Slashdot Mirror
Successful Startups and Patents?
An anonymous reader asks: "Is there a strategy for software startups wanting to make it big with a product in this world of software patents? Particularly, how can a software startup, wishing to take its product to the international market, hope to compete with the large software companies and their armada of patents?"
"No matter whether or not the startup has an innovative product, a patent or two, a little or a lot of funds, it seems that if the large software companies decide that they wish to crush them with law-suits regarding patent infringement, valid or invalid, they can tie them up long enough in court to drive them out of business (via bankruptcy, or to cause them to lose any market lead they may have had). Can anyone suggest a strategy (or discussions on strategies) that could help a software company thrive, without playing the same game? Or are those days over — and the current crop of software companies are so well entrenched (thanks to the patent system) that there is no hope for a a new player?"
For tips on what not to do.
Is there a strategy for software startups wanting to make it big with a product in this world of software patents?
There're tons of strategies a small company with a patent portfolio could pursue.
Sincerely,
Darl McBride, SCO
on Trolling...
The best defense against being driven out of business is a creating stable financial structure for a business. It has become fashionable these days to fund a business via debt instead of equity due to certain tax advantages. However, that also exposes a business to much more risk: if a company's revenue stream falters, they're much more vulnerable to going bankrupt. However, if a firm uses profits rather than debt to grow the business and puts aside a large pile of cash to absorb variances in the marketplace, then their future is secure. This sort of thing is difficult for a startup to achieve, since you are in effect, creating something from nothing. However, large corporations only try to crush you once you're perceived as a threat, and patent trolls tend to sue you only if you've already made a lot of money. So the solution for the software startup is to make sure you keep costs down, overhead low and reinvest profits in the business. Stay off the radar screens of the majors that you fear would crush you until you have a warchest where you could survive disruption of income. And finally, diversify. If all of your revenue comes from one product which turns out to be patent-infringing, you could easily be toast. However if you have multiple product lines which don't all depend on the same technology, then your firm will take a hit to revenue if one gets tied up in court, but it should be survivable based on the revenue from the others.
If we leave apart the issue of patents and infringement and if the startup does make a big name for themselves, they will be acquired by biggies who are on an acquisition spree just to keep up with their competition! A lot of these have happened, famous ones being Flickr and Konfabulator being acquired by Yahoo!, YouTube by Google, and the amazing of all deals was the historic and legendary acquisition of Hotmail by Microsoft...
- Yes, but does it run Lunix?
Companies use to change *just enough* to create a possible gray area stealing ideas/patents/etc ... As long as your legal fees still leave a nice net profit .... Bob's your Uncle!
I worked at such a company for 2 years, back in the 1980's.
-- www.globaltics.net
Political discussion for a new world
I submitted link to Paul Graham's new essay on startups, but probably it is not home page worthy story so I'll write it here again.
A bit off-topic but worth having look at it.
Why Not Not Start a Startup?
http://paulgraham.com/notnot.html
(March 2007)
He discusses about reasons why people don't start a startup and tries to explain that most of those reasons should be ignored.
In no particular order, here are some general strategies. These strategies can be applied to non-software products as well.
Find something at least 17 or 20 years old and clone it. Minimize any changes that aren't old or obvious. (color pong anyone?)
Build using existing components to reduce risks. (API's, scripts, etc)
License from someone else's portfolio. (Unisys GIF patent)
Learn about patents, research them like mad, create a product that doesn't infringe (PNG)
Create an invention, patent it, license it to others (like NTP)
Pay someone else to create and accept liability (can't think of a good example of spec developed software here).
Build something and cross your fingers (RIM: Blackberry)
Build and sell where patents don't apply.
Am I forgetting any? If you don't have money, want to create a unique product, and don't have faith in crossed fingers, you're going to have to learn about patents or restrict your market pretty heavily.
The world is made by those who show up for the job.
Get a userbase. Grow really fast. That's a serious competitive advantage - much more than a few patents.
Don't accept fiat currency. Only accept payment in gold.
I will have a sig when the market demands it.
Typical Slashdot defeatist attitude towards patents.
