Slashdot Mirror

IIRC, the irony is that LTE is used regardless, and there isn't any 3G anymore.

I wasn't sure if the 3G indicator I see in some rural areas was real, but it seems the 3G shutdown hasn't come yet.

Here's what we used at Clearwire for OOB management of radio sites.

http://www.digi.com/products/w...

Digi International Inc. makes a line of routers called the Digi Transport that have 2g/3g/4g/lte options. They aren't really positioned for home users, they are a bit high end/commercial/industrial. You can find them via google or here's a link to the Digi store. http://store.digi.com/index.cf...

For a non battery-assisted tag at 30m through free space with a non-pinpoint beam antenna* you'd be hard-pressed to get 10% [yes, 10%] read for a tag that's < $1 and small. At 100m, no chance. I hate to say it, but I think your idea of RFID as a sensor net is DOA both from the physics involved and the expense (RFID readers/antennas/coax/etc are not cheap).

Look up the power link budget of RFID - the fact that they work at all is astonishing. The power that the RFID reader gets back from the tag falls off as the 4th power of distance - that a wicked barrier. Add to that the directivity issues, and you're talking impossible.

* To read tags scattered across a building, you need omnidirectional antennas, or at least a small number of moderately directional antennas. This, however, is diametrically opposed to the need to read tags at a distance - the more power you focus into a beam, the less there is for reading tags outside the beam.

ZigBee modules are currently $22.50 for the SMT version and $17 for the DIP version. These are essentially small sensor platforms ready to go. They have a UART on board (their interface to the non-RF world is via the serial interface - you can use them to transfer RS-232 data from any device). They have A/D converters and digital IO lines on board, and can be configured to sleep, wake up, take readings, send data to the 'mother ship' (which is just another ZigBee module - no expensive reader needed) and go back to sleep. They can operate off of small batteries (3-ish volts, rx & tx current about 40 mA, sleep current in the microamp region). This is exactly what you want, with the added ability to use always-on nodes as mesh repeaters as well as sensor nodes. Suddenly your ZigBee-enabled lightswitch (that's mains-powered) is acting as a repeater for your solar powered swimming pool wave height sensor - with NO explicit configuration! They even have a USB stick version that appears as a standard COM port to your computer.

PLEASE have a look at the Digi zigbee modules - they're extremely well thought out, and have lots of features in a small, efficient package.

PS - They have higher power ones and have ones with antenna connectors if you want even more flexibility. They have data collectors/routers as ready-to-go products that can publish to the web and be managed over the web if you don't want to roll your own.

For a non battery-assisted tag at 30m through free space with a non-pinpoint beam antenna* you'd be hard-pressed to get 10% [yes, 10%] read for a tag that's < $1 and small. At 100m, no chance. I hate to say it, but I think your idea of RFID as a sensor net is DOA both from the physics involved and the expense (RFID readers/antennas/coax/etc are not cheap).

Look up the power link budget of RFID - the fact that they work at all is astonishing. The power that the RFID reader gets back from the tag falls off as the 4th power of distance - that a wicked barrier. Add to that the directivity issues, and you're talking impossible.

* To read tags scattered across a building, you need omnidirectional antennas, or at least a small number of moderately directional antennas. This, however, is diametrically opposed to the need to read tags at a distance - the more power you focus into a beam, the less there is for reading tags outside the beam.

ZigBee modules are currently $22.50 for the SMT version and $17 for the DIP version. These are essentially small sensor platforms ready to go. They have a UART on board (their interface to the non-RF world is via the serial interface - you can use them to transfer RS-232 data from any device). They have A/D converters and digital IO lines on board, and can be configured to sleep, wake up, take readings, send data to the 'mother ship' (which is just another ZigBee module - no expensive reader needed) and go back to sleep. They can operate off of small batteries (3-ish volts, rx & tx current about 40 mA, sleep current in the microamp region). This is exactly what you want, with the added ability to use always-on nodes as mesh repeaters as well as sensor nodes. Suddenly your ZigBee-enabled lightswitch (that's mains-powered) is acting as a repeater for your solar powered swimming pool wave height sensor - with NO explicit configuration! They even have a USB stick version that appears as a standard COM port to your computer.

PLEASE have a look at the Digi zigbee modules - they're extremely well thought out, and have lots of features in a small, efficient package.

PS - They have higher power ones and have ones with antenna connectors if you want even more flexibility. They have data collectors/routers as ready-to-go products that can publish to the web and be managed over the web if you don't want to roll your own.

