Slashdot Mirror
The Cult of the NDA
Anonymous Coward writes "After looking at hundreds of business plans during the tech boom of the late 1990's, and starting my own company two years ago, I've long been bothered by the near obsession with secrecy shown by many tech startups. This is especially striking considering how few startups are actually pursuing unique ideas. I finally wrote an article about this, The Cult of the NDA, where I argue that too much secrecy can actually hurt a company's chances. Open-source startups, anyone?"
You can be the first one, you know.
Free Software: the software by the people, of the people and for the people. Develop! Share! Enhance! Enjoy!
But I signed an NDA saying I wouldn't. Sorry.
How would you like to know that six months from now something is going to come out that makes the product you're selling obsolete?
Welcome to the computer hardware industry. There is something bigger, badder, and better just around the corner.
Intel's Roadmap, AMD's Roadmap and Apple's product line come to mind
As a rock-in-roll Physicist once said, No matter where you go, there you are.
It's called a bluff. By denying something, changing the topic, etc. you can make people think you have something. It's comparable to taking the 5th at trial. People KNOW what's going on, but it can't be legally used for or against you.
It's just like UFO's. At first the Govt said experimental aircraft were just aliens. Then it bit them on the ass. It still worked though. People weren't talking about the SR71 or the U2. They were too busy building a better tinfoil hat to protect them from martian rays.
if your company makes spellchecking and grammar software, send me a copy. I can use it for posting on /.
New year Resolution: Don't change sig this year
Don't you have any desire to present yourself as intelligent. If you are trying to sell your writings as intelligent, shouldn't you do a little proof reading first.
You don't talk about our business plan!
The second rule of our business plan...
The company I interned for over the summer had everyone under NDA: Our subject experts, recruiters, even us low level interns doing the real work. Our NDA was more for the fact that any of us interns couldnt run out and do the same thing, not that i'd want to. While It seemed like nothing at the time, even telling people what i did over the summer becomes a process of "umm, well, i can tell you that i programmed..." and them not understanding. But it was fun, and i wrote mountains of PHP that are being shown at trade shows.
"Something's wrong with you...and I hope we never do meet again." - Deftones When Girls Telephone Boys
First of all, even if you don't believe that a given company's ideas are "unique", chances are far better than not that they DO. Amazon.com probably honestly believes that "one-click shopping" is a unique idea, and that they deserved their patent on it. And you can bet your buttons that every programmer, managers and janitor who worked on that project signed NDAs out the wazoo.
As the SCO debacle should amply demonstrate, today's corporate culture is not about who's doing what uniquely, or even who "owns" what, but who can best convince/bribe a judge and/or jury. The business plans for many corps seem to be "Try to make money the old-fashioned way (i.e. selling useful products and/or services), and if that fails, sue somebody." To do that, you need reams and reams of paperwork, both to demonstrate that you were "duly diligent" in covering your butt (this is where the NDAs come in) and to document every little thing you've done. (Hence taking minutes of meetings, keeping archives of email, and other time-consuming corporate activities).
Honey, I shrunk the Cygwin
So many companies were being started so fast on such simple ideas that if someone else knew what you planned to do they could start up a company before you, or at the same time and compete. The lack of ideas and simplistic business models are also why the dot com boom ended so fast. There really was nothing special about it and it was easy to compete/use the same technology to accomplish the same things.
Ha! As if that towing company you started is really pursuing a "unique idea" ;)
NDA's are a very different animal in a large, established company. They don't just cover new technology, but product plans, personnel changes, financial information, employees' personal information, vendor's and customers' proprietary information, and all manner of things that a company has a duty to keep confidential.
Whenever I've shopped an idea around for funding, I haven't been to tight-lipped about what the idea was, because I've found that for the most part, people like their own ideas, and just aren't much inclined to steal mine. Getting them to back my idea is a lot of hard work. The real task in getting an idea to market is to convince the backers and the other participants in the venture that you have the right team to develop and deliver the idea.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
NDAs are probably most useful in a society in which just about anything, no matter how trivial or obvious, can be patented. That seems like about where we are right now.
Letting your competitors know anything about your business model gives them opportunity to undercut and end-run.
Your only real profit may come from being first-to-market and unique for a short time; the rest of your company's life will be spent breaking even or dying.
Secrecy is essential.
I would comment on this issue, but my NDA explicitly prohibits me from doing so.
Quid festinatio swallonis est aetherfuga inonusti?
Africus aut Europaeus?
Secrecy under the right circumstances, and with just a tiny little tease of information, can turn into a whole pile of hype.
Many of the most successful technology companies were not the first in their markets, but successful followers which learned from the mistakes of earlier trailblazers.
To take the most obvious example, Microsoft was not the first software company, the first OS vendor, or the first productivity software vendor.
You get the idea. More often, the long-term winners are the companies with the willingness to adopt good ideas from other places, and the flexibility to learn from the mistakes of the pioneers. Making mistakes can be very expensive, and being able to avoid certain mistakes can be a competitive advantage.
So what you're saying is that Microsoft's prominence in the world today has nothing to do with their illegal business practices. Okay, got it. Thanks for setting me straight.
Man, you got a lotta guts coming here and saying that Microsoft is the model we all ought to be following...
GMD
watch this
What was he towing?
eTow.com, your one stop source for all your towing needs.
That was the biggest problem with the dotComs, too much specialization.
"For a successful technology, honesty must take precedence over public relations for nature cannot be fooled." -Feynman
g#d freakign damit..
Ideas are not unique..
the business implementation and excution is what makes the idea a unqiue buisness!
which measn NDAs unless the business is fully established are useless..
Don't Tread on OpenSource
I've worked for several different companies, manufacturing, not software. ;)
And I must say, this is everywhere, and its a real annoying problem.
You sit through the same mettings no matter where you work.
Secrecy is important.. no photos, dont talk to family about what you make.. dont say this, dont answer the phone and talk about this.
And I agree. GET OVER IT
So what if other people make the same things you do, so what if they start making it cheaper. Whoever makes the best product, will keep making the product, the rest, can make whatever else they make and SO WHAT.
If everyone concentrated just a little harder on actually making sure you produced something that people would like, maybe your budget wouldnt be so huge with the 5 layers of people and meetings telling everyone to keep their mouths shut.
The second problem being how companies look at market share. I once worked Kraft foods, pizza, and they had at the time, 19% of the market share.
Problem, they were constantly whining they needed more.
