What to do with an idea?

[daqq]

Hi guys,

I've been doing some research in my free time into a measurement technique - basically, it's a lot of math applied creatively to a hardware setup. I originally did this to see if it was possible and thought it would be fun. The hardware involved is fairly simple and is used all the time, but is not used as a means of measurement. With the math it's possible to get some valuable data about the "measured" thingy.

I've constructed a hardware setup, did the measurements, did the software, the theory checks out, it works. Now I'm at a loss as to what to do with it. I don't think that I can commercialize it myself (as in sell the device) with much success - it's a technique that would be interesting to very specific areas of the industry, or maybe to a semiconductor maker that would cram the whole setup inside an IC. So kickstarter is not likely to work 

So, I came up with the following options on what to do:

Open full publication would be possible, however, there is very little to no profit - I'm not a member of anything, so a publication is nice, but at the end of the day it remains that.

Patenting is dubious - I've been looking into it and software and math patents are iffy at best and are very different depending on your area. Also, the money involved is well beyond my means.

Create the software setup, sell it and hope no one goes into reverse engineering and copying it without permission. I've no clue if you can copyright this kind of thing or how.


How would you go about selling/protecting this kind of idea? The way I see it I can't just come up to some company and tell them the whole idea, unless I have it protected, lest they use it without so much as a thank you.

Any suggestions? How would you go about this?

Thanks,

David

[Mechanical Menace]

Your only real option to keep it private and commercialise it is to obfuscate your implementation amazingly well and treat the maths as a trade secret. It won't stop reversing efforts if anyone is interested in the technique but if you're charging reasonable rates would make it more likely for people to license the idea under NDAs.

Another possible solution would be to release the implementation under a dual licensing model if you take the FOSS route.

[Galenbo]

... The hardware involved is fairly simple and is used all the time, but is not used as a means of measurement. With the math it's possible to get some valuable data about the "measured" thingy.

I've constructed a hardware setup, did the measurements, did the software, the theory checks out, it works. Now I'm at a loss as to what to do with it.

Like you already know, ideas are worthless, it's the implementation that counts.

Sadly the implementation also counts in money. Kickstarter, patents and selfimplementation really cost money.
But what about the market study? Is your product really wanted? Is it really working? Is it better than the alternative? Is it worth the money?

If it really works, you can freelance yourself to do the measurements for others. Build a bulletproof foolproof version and market yourself to companies.
Start low, as low as possible to pay for the rent and the gas. Have fun, make contacts.

[LabSpokane]

You sound like someone in need of a development  grant. I'm not sure about the market in Slovakia. But in the US there are oopportunities to do this through finding an interested party either in government or academia.

Most countries have a first to file patent system now. You could file fairly cheaply then pursue funding with interested parties. A well written non disclosure agreement can protect your interest while working with a reputable company interested in purchasing your idea.

Don't despair just yet. There are many companies that don't like doing fundamental development. If you've conquered that, your idea may be more valuable than you think.

[donotdespisethesnake]

There is almost zero chance you will ever make any money out of this idea, unless you can exploit it yourself.

It seems odd that anyone thinks they could get money simply for an idea, but the practical situation is that there is simply no market for ideas, other than ideas that are patented.

[atferrari]

To a similar question to yours, I recall the suggestion of doing an intensive search to see if anyone before, did something, if not equal, similar, that could have abandoned for any, (still unknown to you) reason. It made sense.

I liked the suggestion of you providing service by putting your creation to use.

[Tinkerer]

So it involves math techniques and software you have written. Why cant you tell us more about what you are measuring? No one here will glean a way to copy you, from you telling everyone what it is you are measuring. If they can, its likely something thats been done before.

[Galenbo]

... If they can, its likely something thats been done before.

It's even maybe something that's already patented.

[EEVblog]

How would you go about selling/protecting this kind of idea?

