Hardware On Rails
With the Open Source world riddled with opportunity for building software solutions for all manner of problems, Jono Bacon and Stuart ‘Aq’ Langridge explore whether there is a market for hobbyist plug and play hardware projects that don’t require you to string together capacitors and resistors but can do genuinely cool things such as the BaconSinkVision.
Remember, we are the very start of the conversation! What do you think? Do you have similar needs to Jono? Do you think it would be fun or useful to be able to build these kinds of home-brew projects? Do you think a market exists for this kind of need? Do you know of a company who is already doing this? Do you want to build Jono the BaconSinkVision because he isn’t clever enough? Share your thoughts in the comments below…
49 Comments to “Hardware On Rails”
Leave a Reply
sorin7486 (83)
Derek (72)
marxjohnson (61)
Flamekebab (50)
b1ackcr0w (49)
Hessiess (42)
James Duncan (42)
mg (42)
tola (40)
Shane Fagan (39)
Jono,
Thought you mentioned in an earlier Shot that you weren’t plugging Ubuntu, yet the brand placement appeared a few times today.
Not brand placement, just explaining what I would personally like to run: there is a difference.
Much like you work with a sysadmin or whoever, you should just get a harware guy to help or do the hardware end and you do the software.
Win-win.
Forgot to mention, though
I enjoy the show… keep it up!
Thanks so much, WJG!
Aq, never say the show’s name that way ever again. Ever. Jk, actually it was kinda funny.
Good shot. The possibilities with hardware are as infinite as they are with software. It would be great to see as many open hardware projects as there are open software projects.
btw, check out http://www.buglabs.net/
I’d love to see this sort of component based high level hardware (I also get bored while doing the washing-up). A few years ago I was looking at buying a high-end stereo for my car, it acurred to me that a custom touch screen device would be much cooler, unfortunately I didn’t have the hardware knowledge to build such a thing.
I’d like to see better interoperability between hardware too, for example it’s not very easy to hook my iPhone up to a tv. Connection of smartphones to a larger screen or other peripheral devices such as keyboards should be much simplier than it is as they are effectively handheld computers.
I have a potential startup idea for a screen capture device for software debugging, again I don’t really know where to get started on the hardware side.
Pete
ps. If your all about online communities why don’t you have the facebook connect plugin for the comments?
I second the Facebook Connect request.
We are aware of how cool this could make things and have discussed it. It may well show up….
Having a common interface for all mobile phones would be a good thing, however this is currently dominated by proprietary connections. The reason for this is so the providers can create vendor lock in, and force you to buy more products from them, The same problem use to be a major problem in software, but is less so now, thanks largely to reverse engineering efforts. Vendor lick-in in hardware should also work itself out over time.
Wow, odd that the subject of hardware hacking should arise today. Just yesterday I happened across the coolest, cheap, easy and massively useful hardware hack EVER! (Jono, you might be doubly interested in this, aparently it works for audio rack gear as well) I posted it to the superb Linux Outlaws Forum
http://linuxoutlaws.com/forums/viewtopic.php?f=3&t=2629&sid=1a79ec7ff64558468f21fb0609909a09
It struck me that you were describing the difference between Lego and Meccano (Erektor in the USA). HW hacking is like Meccano and SW is like Lego. HW is nuts and bolts, whereas SW is building blocks.
it sounded altogether like JonO wants hardware lego bricks. Anyone can easily build something.
The step above the arduino in terms of something that will run a ‘normal’ OS is the beagle board http://beagleboard.org/, it’s essentially just a low powered small computer and to take you example you could just tag on a usb touch screen and get it to run youtube etc…
Arduino is really, as Aq put it, something more simple, there are OSs for arduino (eg http://www.skewworks.com/pyxis/) but it’s never going to run youtube. It makes programming electronics much simpler with its shield style interfaces but there is going to come a point where you come up with something that’s not a shield.
Even beagleboard doesn’t seem to do what I want though: I don’ see a set of common input and output components, unless I am not seeing them?
It has USB ports on there, man. What else need ye?
There is a market for hardware hobbyists, but it’s a bit expensive. Check the Car PC folks, for example, they build very cool stuff. You can buy small touchscreens, embbeded components and things like that.
I forgot to mention, the problem with this kind of thing is the price… things are a bit expensive.
The expense is really a put off. Especially for something that really may not fill an actual need. I can barely find actual neato-hacko accessories for my netbook.
Where I could really see something like this taking off is in the mobile arena. The devices are already sort of limited, I’m not particularly fond of touchscreens for extended use, and the potential here could be pretty amazing.
I wonder if the price could come down low enough? What do you think – with the netbook craze happening, I wonder if the components well get low priced enough.
http://gumstix.com/
Yeah. Nevermind that. I was listening, not reading. But the market is certainly there.
I really don’t think there are many niches that are unfilled in terms of hardware. Think of anything you might want to do; a product with all the necessary components probably exists.
Your real problem is that the software tends to be unhackable. If a hardware vendor were to release a handful of beige box hardware form factors (a router, a media player, an audio player/recorder…) with a few pluggable components (WLAN card), they’d soak up most of the desire for building your own gizmo.
