I see trouble. Lots of stuff to consider.

Wireless Design and Protocol Discussion

Moderators: seaton, strogg

I see trouble. Lots of stuff to consider.

Postby BertIvon » Thu Sep 27, 2007 12:20 am

I’ve been working on this post for the past couple days - lots of info but it needs to be considered to move forward.

I saw the link here on Strobist and I’ve been watching this idea since it sprung up. I’m pretty excited about it and good to see so many people on board, but I’ve got some concerns that haven’t really been considered or have been ignored or given incorrect information to. I see us hitting some serious snags with all of this - we need to deal with some stuff before charging forward on this - specifically the FCC and radio licensing and the issue of other people’s patents.

I’m an EE and did some consult work a couple years ago for a company building wireless sensors for home security systems, so I’m pretty familiar with that stuff - specifically the legal side of it. I don’t have much time to assist these days so I’m going to dump what I know here - hopefully I can comment here and there in the future. Right now I evaluate and write consulting reports for companies considering new technologies so this is what I do all day.

First, nobody has really taken a serious look at the legal side and cost of licensing this thing. I’m reading a lot of false information in the posts here so far and it’s scary.

A big caution to the idea of what “Pre Certified” radio products are all about. This does NOT mean you can just build with them. It doesn’t matter what the product or what you build it out of, if you assemble one, and it emits a radio signal of any kind, and offer it in exchange for any money, it MUST be FCC certified in the U.S., as well as the same type of cert for all the other countries you’ll send it to. It’s the same with Zigbee, Blutooth, or the standard FSK and OOK type stuff - if it makes a radio signal, the finished packaged unit MUST be certified before sale, and that’s the same for all countries that I’m aware of.

The “pre cert” stuff basically means if you follow the manufacturer’s instructions for building with the product, it will generally pass the testing process (compared to building a radio transmitter from your own components). It’s just a way of saying “It’ll be easy to get it certified”. Lots of people think this means there is no cert required - big mistake and it’s going to get someone into big trouble.

Know that even the pre cert stuff is usually built to put out a higher power signal than is legally allowable - that’s because they assume you’ll use a less efficient or compact antenna and they want to make sure even with that less efficient design you can still get full legal power out of it. If you hook up a 1/4 wave or 1/2 wave whip antenna to any production radio transmitter module on a good ground plane, you may generate as much as twice the legal broadcast power. That changes with the slightest change of the board, the exact orientation of the antenna and a ton of other variables - that’s why the finished packaged unit must be tested and certified - not just the radio module.

The same with the European Union states - I read in another thread here of using Zigbee “pre certified” gear as that “won’t need testing” but this is totally wrong, and you’re going to get yourself hit with huge fines thinking that way. Same with the post from Jon saying that as long as the device is under the required levels it’s okay - as I understand it, this is not okay if you’re going to sell it for any amount of money - it must still be certified in its exact sales production configuration, and even if it didn’t require “certification”, you still need to pay someone to test the finished assembly to make sure it is under the limit.

I’ve done the certification and testing process before and basically, you have to submit the exact product using the exact schematic on the exact PCB board as you plan to sell for testing. If you change one resistor (even if it doesn’t effect the radio) then it must go though the entire certification process all over again. The cert lab also shoots high res photos of the internal of your device which can be used against you down the road if someone thinks you changed something minor - if the version you’re selling isn’t exactly the one in the picture from the test lab, then you’re in big trouble - EVEN IF that new version has the exact same radio characteristics as the tested version. This makes the “open source, user configurable” thing complicated - unless the only user changes are via software. You are permitted to change software once a unit is tested as long as it doesn’t effect radio power.

I called my contact at the test lab we used on Monday and asked him some q’s - he had some strong warnings for everyone here that need to be taken seriously. He didn’t have an actual reference off his head for me, but he said the fine in the US for selling an un-certified radio transmitter is $100,000 per occurrence. He also stressed that the FCC in the US and the other agencies in other countries aggressively pursue this stuff - they don’t issue warnings and they follow up on every tip - basically if you’re skipping the certification step, whoever is deemed responsible is going to be in debt to the government for the rest of their life - that’s serious. I told him that sounded pretty extreme and he agreed it was - but that’s how countries keep their airwaves clean and make sure radio gear works without anyone coming along and broadcasting signals out of spec.

Cost in the US for standard FCC certification is between 5k and 10k, and you can expect to pay a couple thousand in every other country for the paperwork filing even if the US did the testing for you (most labs can test and some even certify for other countries - but the filing fees in those countries still apply). And if the other country requires different frequencies, then you need to re-test and re-certify the whole thing for the new frequency, paying another large testing fee. Testing fees also tend to go up for higher frequencies and different protocols. I don’t know and forgot to ask, but I’ve heard that to certify anything in the 2.4 Ghz frequency hopping spread spectrum band can be more like $30,000. That sounds high - I’d like a reference on that.

You don’t need to test if you’re using a complete built module sent from its manufacture with its own power supply input, its soldered to a board and has an antenna already attached, but its iffy what that actually means. If you find a complete assembled unit with its own circuit board, housing, and permanent antenna that has an input line for data, and that unit is FCC certified, you can probably use it and just feed it data - but it’s going to be big as you’ll have to have a product with 2 circuit boards connected by a line if nothing else.

The best I think we can really expect to do with this project is to work up a blue print that people can use to make their own. If I give you a design and you buy the radio chip yourself, then I’m off the hook, and legally, you can build your own even if its out of spec if it’s for your own testing and development purposes. The minute we solder a radio chip to a PCB, it could be construed as selling an un-certified assembly, and the fines along with it. Who is going to offer to distribute the boards knowing they may be personally liable for charges like those? I don’t expect we’ll get many volunteers.

This means that people are going to have to solder their own surface mount components - which isn’t easy for a novice. Those components can only take heat for a very short time as there are no “legs” to solder - which means people are going to kill a lot of expensive radio modules just trying to assemble the thing. You need a high wattage iron to get quick heat and wick solder - the $10 iron from Wal Mart is going to kill the component before getting the solder hot enough to wick under the contacts.

