but offers very few details. Even if someone said to change all of your systems to XYZ using ABC it wouldn't really matter. You can't base a purchasing decision on a few paragraphs. I certainly don't want to draw up a diagram of how your architecture should work and toss out a handful of applications.

The bottom line is that you should know at least some of these details. What are the pain points with whatever and certainly not detailed plans on the horizon.

Here is my two cents....

Come up with a consistent approach to your operating system selection and configuration. Ensure you have the capabilities to deliver a clean and automated of said services. With only 10 individuals it will really will become a painful support paradigm if you continue with some haphazard configuration.

As far as software selection.... because I know virtually nothing about what you currently use or specific fields this is in regards to... I want you to find the most expensive application that does a single 10th of what you want it to do. Buy lots of this software and pray they release the features you need in the next release.

mac mini's with virtual machines running win7

[jsepeta]

computer hardware is probably a pain to procure at a high school, so i recommend the relatively inexpensive Mac mini. we're planning on converting our XP lab to Mac Minis running Windows 7 in a virtual machine (Virtualbox) which means our computer hardware won't be a limiting factor when selecting the software we teach students in our lab. Mini's are as much power as you'll need, and this makes more sense than iMacs when you factor in the cost of 22" or 24" LCDs. and by running Virtualbox, you can even set up multiple vm's so you can test out new versions of software without having to perform complete rebuilds if some microsoft update hoses the system. hell, you can even add some linux to your environment should their be some cool engineering or programming tools that would otherwise be too costly on the microsoft or apple platforms.

yep, you can buy a pc cheaper and of course you can run Linux for free, but it will probably help your students the most if they get a little bit of experience with multiple operating systems since once they graduate from college, they'll probably be using OSX 10.7 or Windows 8. running XP is a nightmare because of the security holes AND because Microsoft has already started to eliminate XP, say 2 years ago when they first discontinued it.

having dealt with apple dealer to school sales since 1991, I think the choice [Mac or PC] is a false choice. And since there are no viruses or malware that run on OSX, the schools we support who run OSX spend a shitload less on support costs, which can quickly suck up your budget, your time, and your patience in a school environment should you be running XP and get zapped by malware. since running vm's is easy, it's become a preferred way to quickly switch a lab from one group of students to the next.

It's gotta be one of the many UNIX variants

[moria]

I work for a research lab in a university and we do a lot of scientific computing and webapp development. Here it is UNIX variants and only UNIX variants. We use Debian Linux on our clusters, Mac OS X or Debian Linux on my Mac Pro or Mac Mini desktops. Knowledge about C/C++ and scripting languages is very important. We are recently interviewing candidates for an opening, and it is very sad to see people who cannot code without IDE and who think building the binary is equivalent to clicking the little button on the toolbar. If education needs to do one thing, then that should be to give students a broader view instead of limiting them to some false impressions. In that sense, UNIX is a much better tool because of its rich history and active development.

Doesn't matter

[swillden]

High School seniors are between 4 and 8 years away from working in an engineering field. That's enough time for things to change considerably, and even if it weren't, the operating system really doesn't make that much difference. If you could give them some experience using the apps that will be relevant to them, that might be a little more useful, but that space is so broad that there's no way you could know what will be needed.

I'd make sure you pick a platform that runs the software the teachers want to use for classes. If that software is available on multiple platforms, then pick the one that is most cost-effective, considering acquisition and maintenance both.

Choose it based upon Applications.

[HockeyPuck]

In our district Freshmen take Earth Science, Sophomores take Bio, Juniors Chemistry and Seniors take Physics. There's also some techy electives such as Intro to Programming, Computer Animation/CAD and an Intro to Computers (teaches the basics of how to use a computer, browsers, word processing, etc...)

Check out the applications that your those that set the curriculum want to use. Some software suites are available for one platform and not another. You can't just say, "We're using OS/2 and that's the way it will be!" As you'll have 10 department heads yelling at you that there aren't any XYZ applications available to it.

Also, who says you have to have 5k PCs each with it's own disk, OS load etc.. Why not look at Virtual Desktops (vmware view with dumb terminals/thin clients in the classrooms? The Unix folks have been doing this for years, but this solution is pretty slick. We've deployed it for all the staff as they only use a dozen or so standardized applications.

Btw, I'm an ex-mainframer and managing 1 mainframe and 5000 dumb 3270 terminals is much easier than 5000 desktops; and speaking from experience managing a couple of large X86 servers and a 100 thin clients is very similar.

What career path do most of your students pursue ?

[tuxidriver]

It will depend heavily on what path your students pursue.