Keep it trade secret and protection can be had by many means. None are foolproof to reverse engineering of course, but given that its niche the odds of someone wanting to do that it small.
Build decent protos, offer it for sale on a formal looking website (make it look like you are a big company, instead of one guy in garage) and then approach people and/or wait until someone comes along who needs it.

Patents are waste of time and a huge amount of money. They don't protect your circuit, in fact they make it easier from someone to copy it. All a patent does is give you the right to sue someone if they do, and that's costs $1M+

[EEVblog]

The way I see it I can't just come up to some company and tell them the whole idea, unless I have it protected, lest they use it without so much as a thank you.

If your idea is as simple as giving the game away by telling them what it does and not how it's done then I'm afraid your idea isn't worth anything.
If the secret sauce is in the how then you have something to play with. Don't tell them that until they sign an NDA/non complete clause.

[janoc]

You sound like someone in need of a development  grant. I'm not sure about the market in Slovakia. But in the US there are oopportunities to do this through finding an interested party either in government or academia.

Ehm, market in Slovakia for something like this is pretty much nonexistent - tiny, 5 million people country with very little advanced R&D industry. In fact, most of our industry is around assembling cars for others thanks to low wages and cheap workforce. Slovakia is the largest car manufacturer in Europe, I believe. VW, Peugeot/Citroen, Kia - all have big assembly plants there.

And getting a grant - that's pretty much a hopeless adventure because of the corrupt, impenetrable and incompetent state bureacracy. Heck, even the universities and Slovak Academy of Sciences are unable to get grants through the government or only with insane difficulties. Private R&D grants are practically unheard of - companies will rather sponsor sports teams and political parties where they could get kickbacks and business from the state than science ...

And to Daqq - your have pretty much two options:

1) If you don't want to monetize your invention outright, but don't want someone else to profit from it without you seeing a dime, then do publish it, ideally in a large journal or conference catering to the field. That establishes prior art and nobody could patent it anymore, potentially preventing you from commercializing your own idea later. If it is something that is actually missing in the market, then it is likely that someone else is working on it too and it is possible they will try to patent it, not being aware of your work. You could end up locked out of your own idea.

E.g. I have a published paper from 1998 or so on multi-touch interaction, years ahead of the work of Johny Lee, Microsoft, Apple and their patents. I published it only in a small local conference back in the day, so if I wanted to actually get over their patents, it would cost me a ton of time and money now ...

Publishing is a much cheaper and easier option if you want to defend against patents than trying to get a patent in every country in the world - that costs a lot of money for both getting the patents and the fees to maintain them. Moreover, the only thing it grants you is a paper that says you can sue the culprit. Good luck trying that against some Chinese or Korean cloner or a big megacorp with deep pockets ...

2) If you want to commercialize it, then do as the others said - keep it secret and build a solid implementation. Then you could look for investors or a company to sell the license to. Or you start a company manufacturing your gizmo yourself. Having only an idea is useless - investors have "pie in the sky" ideas a dime a dozen, that won't convince anyone to put money on the table today.  Well, there are exceptions, but you probably don't want to go there if you have a genuinely useful and working gadget ...

 

[Corporate666]

Few thoughts...

-Be VERY careful if you apply for a "provisional patent" with the intent to follow it up later with an actual application.  Anything not included in your provisional is not allowed to be in your real application.  So if one of your claims included "blinking an LED" and your real application claimed blinking an LED at 5Hz, that isn't in your preliminary app and that's a problem.  In fact, a provisional application needs to be equally as detailed as a regular application, to the point that there isn't really any benefit from filing a provisional - just file the real one.

-Most companies of any significant size will not sign NDA's or talk to people about new products/ideas.  Just something to keep in mind... you should be able/willing to describe your idea pretty comprehensively without giving away the secret sauce or you will run into a lot of roadblocks to monetizing it.

-If it's a niche thing, then think about the actual commercial value.  Is it actually better, or just different?  What specific problem are you solving and/or what specific value are you bringing to users or implementers? 