Second big problem - I’m not seeing any discussion of existing patents. I haven’t looked into it, but it would be good to discuss. You need to ask why hasn’t some other company come out with a cheaper PocketWizard yet? I’m sure some company that makes little pocket games could also make a board that would do what a PocketWizard does, pay the $10k to certify it, and offer it at half the price of a PW and still make a ton of money on it.

So why haven’t they? It’s such an obvious need and it’s not filled. Why?

I don’t know but I’d guess PW probably has some patent on how they’re triggering flashes. If they were the first people to actually build a radio flash trigger and patented it - then you’re asking for more trouble in this. There are the generic triggers but they’re not so reliable - I wonder why not? I would guess they are using some other way to trigger that’s technically outside the claims of the patent PW is using - which probably isn’t very reliable, but they won’t get sued for it either.

I can’t imagine how PW can have such a monopoly with a product that could be produced so inexpensive but yet nobody is actually making one.

Anyone here that owns a PW - can you look very carefully at the unit and all it’s documentation and see if it notes anything about patents pending or in effect? That would be good to know. And even if they don’t advertise it, that’s still not to say it’s not protected by a patent. They could be licensing someone else’s patent too - so searching USPTO for “PocketWizard” may not turn results, but there still could be patents in effect.

There’s a thread about repeating IR pulses - it sounds like that’s what that radiopopper thing is all about. I see his press release says Patent Pending on it. I don’t know him but he does live in my town I think - if I ever meet up with him I’ll ask him about that - but that right there is someone claiming rights to an idea and we can’t just go and copy it. No telling what his claims actually are, but if he’s the first person to actually repeat IR pulses by radio and use them to trigger a flash - and he very broadly patents that concept, then we’re all out of luck on it for the next 20 years. That’s what being the first to think up an idea and make it work is all about - you get to make a blanket statement that nobody anywhere can make one like it without paying you.

I assume that thing works like a TV remote control - I know Radio Shack used to sell a repeater for the TV remote, but if he was the first person to break one out of its box and tape it to the side of a camera flash - then it’s a new application, a new “utility” or “novel use” for an idea as the patent office considers it and he can certainly file a patent for a new use of an old technology, and it’s legal and binding as anything else. If Radio Shack still has a patent in effect on it, he may have to pay them (no telling) but he can certainly keep everyone else from selling a product intended for the same use - open source or otherwise. That’s no fun, but that’s the way capitalism works.

You could probably give away plans of how to modify something to do the same thing, but from what I’ve read in the other post about ETTL light pulses - that looks much different than how a TV remote works, which means it probably is a completely new approach that hasn’t been done before, or at least made to work. You can’t get a patent on an idea without either making it work or very clearly describing exactly how to make it work - so unless someone else has already done this or written in detail how to do this, then we’ll probably have to pay this guy to do anything that involves light pulses and radio waves for camera flashes.

In order to build a system that relays pulses like that you’re going to need to program some kind of processor with some code to handle it - which is probably also (if it’s new) covered in this guy’s patent or some other patent - so either people have to program their own processors with code they download for free off the net (which probably requires them to build their own programmer board to flash the chip), or someone has to distributed pre-programmed chips - and if the program on those chips does what this guy’s radio does, or if it does what Pocket Wizard does - then you’re probably going to be blocked by various patents.


I hate to be so negative when everyone is so amped up and excited, but if I were consulting for a company considering something like this, there’s no way they would proceed any further without making absolute sure of the above first.

I hate to say it, but this just isn’t something that really works for the “open source” model. Unless you get some people to grant money with no strings, there’s the obvious financial limits. It’s probably going to cost from $20,000 to $30,000 to really go from here to a working unit between prototype units, a few versions of a PCB, running a batch of PCB’s to give out to users, paying for the testing in a few countries and a few consults with attorneys to make sure nobody gets sued.

Maybe the best idea is to actually form a company to do this, where interested people make an investment - an actual commercial company with the intention of creating a better and less expensive system. Get 30k in the pot, elect some officials, and seriously consider the legal and design aspects of it all, then produce a unit for just enough profit to cover the expense of development.
?Let’s say you have 100 people willing to chip in - so that’s $300 a person, and let’s say you sell the radios for $75 each (so $150 for a pair) - if they cost $35 a unit to manufacture, that’s a $40 per unit “profit” to pay back to the investors - which means you need to sell 750 units for the people offering the money to get it back - and those people still have to pay another $150 to get a set for their own use.

And you’re going to need to run 500 or so of them at a shot to get that price down - building them one at a time will cost 2 or 3 times as much, so now you need someone to front the bill of $35 a unit x 500 = $17,500 and trust that the units will sell - or get 500 people to pre-order them knowing they haven’t even been manufactured. Not likely.

Getting people to stop talking about an idea and start mailing checks for hundreds of dollars is where it all breaks down - making big payments on something that may never materialize. So you’ll probably need to get the money from an investor source, and that investor is going to want to make money on the deal - which means you can’t just sell your radio for cost anymore, you have to start making money on the deal. And unless you’re selling thousands of them, you have to make quite a bit of money on each unit to justify the whole thing to the people holding the money.

Any way you cut it, it’s going to cost some actual money at some point and you’ve got to ask if the money and effort put into it is justified by the reduced cost of owning this radio, or if you can build in some seriously improved new features that the big companies haven’t thought up yet.

This is a very novel idea and I’m not saying it can’t work - but it’s not going to be as easy as soldering some chips together and getting some programmer to write us some free code. I’m just trying to be the voice of reason here - I’d hate to see the threads popping up in six months or a year that the FCC hit people with fines or that the effort is now being sued for some infringement.

That’s all I’ve got time for - 3 days to write this but it will be good to see if there are ways to overcome this stuff. This is my contribution to the project. Let’s take some pause now and see if this is really feasible before getting too carried away. I’ll check in here from time to time and see if anything new is happening.
BertIvon
 
Posts: 3
Joined: Wed Sep 26, 2007 8:44 pm

Re: I see trouble. Lots of stuff to consider.

