Why Google Choosing Arduino Matters

ptorrone writes "Earlier this week at Google I/O, Google announced the Android Open Accessory kit which uses the open source hardware platform, Arduino. MAKE magazine has an in-depth article about why Google choosing the Arduino matters, why Google picked Arduino and some predictions about what's next for Apple's 'Made for iPod' as well and what Microsoft/Nokia/Skype should do to keep up."

Comments

As someone who tried this...

[errandum]

It already existed (an android-arduino "interface"). It only matters because google is behind it now (with an official API), but whoever wanted to do stuff before already could.

Re:As someone who tried this...

[alostpacket]

AFAIK you need to have the accessory approved by Apple to connect via the docking port and there is a NDA/Licensing agreement that requires you pay Apple a certain amount for each accessory sold. You also have to purchase a chip from them to integrate your hardware. And they require you to submit your financial records/bookeping so that their auditors can be sure you are paying them the fee for every unit of your hardware you sell. This doesnt seem to apply to all accessories, but it does seem to be a real problem for a lot of them. http://www.popularmechanics.com/technology/gadgets/news/4272628 Also, if you dont, Apple will sue you: http://arstechnica.com/apple/news/2011/04/apple-gets-go-ahead-to-move-against-unauthorized-accessor-makers.ars

So, yeah, this could have real impact. Going out and grabbing an Arduino board vs all that draconian stuff is gonna be interesting.

Obviously I have a bias here being an Android app dev, but I believe the two approaches to accessory development are vastly different. And just because Apple has a huge lead out of the gate does not mean they will retain it.

Re:As someone who tried this...

[sonamchauhan]

Personal anecdote: while the docks fit, the ports are not (fully) backwards compatible.

My IPhone 3GS (~2-years old) refuses to charge from an audio base station bought for my previous-generation IPod (~ 4 years old) -- placing it in the dock pops up an error on the phone that says "Charging is not supported for this accessory"

Re:As someone who tried this...

[garyebickford]

If we compare this to the car industry (as it used to be before it got all digified as well), there is/was a big industry making aftermarket parts - everything from brake shoes and taillights to radios. AFAIK nobody ever got sued by Ford for making a Ford-compatible steering wheel. I think the car makers basically felt that the accessory market (i.e. 'bells and whistles') helped their market. They were never particularly thrilled about aftermarket replacement parts, but they didn't stop it. Folks had, and still have, the choice to go to the dealer or go to NAPA - or JC Whitney. And sometimes it's better going to the dealer. Of course, while it's under warranty some things still have to be done by the dealer - but in most states the car makers can not disallow the warranty under if you get your oil change done by someone else.

Of course, that's changing nowadays. Here in MA, Toyota successfully fought off an attempt to pass a state law requiring car makers to release the computer repair codes to third party repair shops, so they could hook up their expensive diagnostic machines and find what was wrong. (I don't recall if this was a legislative thing or a court thing.)

While I agree that Apple may have the right to charge a toll for everyone crossing their bridge, I disagree that it's a good idea. Case in point was the recent article on /. about the demise of independent music because of Apple's 30% rake off the top. Another case in point - I haven't bought an Apple product since 1996, so that's about $30,000 worth of business they haven't gotten. I published software for the NeXT, and had Macs through the early 1990s, but I don't want to be locked into either them or MS. I want the on-ramps to the highway to allow ALL traffic that fits the lanes - I don't want separate ramps for MS, Apple, Google or whatnot.

Re:As someone who tried this...

[Anonymous Coward]

"Businesses who make money selling Apple connectors have to pay money to Apple. It's not onerous, it's business. "

The key point you are ignoring is that under Apple's system, Apple holds *all* the cards. If you invest millions of dollars developing a new innovative accessory and they think it is a threat to them, or if they decide they'd rather sell it under their brand, then they will shut you down and you have no recourse thanks to the agreements you have signed. Yes this is "just business", but that don't mean it is *good* business and Google have just upped the ante fairly significantly. People used to dismiss Linux as an embedded platform all the time with a similar argument to yours and now Linux completely dominates the market, so the idea of open in the device space is already proven.

Re:As someone who tried this...

[alostpacket]

Well, perhaps the draconian characterization was a bit inflamatory, but I don't find your "it's just business" arguments in the least bit compelling either.