WTH for! So you got 19% and a pile of garunteed income, STICK WITH IT, and stop being such a whiney bunch of greedy money grubbing corporate a-holes.
*There's Klingons on the starboard bow, scrape em off Jim!*
Yeah okay, I'll bite : my company, Acme Inc., makes a device that creates unlimited amounts of electrical energy from common household dirt. I was about to patent it and show it to big investors under serious control. But now I'm convinced : where should I upload the blueprints for all to see ?
...
I mean come on, I know it's Slashdot, but let's be serious for a minute
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
Consider the case where your business opportunity has a very limited time frame.
1. you have knowledge that circumstances in an area are about to change (Airport being built, large employer moving out).
2. Your'e business idea is fragile and competitors could concievable kill it still born if they knew about it. ( You have a customer list, that a competitor might litigate against you using is an example)(SCO's business plan is another example)
3. You feel you really are doing something novel and you dont want everybody and their brother doing it. (Ever notice that hollywood will make half a dozen movies at a shot that all seem to be the same idea ?)
More importantly most people don't like blabbing their business around and if they have to tell others what their business, and they certainly don't want their potential investors going out and blabbing.
The idea is somewhat linear, just applying software on a large scale to an area which has never had it before, but the execution is vastly more difficult.
Through luck and creativity, it appears that I've found various ways to execute this. Two and a half years to build one part of the software, another couple of years of building and testing, two years of working for "the man" to pay off my debts run up during the first four years.
Now I've got this fairly decent product which I've just started to roll out to some large customers. And I've had many people sign NDA's along the way. If they are violated, would I have the money to pursue the violators? No, of course not. So that makes them worthless, right?
No, they're not, because people don't know that I don't have the money to fight. So NDA's are just a harmless bluff for me and probably everyone else. But in the interests of thoroughness I should use them.
Yes they're useless. Yes you should use them. Not everything has to be useful to be used.
The baby's fine -- please stop sending business cards.
Just for comparison's sake:
W .h tml
/EJS
"Full Disclosure on Full Disclosure"
http://www.nytimes.com/2003/09/27/opinion/27BLO
"Confidentiality agreements were once primarily used to protect commercial secrets. More recently, celebrities have adopted these contracts to protect themselves against checkbook journalism and embittered assistants. This isn't such a big deal. But increasingly, confidentiality agreements ban their signers from revealing information that furthers more meritorious public debate. The Catholic Church, for example, used them to silence victims of sexual abuse by priests, possibly allowing that crime to continue longer than it otherwise might have.
These agreements aren't made merely out of a concern for privacy. Confidentiality agreements have become a tool used by the rich and powerful against people who can't afford to turn down a job, as a way to stifle public discussion of embarrassing issues, and as a means of ensuring that a whistle-blower can't throw a wrench into the image-making machinery of a public figure."
Read the rest of the editorial to fully appreciate his view point.
While the NDAs discussed in the Slashdot article and NDAs discussed in the Times editorial are different beings (one to protect potential business plan secrets, the other to protect public debate), they still point towards a disturbing trend to use these things in almost all circumstances where they can possibly be used.
Make discussion and conversation illegal, and you've just halted the exchange of ideas.
Just some food for thought.
>I've found that for the most part, people like their own ideas, and just aren't much inclined to steal mine
I can't remember who said this:
"Don't worry about people stealing your ideas. If your ideas are actually good and helpful, you'll have to cram them down people's throats at gunpoint."
Don Lancaster has made the same point about secrecy as the article did. There's only one smartest person in the world, the odds are overwhelming that it's not you, your idea will have occurred to someone else, and the way to make money is to kill bad ideas quickly.
If you prescribe to the beliefs of the FSF and do not believe in intellectual property the only conclusion about the topic of NDAs that you could make is that they are immoral. Information is by its very nature free. Nearly ALL encryption is breakable*, which to me proves that information is inherently free -- all of our manmade attempts to confine information have failed or will at some time fail given adequate time and technology. When a word or idea is expressed it can not be taken back - it belongs to the public. The whole concept of encryption is completely backwards. As an altuistic society we need to look deeper within ourselves and understand that only through cooperation can we succeed as a society. Encryption (and the secrecy of NDAs), though I admit I find it to be a fascinating subject from a purely technical standpoint -- I have read and researched quite a bit on the subject, is just another man made technology which makes the claim to be 'progress' which actually represents a social and spiritual regression. NDAs signify our distrust of others and is yet another thorn in the side of humanity and society which is preventing us from attaining some higher understanding or spiritual enlightenment if you will. *The only case I can think of where encryption would be unbreakable is when some data is encrypted with a one-time-pad and somehow the key has been completely lost from anyones memory.
I'll start with the obligatory acknowledgement that a ton of bogus patents are issued.
But some patents are legitimate.
If a technique is publicly disclosed prior to filing a patent, then a one-year timeclock starts in which you must file for a US patent on the technique or lose the right to file. In addition, the ability to file is lost immediately in Europe.
So, NDAs are very useful if one intends to file patents.
The NDA has taken the rap for being super-secretive. However the NDA covers more than just technology or a product. It also covers expressions and variations on ideas. It's not a bad idea at the beginning (before you produce it), because the NDA allows you to feel free to openly discuss things with others, without the possibility of them running away with your variations on thought processes. Maybe it just provides a bit of mental ease.
Rarely is somebody going to run away with your idea anyways, because they already have their own.
The funniest is when you were already toying with an idea, and then somebody else asks you to sign an NDA, and then you find out in the meeting that their idea is very similar to yours. Then what do you do?
The first rule of NDA Cult is - you do not talk about NDA Cult.
The second rule of NDA Cult is - you DO NOT talk about NDA Cult.
Third rule of NDA Cult, someone yells "Stop!", gets an attorney, or finds shredded documents then deny all knowledge.
Fourth rule, only two guys to a fight(When beans are spilled).
Fifth rule, as many NDA's at a time, fellas.
Sixth rule, no wires, no cell phones.
Seventh rule, NDAs will go on as long as they have to.
And the eighth and final rule, if this is your first night at NDA Club, you have to shut up about what you see.
... a device that creates unlimited amounts of electrical energy from common household dirt.
You must not be married: if you were, you'd know that the spark goes out, and there'd better not be any dirt in the house.
-kgj
At my last position, I was in charge of winning new business, particularly at startups. The evaluation process was challenging to say the least, in terms of funding, management experience, etc..
