Hide Techdirt is off for the long weekend! We'll be back with our regular posts tomorrow.

IP Lawyer Says: 'Stop Wasting Money On Patents'

from the good-for-him dept

Erik Heels, an intellectual property lawyer and a regular Techdirt reader, has put up a nice blog post, explaining why patents rarely make sense for startups (especially if the patents are for software). He notes that, in most cases, filing for a patent is “a waste of time and energy,” not to mention money — not that “your money and time would be better spent hiring programmers, marketers, and a sales force.” Indeed. Unfortunately, lots of startups think they need patents — often erroneously claiming that VCs won’t invest without patents. But as many smart VCs point out, having patents for a startup is usually pretty useless. Startups live or die in the marketplace with a product — and that product is rarely going to wait around for a patent. Focus on building a business, not wasting time and energy on useless patents.

Filed Under: , ,

Rate this comment as insightful
Rate this comment as funny
You have rated this comment as insightful
You have rated this comment as funny
Flag this comment as abusive/trolling/spam
You have flagged this comment
The first word has already been claimed
The last word has already been claimed
Insightful Lightbulb icon Funny Laughing icon Abusive/trolling/spam Flag icon Insightful badge Lightbulb icon Funny badge Laughing icon Comments icon

Comments on “IP Lawyer Says: 'Stop Wasting Money On Patents'”

Subscribe: RSS Leave a comment
Bill Pickett (profile) says:

Explain to me why this is wrong.

I’d like to understand in effect how software patents work. My totally bogus ideas so far are: if you are a big company you can afford to buy patents using a shotgun technique. Patent examiners and under-knowledged and under-staffed so they will approve practically anything. So you collect your portfolio of whatever sticks against the wall. Then if you happen to conflict with another big company you use your own arbitrary patents as a kind of currency and cross-license. If you conflict with a small company or individual you crush them with your lawyers – which is another kind of currency small players lack relatively. Big companies get to only have to deal with big companies. Small companies and individuals sometimes prevail but the deck is completely stacked against them due to the weight of cross-licensing and ability of big players to drag things forever through the legal system.

So, am I just being a cynical idiot or is this roughly how it works?

Bill Pickett (profile) says:

Re: Explain to me why this is wrong.

Weight of cross-licensing means in this context the small cost of doing business to a wealthy company to buy a currency of patents that can be spent for like material. Getting this kind of currency as a small player in trade for dollar currency is not as affordable. And of course patent currencies can be spent without losing them until the monopoly expires.

Derek Bredensteiner (profile) says:

Sir, you have a firm grasp of the reality of software patents.

To be fair, I think the PTO does reject a good portion of what comes through, but they also do approve an equally large portion that’s full of the same crap. The net result is still exactly as you’ve described, the shotgun approach works very well.

“Big” companies also lose out pretty bad (e.g. Microsoft, Blackberry), so it’s not all a one sided affair of a big guy versus little guy. I’d say it’s more Patent trolls vs. Everyone else, most of the time.

But overall, best summary I think I’ve heard.

lawgeeknz (profile) says:

Not only that

Agree entirely and I’m giving the same advice to my start-up clients. Not only that, but the cost of then defending the patent or taking infringement action is a factor of 10-100 times more than obtaining it and takes many years longer (particularly if you have to do that in different jurisdictions).

Unless you’re in the troll business, it’s just not viable.

Anonymous Coward says:

TFA says that the max spend on obtaining a patent is about $34K. How many/much “programmers, marketers, and a sales force” does that really buy? Unless the folks you are referring to live in a 3rd world country then you are not even talking about 1 person for an entire year, you would be talking about 1 contractor for a few months.

If you are operating in a similar space as one or more large players that have patents then I say the $34K is well spent as a defensive move if for no other reason. This only matters if you are also successful in the market place – otherwise you have nothing to defend and would not be a target.

Mike Masnick (profile) says:

Re: Re:

TFA says that the max spend on obtaining a patent is about $34K. How many/much “programmers, marketers, and a sales force” does that really buy?

If you’re only getting one patent, perhaps. But most of the time you end up applying for a group of patents. It adds up quickly especially when that money can be better spent elsewhere.

Mike Masnick (profile) says:

Re: Re: Re: Re:

Really, what is that opinion based on? Let me guess, most startups apply for just enough patents for you to prove your point. Take me to your data.

From personal experience. I don’t know of any company that applies for just one patent. Every startup I’ve ever dealt with has ended up applying for many patents. If you have data to suggest that most startups apply for one and only one patent, I’d love to see that because it would suggest that somehow I magically only deal with abnormal startups, which I find unlikely.