Apple's use of a proprietary, non-standard, and patented connector appears from my perspective (admittedly I am not hardware expert) to be a bit of rent seeking. USB and FireWire have been around a long time and Apple has, since the introduction of it's MFi program, attempted to add the restrictions to headphones. Would you think it OK for them to charge a tithe for using the 3.5" heaphone jack? (They don't unless you want to use the MFi logo currently, but you get my point)

Those that can't get their products to sell enough to be worth the cost of entry aren't going to be missed, pretty much be definition.

That's a really bad way to look at the world. First, it's not really true due to the fact that consumers don't always reward the best products. Often they "go with what they know." Further, cost of entry being artificially inflated doesn't help anyone but the company inflating. If you read the first link I posted you'd have come across a good example of this where speaker/dock makers were saying they could have made better speakers if they didn't have to pay so much in fees to Apple. The original MFi fee was 10% of every accessory sold (it's now a flat fee structure per unit sold AFAIK). In a lot of industries 10% is the profit margin, especially in the competitive accessory market. Anyways, here's the relevant quote:

"If we didn't have to pay Apple for the dock and auth chip, we could have made a much better speaker for the same price," said an official at a major electronics maker, who, like several sources for this story, requested anonymity because of fears that speaking with the press could jeopardize his company's relationship with Apple.

Now if Apple's connectors were truly revolutionary here then maybe there's a case to be made. However, I find it far more likely that the patented connection there is novel at best. It's also bad for the system of capitalism and innovation in general to let the big, entrenched players construct artificial walls to entry.

Your game analogy on the other hand is an interesting one. I'm not sure you are applying it correctly though. Surely many more PCs are sold in a given year than xboxes, and that's clearly because they are more general computing devices with standardized connection ports beaing a commonly touted feature. Also, MS, Nintendo and Sony have all dabbled with, or currently have, alrenative ways of working with their products (thinking software/mini games/web broswers/OtherOS etc). None of them are great, but it's there, so they have clearly considered using it as a wedge to drive against competitors.

So if you limit your analogy to games, (an artificial reduction of the analogy), then yes, it sort of provides an example of this. However if you look at the bigger picture then things aren't so clear cut. I do agree that clearly this is done elsewhere in business and much money is made off of it and that it is unlikely to disappear overnight. I just was saying it will be an interesting competition between business models. But it's far from clear who the winner would be. And I don't condone MS/Nintendo/Sony for behaving that way nor car manufacterurs with their attempt at proprietary computer codes of OBDI stuff. Interestinly too, the same guy in the speaker quote says this at the end:

The same official who was concerned with speaker quality explained that "Apple sales are predictable," making the market for third-party products equally simple to measure. Apple works closely with the companies to provide forecasting and stock availability numbers once CEO Steve Jobs has unveiled a product. "It's an expensive relationship," said the company's product manager, "but a profitable one."

So that does show that it works -- clearly there

Re:As someone who tried this...

[LBU.Zorro]

How am I ignoring that? That's the very foundation of my point. It's Apple's system. If you want to play with them, you abide by their rules. It's in Apple's interest that their rules don't drive away third party manufacturers, while simultaneously making sure third party manufacturers don't piss in Apple's pool. That's why Made for iPod exists. It's not a profit center, it's mainly a way to make sure MFI products work well.

You're making an assumption that isn't actually true, or at least isn't completely true. It is in Apple's interests to keep some third party manufacturers, but as with the Apple AppStore itself it isn't in Apple's interest to keep all of them - it certainly isn't in their interest to keep anyone who competes too closely with their own branded and thusly profitable peripherals. You are correct, MFI isn't a profit center for them, however the iPod, iPhone and all iDevices that use the connector and play with the MFI certified peripherials are a profit center (and Apple peripherals are also profitable for them), therefore close control of MFI and connecting to 'their' devices is what they're after.

To an extent I'd agree with you, MFI does tend to make the peripherals work well with the devices (although like anything the older peripherals are useless), but control is the purpose, not just a good working experience.

Wow, that's pretty scary! I suppose you can cite an example of this happening, right? I mean, if not, and the dock connector has been around for the better part of a decade, you'd think maybe it's not something you need to worry about.

I half agree with you here, but not totally. As far as I know the situation is similar to that of the Apple AppStore, in that you're under an NDA and you can't really talk if you want to keep doing business with Apple. If you don't care about doing business, then you can risk breaking the NDA, but as there's a significant difference between hardware and software it's not likely that you'll be a straw-man that's not worth suing.

Based on Apple's tactics with the AppStore, it's more of a case of a track record than no evidence. I wouldn't be surprised to find out that Apple has done it, after-all where's the drawback for them? It's much easier to control small numbers of large players with a lot to lose than large groups of small players with nothing to lose.