To back up the AC, his post is completely correct. FAA Advisory Circular 91-57 allows individuals to fly remotely controlled aircraft up to 400 feet in altitude, within visual range at all times. This does not apply to university craft, craft owned by a business, or craft owned "by the public" such as police, military, or any city/county UAVs. You have to be able to demonstrate to the FAA that you are flying solely for entertainment/education and that you are only spending your own money on it, and not making any profit off the venture. In fact, it is against FAA regulations for non-individuals to even fly remote controlled aircraft under this AC (I know this based on a very tense and bitter fight with FAA officials over an R/C club on a military installation using donated money to purchase aircraft).

There exist a wide range of technologies for sending telemetry from air to ground over the 2.4GHz and 900MHz ISM bands. (a few examples). There is no need or reason for a ham license whatsoever since it's only purpose is to allow you to fly beyond visual line of sight. The claim that it is necessary is flat wrong.

On to the "anyone can build one" discussion: it's absolutely true. I've built a total of 6 UAVs over the last 5 years. Average cost is $1k if you don't crash it during testing. The typical route is to take an ARF model, add a commercially available autopilot, hook up the servos, put in a bigger engine, battery packs, etc. On-board computers (not counting the autopilot itself) are almost never needed because these vehicles have to be operated line of sight anyway. If you want to do on-board computation, there are a wide variety of options from Gumstix, to pc104 form factor systems. All telemetry is usually handled on the 900 MHz ISM band which I have personally flown as far as 15km (in a UAV-approved range with all of the proper approvals) using a Yagi antenna that was self-steered from the telemetry data. The link still had plenty of signal left on it, but the UAV didn't have the legs to go further and still complete its mission.

Moral of the story: learn the rules and stick to them. The FAA does not accept "but I heard someone did it on /." as an excuse, and they have a tendency to not only shut down the idiot that didn't follow the rules, but every other person conducting UAV research who is even remotely associated with that club/institution/location.

I just became aware of another product like this. Our area uses Itron or Schlumberger automated electric meters. These meters already have a radio transmitter in them. I got the Digi ERT/ethernet gateway http://www.digi.com/products/wireless-routers-gateways/gateways/ert-ethernet.jsp#overview and set it up for my meter. There is an app for both iPhone and Android that provides energy monitoring via the Digi Energy Day Trader website. If your house has a newer smart meter, there is a different Zigbee gateway that works for those meters. I just got mine set up last week, and it is working great.

There are many little ARM boards, some of which are priced as low as $39 in quantity 1. These are useful for applications where the ATMega in an Arduno is too limiting.

The choice of peripherals these guys made is unusual. With a USB port and an HDMI port, you can build a game machine, which is probably what they had in mind. Most such boards are more suited to embedded applications, and have I/O - digital TTL ports, Ethernet, LCD drive, etc.

A problem with these minimal machines is deciding what to put on them. The lowest-price devices tend to have too little of some resource and too much of something you don't need. This leads to a proliferation of little embedded boards with slightly different options, which runs the cost back up.

For hobbyists, the Leaflands Maple may be interesting. It's an ARM board in the Arduno form factor. It's compatible with Arduno daughter boards ("shields"), and has some commonality with the Arduno development environment. Not enough memory to run Linux, though.

The $25 price is a vaporware price - they're not actually shipping. NXP is shipping LPCExpresso for "under $30", and that includes the entire tool chain (Eclipse, GCC, JTAG debugger, etc.)

They sell to anyone, or you can buy their modules from Digikey.com, etc.

The cool thing about the modules I use is that they're intended to be rs-232 modem replacements. As such, they act as a transparent comm channel by default, but can also be put into an API mode where you can send explicit commands and data to any node(s) simultaneously.

The other cool thing is that they have on-board A/D converters and digital I/O lines that can be sampled explicitly _or_ the whole module can sleep for a specified period, wake up, check for data, take some samples, send them, and go back to sleep. This is excellent for remote battery-powered endpoints.

Take a look at the product manual for their XBee ZB OEM modules. If you're interested in playing around w/ ZigBee, their starter kits can't be beat - they include several different modules, prototype boards, ac adapters, cables, etc. Everything you'll need to get started.

Be careful of their product names - they have a ton of products with very similar names. For ZigBee you want XBee ZB or XBee-Pro ZB, NOT XBee Znet 2.5 or digimesh or 802.15.14 or anything else. The -Pro means high power, btw. 100mW vs 1mW.