Keep in mind that ideas have no value.  It sounds like you have more than an idea... you've reduced it to practice and proven it out which does have value, but nowhere near what an implemented and commercialized device has.  A normal licensing fee would perhaps be 1-10% depending on content.  So if your idea is a small part of a product (like a way to measure current that Agilent uses in their new meters), then your licensing fee would be tiny, perhaps a fraction of a percent of the device cost.  If your idea was a way to generate waveforms used in a waveform generator - that's what the device is all about and your fee would be greater.  If you had the patent on chemistry and construction of an X-Ray "LED", then anyone making such LED's by that method would pay a license and it would be on the high end since it's essentially a full implementation of your protected idea.


And I disagree with people who say patents are worthless... it will cost $10-20k USD to get one, but a solid patent most definitely does have value and can easily be sold/assigned to someone.  It also proves out the idea (although an untested patent isn't too valuable compared to one that has been tested in court).  But a patent will keep small players from copying your idea in most cases, and while the big guys might not give a shit and just willfully infringe, in a way that gives your patent more value.  Consider... you approach a patent portfolio company with interest to sell your patent.  Would you rather have nobody using it/infringing... or would you rather have IBM infringing with every chip they fabricate?

[daqq]

Hi guys,

First, thank you all for the advice and opinions! I really appreciate this!

Your only real option to keep it private and commercialise it is to obfuscate your implementation amazingly well and treat the maths as a trade secret. It won't stop reversing efforts if anyone is interested in the technique but if you're charging reasonable rates would make it more likely for people to license the idea under NDAs.

Another possible solution would be to release the implementation under a dual licensing model if you take the FOSS route.

Huh, haven't thought about dual licensing! Thanks, I'll look into it.

But what about the market study? Is your product really wanted? Is it really working? Is it better than the alternative? Is it worth the money?

The technique works and it is slightly better than the next alternative in a specific set of applications. As to it being worth the money, I don't know 

It seems odd that anyone thinks they could get money simply for an idea, but the practical situation is that there is simply no market for ideas, other than ideas that are patented.

I don't think that. I'm well aware that I'll either have to protect it by some means (if possible) or give it away.

To a similar question to yours, I recall the suggestion of doing an intensive search to see if anyone before, did something, if not equal, similar, that could have abandoned for any, (still unknown to you) reason. It made sense.

I did that - as far as I know it is unique - it's an improvement to not exactly the most popular area of industry. I've searched through google patents and found nothing. The net mentions nothing as well.

So it involves math techniques and software you have written. Why cant you tell us more about what you are measuring? No one here will glean a way to copy you, from you telling everyone what it is you are measuring. If they can, its likely something thats been done before.

My apologies, I'd rather talk on vague terms until I figure out what to do best

And getting a grant - that's pretty much a hopeless adventure because of the corrupt, impenetrable and incompetent state bureacracy. Heck, even the universities and Slovak Academy of Sciences are unable to get grants through the government or only with insane difficulties. Private R&D grants are practically unheard of - companies will rather sponsor sports teams and political parties where they could get kickbacks and business from the state than science ...

I see you know our country well 

Patents are waste of time and a huge amount of money. They don't protect your circuit, in fact they make it easier from someone to copy it. All a patent does is give you the right to sue someone if they do, and that's costs $1M+

To be honest that's the vibe I'm getting from the patent system - it has gone very far from protecting individual inventors but serves more like a battleground between megacorporations who have the resources to protect something and fight for it. The best a small individual can hope for it getting bought out by one of these. Oh, and then there are patent trolls.

If you don't want to monetize your invention outright, but don't want someone else to profit from it without you seeing a dime, then do publish it, ideally in a large journal or conference catering to the field. That establishes prior art and nobody could patent it anymore, potentially preventing you from commercializing your own idea later. If it is something that is actually missing in the market, then it is likely that someone else is working on it too and it is possible they will try to patent it, not being aware of your work. You could end up locked out of your own idea.