I've done a mixture of hardware design and firmware development for both storage peripheral companies and IC houses. What I mostly see is:

• Embedded development: Roughly a 50/50 mixture of Windows and Linux. Most compilers for embedded applications (e.g. Green Hills and ARM) are available for either platform. The same compilers are not available for Mac.
• Digital IC design: Linux and some Solaris on big iron. In my experience, there is 0 use of Windows in this space. Most companies appear to be moving towards Linux.
• IC verification (validating the design prior to producing the reticules used to manufacture the chips) is also 100% Linux with 0 use of Windows.
• Analog IC design: Also strictly Linux and Solaris on big iron. Again, no use of Windows in this space.
• Board level electrical design: Mostly Windows.
• Mechanical design (Solid Works, etc.): Mostly Windows. One company I worked for used IRIX on Silicon Graphics workstations for 3D modelling, although they did eventually transition to Solid Works on Windows.
• The modeling work I've seen/done (modeling Mueller-Muller clock recovery, Viterbi decoders, LDPC decoders, etc. used in communications systems) has been a 50/50 mixture of Windows and Linux (in some cases with the models developed with GCC and written to be portable across platforms). I believe this is mostly due to the compute resources that the companies I've worked for had on hand.

I have yet to see any significant use of Mac's, except as clients to log into Linux workstations. Almost all IC design and verification is done on some POSIX compliant OS because of the the requirements of the tools. IC houses I've worked for generally have large numbers of 32, 64, and 128 way multi-processor systems with huge amounts of RAM. Windows XP is simply not able to take full advantage of these large systems and the tools require this much horsepower to be effective. I also have noted that many IC designers generally seem to prefer the power of a good CLI over GUI point-in-click file managers. There is also a lot of scripting in these environments, mostly in Perl (although I've also need shell script and Python used). Linux and similar operating systems lend themselves more for this sort of work.

As for tools, I would suggest that you seriously look at trying to give your students at least a taste of such tools as MatLab, MathCAD, AutoCAD, and S. There are free equivalents for MatLab such as Scilab and Octave as well as Python packages such as SciPy, NumPy, and MatPlotLib (which I sometimes use for modeling). I know that languages such as S+ (or the free R language) are sometimes also used for statistical analysis. If you want to give your more advanced students a taste of chip design, consider the free offerings from Xilinx along with a few of their FPGA evaluation boards (available through DigiKey).

I hope this helps.

Re:Free OS, free software

[MagusSlurpy]

Then you'd better start writing all the software to control the various scientific instrumentation I use, because it all currently requires proprietary software running on the recent Microsoft OSes (that Oxford NMR actually does have a Linux client available, but the PC controlling it runs XP for ease of file transfer).

Any research relying on results produced by close-sourced software is voodoo.

Well, then 98% of published chemical research is voodoo. Companies aren't going to write open software to control the $750K spectrometer they just sold you, and to be perfectly honest, I don't think I'd use software off of Sourceforge to control an investment of that type, anyway. Nd-YAG lasers don't grow on trees, unfortunately.

Why not?

[DiegoBravo]

> Companies aren't going to write open software to control the $750K spectrometer they just sold you, and to be perfectly honest, I don't think I'd use software off of Sourceforge to control an investment of that type, anyway.

I'm not a chemist, but I think your investigation is not about controlling the spectrometer, but the resulting spectra. So I think it would very interesting and potentially productive if you have the source code of the software that transforms/filters/enhance/displays the output data.

BTW, I don't believe the people at CERN will rely on some close software for tracing their particle collisions.

Re:Why not?

[jythie]

Labs that build their own equipment from scratch tend to stick with OSS stuff, labs that buy pre-built instruments tend to use windows based control software. That is actually one of the splits that makes answering the OP's question in any useful way impossible. What OS scientists and engineers use is pretty heavily dependent on their needs, and needs vary wildly.

For chemistry, biology and physics. . .

[MagusSlurpy]

. . .I would recommend Windows, Windows, and (not strongly) OSX.

There is no question in my mind that Windows is the way to go for chemistry software, as I've now spent almost ten years at three different universities working my way to a PhD (almost there!), and besides the occasional foray into Linux (control software for two different brands of NMR), it's been Windows all the way (and the NMR software was available for in a Windows client, also). I could post a list of all the instrumentation I've used, but trust me, it's long, probably around twenty-thirty instruments now.

From my undergrad experience:

I haven't used as much software earning my bio degree, but we mainly used statistical packages, and they all ran on Windows - the SEM (the only instrument I used in that department) ran on XP, too.

I only had a year of physics as required for the chem and bio degrees, but the physics department uses Macs for the computer labs and the classroom computers - supposedly there are a lot of interesting software packages available, which I never used. The instrumentation I had the opportunity to use (the Mossbauer spectrometer and the x-ray diffractometer) both ran on XP, though.

Re:Free OS, free software

[Chryana]

Given the shortage of manpower the OP mentions, I think he could use the administration tools that come with Windows, and therefore should buy licenses for it. I'll even go as far as to say that to base such an important business decision on some idealistic views of how a computer science lab should be ran would be irresponsible, and worthy of being fired.

Any research relying on results produced by close-sourced software is voodoo.

The validity of any research is confirmed by the ability to independently reproduce its results, not because you can check the code which is used to generate the research data.

Re:WetWare 1.0

[Omnifarious]

IMHO, teaching programming without a computer is like trying to teach math without using numbers. I mean the arabic numbering system is basically a shorthand way of writing down polynomials where 'x' is always 10. The numbers have a reality quite apart from their representation and getting that is one of the most fundamental and important ideas in math.

But really, starting there is a bad idea.