JMG says:

Re: Re:

TFA says that the max spend on obtaining a patent is about $34K. How many/much “programmers, marketers, and a sales force” does that really buy? Unless the folks you are referring to live in a 3rd world country then you are not even talking about 1 person for an entire year, you would be talking about 1 contractor for a few months.

This response is kind of misleading. Sure, you may not be able to hire a quality full-time programmer with $34K for more than a few months (or hire marketing consultants for much longer). But you can put that money towards other business-building endeavors (e.g. hire a quality part-time programmer for 6 months, buy access to databases or libraries that may be helpful to you, etc). Buying something on the cheap doesn’t mean it should be bought at all; just because I can’t afford to hire Donald Knuth doesn’t mean I should hire the kid who got an B- in high school computers to do my coding.

Patents should be weighed in terms of benefit-cost ratio, just like any other business decision. If spending $34K on something that may never turn you any additional revenue, customers or high quality work force is what you want to do, go right ahead.

Snidely (profile) says:


VCs like a startup to have patents because it is a type of insurance against the failure of the startup. If the business can’t survive in the competitive market (and most can’t), the VC might be able to recover some investment by selling the patent or suing whoever figures out how to successfully market the product. The problem lies in the fact that suing can make as much or more money than most startups will generate in years. The live or die nature of the startup is then skewed.

Gene Cavanaugh (profile) says:

Patents are useless?

Basically I agree with this blog, but as is true in far too many cases, it states a valid point and then “pegs” on an extreme “solution”.
The patent system (which is at least 95 percent “large entity patents”, costing a “minimum of $14K”) is seriously broken.
Most startups should use patents judiciously, if at all (yesterday I advised a client who wanted to file “several” patents with me, and has the money ready, to use a trademark instead – in his case, he really doesn’t need even one patent (though I am an advocate, not “a decider”, so I will do as directed if it comes to that).
But then we jump to an article that strongly implies patents are never useful! Duh!

The reason the founding fathers put the provisions for our system in the US Constitution for protection of innovative rights? Small inventors with TRUE, VALUABLE inventions had routinely been savaged by wealthier people! The problem is less clear, now, because we HAVE a system, and avarice is curbed!

Even so, in most cases, a small startup does not need patent protection, and IMO business and software patents should not be allowed!

Don’t even get me started on copyright!

Anonymous Coward says:

I must ageee with Mr. M that most VCs with whom I have dealt to not view the absence of patents as a deal breaker. When the existence of patents did come up it was merely in the process of the VCs going through their Due Diligence list.

I do not know the type of ventures Mr. M has been involved with, but those with which I have been involved invariably involved very sophisticated technology first created in the course of internal R&D and government contracts wherein the funds expended for research, development, productization, support, etc. have run in the 100’s of millions, and on occasion in the billions. Beyond these sunk costs, know how, show how, manufacturing complexity, extremely tight system specs were areas in which VCs expressed considerable interest and ascribed considerable importance. Why? Because reverse engineering would likely be impossible, and even if it could be done taking the product to the next level would involve exorbitant amounts of money and time.

Moreover, VCs paid particular attention to the qualifications of those who would actually run the venture, the availability of access to relevant technological expertise, business plans, etc.

In sum, it does not hurt to have one or more patents in hand, but there are many other more important factors that govern the “thumbs up – thumbs down” decisions by VCs. Even so, it is nice to have a patent(s) in hand should a situation arise where their presence can be beneficially employed.

Jose_X (profile) says:

Discussion and suggested changes to patent law (focus on sw)

Patents are bad, let’s consider some changes for the better. In particular, the bottom half of this comment lists some reasons to avoid software patents as well as ways we might change the system to make incentives less harmful to society. Also, let’s contrast today with the time when Article I, Section 8 of the Constitution was written.

Large companies gain from patents (even amid trolls) because collectively they can own the market amongst each other. With software, it is particular important to them to stop open source being used by a great many people against them. Open source lowers the bar to competition significantly and pressures profit margins also significantly. Hence, as a weapon against such competitors, patents help a lot even with loses to trolls. [This would be the case at least if trolls remain small in number.]

As for the cost trolls exact on large companies, as was mentioned in other comments above, trolling is a tough costly process, so many don’t go through with it.

>> The reason the founding fathers put the provisions for our system in the US Constitution for protection of innovative rights? Small inventors with TRUE, VALUABLE inventions had routinely been savaged by wealthier people!

The wealthy always have advantages. They tend to have strong levers and don’t like to put their money at risk or keep it idle while others eat into it. Today, they are using patents against the small product-producing inventors (and want to outlaw the trolls).