How would an individual go about this? Where would be the best place to publish it? I honestly haven't a clue. IEEE? My webpage and spray the internet with links?

Consider... you approach a patent portfolio company with interest to sell your patent.  Would you rather have nobody using it/infringing... or would you rather have IBM infringing with every chip they fabricate?

Assuming I'd patent it in Slovakia (for starters) would there be any actual interest from a patent portfolio company?

[Corporate666]

Assuming I'd patent it in Slovakia (for starters) would there be any actual interest from a patent portfolio company?

Slovakia (BEAUTIFUL country by the way and wonderful people and beer!) is a member of the EU and a signatory to the EU Patent convention (I forget the name of the actual law) - so you can get an "EU Patent" which, depending on the laws in Slovakia, might be able to be filed in Slovakia rather than in Germany.   An EU Patent should be able to be 'extended' into a USA patent, or at least will provide you with a priority date.  There are some specific rules the WIPO has authority over and my memory may be slightly incorrect on the matter, but at least a US patent can be leveraged to extend to a EU patent through WIPO rules and the reverse is also true.

Having a patent that covers the EU + USA is much more valuable than something local or solely national. 

As for what such a patent would be worth to a portfolio company, it all depends on their assessment of the market value of the patent.  The market value will largely be dictated by how solid the patent is, whether it's been tested in court, who would be interested in licensing or owning the patent and whether it can be extended into new areas.  But if you came up with a new way to generate waveforms that are much more accurate and can be done for 1/100th the cost... such a patent would be valuable to folks like Keysight.  If the patent was a way that a specific piece of equipment could do one task differently (but not necessarily better), it would have limited value.

You could contact an IP attorney in Slovakia and get their opinion - and you can probably do some Google research and get a feel for what the idea/technique may be worth.

[Smokey]

...
It seems odd that anyone thinks they could get money simply for an idea, but the practical situation is that there is simply no market for ideas, other than ideas that are patented.
...

No money for ideas eh?

Soap Router...
https://www.kickstarter.com/projects/soaprouter/soap-first-smart-router-w-touch-display-powered-by

Solar Roadways...
http://www.solarroadways.com/intro.shtml

......

Etc...
https://www.eevblog.com/forum/crowd-funded-projects/list-of-dodgy-crowd-source-funded-projects/

[Corporate666]

...
It seems odd that anyone thinks they could get money simply for an idea, but the practical situation is that there is simply no market for ideas, other than ideas that are patented.
...

No money for ideas eh?

Soap Router...
https://www.kickstarter.com/projects/soaprouter/soap-first-smart-router-w-touch-display-powered-by

Solar Roadways...
http://www.solarroadways.com/intro.shtml

......

Etc...
https://www.eevblog.com/forum/crowd-funded-projects/list-of-dodgy-crowd-source-funded-projects/

Those aren't ideas - they are both implementations of an idea... big difference.

The kind of idea people are referring to are more like "we should build a high speed train system between A and B.. who will give me a billion dollars to do it?" vs... "here is a train I am developing that will be faster than all others, and will connect A and B.  Who will give me a billion dollars to make it?".

[miguelvp]

Yup, Jules Verne didn't invent space travel, nor did Asimov invented the robot. And of course other Authors probably precede them as well but those didn't invent a thing either.

[janoc]

Slovakia (BEAUTIFUL country by the way and wonderful people and beer!) is a member of the EU and a signatory to the EU Patent convention (I forget the name of the actual law) - so you can get an "EU Patent" which, depending on the laws in Slovakia, might be able to be filed in Slovakia rather than in Germany.   An EU Patent should be able to be 'extended' into a USA patent, or at least will provide you with a priority date.  There are some specific rules the WIPO has authority over and my memory may be slightly incorrect on the matter, but at least a US patent can be leveraged to extend to a EU patent through WIPO rules and the reverse is also true.

