Hacker News new | past | comments | ask | show | jobs | submit login
Myths about Patents and Trademarks in startups (hitstartup.com)
62 points by samdung 13 days ago | hide | past | favorite | 33 comments

Very software focused. In biotech, patent protection of composition of matter (and less so process) is important to ensure the ability to capture economic value of your research.

But keep in mind the safe harbor clause (Safe Harbor provision of 35 U.S.C. § 271(e)(10)) allows any other company/lab to use your compound/process "in support of submissions to the FDA". This is taken very broadly in the field to mean if you are pre-commercial you can basically use anything you want, no one will sue you because if you're starting-up, everything can be claimed to support an eventual filing. Plus, if you have no revenue a patent fight has no value to the other party.

Obviously every marketed drug has IP around it preventing generics until the patents expire. A lot of inventions that lead to commercial products are actually derived from academic institutions. The Bayh–Dole Act did absolute wonders for the US in terms of biotech innovation. Pre-Bayh-Dole, any research using federal grant money had to go through a Byzantine process to determine IP ownership with a lot typically residing with the US federal government. Post-Bayh-Dole, it simply stayed with the inventor/inventing institution. Some people now consider this bad because evil drug companies are "privatizing the gains", but without that economic payout, they wouldn't have invested to bring the invention to market to begin with.

You could argue it moved the large biopharma companies to become development/networking/marketing companies, which is a fair point, but it simply moved the innovation to universities and nimble start-ups--which is better anyway for actual disruption and more equitably spreading economic gain.

Author here.

I agree that it's more software focussed and I would be adding the role of IP in other industries in my upcoming updates.

But, the main reason I explicitly didn't go into Biotech/Pharma is because living in India which has been at the receiving end of the US Pharma's patent protections and on the other hand witness generic medicines save lives everyday; I didn't want my biases to affect the content.

Totally understandable. There is a minority of non-CS commenters on HN, so I always try to expand the conversation. Wasn't meant as pejorative.

Re your comment on generics saving lives, yea, this is a pickle question between sovereigns in negotiating trade agreements. The US pharmaceutical behemoths make too much money in my opinion. The inability of Medicare to negotiate on drug pricing was an insane give away from Bush that effectively allows drug companies the ability to set their desired profit targets. But with other nations catching up in terms of Quality and manufacturing tech transfers, most small molecules will be easily riffed. The era of blockbuster small molecules is coming to an end with a lot of the space/targets explored. More complicated biological and gene/cell therapies are the next frontier and it's going to be a lot harder to riff these technologies because "the product is the process" a lot of time. It's a lot harder to steal process without hiring know-how.

W.r.t India, it has mostly accepted the patent obligations of US Pharma as it gets to export generic medications to US and I think 50% of Indian generic medications get through FDA approval.

But this was pre-COVID, not sure how new found gold rush in Pharma industry worldwide would affect the future of Generic medications and patentability of pharma products. It could go either way of improving manufacturing and availability of generic medications (or) Pharma behemoths could take advantage of current situation and derail generic medicine development indefinitely.

Generics ARPU is still above $300 / year. New generic only requires 24 people to get FDA approval.

On the contrary, the question I had while building my product was - “Am I infringing someone else’s patent?”

That was years ago. Now, I focus on value creation first and let the product succeed before venturing into patent research.

Hardware might be much tougher though. Would be interesting to see someone with hardware experience to share their real life experiences.

Author here.

Even patent trolls wait till you are able to pay if you infringe their 'supposed invention' i.e. In other words focussing upon growth is a rational decision, when you get big you'll know if you are in the radar of trolls or genuine inventors.

This applies only when you truly don't know whether you're infringing IP, if you're explicitly copying an invention; unless you are in a country which has no regard to international laws and actively encourages its citizens to copy inventions of other countries, you are at the risk of patent litigations.

Trademarks help in domain name disputes. Your registered business name only applies to that state. It's not dumb to setup a trademark for your company name after it's used in commerce. For small business it's affordable as well based on the fee structure at the USPTO.

*Not a lawyer.