One particular company, involved in one of the latest and greatest tech crazes, had requested that we design ASICs (custom chips) for them. At our first meeting, they made it known in no uncertain terms that we (the vendor) were not to ask any deeply probing questions regarding the end product, or they would promptly and henceforth terminate all discussions. This was the first clue, since we already had a comprehensive NDA and it's a very atypical request.
At one point in a subsequent discussion they wanted to include an embedded processor from one of the well-known embedded processor providers. They claimed another vendor could run it at some number of MHz, and asked what we could do ours at. I replied that I wanted to understand some basic things regarding the processor's use - MIPS rating, types of ops like multiply-accumulates, any add-on DSP functions, etc.. The response I got: "Well, we're not sure, we just want the fastest one." I said "we are quite comparable in process performance and can rework it to at least as good a standard depending on your needs." The impression from other engineers working with me was that they had no clue what they were doing.
Needless to say, soon thereafter they had undergone a major reorg and we didn't hear anything until many months hence, when a different individual with very specific requirements came by and who was very easy to make a business case for. As we found out, the thing that made them "special" had nothing to do with us directly anyway.
The point I'm trying to make is that, in a rush for secrecy, you can end up hiding a lot of the issues from vendors, customers, and investors. Most startups that I have dealt with are basically taking one or more old ideas and adding their "special sauce" to the equation. That "special sauce" needn't be revealed, but if they want some cooperation and funding (and ultimately survival), they should be a little less secretive. At best, you will appear incompetent, and at worst, you will look like so many swindlers in the business world before you.
Without an NDA, what's to stop the VC going to another startup, and saying 'Your idea sucks, but here's a better one, and i want 60%'?
A handshake?
This sig left unintentionally blank.
I worked for an "open source startup" a few years back. We were producing accounting software for a specific market niche. No NDAs and public betas of our upcoming products. Worked alright, except our competition always seemed to have our latest features in their latest releases. Wonder how they knew? Seriously, it would be nice if there wasn't a need for NDAs, but truth is, unless you hold something back, your competition will eat you alive.
The Case Against Patents
What does he say about NDA's? Publish your ideas in trade journals ASAP!
He's a wise old man. Go read his whole site. It will do you, and the economy good.
roflmao
Interesting thing about IP and stuff. A paragraph, a sentence could put tens of thousands to work or just the opposite.
"Don't worry about people stealing an idea. If it's original, you will have to ram it down their throats." -Howard Aiken
Having said that, I have seen two large organisations work themselves into NDA deadlock over something that both of them agreed would be mutually beneficial. In large organisations, NDAs also place an extra cost on any simple transaction, as the legal departments go to work.
It occurs to me that start-ups use NDAs as a sort of mixture of puffery ("see, we're big, too") and cargo-cult behaviour ("the big companies use this, so if we behave like them ...").
Open source startups? Simple: Red Hat is just one of such examples.
Less is more !
I often have to sign an NDA, sometimes just to get a job interview. If the terms are reasonable, I have no qualms about this: the agreement is just a written form of an implicit agreement I see as part of my professional ethics. If somebody trusts you with sensitive information, it is simply wrong to be careless about passing that information on.
It occurs to me this argument is partially about the attitude gap between the open-source (or "free") software community and the closed-source (or "commercial") software community. Thing is, these two communities don't have to be enemies. Yeah, some OS people think that commercial software is evil, and some commercial software people think that the OS movement is economically clueless. But the reality is that no one model is the best possible one for all kinds of software. Some projects will prosper if they're driven by volunteers who just want to advance the state of the art. Others will only succeed if they're driven by well-capitalized entrepeneurs out to make a buck. Neither model is likely to go away, and I predict that more and more companies will come to rely on both.
I was just amused to see this paean to openness being credited to "Anonymous Coward."
An NDA is a tool. Properly applied, it can be a very powerful protection as well as a marketing tool. Overused, and it becomes useless and unenforceable.
Regardless, personal integrity is the key to real secrecy: don't hire/work with/partner with people you don't trust. And that trust has to extend to the decision to share information.
-- $G
I work for a small research and development company, and yes, we have very strict NDAs. You need one just to get into the building. But in our case, we need them because we actually do develop new ideas. And I don't think that saying that all these tech companies aren't developing new ideas is giving them a fair shake.
We are much more of a tech company though. We develop new technology, but we also do a lot of design projects, for instance the SmarTruck 2
While secrecy in some areas clearly helps, IMHO - the real thing that won the cold war was the fact that intel (and many other US industries and products) could doubble it's chip speed every 18 months. The russians just couldn't compete against that kind of growth, that kind of economics even though the processes and technology wern't government secrets.
This is even more so with terrorisim.
This reminds me of the timely adaptation of the corporate debacle by Scott Adams, in The Way of the Weasel (or was it Dogbert's Management Manual?), where Adams maintains that the only way to tell if a project is succeeding is by counting the meetings.
/. might fall under this assumption, although I doubt anyone would require an NDA for that.
Doesn't figure what the project is, or if it's making any money. Just how many meetings and if there is more room to gloat over a budget for staplers or more meetings at JJ Muggs, or insert any other yuppie bar & grill.
I'm guessing it doesn't figure what was even said at the meeting, or who was too stoned/drunk to speak clearly, or who was even there. Just that one took place and important meeting stuff happened.
NDAs might also fall under this assumption that more clutter means more success, but maybe that is only true to especially idiotic judges and juries.
Sometimes the more that is said, much like this comment of mine, the more it is believed that something actually occurred that was tangible and worth something to the audience. Something worth karma at
File a patent on all new ideas. If you're lucky, one of your visitors will implement your idea and you can collect. The other reason you want to file a patent, is defensive. Otherwise, the company that just visited might file their own patent, and when your device is ready to go to market, they'll shut you down.
The other comment I have to the article, is that it is not always such a good idea to have something new and unique. In the very early phases of funding, VCs or angels have no idea if it will fly or not. If they can find at least one other company with the same basic idea, they consider that a validation of the market, and thus, the investment to be less risky. Being first on any market is such a two edged sword. Anyonme remember the Apple Newton?
-- Another senseless waste of fine bytes.
A number of startups that I've dealt with are wrongly convinced of the great importance of what they're doing. They sincerely believe that IBM and Microsoft are spying on them and tracking their movements, desperately trying to learn their secrets.