Yep, you can do that. However, be prepared to pay really big bucks for each additional country where you want the patent to be valid. It costs a lot of money, And even then, that is just the filing and upkeep of the patent. You will still have to enforce it.

There is no extension to the US - the only non-European patent union (basically EU) states where the patent can be "extended to" are Bosnia and Herzegovina and Montenegro. If you want an US patent, you must patent there. It might be easier because of having an EU patent, but that is not guaranteed, probably depends on USPTO.

The other option is to file for an international patent, but that costs an enormous amount of time and money.

The English version describing the procedure is here:
http://www.upv.sk/?information-for-patent-applicants

You could contact an IP attorney in Slovakia and get their opinion - and you can probably do some Google research and get a feel for what the idea/technique may be worth.

Actually, the Slovak patent office (http://www.upv.sk/) is likely the best first point of contact - they will explain him what is he going to actually get if he decides to patent and how much it will cost.

[janoc]

I see you know our country well 

Tha't because I am Slovak, only living in France atm

How would an individual go about this? Where would be the best place to publish it? I honestly haven't a clue. IEEE? My webpage and spray the internet with links?

It is not that hard. You  find a scientific or industry journal relevant to what you are doing and submit an article. It goes through a review process by the editors, you may get rejected few times, but once the paper is sufficiently good, it will eventually get accepted for publication. The exact procedure depends on the journal, whether it is a scientific journal or more an industry/engineering oriented one. It is best to check the published articles in the journal you are targetting in order to have an idea about the quality and form they are expecting.

Where to find the journals - IEEE, ACM are a good start for technical publications, you could also contact the Slovak Technical Library (Slovenska technicka kniznica), they likely could advice you where to publish or at least give you access to some of the journals. They are typically paywalled and it costs an arm and leg to get access ($30+ for a single article is normal!) unless you are associated with a university that has a campus subscription.

The other option are conferences - you submit a paper to a conference, if it gets accepted you go there and present it. Then it typically gets published in the proceedings which have their own ISBN number and are indexed and searched by the major scientific search engines.

Another option is to team up with someone who has experience with scientific publishing - journal/conference papers usually have a quite formalized style and if you have never written one, it could be a bit daunting. Collaboration with a university researcher could be worth considering there - they have a lot resources and experience on the publishing side.

If you don't want to go this route, you could always publish in an industry journal that has a reasonable coverage (i.e. not only something national). Those usually have much less strict criteria for paper acceptance and are still a decent option for establishing the prior art.

[daqq]

A bit of an update: I've decided to publish it.

Would you recommend a publication in a magazine like Circuit Cellar or Elektor? Or rather some academic publishing paper - I don't have experience with those at all?

There's a fair bit of math, pretty simple actual eletronics.

[janoc]

Go for an academic journal if you can. Those are regularly searched for prior art when someone applies for a patent.

You can find a list of journals that could be relevant on the IEEE website, for example. Find one that is in the field of your invention and they have guidelines and everything on how the submission should be made.

[TAMHAN]

Hello,
sorry to "pop" up like this - but I also live in Slovakia. Bratislava, to be exact.

And I have both media contacts - am a frequent contributor to Elektor amnd Heise, among others - and, um, might also be interested in manufacturing test gear.

If you want to meet, let me know via an email to tamhan at tamoggemon point com and call me: 0940301347. We could meet in Trafo or something, or you could come to my lab (close to Sancova).

[daqq]

Hi guys,

So, Elektor rejected the article because it was too mathy and too niche, so there's that  Circuit cellar ignored 2 emails and one facebook message, so that's that. I'll try something more academic then.

TAMHAN, thank you, but I will have to decline the offer.

[TAMHAN]

Hello,
all my best in future endeavours. You know where to find me - just PM or email if you ever change your mind.

Tam

[Tomorokoshi]

Another approach to get some value of another kind. If you can publish it somehow, the value is that you can list the publication in your resume.