Author here.

Trademarks can also help to claim App names in Appstore[1]. But in several countries(even federal), registration/incorporating applies to all the states. Since, Appstore require DUNS which has company name, may be Apple would respect incorporation too especially if the app is country specific.


Excellent article, I'm bookmarking it to help my customers to realize most of the time they are loosing their time and money with patents

I strongly urge you to not take legal advice from any source other than your attorney.

> Besides, in most countries just incorporating/registering our startup as a company would prevent anyone else from using the same name for their company.

So, the question is. Is it good/bad to incorporate using your product/service name as a corp. name? What are the pros and cons?

(Because even if you do, your product will probably end up being called something else in the end)

Author here.

>Is it good/bad to incorporate using your product/service name as a corp. name? What are the pros and cons?

Apart from intellectual property, Unless there is a specific reason not to name our startup with our product name, it's always a good idea to have the same name for our startup as our product in terms of basic branding etiquette.

In countries where duplicate company names for the same product are not allowed, this will serve as an added protection. But it doesn't prevent someone from brandjacking by typosquatting, neither does trademark prevent it, Facebook has enough trademarks as legally possible yet it goes to the length of decloaking domain names to prevent this[2](Not showcasing this as a positive practice).

I've written more about Incorporation & App names here[1].



This is also wrong and slightly English common law centric. A company name does not prevent someone to open a company in the same name in another state. In English common law, using a trade name gives you some protection, using a trademark gives you more protection. Trademarks are for classes. If you sell software under the name cherry cake, nobody is preventing someone to sell shoes or a car with this name since it is a different class.

If your company looks like it may be successful, trademark squatters will trademark your logo/name in non-US countries before you get there.

I am personally aware (clients) of several times this has happened. Settlements to get your TM back start around $100,000.

Patents are expenses that a startup should not spend. There are too few upsides. And if you need to defend yourself against a bigger player or patent troll, you are bancrupt anyways

Says person who never started a venture backed startup. Ignore this poster. VCs want to invest in companies with moats. Depending on the domain, you may require one or more patents and potentially a Freedom To Operate.

There are plenty of upsides. Patents are assets that cost much less to obtain than their value. Even small companies should be able to afford one patent application per quarter.

Put IP protection in the budget just like any other cost-of-doing-business expense.

The article does not seem to address, should one patent a method or algorithm that defines your product as a startup before launching the product?

Author here.

From the content,

>Most countries don’t allow patent for a software[1]. Economic theories, methods of doing business, mathematical methods or computer programs as such are not patentable inventions in several countries.

The hyperlinked content from WIPO details about patentability of softwares/algorithm(Link doesn't seem to be working now, so including archive)

In short, in USA -

"According to the US Supreme Court , the Congress intended the statutory patentable subject matter to include anything under the sun made by man,” but the laws of nature, natural phenomena and abstract ideas are three specific areas which are not patentable."

So, you can apply patent for the application of that algorithm.


Truth is, it’s really semantics. A patent attorney knows how to describe an invention to pass most eligibility requirements for stuff that most software-related inventions.

If it's the "before" you're questioning ...

You can't do it afterwards: USA has now aligned with the rest of the World as a first-to-file patent system.

Your product, if available to the public, would be a prior art citation against your patent.

Incorrect. There is a 1-year ‘grace period’ for an inventor’s own work. And if someone ‘stole’ your idea and tried to patent it before you, there is a process to invalidate it because they derived the patent from your work.

I think incorrect is a bit strong, but you're right. ;o)

Isn't there another slight proviso, before the USPTO, that a third party disclosure is still a valid prior art citation, in the Grace Period if its source is not the applicant/inventor?

Also, outside USA (UK & Aus at least) there are specific international trade fairs that are recognised and disclosure at these gives a sort of grace period similarly to USC grace periods.

Like you say, there's account for abusive disclosure in various laws; also more countries accept grace periods for utility model patents than do for regular patents AIUI.

This is the problem with reducing complex laws to broad brush advice, lack of vital nuance; there's lots of geographic variation too.