The Znet 2.5 modules are the same hardware as the ZB ones, the only difference is a firmware flash.

Support page for XBee OEM modules

Their store

One of their starter kits

starter kit at digikey.com - only $129.

They sell to anyone, or you can buy their modules from Digikey.com, etc.

The cool thing about the modules I use is that they're intended to be rs-232 modem replacements. As such, they act as a transparent comm channel by default, but can also be put into an API mode where you can send explicit commands and data to any node(s) simultaneously.

The other cool thing is that they have on-board A/D converters and digital I/O lines that can be sampled explicitly _or_ the whole module can sleep for a specified period, wake up, check for data, take some samples, send them, and go back to sleep. This is excellent for remote battery-powered endpoints.

Take a look at the product manual for their XBee ZB OEM modules. If you're interested in playing around w/ ZigBee, their starter kits can't be beat - they include several different modules, prototype boards, ac adapters, cables, etc. Everything you'll need to get started.

Be careful of their product names - they have a ton of products with very similar names. For ZigBee you want XBee ZB or XBee-Pro ZB, NOT XBee Znet 2.5 or digimesh or 802.15.14 or anything else. The -Pro means high power, btw. 100mW vs 1mW.

The Znet 2.5 modules are the same hardware as the ZB ones, the only difference is a firmware flash.

Support page for XBee OEM modules

Their store

One of their starter kits

starter kit at digikey.com - only $129.

They sell to anyone, or you can buy their modules from Digikey.com, etc.

The cool thing about the modules I use is that they're intended to be rs-232 modem replacements. As such, they act as a transparent comm channel by default, but can also be put into an API mode where you can send explicit commands and data to any node(s) simultaneously.

The other cool thing is that they have on-board A/D converters and digital I/O lines that can be sampled explicitly _or_ the whole module can sleep for a specified period, wake up, check for data, take some samples, send them, and go back to sleep. This is excellent for remote battery-powered endpoints.

Take a look at the product manual for their XBee ZB OEM modules. If you're interested in playing around w/ ZigBee, their starter kits can't be beat - they include several different modules, prototype boards, ac adapters, cables, etc. Everything you'll need to get started.

Be careful of their product names - they have a ton of products with very similar names. For ZigBee you want XBee ZB or XBee-Pro ZB, NOT XBee Znet 2.5 or digimesh or 802.15.14 or anything else. The -Pro means high power, btw. 100mW vs 1mW.

The Znet 2.5 modules are the same hardware as the ZB ones, the only difference is a firmware flash.

Support page for XBee OEM modules

Their store

One of their starter kits

starter kit at digikey.com - only $129.

They sell to anyone, or you can buy their modules from Digikey.com, etc.

The cool thing about the modules I use is that they're intended to be rs-232 modem replacements. As such, they act as a transparent comm channel by default, but can also be put into an API mode where you can send explicit commands and data to any node(s) simultaneously.

The other cool thing is that they have on-board A/D converters and digital I/O lines that can be sampled explicitly _or_ the whole module can sleep for a specified period, wake up, check for data, take some samples, send them, and go back to sleep. This is excellent for remote battery-powered endpoints.

Take a look at the product manual for their XBee ZB OEM modules. If you're interested in playing around w/ ZigBee, their starter kits can't be beat - they include several different modules, prototype boards, ac adapters, cables, etc. Everything you'll need to get started.

Be careful of their product names - they have a ton of products with very similar names. For ZigBee you want XBee ZB or XBee-Pro ZB, NOT XBee Znet 2.5 or digimesh or 802.15.14 or anything else. The -Pro means high power, btw. 100mW vs 1mW.

The Znet 2.5 modules are the same hardware as the ZB ones, the only difference is a firmware flash.

Support page for XBee OEM modules

Their store

One of their starter kits

starter kit at digikey.com - only $129.

ZigBee, which, as the article states is the key to this system, is a protocol that runs over a wireless mesh network. I use XNet ZB modules for my tinkering.

i'm looking at smart meters that use zigbee modules from digi:
http://www.digi.com/products/wireless/zigbee-mesh/xbee-zb-module.jsp

I won't be dealing with the hardware, just talking to the devices on the zigbee network so hopefully the whole thing is invisible to me.