Postby JonSenior » Thu Sep 27, 2007 1:22 am

I confess I've not taken as long over this response I you have over the post, but a few comments in return. I have heavily snipped your post in order to address those areas I have experience in.

BertIvon wrote:A big caution to the idea of what “Pre Certified” radio products are all about. This does NOT mean you can just build with them. It doesn’t matter what the product or what you build it out of, if you assemble one, and it emits a radio signal of any kind, and offer it in exchange for any money, it MUST be FCC certified in the U.S., as well as the same type of cert for all the other countries you’ll send it to. It’s the same with Zigbee, Blutooth, or the standard FSK and OOK type stuff - if it makes a radio signal, the finished packaged unit MUST be certified before sale, and that’s the same for all countries that I’m aware of.


I've followed up some of the stuff on European regulation vs FCC. There are clear differences. FCC is the highest standard to adhere to.

The same with the European Union states - I read in another thread here of using Zigbee “pre certified” gear as that “won’t need testing” but this is totally wrong, and you’re going to get yourself hit with huge fines thinking that way. Same with the post from Jon saying that as long as the device is under the required levels it’s okay - as I understand it, this is not okay if you’re going to sell it for any amount of money - it must still be certified in its exact sales production configuration, and even if it didn’t require “certification”, you still need to pay someone to test the finished assembly to make sure it is under the limit.


This is why I having been saying all along that we don't want to produce a kit. Make a design. Even run a batch of PCBs (Don't put a drop of solder on them).
This makes the “open source, user configurable” thing complicated - unless the only user changes are via software. You are permitted to change software once a unit is tested as long as it doesn’t effect radio power.


I think the idea was to create a decently flexible base design that has the capability for funcionality upgrades in software.

He also stressed that the FCC in the US and the other agencies in other countries aggressively pursue this stuff - they don’t issue warnings and they follow up on every tip - basically if you’re skipping the certification step, whoever is deemed responsible is going to be in debt to the government for the rest of their life - that’s serious.


The eBay remotes suggest that actually this is not the case. At least... that they are aggressively pursuing large corporations that break the testing rules. In terms of being noticed as an end user... we'll be putting out our signal (at whatever level) for a few of a milliseconds and sub at (probably) sub-1/s intervals. Unless it makes all TVs jump, or opens garage doors, I doubt that any end users will have much grief from the FCC.

You don’t need to test if you’re using a complete built module sent from its manufacture with its own power supply input, its soldered to a board and has an antenna already attached, but its iffy what that actually means. If you find a complete assembled unit with its own circuit board, housing, and permanent antenna that has an input line for data, and that unit is FCC certified, you can probably use it and just feed it data - but it’s going to be big as you’ll have to have a product with 2 circuit boards connected by a line if nothing else.


Funnily enough that's exactly what the xBee and Sparkfun Nordic modules provide. Provide power and link the data lines to an MCU. Board is provided with solder tags or an SIL plug. It's a stand-alone daughter board. That simplicity appeals to me and if it means it's FCC compliant then so much for the better.

The best I think we can really expect to do with this project is to work up a blue print that people can use to make their own. If I give you a design and you buy the radio chip yourself, then I’m off the hook, and legally, you can build your own even if its out of spec if it’s for your own testing and development purposes.


I with you here.

This means that people are going to have to solder their own surface mount components - which isn’t easy for a novice. Those components can only take heat for a very short time as there are no “legs” to solder - which means people are going to kill a lot of expensive radio modules just trying to assemble the thing. You need a high wattage iron to get quick heat and wick solder - the $10 iron from Wal Mart is going to kill the component before getting the solder hot enough to wick under the contacts.


Actually, if we use the modules, then we can remove the need for surface mount. It'd be nice for compactness sake, but it'd be relatively easy to produce two PCB designs, one using SMD the other, through-hole.

Second big problem - I’m not seeing any discussion of existing patents. I haven’t looked into it, but it would be good to discuss. You need to ask why hasn’t some other company come out with a cheaper PocketWizard yet? I’m sure some company that makes little pocket games could also make a board that would do what a PocketWizard does, pay the $10k to certify it, and offer it at half the price of a PW and still make a ton of money on it.


Principally; This is specialist equipment in a specialist market. Not of much interest for a company seeking a fast buck. That said. Elinchrom have done pretty much that with the Skyports.

I will have a hunt for some patents, but the existance of a patent does not make it defendable. And I do have some experience in this area. The US patent office in particular has a reputation for granting pretty much anything and leaving it to the lawyers to argue over in court.

Here follows a discussion of the technology:
The use of RF to transmit data and trigger devices is prior art. See remote controls for garage doors.
The use of encoded RF to prevent false triggers is also prior art. Garage doors again. There are specific devices which serve this purpose.
The use of a camera synchronisation to trigger a flash by cable. Very much prior art since it (AIUI) predates hot shoes.
The use of RF to remotely trigger a flash. RF as a replacement for cables is a given. Hence "wireless ethernet".

In the event that the manufacturers of the PW decided to sue (civil proceedings), they would have to have someone to sue. Since the design would be public domain and all users (as discussed above) would have built their own devices, this would involve a law suit against each individual. No company in their right mind would take this approach since even a win would result in a loss of many thousands of pounds.

So why haven’t they? It’s such an obvious need and it’s not filled. Why?


I don't know that it is an obvious need. It is for us. We've used the eBay triggers, and stared longingly at the PW product specs but as I said earlier, this is a specialist market in a specialist market. I doubt that the user base is big enough to tolerate more than one manufacturer at or near this price point.

The radiopopper thing would be more of an issue, but I recall seeing devices which did the same thing, but in a form factor designed for household IR remotes. Patent pending has even less strength than a granted patent. I know, I've been patent pending on a computer security product for more years than I care to think about!

I hate to be so negative when everyone is so amped up and excited, but if I were consulting for a company considering something like this, there’s no way they would proceed any further without making absolute sure of the above first.


You're right to bring it up, but I don't see that it holds the relevance for us that it would for an attempt to form a company.