Google hasn't "upped" shit. If you want to sell your product to many hundreds of millions of people, you pay a small pittance to Apple. Just because Google's system doesn't require a license is pretty much inconsequential to whether or not one will make products targeting the iPod dock connector. You don't create products in a vacuum. You need customers. Apple's customer base is far more lucrative than Google's.

In your opinion Google hasn't "upped" shit, in other people's opinions it might have - certainly it has interested me, a lot more than any hardware integration with an iDevice would do. The important point to realise with this is how many people might now tinker with Android and Arduino - people were tinkering with Arudino and Android already, and this makes it official (or at least easier for some). Now, I agree, it's unlikely that any one of these will suddenly mass-produce something, after all they're just small tinkerers. However it is very likely based on the number of people who will look into it (assuming there is a large number for a cheap, open and easy to use platform) that a lot of things that are good will come out of it. And this is where Apple may lose out, after all if you create a nifty hardware dongle, release the app into the Android Market and you're good to go - show me the speedy lifecycle of Apple hardware and apps?

Yes some of them will be rubbish, some of them won't work, a lot will be incredibly unique and useful for only a handful - but it'll be more than you'll ever get out of a iDevice unless Apple removes the barrier to entry. Show me the

Re:As someone who tried this...

[alostpacket]

Well you seem to be missing my point, not sure if I wasn't being clear or what. I didn't deny it's profitable, I posted that you were correct on that part of what you were saying, but you seem to be upset about it anyways? Anyways, I'm not here to argue which is better, Android or iOS, that seems a huge waste of your, and my time.

But I am confused how can you really think that Android is no threat to the iPhone when it just surpased iOS in US market share. It hasn't caused the iPhone to fail overnight, but you're crazy if you think Cupertino isn't totally stressing how to move forward in the the next few years. In 2008 no one would have predicted Android taking #1, they would have thought it laughable, myself included. But here we are in a fiercly competative market, and it's great.

Anyways...My two main points are this: 1) Apple's proprietary connector *seems* like fake invention in order to extract license fees. And thus the entire mess they make people go through, (especially sending them your financial records and the per-unit tithe) seems like fruit from the poisonus tree. This is not to say that the ecosystem has no benefit. And of course I don't think they are required to license things for free -- but I don't think it's valuble when companies make up fake products in order to get a license fee or patent. It's not 100% clear to me that this is what is happening here though. I readily admit I'm not a hardware engineer, nor have I reviewed their patent(s) -- but it's hard to imagine technical reasons why USB and Firewire would not have sufficed at performing the job. And you havent really provided any evidence to contradict that. Do they have to use a standard? Of course not. But is it good for consumers when companies rebrand existing technology and throw a shiny new patent on it?

Regardless -- the point is -- it's clearly and obviously a higher barrier to entry. And one that *can* work for certain types of businesses and/or certain markets with dominating players. But it's not the *only* one.

2) On the other hand, Google's open-API approach (in this instance) vastly contrasts the business style of Apple, and could make for very interesting market competition in the years to come. If you really think this will have no effect, see above point about how much the G1 and Android were a total joke compared to the iPhone in 2008.

Rest of the post I will just try to address quickly as this conversation is draggin on at this point.

- The consumer argument is offtopic -- both ecosystems will provide ample accessories.
- You contend there is, and will continue to be, substantially less money to be made in the Android accessory market but have nothing, not even an anecdotal fact or statistic to back that up.
- Most Android devices use microUSB
- Coby is a bad example when discussing Android and iOS etc. I'm not saying go buy a flip phone or tin can and some yarn or some crap. All things *basically* are equal when discussing high end smarphones. They have their pros and cons. But I agree that it's obviously not the only thing to consider.
- With regards to Apple fixing the fee structure, the flat fee itself is more than the cost of some USB cables, so it could easily be a much higher percentage. I didn't say they were fixing it, just that it has changed. You just assumed they made it better.
- Entering into deals with Apple *is* risky, they can kick you out of the garden and clone your product at any time (same is true for anything built upon a layer of someone else's technology) However, using open standards makes this far less likely to happen.
- If the connector has no real tangible value beyond providing a road to a patent, then it benefits no one but Apple. Consumers are paying more for less and/or companies need to shave off the margin.
- The speaker/dock maker's statement was not meaningless because you have to make you product fit within a certain price point to sell. Adding 10% to your cost could mean that

Re:As someone who tried this...