People get excited and enthused by results. Nobody is going to be excited and enthused by a set of principles that don't have any connection to anything else they know. Getting people excited about learning is the biggest part of the battle.

Teach the kids to learn...

[0100010001010011]

You're going to run into one of two problems.
1) By time the kids grow up everything in industry will have changed.
2) You can't afford what they use in industry with a HS budget, even the [college] student licenses.

I'm a mechanical engineer. I make my living using Matlab, Simulink, CANape and some internal company programs.

I went to HS with Windows ME (with MacOS 7/8 at home) I honestly don't ever even remember using them. Our "Physics Lab" was an Apple II running some highly custom software and hardware. (Running lasers to time ball bearings going down ramps and such). I learned the basics of programming with TI-Basic. In college I picked up Java, C, & Matlab/Simulink.

Now I run 10.6 at home and XP at work. Something no one could have predicted back in the day. Teach the kids the basics. If someone 'gets' how to program, it doesn't matter. If someone 'gets' chemistry, it doesn't matter if they're drawing them on paper or in some 3D model.

And I haven't priced a student's version of Matlab recently, but I know my seat at work runs 20k. Simulink doesn't make too much sense until you've had DiffEq. I haven't used Octave enough to know how compatible it is. CANape... well you'd need quite a bit of money for the stuff to run it on. There's a reason there are a half dozen solid modeling programs, because companies use different ones. And with my short time with most of them, they're completely different. AutoCAD, CATIA, ProE, SolidWorks, etc.

Re:Teach the kids to learn...

[Anonymous Coward]

Octave is MatLab enough for someone who is just learning.
For highschool or first year undergrad stuff it can be considered matlab without the fluff
ie. graphical interface/array editor, built in clicky menus and ezplot. I can't think of how this would be a disadvantage when it comes to teaching people how things work.
I've heard that its floating point isn't as good (second hand), but I've never run into any problems (in undergrad physics).

Re:Teach the kids to learn...

[atomic777]

There is no platform that will satisfy all objectives; arguments can be made for Win, OSX and Linux.

Of course, my vote would be for Linux. Let's remember that this is for a high school. Octave is more than capable of serving as a matlab replacement.

R has now supplanted S-Plus as an industry-standard (at least academia-wise) statistical programming language, one I also use frequently.

Between octave and R, and the other general purpose programming languages that are a breeze to develop with in a Linux environment, there is a great deal of important scientific work you can do with free software.

Linux is also the only platform that makes sense when you start needing to crunch lots of data on many servers, especially with a small budget. Linux is standard on all academic clusters I have seen. Give these students the skills to manage data crunching on a small cluster of linux machines and you will do them a tremendous favour.

If you have some tools which are proprietary and specialised, you can easily set up a couple of windows/osx machines for their use specifically. But it's hard to beat the value of Linux as a general purpose scientific platform.

Re:Teach the kids to learn...

[farrellj]

The other thing to think about it security. I used to support a medium sized school district years ago running Novell Netware and IBM's AN/ICLASS software. It was the most challenging environment because Murphy LOVES high school computer systems! Things will break if they can, and students will break them if it's possible. Viruses were rampant, and more often than not, the students knew more than the teachers. Now, I know that the last one has gotten somewhat better, but it still is a problem.

Just from the virus problem alone, I would recommend that people use Linux because it takes a lot more to crack a well secured Linux system(s) than a Windows or Mac...as various security competitions have shown. Another good one to consider is that you don't have to worry about people stealing licensed software, or the licensing information to run the softwares at home, there by eliminating a possible legal liability.

ttyl
          Farrell

LOL - You actually asked slashdot what OS to use!?

[starcraftsicko]

You say you want to choose the OS for your HIGH SCHOOL science lab based on what your engineer wanabe students may actually use when they make it to industry. Good Grief!

XP is already EOL and DISCON. They won't be using that in 6 years.
Win7 will have been replaced by at least 2 subsequent versions and will probably be DISCON.
OSX 10.6 will be replaced and DISCON, will be actively unsupported by Apple.
Whatever version of Linux you choose will have forked 600 times by the time they get out of college. Whichever one you pick now will be wrong.

The (wrong) choice you make today will have absolutely no impact on your students' preparedness for real-work in 6+ years.

SO:
Find the applications you want to use. Choose an OS that runs them all.
OR
Ask the IT guys where you work to choose. They have to support it, they know what they know how to support best.
OR
Load an old Slackware Distro and make the IT guys hate you. Make your students write the software they'll need. Then they'll really be prepared.

Protip: When you ask SLASHDOT what OS to use for ANYTHING, the consensus answer is going to be "well, you could use linux..."

Re:Absolutely disagree

[Shrike82]

A calculator is a tool to make doing maths easier - you don't need a calculator to do maths. Have you tried programming without a computer? It's your analogy that fails. Also, since I actually teach first year university students programming I can say with some authority that presenting them with theory alone is destined to fail. You can actually see their eyes glaze over as you dive into the second hour of a lecture about what classes are, what a method is etc.

However, you mix that up with demonstrations of a HelloWorld program, a simple GUI that does something pretty or whatever, and they stay interested.