I hate to say it, but this just isn’t something that really works for the “open source” model. Unless you get some people to grant money with no strings, there’s the obvious financial limits. It’s probably going to cost from $20,000 to $30,000 to really go from here to a working unit between prototype units, a few versions of a PCB, running a batch of PCB’s to give out to users, paying for the testing in a few countries and a few consults with attorneys to make sure nobody gets sued.


I think that you're grossly over-estimating the development costs here. For a large company with full-time research staff then perhaps there would be some relevance, but not to a loosely affiliated group of individuals from around the globe. The costs will be in the order of components + PCB production + PCB production (When we bugger up the design for the first one). BatchPCB will produce cheap PCBs courtesy of our friends in China. Parts can be sourced (generally) locally. Time costs are not really relevant to anyone but the particularly anal since we are all doing this in our spare time. Given the legal framework in the USA and UK (The only ones I really know about) with regard to IP I find the idea of consulting an attorney to check for liability laughable. Patent protection boils down to looking scary enough to dissuade the competition. Defending a well-written patent is hard work. Defending the average patent is nigh on impossible.

Maybe the best idea is to actually form a company to do this, where interested people make an investment - an actual commercial company with the intention of creating a better and less expensive system. Get 30k in the pot, elect some officials, and seriously consider the legal and design aspects of it all, then produce a unit for just enough profit to cover the expense of development


I think is actually completely the wrong way to approach this. Simply creating a company diverts valuable pocket money away from prototype boards and radio modules. The whole point is to create a design which "belongs" to the community. If that goal is kept then anyone with a soldering iron can make one of the company's products and profits will fall short of the mark. If we're just after another closed source trigger, we should email Elinchrom and PW and offer our suggestions and wait for them to make something of it.

This is a very novel idea and I’m not saying it can’t work - but it’s not going to be as easy as soldering some chips together and getting some programmer to write us some free code. I’m just trying to be the voice of reason here - I’d hate to see the threads popping up in six months or a year that the FCC hit people with fines or that the effort is now being sued for some infringement.


I'll write you some code. Others here have already offered to do the same. We have a "brain trust" of EE talent that should be able to muster together a design for linking an MCU and a transmitter. As I pointed out earlier, if the FCC was inclined to pursue action against individuals using untested trigger units, then we'd already be seeing the posts for the eBay triggers. A quick look at the ratings of any of the eBay sellers shipping these things should give you an idea of the sheer number of units that are already in use.

A lot of what you have to say would be disturbing and pose a real threat for a genuine company based in the USA and wanting to develop a real product for sale. A large majority does not apply to an international open source group who's final product will be entirely intellectual. More specific references to case law or documented examples would be nice, but before anyone allows their enthusiasm to be too heavily dampened I would point out that your post consisted of supposition and heresay. Good points, needed mentioning, but in my opinion (and I am not a lawyer), not largely of relevance to us. I await contradiction.

Jon :wink:
JonSenior
 
Posts: 60
Joined: Fri Sep 21, 2007 7:27 pm
Location: Paris, France

Postby BertIvon » Thu Sep 27, 2007 1:52 am

Thank you Jon - you make good points also.

If what we're after here is purely a do-it-yourself product that we're just providing the instructions for, then no - there is no worry about any of this. You're right that nobody will get invidually sued or hit by the FCC (again it's totally legal to do it for yourself even under the FCC, as long as you don't sell it to anyone else). If people are willing to solder their own units together off a shopping list provided here - then that would work great. I think that would be a true to the "open source" idea anyway.

It's true that most people will be able to solder existing self contained modules - they'll be larger than individual chips but much easier to work with. I don't think anyone will complain much if this thing is the size of a cell phone, which it could be that small even with pre-fab modules.

And good to know about the prior art ideas on the radio trigger - who knows, maybe PocketWizard doesn't have any patent claim on their product and nobody has challenged them. It's possible.

The ebay triggers - I'm sure they are certified - they've got to be. From every warning I've gotten from anyone who knows, they'd be crazy not to certify them. Either that or PW is loosing lots of money by not making a simple phone call to report them.


There seemed to be more talk in the original dicussions of building some kind of assembled kit or a kit needed to be soldered together - as you point out that again is a very bad idea because it does pose these risks (unless we have a member in some odd country off the map that will ship them internationally with little fear of reprisal - which by the way may be the way the ebay triggers are getting to the market).

As long as it stays to creating do-it-yourself plans - this could go a long way. There could even be 7 different versions of what you could build for different needs - that would be cool as it costs nothing but time to put up the plans for the public.

I'm also interested to see more solid references to what I've covered - I admit I didn't look it up myself, but I talked to some people who's jobs depend on it and I trust what I'm being told.
BertIvon
 
Posts: 3
Joined: Wed Sep 26, 2007 8:44 pm

Postby strogg » Thu Sep 27, 2007 2:30 am

BertIvon wrote:If what we're after here is purely a do-it-yourself product that we're just providing the instructions for, then no - there is no worry about any of this. You're right that nobody will get invidually sued or hit by the FCC (again it's totally legal to do it for yourself even under the FCC, as long as you don't sell it to anyone else). If people are willing to solder their own units together off a shopping list provided here - then that would work great. I think that would be a true to the "open source" idea anyway.


What would be the legality in regards to FCC licensing if the units were sold "at cost" as opposed to being given away or sold at profit?

IE: we only charge as much to pay for the parts and the PCB layout
strogg
 
Posts: 19
Joined: Thu Sep 20, 2007 8:31 pm

Postby seaton » Thu Sep 27, 2007 3:11 am

Great discussion guys and very valid on all fronts,

I know everyone will have different requirements and what they wnat out of this project as they all have different levels of need and technical expertise. What I see as our main goal is a "reference design" for a hackable trigger that is not only open but is both reliable and affordable for the hobbists.