[drinkypoo]

The situation as I understand it, not being a lawyer but having done a fair amount of research on the subject, is that you may freely reproduce the part so long as it is covered by neither copyright nor patent and you don't reproduce any trademarks normally on the product, so you may have to do more than a simple mold and cast job even to just make a simple cast part.

In practice a whole car is covered by a design copyright but not a single fender. You can sell all the bodywork at once but not a complete car wearing all of it. And even that is OK if you buy the bodywork from the actual manufacturer. Buying aftermarket bodywork and selling a car that looked just like another car might land you in court, but I don't know that it's ever been fully played out. For example the GT40 is being made in the original [body] design by multiple manufacturers.

The situation vis-a-vis secret codes is that there are standard and non-standard codes, the standard codes are mandated in the specification but access to the non-standard codes is not. Further, IIRC only the powertrain codes must be implemented so body codes could just be undelivered unless you send a special command. This has led to a whole bunch of OBD-II snooping.

Finally, the automaker (or anyone else) cannot deny you warranty protection for a replacement part unless they can show that the replacement fails to meet specifications. In the specific case of oil the oil is graded so you only need to buy a lube of the proper grade. In any other case they're going to have to provide specifications in court to prove that your part fails them.

"magnetic core memory" extension board

[ciaran_o_riordan]

A friend recently made a "magnetic core memory" extension board for an Arduino:

http://www.corememoryshield.com/report.html

Just an example (with pictures) of what can be done with these things. (Magnetic core memory was the main form of non-volatile memory for computers from the 50s through to the 70s.)

Re:"magnetic core memory" extension board

[betterunixthanunix]

Magnetic core memory was the main form of non-volatile memory for computers from the 50s through to the 70s

It was also the most commonly used form of RAM; I have an old Fortran textbook that says something to the effect of, "Semiconductor memory will probably become popular over the next decade." It is also the reason we still speak of "core dumps."

Re:"magnetic core memory" extension board

[TENTH SHOW JAM]

There is nothing intrinsically special about Arduino in the same way there is nothing intrinsically special about Ubuntu. The thing that makes them special is the communities that build around them helping each other.

This means that getting started with Arduino means I get to do cool stuff with microprocessors sooner. I actually implemented a link for fire panels over Ethernet using Arduino and some basic programming knowledge. I could have used a range of other systems to do the job but I selected Arduino because I could buy some of the "Shields" off the shelf and was able to make the rest using prototype boards.

Time to market 3 weeks. Experience before with Microprocessors 0 weeks.

Re:"magnetic core memory" extension board

[GumphMaster]

AQS-901. Airborne, real-time acoustic processing. For added flavour we patched binaries on-tape with paper tape patch reels.

Yes, but

[Zerth]

Why are they charging nearly $400 for something that you can buy separately for $200?

Re:Yes, but

[Samantha Wright]

Stop the presses! I've got a $1600 version that comes with a 3-picosecond Oracle support contract!

Re:Yes, but

[errandum]

200$? An arduino costs 20-30$ O_o.

Re:Yes, but

[Jason Earl]

Probably because Google doesn't want to get into the Arduino hardware business, but they do want to make getting the proper board as easy as possible. If price is not an issue you can get precisely the right board from directly from Google. If you are on a budget, you can do a bit of research and save yourself some money. Everyone wins.

Re:Ideal for commercial applications?

[icebraining]

Well, you can use the same chip (ATmega2560) for any commercial application, so you can use the Arduino for prototyping and then reuse the code for the final product.

Re:Lame

[Naurgrim]

I see I did not get in before the Arduino haters. Yes, I know, it's a simple board, I understand that it's not a PIC or whatever embedded thing you prefer. I accept that. But it's a nice, easy to use board. It's fun. You can do stuff quickly with it. It's good for quick little things. I'm sorry that us Arduino users don't measure up to your expectations. I'm not going to tell you that you are wrong for your embedded choices. Can I get the same courtesy?

Re:Because it's too weak to "tether"

[maxume]

Don't most Android phones have the bits needed to watch TV over the network connection built right in?

Not *just* Arduino

[brian.swetland]

Keep in mind that the *any* device that supports USB Host mode can be an Accessory. There's a full open source reference implementation for Arduino, but the protocols are documented and open and you can implement it on any hardware you like.

Docs and Specs: http://accessories.android.com/

Google IO Talk: http://www.youtube.com/watch?v=s7szcpXf2rE

Re:Because it's too weak to "tether"

[hitmark]

Depends. It could be that the part will basically be used to repackage data so that a on-phone app can deal with it.