Patents and their applications are published. It is this publication more than the grant of the monopoly itself which drives innovation. By looking at how your competitors do things you can learn what they do right and what they do wrong.
If they do everything right and you can't think of a better way, consider getting a license from them.
If they do everything wrong and you can think of a better way, bingo! There is the nub your business idea.
Do not bury your head in the sand hoping patents will go away, go out there and look at them and learn from them. Patent databases essentially hold the vast majority of human ideas.
You can always emigrate to a country without software patents...
(Note: all this IMHO, I am not an expert) I have a gaming company, I'm at my second successful startup. Frankly, one cannot do anything against patents. It's more or less like russian roulette, you get the bullet one chance in six (true specs for a 6-shot pistol, not in reality ;) ) to bite the bullet.
;)
That said, most tools, I get from open source. I carefully read the licenses and respect them precisely. GPL is good for tools, as long as it's not on the end-user computer. LGPL has the same limitation. ZLib license is great, as well as most MITs. All in all, read your licenses. And yes, you have to purchase fonts
On what you are trying to innovate upon, I suggest you take a good hour to look up google, and patents databases, and try to see if you can do it or not. If you can get a legal guy to do it and give him/her a grand, that's your choice and it's a valid one. However, even Microsoft and Apple are permeable to patent lawsuits, so it shows it's really ludicrous to think you can be 100% safe. Thing to do is be "good-enough" safe.
I took a good week total in past 3 years to look up patents, licenses, read legalese. Not that bad and I feel safe. I paid twice for very precise things too.
just friggen' do it. Keep in mind one thing though. You start a business to make money. If your just want to gain cred for you skillz or code something new and cool, then get a job or join/create an open source project.
If you want to run a business, then put in the effort to figure out how you'll make a profit. Beware your desire to build something new because it might not be new at all. Take some time up front to research your idea. Lots of guys hunker down for 6 months or more creating something that is already common.
Your main point appears to be that a mega-corp will wander in and sue you into submission for patent infringement. If that happens, then there are a couple options. First off, realize that you are unlikely to be sued unless you are actually doing enough business for it to be worthwhile/threatening. Treat every nastygram as an opportunity. Also, realize that every successful business should expect to be sued by someone for something. It's a symptom of having money.
Strategies:
1) You have no patents. So, ask them what it is they want. You popped up on their radar somehow. Maybe they want to buy you or prevent someone else from buying you. If you want to fight, contest their patent. Find some prior art. Find a very similar patent and license it.
2) You have your own patent(s). Same as above, but threaten 'em back if it makes sense.
I am a lawyer, but not yours. Anything I tell you might be a total lie intended to benefit my clients at your expense.
To be successful, the product has to actually be good enough on its own.
The fight back against big brother lawsuits, be able to prove invalidity, and after the discovery process shows that, file charges of barratry against the lawyers. Lawyers are only confrontational when it comes to clients. When it's their own ass on the line, they scurry like cockroaches when the light is turned on.
"I may be synthetic, but I'm not stupid." -- Bishop 341-B
Ignore Microsoft, he says, because "Microsoft is Dead."6 1
His take: http://www.paulgraham.com/microsoft.html
Mine: http://www.networkworld.com/community/?q=node/135
A previous poster made a very good point: nobody will knock on your door until there is money involved. You MUST do the basic searches and preserve evidence that you have done so - in any case, not having bothered is not going to help you in court.
However, when that fateful letter eventually arrives you should have an idea what you want to do. There are 3 reasons why someone would get you into a patent suit:
(1) you're a threat to them, and they want to shut you down
(2) you're making a profit and they want their 'protection' money (let's call the racket like it is)
(3) they would like to buy you, but not at current value.
Whatever answer you decide (sell, license, close) - remember that such a suit is never started for fun, so there is an assumption that you're worth money so you are in a strong position to negotiate, even though they don't want you to realise that.
Good luck!
Insert
Talk to these guys.
Do not bury your head in the sand hoping patents will go away, go out there and look at them and learn from them.
And then when the judge finds out that you did actually research patents, the patent owner has you by the short curlies.
Damned if you, damned if you don't.
http://www.maxineudall.com/2010/02/should-economists-be-sued-for-malpractice.html