Patents that harm few people in practice is one thing. Patents that hand-cuff a whole shipload is another. Patents that can greatly stall progress and abridge individual rights for a very long time relative to the speed that industry could otherwise move is even worse.

Translated this means that with today’s higher level of education; low bar to entry in software (cheap powerful computers, $0 to copy/borrow, retool, and distribute as software products are information and not beholden to laws of physics as are what they represent/virtualize); terrific collaborative possibilities (cheap accessible Internet and tools); and high rate of growth and innovation possible, long patent monopolies in software in today’s world have a much higher cost than would a traditional patent two centuries back in a typical field of endeavor then.

Let’s consider a series of questions about the effects of our evolving patent system (focusing on software today vs. other patents back in the late 1700s).

20 year monopolies are even worse today.

Was that patented invention back then likely to be reinvented within 20 years? If there were few inventors, then likely yes but not nearly as quickly as would be the case if there were a lot of inventors (as is the case for software today).

Greater stifling occurs today.

Were a lot of creators likely to be hand-cuffed? As a result of (a) the level of education and opportunities that exist today and (b) the low entry bar to software, the more participants that exist and the more collaboration that is otherwise possible, the more people are likely to be hand-cuffed. Today, with software, we don’t have the situation as we would with most patents granted two centuries back; many more are hand-cuffed with software today vs back then.

Society will lose out today on very large possible savings.

Would society back then have risked having patents remove access to a very large amount of powerful consumer products and business tools at $0 (as would be the case today when FOSS is killed in favor of a patented monopoly high cost product)? No, back then neither software nor anything at all like it existed, at least not of the things that were being patented: eg, business methods, fiction, and mathematics were not being patented. Software patents that are used to hurt FOSS would carry a very high cost to society by removing lots of very inexpensive access, perhaps very high relative to any potential loss possible by a patent taken out two centuries back.

Is a monopoly subsidy efficient?

Is it efficient to allow a single person to try and build a business from scratch that otherwise might not be possible, rather than allow existing infrastructure and established know-how be utilized? No. We want to reward the inventors but not stall progress. This applies as much then as it does now.

Lower the duration of any granted monopolies so as to limit damages and unfairness (limit case being 0 years).

We should keep in mind that few can copy that quickly, especially with copyright protections in place, unless they already were sniffing in the same area or had prior made significant advances. It makes no sense perhaps 99% of the time or higher to give anything exclusive for a long 20 years to one over all others. That 20 should be a much smaller number.

Could we have done better then (as we could now) with some other incentive besides a monopoly subsidy?

Well, a monopoly encourages sloth, horrible pricing, and suboptimal implementations (through lack of competition and copying/leverage/collaboration). If patents are easy to get on non-revolutionary things and/or are too broad, then we acquire all of these liabilities (sloth, high price, low quality, etc) for very very little gain (slightly speeding up introduction of invention concept).

Renewable Tax Credits as a less damaging and more incentivizing alternative to monopolies. Renewable tax credits are not disruptive of progress and individual rights. These could be renewed if certain conditions were met, eg, conditions which might discourage you sitting on your hiney. This sort of incentive would solve the troll problem in avoiding the powerful patent monopoly grants.

Let’s consider renewable tax credits on R&D for being the inventor of record. These would give you an edge over other inventors that were slower. This would motivate you to speed up discovery/inventing. This would help you if you lose out on first mover advantages to more efficient rivals or if you prefer to focus on R&D. [See below, prizes and mandatory royalties as way to augment this, at least in some cases.]

Renewable Tax Credits for Taking a Product to Market After Inventing would benefit you if you can come up with a decent product, even if others copy (or you copy them). This means you are motivated to carry the invention over into a product society will want. These renewable tax credits give you an edge to help make up larger competitors that you beat to the invention punch. Since in a market with many competitors and with many fast evolving products (eg, software) everyone else will collectively out innovate you, you really gain from copying them more than any single competitor gains from copying you. Here you also benefit if you lose first mover advantages. Renewable tax credits might be conditioned on you not sitting on your hiney.

Renewable tax credits to the “first” is not such a lopsided way to reward one group over everyone else.

We can add other tools besides renewable tax credits; however, note that in the late 1700s nothing of this level of sophistication was available! The Constitution provided a backstop instead of good law at that point in time, but at least the backstop was there if in fact *progress would be promoted*. Over time, presumably we would come up with good law that would not stifle or carve out markets for 20 years out as is the case with patents (a bad law) today. The key according to the Constitution has always been not to grant monopolies if progress would not be promoted.