I'm going to have range issues though

A nice follow-on to Wireless Network Modded To See Through Walls , it seems like pairing ZigBee with some cheap GPS chips (say, SiRF Star III) would pretty much do the job. Maybe you could put three of them in there for failover to satisfy reliability requirements, the whole thing would still come in under two hundred bucks for a prototype. :)

I actually worked on a small project that involved deploying headless desktop-based Debian servers to locations all over the continental US.

We'd fed-ex the boxes, and most of the time, they'd show up with a hard drive detached, or worse (one fedex ground shipment from CA to FL knocked the RAM right out of the slots on one box). What was worse, we didn't have any technical expertise on site to speak of, so even simple problems were hard to diagnose.

The ASUS motherboards we were using happened to have serial ports, and the BIOS also happened to natively support pumping text-mode input/output over said serial interface -- so you could edit BIOS settings, tweak bootloader settings, put the machine in single user mode and fsck the whole disk, etc. etc. all over serial.

We experimented with plugging these things into serial-over-ip devices; specifically, one like this one -- although I think we paid about $60 each. Results were mixed. For one, it was pretty painful getting things operating at a reasonable serial bitrate (especially for curses-esque interfaces like the BIOS settings interface -- characters were getting lost), and making them reliably accessible over IP wasn't easy either. You could configure these things to "phone home" when they were powered on, but the configuration interface and documentation was pretty bad.

If I recall correctly, KVM over IP devices were a bit more pricey.

So, long story short, when it comes to low-cost remote server management, in my experience, there's something of a lack of quality offerings.

DIGI has been selling these devices for years. Others did before them;

http://www.digi.com/products/cellulargateways/

When you have a remote office out in the middle of no where and you need Internet, sometimes this is the only option. This or VSAT/DirecWay.

Actually ZigBee was designed for interoperability. Check out ZigBee Application Layers:: http://ftp1.digi.com/support/documentation/90000976_C.pdf I think part of the problem with automated light switches is they're high liability and low volume. It burned down your house? oops sorry.

I have to agree with you about ZigBee. If you're into tinkering with electronics, you can get XBees for about $20 a pop http://www.digi.com/products/wireless/zigbee-mesh/xbee-zb-module.jsp I have used them for a few projects and found them to be very reliable, well documented and having a good support community. There's also a fair amount of open source software out there for interfacing with them. However if you want a commercial solution, it's Control4 and you're going to pay out the wazoo. It's too bad someone can't step up with a low cost alternative. The RF chip itself is under $10 in volume.

Again, check these out. They're cheap and very functional.

If it's not too late, I'd consider switching to these. No need to pull wires, they self-organize into a mesh network, and have a simple packet-based API. They also have a few digital and analog inputs onboard, so in may cases you don't need any additional circuitry. They can also be configured to awake from sleep every 'n' seconds, take some measurements and report them, and then go back to sleep. Perfect for what I want to do.

ZigBee supports various levels of security that can be configured depending on the needs of the application. Security provisions include:
* 128-bit AES encryption
* Two security keys that can be preconfigured or obtained during joining
* Support for a trust center
* Provisions to ensure message integrity, confidentiality, and authentication.

ZigBee supports various levels of security that can be configured depending on the needs of the application. Security provisions include:
* 128-bit AES encryption
* Two security keys that can be preconfigured or obtained during joining
* Support for a trust center
* Provisions to ensure message integrity, confidentiality, and authentication.

Not sure about the software, but in the video I clearly saw they're using XBee modules for RF comm on some of their projects... peer-2-peer, mesh networking, yadda-yadda.

http://www.digi.com/products/wireless/point-multipoint/xbee-series1-module.jsp

You can get cool OEM boards here, and a cool dev kit with 5 different modules here.

You can get cool OEM boards here, and a cool dev kit with 5 different modules here.

Your environment sounds like a single machine is not your entire problem. You may want to look at some of the console servers that are available. They are designed for what you are describing (I have used Digi before):

• Digi
• NTI
• ThinkLogical
• Raritan

My answer: ZigBee! They're mesh topology, so you don't have to have line-of-sight to the coordinator. They have interoperability as part of the ZigBee spec by using defined profiles.

These specific devices are essentially rs-232 devices with some A/D and digital I/O lines. The end device incarnation can sleep most of the time and awake to take samples. You can run a 'sleepy' endpoint on batteries for 1 yr+. With a decent antenna, you can get pretty far on 1-2mW, and if you've got power available, they sell 100mW versions. I'm setting up monitoring using these beasts, and hope to use the same protocol for my home entertainment system.

Wouldn't one of these be even more of a worry? This thing is barely bigger than RJ-45 connector. And it even runs linux.