To most of them, I would say: you have only a few lines of crappy code and a silly idea. Microsoft and IBM don't care about you, and neither does anybody else. When you have some innovative complicated algorithm, or a large code base that's difficult to duplicate, only then should you worry about protecting your IP, not before. Having IP worth protecting is the difficult part; keeping secrets is comparatively easy.
I agree wholeheartedly with the message. Too many startups are under the impression that it's the product and the protection of the product that is to make their company sucessful. Ask yourself if you could make a better hamburger than McDonalds? It's the system that makes a business... how well the business is managed, how sales are produced.
Of course the product matters, it has to be something that the market needs or thinks it needs after you show your clients how much it will improve their lives. Produce the best product you can but really more companies should focus not so much on the NDA but on developing their business.
I helped start up an Internet business in '96 that sold for three-quarters of a billion dollars in 2000. We produced innovative products but so did our competitors however we produced a system that allowed us to have better customer service, instant reporting and faster turnaround times. It was the models that we created that attributed to our sucess far more than our products. Of course most people reading this will say that it was just an issue of timing and they are probably right with respect to the purchase price of our business but not the fundamental growth that we produced.
Football Sports Contest - Win $500 for having an e
I love Open Source and the GPL. Most of the solutions I design rely heavily on the work of others. But I also know that if I sell it / distribute it, all the changes, all the secret and unique little ways I have chosen to leverage the Open Source, well, it too becomes Open Source, and then the guy down the street steels my code, my business, and my business plan, and I have to give the Volvo back. So how do I deal with this and still reap the benefits of other's labors? HOSTED SOLUTIONS!!!! YEHHHHHHH! The software ain't for sale, just the use of it.... And I'm still "clean" in the eyes of Open Source / GPL (legally, that is...)
That's the longest and wackiest anti-IP rant I've seen in a long time.
Naive is a word that comes to mind.
Did it ever occur to you that we're not all one big happy family of earthlings all trying to get along?
There *really* are people out there that will kill you, rape you, rob from you, and then blame you afterwards.
If you live some place where that's not true, let me know. I'm moving there right away.
... since the domain registration info for the blog is so readily available?
It's nice this time of year. Pigs fly, microsoft offers a competative advantage and the politicans place the responsibility of their job foremost.
Well - I have tried a few times to get funding for ideas that was partly based on Open Source. The problem is not the engineering side but the VC side of it. I have yet to find a way to convincingly explain to a VC that by using open source I can cut the development cost down to 1/10th of what it would have been if I was to develop everything from scratch, and that it actually will lead to a better quality "version 1" product. But VC's seems completely focused on the IP of the product. Without that it seems that even a good business based on knowhow and delivery of services are unworthy of investment.
I for one would love to hear some good arguments that could convince VC's to invest in Open Source based projects.
COOTYS RAT SEMEN... err, no. Make that:
TOO MANY SECRETS!
NDA's have become so stupidly overused because groups of 20 year old guys, with dreams of millions, who think that a business model is that nude woman wearing a tie on their desktop, want to feel like a real company.
I've been asked to sign NDA's for some pretty stupid things - "We need some audio for this game - it's a first person shooter... but it's set on the moon!" "Our system actually lets the average end user change the text on their websites!" For fuck sake.
The idea of "trust no one" is ugly and insidious. I have recently switched to Linux and it took me a VERY long time to understand the Open Source model. I still can't believe I can change the source if I want.
Tell others about open source and their reactions range from "You computer hippies" to "You terrorists" (Ok, mostly its "stop talking about computers you nerd, this is a pub").
People get pretty protective of their latest greatest idea. Which is understandable. The main problem with keeping secrets is that YOUR IDEA IS NOT ORIGINAL. That's not a bad thing but really, just do it and stop trying to stop others doing it.
Well way back in the previous century, in one of my Army Professional Development Courses we were taught that one of the "lessons learned" from the Grenada invasion was "do not sacrifice your mission to operational security". Seems that is a "lesson learned" just about every five years by the military. Also seems that it is *never* on the lessons learned list by startup businesses.
Eve Fairbanks says I drive a hybrid!LOL
Once we come to market, sure, somebody else will be able to make copycat products (limited perhaps by our patents), but we still get the first mover advantage with a new kind of device.
Now in this case, there is genuine uniqueness to the product. With a lot of software companies, their uniqueness is all about 2 or 3 features that somebody else can easily nab and throw into a competing product. In fact, since "manufacturing" in the software business involves clobbering in some new features and releasing a new build, which can usually be done in a few days, the time from reading some documentation to coming out with 1-for-1 matching feature sets is often measured in a matter of weeks, not months or years. I think that's why people in the software business are so paranoid about NDAs - keeping featurization and product details secret until it's on the market.
I think the other reason is many companies don't want anybody outside to hear how ugly and dirty their software is and what a big nasty hack it was to kludge it all together.
So does this change the...
1. [Insert crazy idea or technology here]
2. ??????
3. Profit!!
business model to...
1. NDA
2. [Insert crazy idea or technology here]
3. ??????
4. Profit!!
???
-- Gun
-- Stu
/. ID under 2,000. I feel old now.
... is that the author of the article fails to state how NDAs harm the chances of the technology startup. Basically the argument is that the company will probably fail anyway, so why bother? The simple fact of the matter is that the secrecy is not hurting the company, and there are a number of circumstances where the secrecy can help the company.
While it's true that NDAs can hinder a startup in its early stages, the same kind of thinking often poisons contract negotiations and relationships with initial customers and vendors alike. Living in constant fear (along with the legal bills that come with it) can choke a company with paperwork and expenses when it should be concentrating on solving problems and moving the business forward.
As someone who does business with startups, I can tell you there is no better protection for the company's future business prospects than effecient negotiations followed by outstanding performance. The whole purpose of buying a product from an outside source is to solve a problem more easily than would be the case with building a custom solution. When stacks of NDAs and contracts appear, the balance can easily tip back to an internally developed solution or towards the not-as-appealing but easier to procure (and probably less risky for the project team) product from a bigger company.
it was amply evident that apple would be the platform.
because it was the happening place for postscript and layout, which meant images. images mean film and film for print means thirty to one hundred fifty meg image files.
so therefore afterward and henceforth, itty bitty image processing was a done deal and export of EDLs was the trick to bring the entertainment industry onboard.
was commodore going down the commode about then or did you figure that play would do their own??
onboard scsi always did rock out like wesley willis.
Any preoccupation with ideas of what is right or wrong in conduct shows an arrested intellectual development. (Wilde)
He's well aware we can find out who wrote the article. But because he AC'd, we don't know his /. identity. Just a little touch of anonymity on this end that he might want to keep.