My experience with DIY electronics has been strictly analog. I made a battery box for a pair of electret mics, wired so that (unlike the vendor-supplied battery box) it would drain the battery only when the mics were plugged in. That was satisfying.
I might play with op-amps and such in the future, but I really don’t see the attraction in making your own Linux-running computery device. It’s either the same as building a desktop computer (I do that, but it’s not an interesting experience), or it’s ridiculously difficult, requiring mad soldering skillz that should really be handled by a factory robot.
In days of yore, computers had lots and lots of peripherals hanging off of them. This was simply because the cost of the core system (CPU/storage) was 1000’s of times the cost of the terminal or printer you were connecting to them.
These days the peripherals are sometimes more expensive than the core. A good-enough CPU can be had for around $100, about double the cost of memory, but maybe half or 1/4 of the screen cost. So the equation is different.
However needs and capabilities have changed.
Firstly I have lots of personal peripherals that I might want to hook together dynamically. I want to sync my phone to my PC to my stereo to my TV to my car for instance. I am also not wanting to be locked into a brand or generation if I can help it. From an environmental point of view I would like to be able to reuse/recycle components and avoid disposing of them.
However making things modular and fit-together can be difficult. It could cost 10 – 100 times more to make something pluggable. And there is a disincentive as well if I, as a manufacturer, make things modular, and a competitor might undercut my parts.
Ideally I would like a notebook with a nice solid frame that I could swap in and out CPU or storage or screens as and if they are needed.
I am playing a bit with Arduino. It allows a lot of modularity, though requires me to have a pretty good knowledge of electronics. So it isn’t quite a Lego brick yet.
Not sure if I have come to much of a conclusion here, but I certainly like exploring the possibility of open-ness and flexibility as it applies to hardware.
Although not modular, various bits and pieces do show up here (http://www.linuxfordevices.com/) pretty regularly.
That said, the previously mentioned BugLabs is probably the closest thing around at the moment.
Ok, let’s face it, part of the problem here is you want to run on it. The small electronics world runs embedded systems which are a rather different animal. I have a pogo plug which I can put Ubuntu on, but I have to install that using a serial console, flash some settings, and install off a boot server.
Better for me would be guides on recycling some older hardware. I have a Palm TX and Nokia N800 both of which have wireless and bluetooth and touch screens. The challenge for me is assembling the software to make them act the way I would like them to.
I don’t have any experience of Auduinos or hardware hacking as yet, but I’m not sure I really understand what Jono’s vision is here.
Like Aq said, it’s possible to have a waterproof touch screen. Just get something like a touchbook (http://alwaysinnovating.com/touchbook/) and put it in something waterproof.
If you want something more basic to start from, maybe something like a Sheevaplug (http://en.wikipedia.org/wiki/SheevaPlug) would be a good starting point? Its got USB (the most common modular I/O interface you’re likely to find), ethernet and an SD slot, it runs ubuntu, and it’s small enough to put in some sort of case. It’s also (relatively) inexpensive so makes a good arduino-style “lump” to start from when you need something more towards the microprocessor than the microcontroller.
If that doesn’t provide the kind of thing you’re after, I’d be interested to hear some more examples of the kind of things you’d like to be able to build where your need hasn’t been anticipated by a company.
The Beagleboard would be a better starting point for the device described in the shot, its smaller and has a HDMI/DVI output. USB simply does not have enough bandwidth to drive a LCD display at any usable resolution. The USB monitors which exist all use very heavy, lossy compression which means that they heavely degrade the image quality before displaying it.
Hey, Antonio here, co-founder of fizzPOP. Someone’s already mentioned it, but there is the stuff from Bug Labs that is very modular.
Otherwise, I’m not really sure what your vision is. You can already by components from any hardware shop and make them into anything. Sure, it’s not plug-and-play but that’s because they’re designed to be as open as possible. For example, if companies only made components for Arduino’s, they’d be locking out anyone who doesn’t use one, including the big boys like Sony and Apple etc.
Oh another note, Mitch Altman is coming to Birmingham, UK and other UK hackerspaces this March, so you could meet him and talk about this yourself if you wanted to
The thing with software frameworks like rails is that if you do something which the software cannot do, you can dig into the source code and change it. With hardware that simply isn’t possible because you would have to reprint all or some of the integrated circuits. Such modular hardware would result in an extremely inflexible system, such as exists within desktop computers.
This was a really interesting shot, I would love if there were a way to just put hardware together a bit like the way you might put together our own PC for example… I am a designer and I get ideas all the time that I have no idea of how to implement and try them out which makes me really jealous of people like Steve Jobs who can do that because they have all the people and resource to do just that… What I think need to be done to make this a reality of it is even possible as they said in the shot since I am a noob to is for hardware people to start make some open source hardware to begin whit and then for those people to start up a webpage were they could sell theses hardware components. That way they would make some money and the components could get better and better over time…
Agreed. However, hardware carries a setup cost, unlike software. It costs a reasonable amount of money to set up a company or organisation that can create and sell this open-source hardware, and there’s no guarantee that you’ll sell enough to make that money back. Which is why no-one’s done it…
Does it need to be done? I think it’s a case of a solution looking for a problem.