I'm currently using these critters in that very application: http://www.digi.com/products/wireless/zigbee-mesh/xbee-series2-module.jsp A ZigBee module that can wake up, take a few sensor readings, transmit them to the 'mother ship', and go back to sleep. To me that's the cat's meow in distributed sensor technology - a fully meshed sensor array that can react to external interrupts and transmit data wirelessly and run on batteries for years. Their dev kit contains 5 modules, RS232 and USB dev boards, antennas, etc. Very complete and very sexy indeed.

Could this have been an eight-port serial board? check out: http://www.digi.com/products/serialcards/pcx.jsp

So far, all of the USB-over-Ethernet devices I've found support only bulk and interrupt transfers, not isochronous transfers. That means no USB audio devices, and no cameras. (Some of them mention "digital camera" support, by which they mean "mount the card as mass storage or use PTP", not live viewfinder mode.) Several vendors specifically point out "in this firmware release", leading me to suspect that they plan to add isochronous support in the future. Except these products have been out for years and no such update has happened.

That's a shame, because a pile of cheap webcams at the end of such a device would be ideal for home security, machine vision on a tethered robot, and lots of other applications. Being able to pass USB data over a wifi link (using the integrated wifi in a laptop) would mean one less jack to snap off the motherboard. Looks like it's just not meant to be.

I should note: The USB-over-Cat-5 (or USB-over-fiber) devices DO support all types of transfers, because they're working at layer 1, just boosting the USB signal so it'll survive a few extra feet of UTP wiring. It's only the higher level devices, which wrap USB frames in Ethernet frames or IP packets, that can't support the stringent timing requirements of isochronous transfers.

If a universal power system were widely adopted, all of these bricks could go away. The device makers would have every incentive to not include a brick with each device (cost, weight, package size, etc.)

The almighty buck is an economic incentive only for the brickmakers -- they want to sell lots of power bricks. But they only sell wholesale to the device makers. They don't sell to the consumers, the device makers do. The device makers have it in their economic interest to offer the lowest price, not to sell a brick. If they could save two dollars by not buying bricks, they could drop their prices by one dollar and still pocket one dollar for themselves.

There is already a standard out there: USB PlusPower for cash registers. They've incorporated USB backward compatible piggybacked high-current +5VDC, +12VDC and +24VDC connectors. Several years ago some large retail chain stores refused to accept a half-dozen power bricks under each cash register, and demanded of IBM that they develop a way to power the many peripherals each cash register needs (scanners, printers, mag stripe readers, PIN pads, cash drawers, scales, etc.) NCR and Fujitsu added their support for a standard, and USB PlusPower was the result. All the large-player peripheral makers support it now, too. (Here's a sales document for a USB PlusPower hub for your PC that explains the standard.

From the document: "The USB PlusPower design provides the following voltage and current

• +5 volts DC at up to 6 amps per connector (up to 30 Watts)
• +12 volts DC at up to 6 amps per connector (up to 72 Watts)
• +24 volts DC at up to 6 amps per connector (up to 144 Watts)"

Consumers need to do the same thing, but as of yet have never organized and demanded such a thing. It's considerably tougher to do at a consumer level. Consumers have never organized very well. And there are very few cash register manufacturers in comparison to all the motherboard and system builders out there. There are very few "large customers" that can use their buying power to influence the industry.

Part number info from Digi Connect

Use the numbers to search on Nu Horizons.

This one's not an RJ-45 jack form factor...but it has USB, LCD and uses a faster ARM9 core. Pretty neat!

http://www.digi.com/products/embeddeddeviceservers /connectcore9c.jsp

-AP

Digi already makes a wireless version too:

http://www.digi.com/products/embeddeddeviceservers /digiconnectwime.jsp

A common application for this sort of device is that you can just plug it into an existing device that doesn't have ethernet or wireless ethernet and voila! Ethernet connected device!

For example, say your company makes heart monitors with an RS-232 interface or some other serial or GPIO controllable bus. You can just sit this device in your design and instantly have an Ethernet-enabled heart monitor running with a command line or a web-interface, etc. It's a pretty cool way to upgrade old hardware designs cheaply.

-AP

This is a Digi Connect ME, which has been around for a while. I have one, and it runs uClinux nicely.

Dunno what Kleinhenz is shipping, but I'm gussing it's just the DCME with uClinux flashed onto it. Nothing new here.