Do you want everything you ever said on /. tied to your real name?
All's true that is mistrusted
NDA's may be frivolous, but given the frequency with which companies are dropping lawsuits these days would certainly inspire me to a more conservative approach to what I do and don't let people disclose.
...
Plus, startups aren't so much a dime-a-dozen these days, so it would seem not so likely that a million other people have the same idea you do.
I can't argue with most of the points in this article, but I think it's missing the real reason behind NDA's. If someone else who can do exactly what you can do, only better, comes along and does so, then they DO have an advantage over you.
The logical response to this might be "but then if they can do it better than you, then you're doing something wrong" - which might be true, except that some of your potential competitors might be well-established in other markets, and are interested in venturing into the market you're entering or creating.
2) The first company to capitalize on a new product or idea has a unique and sustainable advantage.
Many of the most successful technology companies were not the first in their markets, but successful followers which learned from the mistakes of earlier trailblazers.
To take the most obvious example, Microsoft was not the first software company, the first OS vendor, or the first productivity software vendor.
IBM was not the first computer company.
Not to get pedantic, but this argument doesn't hold water - just because the first company to an idea doesn't ALWAYS succeed doesn't mean that they don't have an advantage.
NO google wasn't the first search engine, but Texas Instruments designed and developed the first Integrated Circuit in 1958, and they're still very active in that market (did a report on TI a year ago). There are certainly other examples out there. The point is, for every company that doesn't make it despite being #1 to market, I'd be willing to bet there are a nearly proportionate amount of counter examples to support the claim that being first to market is an advantage - just not the end-all of success.
Moo
I think the major reason for NDA in start-ups is because of the vulnerability of them.
When you start a company, you need to build the whole enterprise-system (I don't talk about the ERP, but the "how the company work). So you are not as quick as a well established enterprise. So you may not be able to move fast enough to counter-attack if somebody is trying to copy you. So you need to protect yourself.
Montreal - Best city to live in!
Well, I guess it's hardly surprising that two people involved in the same kind of business should have the same views, but still...
A Good Hard Kick In The Ass, by one Rob Adams. I borrowed it from the library (sorry Rob :) mainly because I thought it was a catchy title... but it turned out to cover almost exactly the same notions as included in this guy's "Cult of the NDA" article (except, obviously, in a lot more depth and with a lot more entertaining anecdotes). There's an entire chapter of the book that essentially just says "You think you have a unique idea you need to keep secret? You don't and you don't."
Adams also insists on what he calls "execution intelligence" being one of the key pillars in a business-that-might-have-a-chance (as opposed to a business that has no chance because it's still hung up on worshipping its own not-unique and not-even-very-good idea).
Some good stuff in the book. Certainly worth a read - even though a few bits of the book are rather amusing from a post-dotcom perspective (the book was published in 2002 according to Amazon, but I suspect most of it was written quite a bit earlier).
Pete.The story goes...
Rodenberry pitched Star Trek at CBS before NBC. CBS appeared to love the idea and kept him there all day discussing plots, characters, the whole shebang. At the end of the day, assuming such interest meant they wanted to pursue his project, he asked "What next?"
They told him they had no interest in Star Trek, but wanted to hear his ideas because they too were coming out with a space exploration show.
So there have been four Star Trek series, what, 8 movies, books, etc. And just one Lost in Space series and one movie.
I have a world changing idea. (Yeah!) It took an enormous effort for my company to build it. (OK.) It has a market waiting for it. (Good, maybe we can make some money.) There are several big companies that have the resources to reproduce our effort, and once we are successful, they will have the motive to do so. (Uh, oh.)
But wait! There is an answer. This great country (U.S.) has decided that software ideas can be patented. Now this is usually done by those big companies to raise the barrier of entry so startups can be killed or at least threatened if they become competitive. Luckily our idea is far from mainstream and requires much specialized technical knowledge to implement. We will be first to market, and our solution answers many issues that are in the news today.
But we cannot advertise the features AND get a patent on those features UNLESS every prospective customer signs an NDA. Once the NDA is signed, the customer is considered part of the development team, and does not count against the application for a patent.
---
This is a true story. In our favor, several of us involved in the effort are consultants to our target market, and have the contacts to create a market for the product under the terms specified above. We are also able to identify that our product is unique. To answer the article:
1) A unique new product or idea is essential to a startup's success.
In this case, yes, but only because we need the patents before a big company decides to clone us. The business plan includes the need for patents, and NDAs are required for the patent application to be successful.
2) The first company to capitalize on a new product or idea has a unique and sustainable advantage.
No, we may be unique today, but that is not sustainable without patents. We are not counting on being "first", we are counting on being "only".
3) I have a unique idea for a new product or service.
Yes, because the idea requires much technical ability to implement, and (so far) no one else has thought of it. And because of our unique skill set, we would probably be contacted to help if anyone planned a similar product.
4) If others find out about my unique idea, they could bring it to market first, and steal the advantage from me.
Yes, but today they would not care. We are protecting it to get the patents that will protect us later when they do care.
5) Therefore, by disclosing my idea only under the strictest confidentiality, I preserve an advantage for myself.
Yes, we need the NDAs to get the patents to have an advantage later.
We also have what it takes to make a startup successful:
- Dumb Luck: We have customers before we have a product.
- Execution: We have a business plan, and several high-level managerial types that believe in it.
- Stay focused: This is the hard one. If we dedicated the resources to the product, it might have been released a few years ago. We actually thought the code was done, and had dates for installing it at customers, and then found a major problem with the platform under it that causes the product to crash under heavy load. Oops. And we have consulting careers that limit the time we dedicate to it. We believe the product can be delayed because the IT world has stagnated. So far, this is accurate.
I spend my life entertaining my brain.
Then I'll direct you to a URL with my comments on this article.
John Kerry is a Joke!
Just imagine: I've got a secret SO big that, if you want to learn about it you must promise not to reveal anything about it. Boy, how important am I!
The more I think about it, the more I realize that things have not changed much since my days at school: the bullies remain bullies, the assholes remain assholes, and all carry on playing the same stupid games.
I'm surprised the complaint is about NDA's and NonCompete's. The author must live in California, where noncompete's are banned (or, don't have much teeth).
Here on the right coast of the US, NonCompete's run wild. It's very common for engineers to be threatened by their previous employers Lawyers for taking jobs at companies in the same industry.