Part of this design will be schematics and software APIs, documented protocols etc, everything we need to build a set of triggers - wouldn't it be great if some of the name brand manufactures started to come on board because they wanted to make sure their product worked with the "open" remote trigger system, blue sky atm I know but possible, given the size of the potential userbase, and if they start getting more and more customers requests, I can just hear the conversations with a sales rep "does xyz camera (or flash) work with the open trigger system, oh it doesn't, but wxy does......" then I'm sure we they will get involved somewhere down the line in some way (fingers crossed anyway) Heck once we get this thing up and running I'll write an open letter to all the major manufacturers asking them if they woudl like to take part in the project, or even just to open up their protocols (whats that Canon? you would also like to donate a 1D MKIII to the project for some development? hehe ;)

As far as commercialization of the product, someone with backing in terms of $$$ will probably get a version of our design FCC approved for the market that is non-geek inclined and the professionals, even though it will cost them $10-20K in FCC approvals, it will still be cheap as the majority of R&D costs have already been done. So someone will probably end up selling kits, great that what we want, a cheaper more reliable trigger, oh and yes expandable, so in the end our project has been successful, we will have achieved what we set out to do, and if people have the time, energy and $$$ to make some money, good for them, Me, well I just want a cheap hackable trigger to feed my Photography and Geek obsessions all roled into one lol

As far as patents well I guess this is always a "grey" area and of this I have no expertise, however, if we create our own protocol from the ground up and it is well documented, also if we use off the shelf components i.e. xbee/nordic etc, then surely thats what we are paying for when we purchase their product, the license to use their IP.

As far as reverse engineering specific camera protocols i.e. E-TTL etc well I guess this is the "grey" area, are they infact patented?, but as far as TTL and manual triggering then surely we should be fine, but in reality what have we done? We are not selling the information, who are they going to sue? In what country are they willing to pursue this. I guess that this is exactly what the radio popper will have to worry about, there would be easier low hanging fruit than an open source project that is not in it for commercial gain. Also this will all be implemented in Software, so if the basic design allows for basic and TTL triggering, what have we done to infringe patents?

If we form a company etc then we could be potentially a target for anything legal, as we would then be a legal entity. We need to stay flexible and mobile so we can adapt rapidly to change if required. Anyway a company would just cost too much $$$ and as Jon said will divert us from the real thing, developing the trigger.

Stephen...
seaton
Site Admin
 
Posts: 134
Joined: Mon Jan 23, 2006 8:48 am
Location: Bunbury, Western Australia

Postby Rudeofus » Thu Sep 27, 2007 9:44 am

First of all, thanks BertIvon, for bringing all this up! I'd rather have people rain on our party from within than having to deal with nastigrams from companies whos patents we violate.
seaton wrote:As far as patents well I guess this is always a "grey" area and of this I have no expertise,

There is no grey area. If we violate patents, we get sued. In case a patent is questionable (I think, a few years ago the wheel was patented in Australia to prove their patent system stinks), we'd have to litigate it, an expensive and lengthy process.
however, if we create our own protocol from the ground up and it is well documented, also if we use off the shelf components i.e. xbee/nordic etc, then surely thats what we are paying for when we purchase their product, the license to use their IP.

Using off the shelf components won't get us off the hook. You can reimplement most patented designs with off the shelf components, that's why patent protection is important. As long as these off the shelf components are used in a novel and non obvious way, patent protection applies.
As far as reverse engineering specific camera protocols i.e. E-TTL etc well I guess this is the "grey" area, are they infact patented?

I don't think wireless E-TTL is patented, since Metz sells units which exactly replicate their behaviour. Also, Sigma sells flashes which support E-TTL over the air (but they don't replicate the exact protocol). I highly doubt that Metz or Sigma pay any royalties to Canon, especially Sigma.
but as far as TTL and manual triggering then surely we should be fine, but in reality what have we done? We are not selling the information, who are they going to sue? In what country are they willing to pursue this.

If triggering a flash over radio wave was patented, we'd have a problem. The fact that so many companies sell these triggers suggests, that no patent is in the way of doing this.
I guess that this is exactly what the radio popper will have to worry about, there would be easier low hanging fruit than an open source project that is not in it for commercial gain.

If we cut into radio poppers profits, they will try to attack this product, especially if we make it easy for them by trampling on their patents. An open source product would be an ideal target for patent litigation, since we're not in it for the profit. Hence, we wouldn't want to dedicate wads of money to patent litigation ... :evil:
Also this will all be implemented in Software, so if the basic design allows for basic and TTL triggering, what have we done to infringe patents?

We don't know until we check applicable patents first.
If we form a company etc then we could be potentially a target for anything legal, as we would then be a legal entity. We need to stay flexible and mobile so we can adapt rapidly to change if required. Anyway a company would just cost too much $$$ and as Jon said will divert us from the real thing, developing the trigger.

I agree, that forming a company won't protect us from litigation. However, not forming one doesn't protect us either. If you host a web page which carries software that violates patents, you may already be in trouble, even if you don't sell this software for money. Violating a patent is not tied to making a profit or trying to make one.
Rudeofus
 
Posts: 27
Joined: Sun Sep 23, 2007 11:28 pm

Postby JonSenior » Thu Sep 27, 2007 10:32 am

Rudeofus wrote:There is no grey area. If we violate patents, we get sued. In case a patent is questionable (I think, a few years ago the wheel was patented in Australia to prove their patent system stinks), we'd have to litigate it, an expensive and lengthy process.


That is frankly untrue. The "wheel" patent was demonstrating a problem with a fast-track system that had been introduced, but there is a patent in the US (not pending AFAIK) which describes a method of swinging on a swing. This involves side-to-side movement by pulling on the chains.

The fact that a patent has been granted does not been that it has any strength in law. It simply means that (Assuming that everyone does their job properly) to the best knowledge of the examiners it represents a novel invention in that field. If you want to enforce the patent you have the burden of proof in your lap, both that the person you are sueing has infringed your patent, and that the patent has validity. If you can show one example predating the patent application which would be "obvious to someone knowledgeable in the field" then the patent is invalid and the case is lost.

Furthermore software patents are valid in the US and (AFAIK) nowhere else. Certainly not in Europe, we're still fighting that one! Even in the US it has been demonstrably hard to prove patent infringement in the case of software.