IIRC, old newsgroup threads when these came out suggest the quantity cost is ~$50/ea, so this product's convenience comes at a bit of a premium.

Field of view would be easy to conquer if you provide several simultaneous streams. That's the approach I'm investigating with stereo pairs of cameras for 3d view. The problem is that most camera makers don't follow the USB spec, so you can't plug in more than one of the same model at a time.

The exception to the rule is the Vista Imaging VICAM, which was originally made in a parallel port version, then bought by 3Com and released in a USB version as the 3Com HomeConnect Camera. It was then bought by Digi/IONetworks and released as the Watchport/V. In all of its USB incarnations, you can run several Vicams simultaneously.

According to the informative reviews I've read, Vista did a lot of things right with this camera. Despite being an old design, the image quality is still among the best you'll find. You can get accessory lens packs, to change the field of view.

Other cameras that support multiple instances are the Logitech QuickCam 4000 and the ADStech "Turbo USB 2.0 webcam". There may be more.

Now, if only I'd thought to do an Ask /. two weeks ago, someone could've done all this research for me. ;)

Field of view would be easy to conquer if you provide several simultaneous streams. That's the approach I'm investigating with stereo pairs of cameras for 3d view. The problem is that most camera makers don't follow the USB spec, so you can't plug in more than one of the same model at a time.

The exception to the rule is the Vista Imaging VICAM, which was originally made in a parallel port version, then bought by 3Com and released in a USB version as the 3Com HomeConnect Camera. It was then bought by Digi/IONetworks and released as the Watchport/V. In all of its USB incarnations, you can run several Vicams simultaneously.

According to the informative reviews I've read, Vista did a lot of things right with this camera. Despite being an old design, the image quality is still among the best you'll find. You can get accessory lens packs, to change the field of view.

Other cameras that support multiple instances are the Logitech QuickCam 4000 and the ADStech "Turbo USB 2.0 webcam". There may be more.

Now, if only I'd thought to do an Ask /. two weeks ago, someone could've done all this research for me. ;)

There also needs to be a way to load bootstrap code remotely. For instance, having a TCP/IP enabled BIOS be able to run TFTP or some other protocol to load a netboot floppy image. Then you could give it a LILO command instructing it where to find a boot image, preferably one on a server in the same hosting center.

Sun hardware has had serial consoles that can boot from the network for years. The syntax for the current OBP (OpenBoot PROM) revisions is here: http://docs.sun.com/db/doc/817-2701/6mibjioqr?a=vi ew

Coupled with with a terminal server/power management module such as this you'll get all those features.

For x86 hardware, some vendors are shipping with serial console capabilities which include network booting, such as Dell's DRAC

Remote floppy boot. DRAC offers remote media access, allowing the server to boot from remote media. DRAC II uses floppy redirection. Administrators can insert a bootable DOS diskette into the diskette drive of the desktop machine and boot a remote server to that floppy. Administrators can then run operations from the floppy, including functions such as flash BIOS to recover servers with BIOS problems.

DRAC III uses Trivial File Transfer Protocol (TFTP) to transfer an image to the card and lets administrators enhance remote floppy performance by downloading floppy images to the memory on the card (see Figure 5 ). Functions on the "diskette" are executed in a DOS environment for 32-bit systems.

So apparently someone with a Lantronix bias, or an anti-Digi bias, has moderated the reply mentioning the Digi Connect ME out of sight, which is a direct competitor to the Lantronix XPort.

Now why would anyone do that? Maybe it's time for me to finally check out this meta-moderation thingy... yeah, I'm a lazy lurker.

Digi has an embedded IP device called Connect ME.
You can get it with GNU tools which is nice if you are used to gcc and gdb.

All consumer electronic equipment has to pass FCC regulations Part 15 Class B (I couldn't find this on fcc.gov, go figure.) As the page at the link says, it's not a guarantee that interference won't occur. However, I would think that having to meet said regulations makes it less likely that the non-metallic cases will make a significant difference. In fact, a lot of the modern cases are made more of plastic than they are metal. Not necessarily transparent plastic, but plastic all the same. I havn't heard of any problems.

The CRTC has similar regulations which I couldn't find, but I'm sure they're very similar to the FCC's. If the device passes the FCC, chances are it passes the CRTC.

Anyway, I often used to run computers with an open case, or no case at all, and never had a problem with EMI. I'm sure there are others here who could confirm this.

"Ya don't run GUI's on data center equipment :)"

Oh, how I wish that were true >X-(