I can see the need for secrecy in the middle of development (I don't want my competitors to know my release and feature schedule, for example). Once my product ships however, at issue is the artificial restrictions placed on other people learning from my product (such as Patents and DMCA).
I own a software outsourcing company based in India. We rarely work on unique or innovative ideas. We do whatever work the client (mostly US based) gives us. Do I need my employees to sign a NDA? You bet!
One of our clients had some work done by another company here. They were stupid enough not to ask for the source code. Some time later one of the programmers (of the Indian company) contacted the US client directly and said he had all the source code and would put it on the Internet he they didn't pay him certain amount. He said this would also expose several security loopholes in the software. My client had to pay him but since then they have started working with us.
Moral of the story: You decide...
I, for one, have agreed not to disclose the location of our new vc overlords.
We actually said no to the first NDA they came up with, which had some pretty rediculus clauses (I can't remember the exact terms, but I think it said something like not being allowed to work at another "computer company" for 2 years or something, and a few other stupid things). Anyways, the real guts of the NDA were to protect the company from people leaving and going to other companies, which I don't think is unreasonable.
Basically, it was the standard sort of "don't tell our competitors what/how we're doing". It also said that employees were not allowed to go work for other companies doing the exact same thing for 6 months (something like that). I'm sure a lot of programmers on /. will cry out at that.. but really, it makes sense.
If I were to get a better job offer from a competitor, it could be damaging to the company I was at, as well as giving the competitor an advantage. I would have inside knowledge of what's implemented, how (a particularly good one), and what's planned. If they were trying to catch up with features, I'd have some insight into how to do it since I'd done it before at the other company (ie, I know what not to do, what customers thought (if there was feedback), etc). Basically, the competitor would get a version 2 of everything, without going through any problems that might have been found with version 1.
The current company would also have to find someone to replace me, and then train them about any specifics of the product. Obviously, whoever they hire won't know the codebase as well as I did, so they'll set back the project timeframes a bit while they get up to speed. That alone could make ir worth it to steal staff from other companies.
While I agree with some of the points in the article about ideas (which /.ers have already talked about quite a bit), they are necessary for some things.
Speak before you think
Blank NDA's with you and each time you get drunk at a bar have someone sign one because you have a great story to tell. After every one has a NDA on each other our fear of reality will go phfft into a black hole and every one will be safe because it their idea gets stolen they will have a legal reason to destroy someones life. Sort of if everyone has a gun there would be alot less crime. But then there will be silence when someone ask if anybody has heard a god joke latly, but that will be good won't it?...Just rambling and someday I won't stop.
I eat my grapes at room temperature, cuz the cold ones hurt my teeth
the bitch is ugly
Here's my thinking in regards to ideas. I think in general it's best to let everyone you can know. Here's why.
Let's say you have a good idea - great, 100 other people probably have that very idea as the top priority to work on right at that moment.
Of those 10 people probably have funding to produce and market the idea and are feverishly working to do so.
Now of course that just puts you on equal footing with them. What's the worst that can happen? Someone beats you to a patent. The best counter? Publish early and often! It's cheaper to produce prior art than patents, and keeps you on equal ground with everyone. It might give a few people a leg up, bet it's better than having someone patent the ball you're all playing with and end the game altogether.
There is of course a time and a place for patents, but a lot of ideas are ephemeral and not worthy of such deep protection for the inventor.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Larry Stark, a professor at Berkeley, tells the story of sitting around with a group of other researchers including Warren McCulloch (of the McCulloch-Pitts neural net model), who was expounding on his current research ideas. At one point Larry broke in and asked "Warren, why are you telling us all your ideas? Aren't you afraid someone will steal them?" to which McCulloch snorted "Steal them?! I can't force them on my own graduate students!"
Right, so what's not to understand? How many companies really want everyone to know that they're working on boring unoriginal projects? Especially in a bubble economy, hype and speculation could mean the difference between a billion dollar IPO and a quiet death. Really, if you don't have anything worth investing in, isn't it better to make people think you might by being quiet and secretive, so you can swindle them and get out of town?
Nothing has changed; snake-oil salesmen never described the ingredients of their medicines, claiming only that they were "patented" or "secret" - no sense trying to sell sasparilla at 100 times the cost, is there?
IBM lawyers are known to be very about NDAs, and IBM personel are only allowed to sign them if they meet certain conditions. Is there a sample NDA posted anywhere that meets with IBM approval? That would be useful whenever a plain vanilla NDA is needed.
Picking up chicks! I've often had conversations go like this:
/. denizens don't get to screw enough girls, so why not kick start YOUR sexual career by talking out your arse some more!
;)
Girl: So, mr boring gentleman, *yawn*, what do you do for a living?
Me: I'm sorry, i can't discuss it, i'm under NDA.
Girl: *conspicuously not yawning anymore* What the hell is an NDA?
Me: Non Disclosure Agreement. I'm legally bound not to discuss most elements of my employment.
Girl: Oh come on, you can tell me!! *twirling hair now*
Me: Hah, yeah well... I suppose you're ok... I can tell you that I provide expert consultancy services for a large national concern. More than that, i cannot say!
Girl: "national concern"? does that mean the GOVERNMENT?
Me: I could tell you, but they'd kill us both... *winks*
And that's the bit where you get to take them home! Come on lads, all I hear are stories about how
ps. I actually *AM* under NDA. Not for the government though
Perhaps the answer to the problem of teenagers dropping bricks from motorway and railway bridges is to sue Tetris.
As a person who have witnessed the start up of at least two tech companies very closely and currently attempts to set up a new company together with some colleagues, I just want to agree with the headline (haven't read the full article yet and won't have any time doing it today) :(
Recognizing that our idea isn't that original (although it contains a quite original mix of other ideas), we have been very open about what we intend to do towards everyone we've been talking to in the business. This has paid out very well so far in the way that potential business partners that we didn't even know of have stepped forward (somebody we spoke to knew somebody who... etc) and approached us, which has opened up fantastic possibilities for us. We today stand on a much more solid ground, with the access to a wast network of complementing competence and established business associates who are interested in helping us getting a good start, since they see that our business can benefit them in the long run.
We are not too worried that somebody with the insight in our business successfully would copy our ideas. We have allready created momentum through this backing and established good relationships with the parties around us. Anybody who would attempt to take over our position in the network would have hard to fit in our shoes.