If we cut into radio poppers profits, they will try to attack this product, especially if we make it easy for them by trampling on their patents. An open source product would be an ideal target for patent litigation, since we're not in it for the profit. Hence, we wouldn't want to dedicate wads of money to patent litigation ... :evil:


This is both true and untrue. If we do cut into RadioPopper's profits then we become a target, but who do they sue? You? Me? Seaton? Depending on your country of residence the patent may not even be valid. Since there is no one entity it would have to be an attack on anyone who makes one and that would not be an easy case to fight. Especially given that the technology and idea of using RF to extend IF clearly and demonstrably existed before. I am not a lawyer and haven't yet seen the patents, but I doubt that the idea of changing the IR source from a TV remote to a flash would count as an inventive step.

We don't know until we check applicable patents first.


Added to my todo list! :(

Violating a patent is not tied to making a profit or trying to make one.


Violating one isn't, but the value of litigation is.

Jon
JonSenior
 
Posts: 60
Joined: Fri Sep 21, 2007 7:27 pm
Location: Paris, France

Non-profit

Postby anton.tagunov » Thu Sep 27, 2007 11:14 am

Non-profit organization is how open source contributors normally protect themselves from being sued.
The non-profit appoints a chairman. The chairman approves any deeds of the project. If anybody gets sued it's the organization.

Are there any other useful endowments that a non-profit can provide?
Less trouble with tax authorities when somebody collects donations to build prototypes?
Last edited by anton.tagunov on Thu Sep 27, 2007 1:28 pm, edited 1 time in total.
anton.tagunov
 
Posts: 19
Joined: Fri Sep 21, 2007 6:52 pm

Postby Thonord » Thu Sep 27, 2007 11:31 am

I'd like to get back to one point Bertivon made and that JonSenior commented. (Sparkfun)

quote unless the only user changes are via software. You are permitted to change software once a unit is tested as long as it doesn’t effect radio power.

Are there any programmable dongles out there? with a USB interface and some free "Input ports" (not USB) ?

Tom
Ppl who agree need normally not reply, those who disagree or have questions do.
Or - just ignore me.
Thonord
 
Posts: 50
Joined: Fri Sep 21, 2007 2:32 pm
Location: Norway

Postby seaton » Thu Sep 27, 2007 3:49 pm

Ok so please bear with me but I have very limited patent knowledge and I'm trying to get my mind around it....

So what happens, if for example we come up with our own novel way of remotely triggering a flash with our own protocols and design independant of existing patents? i.e. A clean room approach. So for example if there is an existing patent for remotely triggering a camera flash, but done differently, i.e. our own cleanroom approach uses a different protocol etc. surely we are not violating any patents? From my understanding (albiet very limited) a patent has to be detailed in its description of the components of the IP it is protecting, so can't just be "remotely triggering a flash via wireless".

I personally know a patent holder that had one of his patent violated in three different countries, by one of his customers he sold his device to, they had copied it and resold it in their own machines, he pursued it in two countries and althougth he eventually won, it was as a huge drain on his resources in terms of time, energy and money because he had to prove the violation in each case, in the third country he elected not to pursue it, so now he protects his IP via copyright in terms of software and firmware which is he says is a lot easier as the onus is on the violator to prove/disprove the copyright violation, not the owner of the copyright.

And as far as software patent violation, look at the linux software in the kernel action between sco and the world. I think it would be very hard to prove, and who are they going to sue?

So if we do violate any patents, and I'm sure we are not about to do it knowingly, then who would the sue in terms of our project? me, yes I have initiated the project but I don't own anything of it, maybe the domain name this site is hosted on. But who has developed it? all of us?, so they sue individuals in the group who knows? What would they get out of suing me? my car? well that would dissapoint me (love my Z3 sigh but I don't get to drive it much with 3 children) but it's not the end of the world, I have my health, my family and my assets are tied up in family trusts so not much to get hold of there either, they can have my creditcard debt if they want ;) And if they close the site and project down then I'm sure that the project will only pop up under a different name on another site possibly in another country.

Really will it be worth anyones while to pursue it in a court of law? well I guess we will never know until it happens, ands surely we can negotioate anything first, as soon as we know we are violating a patent well then we need to re-assess, and look for another way of doing it, or come to an agreement with the patent holder, i.e. fair use for individuals/hobists, if the product is commercially sold by someone then they need to come to an agreement with the patent holder not us, I'm sure we can come to an agreement if required, its only a win win for all involved rather than getting laywers etc involved.

Stephen..
seaton
Site Admin
 
Posts: 134
Joined: Mon Jan 23, 2006 8:48 am
Location: Bunbury, Western Australia

Postby Rudeofus » Thu Sep 27, 2007 6:29 pm

JonSenior wrote:
Rudeofus wrote:There is no grey area. If we violate patents, we get sued. In case a patent is questionable (I think, a few years ago the wheel was patented in Australia to prove their patent system stinks), we'd have to litigate it, an expensive and lengthy process.


That is frankly untrue. The "wheel" patent was demonstrating a problem with a fast-track system that had been introduced, but there is a patent in the US (not pending AFAIK) which describes a method of swinging on a swing. This involves side-to-side movement by pulling on the chains.


In case of the patented wheel it's quite easy to provide prior art. This may not be as easy in case of triggering a flash remotely over an RF link which is used to tunnel flash specific IR signals (I have no idea whether radio popper has patented this, but I'd have in their case, and they mention pending patents). We're trying to do something which is done by only a few companies - chances are that they have marked their turf with broad patents. They certainly won't feel compassion for a flock of hobbyists (sorry, David, for the pun) potentially detroying their market.

JonSenior wrote:The fact that a patent has been granted does not been that it has any strength in law. It simply means that (Assuming that everyone does their job properly) to the best knowledge of the examiners it represents a novel invention in that field. If you want to enforce the patent you have the burden of proof in your lap, both that the person you are sueing has infringed your patent, and that the patent has validity. If you can show one example predating the patent application which would be "obvious to someone knowledgeable in the field" then the patent is invalid and the case is lost.

Then we better back up our design decisions with prior art (older than 17 years or so) and make sure there is no valid patent around upon which we could potentially infringe. Also, if we come up with ideas, we should post them here as soon as possible (or publish them somewhere) so nobody can patent them later on and use the patent against us.