Prizes are one way to reward inventors. The government could create an Inventor’s Aid Tax and use it to pay those that would get something along the lines of a Nobel Prize. Private industry would manage this. The government would give various such judging organizations grants to use for this purpose. These organizations would incorporate under a new section or otherwise simply apply for this privilege (for the conditioned grant money). Note that in our current patent system such an Inventor’s Aid Tax effectively already exists.

If we really want to give a monopoly to “the little inventor”, as some claim is the motivation behind the Article I, Section 8 inventors’ subsidy clause, then pass patent laws to prevent the large from attacking smaller foes. Have safe harbor for things like open source since this adds much to society and patents against it would hurt society. In short, we can prevent certain abuses by trying to code them into the law. This would help were the existing law not otherwise to change much from its current broken state.

Mandatory royalties in sliding scale?

We might instead consider a sort of sliding scale royalty system for a small number of years starting with the year of the first product that makes it to market. This is not too unlike the situation where the patent owner waits years until the market develops and then sues for 6 years of past “damages” plus future years. A difference is that the industry is not encouraged to wait for the 20 years to pass but is encouraged to begin marketing a produce immediately even if for little or no profit. Also, we can assume here we are still avoiding the injunctive powers of patents. The royalty, if applicable, would be on profits (if any) of competitors (details to be worked out later for various scenarios). This would actually be a cap (meaning it could be waived or lowered on a per competitor basis). We could even have the royalty cap amount go down after every year. The royalty values would be determined based on the total number of patents available that might apply to a product. Those seeking royalties would register or advertize as this would affect those that come after. Groups like the USPTO could help here in valuing or judging applicable patents and perhaps in assigning royalty grades/ranges.

Implicit and no fee patents (like copyrights) would be a last resort so that poor industrious software developers are not abused by the wealthy.

If we keep things roughly as they are (ie, a monopoly subsidy) and continue to stifle, then at least make patents be implicit (automatic) and $0, as is the case for copyright. Patent stifling should be accessible to all inventors equally, whether they are starting off rich and lazy or are poor and industriously spend their time working on product details. This way the little guy and FOSS are not completely abused by the wealthy, where the innovations by the former are leveraged by the latter without hesitation but the latter can decimate the former’s creations (sw is complex and would violate lots of patents) or at least stagnate them by patenting in front of the FOSS products. If patents can be used against FOSS, then there needs to be a practical tool for FOSS to have its own innovations be recognized in a manner that is as real and useful as are patents: make patents be implicit and $0.

We might also consider maintaining an official database of inventions which would be organized (eg, by dates) by the private sector and organized as necessary (“lazy evaluation”) rather than as a delay before every patent was granted. This scales since the USPTO is relieved of the task of research to determine priority and scope. The USPTO could then serve to help referee or as low cost expert witnesses. [They could do other tasks as well if some of the above suggestions are implemented.]

Such a database of invention could be used with patent monopolies or instead with some of the above suggestions. The database could be used as people file for inventor tax credit or simply to help support the system. There might be other benefits for registering.

In the spirit of public disclosure (another alleged reason for supporting patent monopolies), revealing all source code to software (or full reproducible blueprints? to products) could lead to another type of tax credit.

In any of these cases, we can still have safe harbor provisions and other special exemptions or rules.

Anyway, to avoid stifling and abridging rights, we simply should not allow software patents to exist. We may or may not utilize the prize/credit or similar examples mentioned above. We could allow some degree of patent monopolies but they should be short and few granted per year, eg, with an upper bound per year and judged by a panel of experts representative of academia and various industries.

AJ says:

Wasting time on applying for patents without lawyers

The whole patent system is rigged up. The patent examiners will barrage you with silly administrative objections (t’s not crossed, i’s not dotted type) for months and then reject the patent on the basis of some adminstrative/procedural/format objections – not technical merits!!

AJ says:

Re: Wasting time on applying for patents without lawyers

Also, the best part is that if you want to challenge the rejection you have to spend hundreds of dollars to go through the appeals process!! You are already guilty of submitting a frivolous patent and then have to spend hundreds of dollars to prove that it’s not the case. Have you also noticed how much coaxing they do to convince that you should hire a patent lawyer to file your patent!! I guess the PTO’s drive for individuals applying for patents is a bunch of crap.

Add Your Comment

Your email address will not be published. Required fields are marked *

Have a Techdirt Account? Sign in now. Want one? Register here

Comment Options:

Make this the or (get credits or sign in to see balance) what's this?

What's this?

Techdirt community members with Techdirt Credits can spotlight a comment as either the "First Word" or "Last Word" on a particular comment thread. Credits can be purchased at the Techdirt Insider Shop »

Follow Techdirt

Techdirt Daily Newsletter

Techdirt Deals
Techdirt Insider Discord
The latest chatter on the Techdirt Insider Discord channel...