Therefore I'm more worried about the parties further away that we have no knowledge of and independently are going for the same business plan, backed by another network the same way we are and are likely to become fierce competitors 5 years down the road...
I have been asked to sign NDAs that would have given all my design work in a broad field to their company forever. I used to cross out most of the agreement and later send them one of my own as an alternative, but now I just tell them to get lost if they need a NDA. Turns out, that the folks that want NDA's spend so much on legal fees they never get profitable anyway.
Use a frienda... pronounced Fren D A
It isn't as legal as an NDA but it is more social and requires a hell of a lot less legal paper. You also have the advantage of breeding trust intead of distrust.
Just make this clear... if you break the FrenDA you will always point out how you got screwed everytime that person's name is brought up in public.
This is pretty much how everyone in our startup works right now. Our CEO even believes in it.
Honestly... how often are NDAs even enforced? They're not. They're crap.
Startups have to focus on shipping code not on suing people.
Kevin
Sure it is a way to get extra value out of a patent, just not a very nice way.
... but counted in months, not years.
The patented idea is yours from the day of application unless someone beat you to it, secrecy is only usefull as a way to play the patent system for all its worth.
Patent applications should really be published, if necessary with a small delay
Far from other people stealing your idea, you may put them off. If you have several startups working in secret on essentially the same idea, they'll probably all come to market within a few months of each other. Specially if, as in most cases, it was an idea whose time had come. So they will fragment the market and none will do well.
If you go public, you will probably put others off starting up projects to do the same thing. If you pull your PR bullshit well, you might possibly drive others to pull out. "If they are ready to go public, they must be ahead od us, so we might as well pull out".
If the idea is really original, you shoudl be able to get a patent on it. And a patent is a form of going public, so once you have made your initial filing, shout up. And a patent is likely to impress the VCs too. But if you can't patent, bullshit. It works for evryone else, so why not for techies?
Consciousness is an illusion caused by an excess of self consciousness.
It seems this discussion has gotten a little off track with so many ideas from /.ers throwing opinions around.
;-)
Here's the track record I've noticed:
Companies funded by VC's or public investors are much more likely to put a NDA in front of every employee, contract worker, visitor, etc. I always figured it was to make the old investor, who only understands the general business model (the money side), feel more comfortable when he sees a company going to great lengths to protect it's information. Think back, if you've worked for a public traded or VC funded firm, all the money wasted to impress share holders. One ISP I worked for spent tens of millions of dollars making their offices look nicer a quarter after we had posted some massive losses. There was literally no hope of breaking even, yet they felt we needed new cubes, offices, fancy plasma displays in the halls, etc. None of this really contributed to the bottom line: offering the best service to our customer base so they wouldn't mind paying a couple of extra bucks a month. (preceeding the buyout, we had 3 consecutive quaters of profits; after the merger and investor mayhem, losses galore)
Private firms, held by an individual or small group of partners, tend to be a bit more inclined to be open with their business, as to build a more honest relationship with the companies they deal with. Since they are small, have less potential capitol in a bind, they are usually more likely to be your buddy without attorney supervision.
Just seems to be common sense.
mini Ask-Slashdot follows:
Actually, I *do* have a unique idea (being somewhat of an analog to a high-speed analog circuit designer, I do not think anyone else has come up with the idea), and it has a *huge* market size. I estimate it at $50B a year.
Now, *that* is my problem when I try selling the idea to venture capitals. They see the number and think I'm mad or something.
I tried to underestimate market size (by truncation, pessimistical estimates etc.), though, but the VC's bid-counters usually find the rounding errors here and there in my math.
Any advice regarding this particular problem?
Those who can, do. Those who can't, consult.
Not the people starting the company. It is a simple idea, VCs finance a company but don't really know if the company has value or not. If you own something and don't understand it, you don't know what you should protect and what you should not care about so make everything secret and you are covered.
Many of the people starting companies in the late 90s that used VC did so in order to flip the company quickly using a FastCompany model:
1. Build it big
2. Hype it
3. Flip it
4. Profit and get out
So the lesson is this: If you have something you want to flip in 2-3 years and will be able to, take the VC and the NDA.
But if you want to build a company do it without VC. But remember building a company yourself it really hard and you will most likely fail. You will most likely fail with VC but as long as you don't personally guarante the company you are losing their money.
Of course, in the process of starting up a new company everyone is filled with fervor, believing that their idea is so good it will be like an earthquake when it becomes known.
Reality is that there are all these other hurdles involved in marketing, getting startup funding, getting more funding, getting a working product, etc.
No one ever thinks about Corporate Death, only Life. When you're devoting yourself heart and soul to bringing to life some brand new idea, the very last thing you think of is an orderly exit strategy of what happens in the event you fail. Something along the lines of a Last Will and Testament that says if no buyers are found for a period of 2 years that the patent and software and all rights are donated to a charity, such as the EFF or the FSF.
And that's too bad, because some genuinely good ideas fall into oblivion until they're rediscovered, perhaps years later, and successfully brought into the mainstream.
"Provided by the management for your protection."
I've rarely read something that so completely matched my experience at a startup. I'd add one more thing: by keeping your idea secret, you delay finding out the reaction of your potential customers to your product. And you may need to make some significant changes to your product to build a successful business. One thing to be careful of is that if you are going to file a patent, especially in Europe, you shouldn't disclose the *details* of your invention until the patent has been filed. Of course, IANAL, and this is not legal advice. By the time anyone's paying attention to what you're doing, you've had time to patent any really interesting idea you have.
The business model of centralized conglomerates marketing the digital rights of a handful of artists is outdated. Both consumers and creators have become disenchanted. A new model, consisting of a distributed network of thousands of creators hosting their content on Open Source CMS and syndicating it to trusted archives and marketplaces, is emerging. In order to build a trusted network of marketplaces supporting common standards for syndicated commerce, the business plan should be shared openly. The transparency provided by Open Source will foster the adoption of open standards for DRM and syndicated commerce. 22surf encourages artist-hackers to download our business plan for building profitable archives and marketplaces with Open Source CMS, change and build on it, and join in the following revenue streams: 1) sell keyword advertising throughout free OSCMS hosting services (blogs, galleries, etc.), 2) sell advanced hosting options/extra disk space, 2) charge 5% on content marketplace transactions, 3) charge 5% on Open Source Arts services marketplace transactions, 4) manage/host media assets of large businesses (record labels/movie studios/etc.), 5) keyword advertising throughout the free hosting services, 6) a friendster/FOAF (friend-of-a-friend) network, 7) a mobile FOAF service which alerts people when friends are in close proximity. All of these revenue streams may be realized with a small team leveraging Open Source CMS such as vvgallery, oscommerce, cafelog, postnuke, and gallery. With common standards for syndication based on RSS/RDF, bands, writers, photographers, and friends will be able to enjoy syndication across multiple archives and marketplaces. Working together, trusted marketplaces utilizing open standards for syndication will prosper. Surf's up!