JonSenior wrote:Furthermore software patents are valid in the US and (AFAIK) nowhere else. Certainly not in Europe, we're still fighting that one! Even in the US it has been demonstrably hard to prove patent infringement in the case of software.

I think software patents won't be a problem for us, we don't try to write a free word processor which can read OOXML :-)

It's good old fashioned hardware patents we're up against, and chances are they exist. Fortunately the internet provides us with a convenient way of searching them, so we might as well go ahead and do that upfront.
JonSenior wrote:This is both true and untrue. If we do cut into RadioPopper's profits then we become a target, but who do they sue? You? Me? Seaton?

They can sue anyone hosting a web page. They can sue anyone selling PCBs or modules. If they can prove you facilitate patent infringement (e.g. by helping someone build something which depends on their patents), they can be after you. If the RIAA can do this, anyone can.
JonSenior wrote:Depending on your country of residence the patent may not even be valid. Since there is no one entity it would have to be an attack on anyone who makes one and that would not be an easy case to fight. Especially given that the technology and idea of using RF to extend IF clearly and demonstrably existed before.

If we can show prior art, we're off the hook, mission accomplished.
JonSenior wrote:I am not a lawyer and haven't yet seen the patents, but I doubt that the idea of changing the IR source from a TV remote to a flash would count as an inventive step.

I've seen patents much more obvious costing big companies wads of money. Interval wind shield wiper, anyone? How novel is forwarding emails to a mobile phone? Or embedding content in a web browser ?
Violating a patent is not tied to making a profit or trying to make one.

Violating one isn't, but the value of litigation is.

Just paying the legal costs can be prohibitively expensive for most of us ...

seaton wrote:So what happens, if for example we come up with our own novel way of remotely triggering a flash with our own protocols and design independant of existing patents?

If we come up with something new and pubish it, it can not be patented anymore by anyone, except the inventor (who has, depending on country, some time to file a patent)

seaton wrote:i.e. A clean room approach.

There is no such thing as a clean room approach as far as patents are concerned. A patent is a novel invention published by the inventor, for which he gets a 17-20 year monopoly in return. Since the subject of the patent has been published, you can not claim lack of knowledge of it.

seaton wrote:So for example if there is an existing patent for remotely triggering a camera flash, but done differently, i.e. our own cleanroom approach uses a different protocol etc. surely we are not violating any patents?

If just a particular protocol has been patented, we're off the hook. If a broader concept is patented, we're in trouble.

seaton wrote:From my understanding (albiet very limited) a patent has to be detailed in its description of the components of the IP it is protecting, so can't just be "remotely triggering a flash via wireless".

If you can show that nobody has ever triggered a flash via RF and then you can prove that it is not obvious to do so, you can patent it.
But I don't think anyone has patented "remotely triggering a flash via wireless". What has most likely been patented some time ago is using a specific modulation scheme or style of data exchange to trigger a flash reliably. If we stumble on exactly this method, we're in for trouble.

seaton wrote:And as far as software patent violation, look at the linux software in the kernel action between sco and the world. I think it would be very hard to prove, and who are they going to sue?

For instance IBM for using linux in their commercial products ? :evil:
That's the reason, BTW, why IBM has committed their huge pool of lawyers to defeat SCOs claims.

seaton wrote:Really will it be worth anyones while to pursue it in a court of law? well I guess we will never know until it happens, ands surely we can negotioate anything first, as soon as we know we are violating a patent well then we need to re-assess, and look for another way of doing it, or come to an agreement with the patent holder, i.e. fair use for individuals/hobists, if the product is commercially sold by someone then they need to come to an agreement with the patent holder not us, I'm sure we can come to an agreement if required, its only a win win for all involved rather than getting laywers etc involved.

Of course you realize, that
  • some companies are going to be mad at us for stepping into their territory and challenging their profitability ...
  • some of us just want to go on with our life and not fight patent law suits in court ...


So I guess, we better do some research upfront :-)
Rudeofus
 
Posts: 27
Joined: Sun Sep 23, 2007 11:28 pm

Postby anton.tagunov » Thu Sep 27, 2007 7:23 pm

Rudeofus wrote:Just paying the legal costs can be prohibitively expensive for most of us ...
don't want to sound like a broken record... :-)
however do I get it right that if there's a charity between people and the cruel world and
the charity is sued too hard it can simply go bankrupt protecting project members from any expenses?
anton.tagunov
 
Posts: 19
Joined: Fri Sep 21, 2007 6:52 pm

Re: I see trouble. Lots of stuff to consider.

Postby chrisb » Thu Sep 27, 2007 7:37 pm

BertIvon wrote:I’ve been working on this post for the past couple days - lots of info but it needs to be considered to move forward.


A big caution to the idea of what “Pre Certified” radio products are all about. This does NOT mean you can just build with them. It doesn’t matter what the product or what you build it out of, if you assemble one, and it emits a radio signal of any kind, and offer it in exchange for any money, it MUST be FCC certified in the U.S., as well as the same type of cert for all the other countries you’ll send it to. It’s the same with Zigbee, Blutooth, or the standard FSK and OOK type stuff - if it makes a radio signal, the finished packaged unit MUST be certified before sale, and that’s the same for all countries that I’m aware of.



This isn't strictly true, otherwise there would be no sense in manufacturers certifying radio modules. If you design a product using an approved radio TX/RX module, then you don't need to register the radio portion with the FCC. You still need to test the whole unit to FCC Part 15, but this testing doesn't require registration.

BertIvon wrote:Know that even the pre cert stuff is usually built to put out a higher power signal than is legally allowable - that’s because they assume you’ll use a less efficient or compact antenna and they want to make sure even with that less efficient design you can still get full legal power out of it. If you hook up a 1/4 wave or 1/2 wave whip antenna to any production radio transmitter module on a good ground plane, you may generate as much as twice the legal broadcast power. That changes with the slightest change of the board, the exact orientation of the antenna and a ton of other variables - that’s why the finished packaged unit must be tested and certified - not just the radio module.



No, pre-certified radio devices are not built to put out a higher power, if they did that, they would fail follow up inspections. They also have to list the antenna that was used during certification. A warning is then placed in the user instructions that use of another antenna will invalidate the FCC certification.