Consider their demonstrated use case of a exercise bike feeding activity data to a phone app that then use that as input for a game.

Re:Ideal for commercial applications?

[Shadow of Eternity]

I'm not Bad Analogy Guy so I'll be a bit more literal: The point of Arduino is precisely that dumbed down programming environment, it brings the concept of basically making something computerised (to a point) to an a MUCH larger group of people than before. Right now there are tons of people out there doing things with these chips, making all kinds of little hacks and projects, that would ordinarily have thought "Hey what if I could do X? Oh, too complicated, what a damn shame" and are instead thinking "An arduino could probably do that".

Now for anyone that really does know coding and how to work chips and whatnot giving them an Arduino and making them use it "normally" is like giving them Duplos, but it's still Strictly Better for everyone to have these kinds of easily accessible solutions around for all the people that DON'T know that kind of thing. Sure a lot of them basically just sit there in easy-mode and never go any deeper but others will learn more in time, and just having it THERE makes the concept that much more ubiquitous.

Re:Ideal for commercial applications?

[hot soldering iron]

You just answered your own question, and you still don't understand?
"it comes with some dumbed down programming environment for people who don't want to use C/assembler."

Bingo!

How many people do you know that were taught assembler in school? I was taught because I was in an industrial electronics program, emphasis on industrial manufacturing and maintenance. I think they quit teaching assembler to CS students in the mid '80s, and quit teaching C soon after, shifting to C++/Java. How many people do you think were programming PCs when you had to flip switches, as compared to just typing it in and hitting enter?

CLUE: Make something convenient to use, and people will use it. Make it necessary, but inconvenient, and people won't. Are you sure you're smart enough to be allowed here?

Re:Ideal for commercial applications?

[aXis100]

I agree 100%. Arduino is like the legos of microcontrollers. I've used many plain Atmel microcontrollers before and there was a steeper learning curve, whereas Arduino has been very quick, cheap and FUN!

There is also the benefit of reasonably standard IO wiring, so that when the community shares projects they are dealing with common hardware layouts.

Re:The end of "Made for Ipod" ?!?

[TheRaven64]

1. This is obviously designed for those who make hardware as a hobby. It's a cool hobby, and every once in a while something that would be useful generally comes out of it, but makers aren't making stuff that the average person wants.

You know what caused the last couple of tech bubbles? The fact that it's really easy to write some software as a hobby, then commercialise it later. Google wants to do the same sort of thing with Android accessories. If 1,000 people make an accessory as a hobby, then one of them may have a really great design. They can then probably get funding to turn it into a shipping product (I imagine that Google has primed some VC funds to do exactly that). Going from working prototype to shipping product is largely a question of having sufficient money to invest.

Perfect for open hardware projects like RepRap?

[Paul Fernhout]

Just posted here: http://groups.google.com/group/openmanufacturing/browse_thread/thread/8d32987e3767c868#

So, is this going to make an Android phone an important part of a lot of open source hardware projects (including RepRap perhaps)?

Note also:
    http://faircompanies.com/diy/view/make-your-own-open-source-android-smartphone/
"Flow DIY is an open source hardware platform so anyone can make a smartphone with the Android operating system and the exact capabilities one is looking for. Its components as well as the final creation by the user are open source, a first step toward the generalization of DIY devices. Interest is growing in personalizing not only software and web applications, but in everyday devices. A legion of DIYers are demanding tools to create increasingly more sophisticated devices. ..."

As I've said elsewhere, with the turnover rate of Smartphones, in two or three years, today's generation of smartphones will be free-as-in-discarded. :-) So, it can make sense to build stuff for them, especially since if they are free-as-in-discarded-beer then they can be free for kids to use for educational things (like instead of the OLPC XO-1). Reference:
    http://listcultures.org/pipermail/p2presearch_listcultures.org/2009-November/006250.html

That's one reason I started working on Android software (and under a three-years-and-its-free-under-the-GPL model that I am still conflicted
about).
    http://www.artificialscarcity.com/

Still, sadly my Google Developer Smartphone died several after I got it and I never got around to sending it in for replacement, so I guess there is an amount of old phones that will not be usable for similar reasons (but I doubt that will be the majority). Also, as people have pointed out, the Smartphone batteries tend to go, making them less useful as they age (although I guess you could hack in some alternative power if you were motivated).

Still, I'd suggest that if one is making an open manufacturing project that requires computing, integrating an Android Smartphone might be an interesting idea.