I believe (and cant be bothered checking in true /. style) that Volvo developed the three point seat belt.
Realising it's safety potential, they then chose to patent it but make it freely available to other manufacturers.
Although it's a slightly different situation since Volvo were almost certainly going to use their invention regardless of what others in the field did.
What company was this? I'd be interested in seeing what you guys came up with.
I didn't think accounting software lent itself to anyone considering open source development, unless it was for political reasons (govt., taxes).
Fuck Beta. Fuck Dice
The entire MBA thing is built on the belief that if one does everything The Right Way(TM), the business will not fail.
That's not the case, of course. This quote from the Swiss(?) army reflects the spirit of "luck": "if the map and the terrain do not match, trust the terrain".
I'd like to see business consultants trusting the terrain instead of trying to fit the terrain into their dream world methodologies....
The NDA here includes a non-compete clause. I think this is common. The wording says that I am not permitted to do anything that will compete for 1 year. This effectively means I would be unable to work for one year as virtually anything can be defined as competing. How many people have signed away their right to support their families?
I've worked at a tech startup or two, and it seems as though they all think the same way. They flatter themselves into thinking that they're "visionaries" who are going to change a whole industry (utter madness of course). Their egos swell up, they puff out their chests, they start walking around like they're geniuses (a legend in their own mind!) and before you know it, even the secretaries are signing NDAs, noncompetes, etc ad nauseum. It's generally total horseshit, of course, but what can you do? People are silly, shallow creatures and they love to puff themselves up. It makes 'em feel like bigshots.
What's worse than NDAs, though, is IP agreements which lay claim to all work a programmer does, whether on company time or off. The incredible arrogance of dot-bomb founders is apparent here. They think they have a right to "monetize" everything done by every one of their employees -- to co-opt not only their own stinky ideas, but the ideas all of their wage-slaves have too!
The world would be a whole lot better if some of these people would catch a clue or two and realize they're just small businessmen trying to start up a small outfit and behave accordingly: appreciate their employees, work hard, show a little humility, and build their business. Of course, that'll never happen...
Farewell! It's been a fine buncha years!
Think an NDA is bad? Try a US security clearance... they can send you to "federal pound you in the ass prison" with very little evidence.
Sometimes it isn't an either-or decision. The whole concept of the NDA is so that you get to choose who has access to (and can use) your ideas and designs. The whole concept of open-source is based around the idea that others (not everyone, GPL doesn't allow commercial entities) to help out if they re-contribute to the community.
I think the key reason some (not all) startups still look at the NDA as a good thing is because it is all too easy to see how some company (M$?) might steal your idea if you just advertise some revolutionary source code online. If M$, as an example, were to take your code and utilize it (say, making something like J++ or C# instead of the original Java), it dillutes and even contaminates the original design idea and the spirit of the work.
By utilizing a custom NDA (instead of boilerplate), you could easily say that this information can only be shared with those that agree to be bound by the GPL/BSD/Apache/whatever license -- thus eliminating the LEGAL ability for other companies (say, M$) to incorporate it into their own works.
Unfortunately, I think that all legal issues are solved by he who can afford the best lawyers, so it doesn't really matter if they have agreed to the NDA or not, there are always ways around the law for those with money.
Malachi
http://www.google.com/profiles/malachid
I've had quite a lot of direct experience trying to raise $$ from the VC community. First, let me say that from my direct experience, few if any VC's will sign an NDA *before* getting initial information from you. It is pretty common for many folks to see similar opportunities in the market and the VC may see 10 plans but invest in only one. They don't want to be sued when the one they invest in comes to market with a product that looks a lot like the product from one they didn't invest in.
As others have said, if you have patentable technology, you need to avoid discussing or disclosing it without an NDA because you'll potentially lose your right to patent. Remember that the VC's have had thousands of folks make this kind of claim, and then have it turn out to be overzealousness and ignorance. Too many people starting a company really know enough about what others in the space have done.
So, VC's are just not going to sign an NDA unless they know you, have made lots of $$ with you before, etc. And probably not even then. This makes it hard for a startup, but you have to get people to the point where they are interested enough to believe in your business and that you may have something worth protecting, so that they'll sign an NDA.
The author is right in saying that too many entrapreneurs overly estimate the uniqueness of their idea and business plan. Contrary to the American myth, rarely is the best idea or technology the winner in the market. If you think I'm wrong you must think DOS is the best OS ever built. This all makes it harder for the small per centage of people who do have a unique idea and technology that does need protection.
I often think that software is a special case. Here a first mover advantage is really important. But then I a bit more. If something can be built by 3-4 folks in a garage in 6 months, then either the Open Source community or a large well heeled software company will knock it off in less time with fewer people (they get to learn from your work). A VC who invests in such technology is nuts. maybe you think you have something patentable. But unless it is the RSA you are unlikely to really be able to make a patent stick or afford the cost of defending the patent. Remember enforcing a patent can cost 100's of thousands of dollars and the attorneys almost never take those cases on contingency.
In the end, being an entrapreneur, raising money and building success is REALLY hard. Everyone thinks they can do it, a bunch of people try and and 99.99995% fail. The author is right about needing luck, talent, timing, the right partners and a million other things to be a success. Is the idea of ordering books on the Web such a brainstorm? It is their execution of the idea that makes Amazon a success. Is building a product for the home market for 1/10 the price of what professional products sell for in the corporate market a brainstorm? No, but Linksys and DLink seem to be doing well.
I know of a company that builr rugged mobile PC's using frequency hopping spread spectrum wireless technology in the mid 1980's. They raised a lot of money (10's of millions of dollars) and failed. They were a decade ahead of their time.
So, if you are an entrapreneur unless you are the one in 100,000,000, all the secrecy will have little impact on your chances of success.
If the country your startup is based in doesn't allow for software patents then you should guard your secrets with an NDA. Even if you're working on hardware projects you might wish to guard the software, see Transmeta for example, their crusoe chip was part software part microcode.
All depends on how radical your product is.