BertIvon wrote:The same with the European Union states - I read in another thread here of using Zigbee “pre certified” gear as that “won’t need testing” but this is totally wrong, and you’re going to get yourself hit with huge fines thinking that way. Same with the post from Jon saying that as long as the device is under the required levels it’s okay - as I understand it, this is not okay if you’re going to sell it for any amount of money - it must still be certified in its exact sales production configuration, and even if it didn’t require “certification”, you still need to pay someone to test the finished assembly to make sure it is under the limit.



European union is the same as the FCC with regard to pre-certified radio devices.

BertIvon wrote:I’ve done the certification and testing process before and basically, you have to submit the exact product using the exact schematic on the exact PCB board as you plan to sell for testing. If you change one resistor (even if it doesn’t effect the radio) then it must go though the entire certification process all over again. The cert lab also shoots high res photos of the internal of your device which can be used against you down the road if someone thinks you changed something minor - if the version you’re selling isn’t exactly the one in the picture from the test lab, then you’re in big trouble - EVEN IF that new version has the exact same radio characteristics as the tested version. This makes the “open source, user configurable” thing complicated - unless the only user changes are via software. You are permitted to change software once a unit is tested as long as it doesn’t effect radio power.



This isn't true, if your lab is charging you re-test fees because you replaced a resistor, then you are being ripped off and need to change labs now. I work in an EMC test lab, if we told clients that they need to re-test because they changed a resistor, we wouldn't be in business for long.

BertIvon wrote:Cost in the US for standard FCC certification is between 5k and 10k, and you can expect to pay a couple thousand in every other country for the paperwork filing even if the US did the testing for you (most labs can test and some even certify for other countries - but the filing fees in those countries still apply). And if the other country requires different frequencies, then you need to re-test and re-certify the whole thing for the new frequency, paying another large testing fee. Testing fees also tend to go up for higher frequencies and different protocols. I don’t know and forgot to ask, but I’ve heard that to certify anything in the 2.4 Ghz frequency hopping spread spectrum band can be more like $30,000. That sounds high - I’d like a reference on that.



Costs for standard FCC testing are in the ball park, 5 to 10 k (US$). I'm assuming that the couple of thousand more per country are for the radio certification? If the radio is already certified in the countries that you are interested in, they shouldn't be any other fees (apart from maybe an administration fee for a test report). Testing fees shouldn't go higher if you use higher frequencies, if you are being charged more then find a different lab. To certify a 2.4 GHZ device with the FCC is about US$500.

BertIvon wrote:You don’t need to test if you’re using a complete built module sent from its manufacture with its own power supply input, its soldered to a board and has an antenna already attached, but its iffy what that actually means. If you find a complete assembled unit with its own circuit board, housing, and permanent antenna that has an input line for data, and that unit is FCC certified, you can probably use it and just feed it data - but it’s going to be big as you’ll have to have a product with 2 circuit boards connected by a line if nothing else.



It doesn't need to have a permanent antenna built onto the board, see my comments above.

BertIvon wrote:This means that people are going to have to solder their own surface mount components - which isn’t easy for a novice. Those components can only take heat for a very short time as there are no “legs” to solder - which means people are going to kill a lot of expensive radio modules just trying to assemble the thing. You need a high wattage iron to get quick heat and wick solder - the $10 iron from Wal Mart is going to kill the component before getting the solder hot enough to wick under the contacts.



I agree with this and I've already made a similar comment. Using an off the shelf radio module isn't going to remove the need for other components.

-chris
chrisb
 
Posts: 7
Joined: Mon Sep 24, 2007 1:04 pm

Postby Rudeofus » Thu Sep 27, 2007 8:41 pm

anton.tagunov wrote:
Rudeofus wrote:Just paying the legal costs can be prohibitively expensive for most of us ...
don't want to sound like a broken record... :-)
however do I get it right that if there's a charity between people and the cruel world and the charity is sued too hard it can simply go bankrupt protecting project members from any expenses?

I don't really think you can get away with doing illegal things by just forming a charity. No more speeding tickets, anyone? :-)

If that charity can be proven to have done illegal things (e.g. violate patents) which made them go bankrupt, I assume there will be criminal investigations which may lead to really bad consequences for the folks representing this charity ... or the folks knowingly doing illegal things.

Think of it like this: Charity A violates patent of company B. Company B suffers from reduced sales (=damages! ) because A undercut their pricing. B sues for damages, A loses in court, goes bankrupt. However, the damage has been done, B is entitled to compensation. They are not getting any, so they will file criminal charges against whoever is responsible (someone apparently led A into owing a lot of money knowing A can't repay). Club fed welcomes someone from A!

Note: I am not a lawyer, but the company I worked for went bankrupt last year, doing nothing illegal. Charges were still filed against the owner, because some people lost money ...
Rudeofus
 
Posts: 27
Joined: Sun Sep 23, 2007 11:28 pm

Re: I see trouble. Lots of stuff to consider.

Postby Rudeofus » Thu Sep 27, 2007 8:54 pm

chrisb wrote:European union is the same as the FCC with regard to pre-certified radio devices.

AFAIK in Europe you are responsible for only selling equipment which follows the rules, but you are not required to go through official testing. Going through official testing might give you some slack if someone else proves your device to violate the rules (because you can claim you did everything you could), but it's not required.
chrisb wrote:if your lab is charging you re-test fees because you replaced a resistor, then you are being ripped off and need to change labs now. I work in an EMC test lab, if we told clients that they need to re-test because they changed a resistor, we wouldn't be in business for long.

I assume you have to recertify your new design, but a reasonable lab will decide that a change of resistor limiting current through an LED won't change the RF properties, so they'll certify it without testing it again. Note, that a resistor might also be there to set the bias current of your output driver, and the RF characteristics would change a lot if you modified that. Simply changing your design without notifying anybody probably won't stick.
Rudeofus
 
Posts: 27
Joined: Sun Sep 23, 2007 11:28 pm

Next

Return to Wireless

Who is online

Users browsing this forum: No registered users and 1 guest

cron