The components to build just about anything that you want are out there, with instructions on how to use them.
There is of course issues around patents that would stop you selling your creations, but that’s a whole other problem!
ah, the instructions are…hard to get in to. Hardware hacking is like programming; once you know what you’re doing, there’s a world of possibility, but it’s really hard to get over the initial hurdle. I’m not sure how that’s best fixed…what would you suggest?
Therein lies the problem with learning a new skill: it’s hard to get into. Learning French, learning to drive, learning how to cook or even learning how to walk are all hard things to learn at first. Even more visual ways of programming like Pure:Data, nodebox and Scratch still have their learning curves.
The solution in each of those cases was to either take care of the a lot of the simple tasks for you (which just about every programming language does) or lots of education.
As more and more people start to use Arduino’s and custom hardware in their work hopefully people will see that it’s just not as scary as it once was. Aside from learning at university/college I’m sure there’s lots of people who have learnt things from the Internet and things like Jono’s recent Lernid python tutorial. More plz!
However, one overriding thing in all of this is one thing: dedication. If you’re not dedicated to learning something you simply wont. If there’s a hacking community near you just ask someone to teach you the basics and I’m sure within 30 mins you’ll be on your way…
Just fitted an old Maemo touchpad device into the kitchen for a very similar purpose to the original idea in this shot. Modular hardware would be of great interest as I often come up with projects to build and end up searching around for a compromise platform to modify.
It appears that IDG listens to shot of jaq as well: http://news.idg.no/cw/art.cfm?id=25B8CB70-1A64-67EA-E4D482E1CEBA3792
Well, of course everyone does. This might be independent invention, mind
Why have a Youtube-playing touchscreen thing anyway? Why not just listen to podcasts? Or are they all dull and boring?
Gerv
All the good ones are really short.
Arduino is great, but you aren’t going to be running Ubuntu on it. The BeagleBoard, however, will run Ubuntu and I believe the Canonical folks are working on adding native support for the BeagleBoard-xM (has the extra memory to do native builds and >1GHz clock speed).
With the Bug 2.0, they gain compatibility with the BeagleBoard by using the same core chipset–and they are doing exactly what you are talking about. Doing more to enable people to build more plug-in modules for the Bug 2.0 seems to me to be the way to cover a lot more use cases. If you do want to do Arduino stuff, the Bug 2.0 has a Bugduino plug-in module–so you can have Ubuntu and Arduino in a nice tight plastic case with a battery.
You might be interested in the write-up I did on building an open source hardware community based on your self-analysis with the LugRadio guys. See: http://www.osbr.ca/ojs/index.php/osbr/article/view/1050/1009
BTW, the Always Innovating Touch Book is also based on the BeagleBoard as are the Gumstix Overo modules (very cool, much smaller than the BeagleBoard). You can also get one of another more than a dozen OMAP3-based modules that will run the higher level bits (some may require additional low level bits to boot into Ubuntu). See http://wiki.omap.com/index.php?title=OMAP3_Boards.
The thousands of hackers on the BeagleBoard means that there are plenty of good starting points for building your next open hardware project off typical desktop software without the great power sucking sounds and space taking of traditional computing hardware.
That’s a really interesting article! And it’s nice to see that our categorisations for why people get involved still live on
Thanks! I would have made a Top Trumps reference, but I think most people wouldn’t have gotten it. I’m still addicted to the Buffy Top Trumps game!
Jono get’s a dishwasher – problem solved.
Modular hardware would be a cool idea. But I can’t see this taking off with regular consumers for obvious reasons. However if some enterprising soul would want to start a firm that offers custom kits or nice easy to use hardware snap-on style modules, I’m sure they’ll find a decent number of hobbyists. Heck I’d buy kit from said firm/person. But I’m already up to my ears with a hobby startup… Anyone else up for the task?
So the answer is right under your nose. (Slack given because it was just before you joined as a host)
http://www.arduino.cc/ (Open source hardware Mate!) and affordable for all, go Massimo! (Love that podcast)
And all the parts your heart desires from: http://www.makershed.com/
Also, many pre-written programs, ready for your compiling pleasure http://arduino.cc/en/Reference/Libraries
FranticFinn
How about apt-get for open source hardware?
http://designfiles.org/dokuwiki/skdb http://gnusha.org/
See also: http://openmanufacturing.org/
I know I’m late to the party but I also concur that I was thinking ‘buglabs’ while I was listening to the shot.
I also have a friend who’s tried to solve this problem from a ‘web integration perspective’. His solution is over at; http://iobridge.com/
It’s more for hooking up the capacitors and servos to the web vs. an embedded system that can run Ubuntu.
And as with the other comments, I think gumstix is an awesome solution except every time I try to go order one I’m overwhelmed with the choices. Which speaks to the comments in the shot about how usable the systems really are right now…