A few comments:

The patent and trademark systems around the World differ.

USPTO is very open to business method patents, and software patents.

In EPO countries software has to have a real World effect, mostly, to be patentable.

Patent agents/attorneys do specialise.

Defensive publication is cheap, no need to get a patent to defend yourself. Though if you want to retain a "secret sauce" as part of your product they could probably help there.

Passing Off is a thing in lots of jurisdictions, like an unregistered trademark. Geography and company size matter in UK/EU trademarks (don't know about others), and you have to trade to retain trademark protection.

>This may help in getting the patent, but it also enables someone else to exploit the same loop holes to file a patent for the copy of our invention by just changing the grammar/vocabulary because our patent was too generic. //

That's not how that works.

I think the takeaway of "capital management might mean patents are a bad fit" is a good one.

In the UK one of the first questions asked of most companies in Dragons Den (equivalent to Shark Tank; an investor seeking "game" show on TV) is "do you have a patent" because they are seeking secure investments. Investor expectations is a big reason to get a patent.

[I'm imagining the author is not in a Western country based on the dropped indefinite articles?? Some idea of their experience would help give authority to the article.]

tl;dr it's a good attempt at an overview, but some things you can't broadbrush. At least seek an overview tailored to your target market.

Author here.

>>This may help in getting the patent, but it also enables someone else to exploit the same loop holes to file a patent for the copy of our invention by just changing the grammar/vocabulary because our patent was too generic. >That's not how that works.

I think, it's incomplete when the prior argument for that is not included.

This is the complete argument -

>Unfortunately, patent attorneys tend to force the inventor to not to be specific in describing their invention in the patent application in order to exploit the loop holes in patent examination; This may help in getting the patent, but it also enables someone else to exploit the same loop holes to file a patent for the copy of our invention by just changing the grammar/vocabulary because our patent was too generic.

That is, if you compromise while drafting patent someone else will exploit it the same way.

>Investor expectations is a big reason to get a patent.

I agree, especially if it's a 'proven' HW startup. Emphasis on 'proven', if the patent is not for a solution which solves a real problem with commercial viability, no investor is going to turn your side even if you get a patent i.e. in other words if you are not sure about the success of your product, patents are just a waste of time and wouldn't bring investors magically.

>I'm imagining the author is not in a Western country based on the dropped indefinite articles??


>Some idea of their experience would help give authority to the article.

Close family member is a patent attorney with 20 years experience, having worked for largest multinational companies and current work being facilitating patent deals(usually worth several millions) in a foreign country (My arguments in this content was validated with that attorney). I have personally applied and later abandoned International patents for my products in my earlier startup (The said attorney was not involved in my patent application due to work clauses).

I appreciate what you are trying do here. Startups can always use more education about IP protection. But IP law is very country specific, and it is one of the most complex areas of law.

Some of your myths/truths are dangerous and may cause companies to irrevocably lose IP rights that they probably should protect.

>But IP law is very country specific

I think that's exactly what I state in explaining the myths, there are 13 occurrences of the word 'country/countries' in my reasoning, I have linked to WIPO and PCT source article where necessary. I don't see a place, where I state universal applicability of my reason; I would definitely correct if mentioned.

>Some of your myths/truths are dangerous and may cause companies to irrevocably lose IP rights that they probably should protect.

That could be dangerous, point me to a specific statement. Much of the criticism by the parent (Which I heartily welcome) comes from picking up a sentence at random but not the prior argument; so they're incomplete.

P.S. I also have trademarks(wordmark/devicemark) for my products.

Agreed. Author is not a patent attorney, I’m willing to bet, and shouldn’t be doling our advice about an intricate and complex area of law that they apparently don’t have a deep understanding of. Approach with caution.

1) Patents are assets. They are important for valuation. Investors care about them and they are important in acquisitions or pre-IPO audits. You don't have to have patents, but not having any will reduce your valuation.

If a tech company is successful, it will have patents. If they do not have enough, they will sprint to get more patents before exiting. Though, if a company waits until they get close to exiting before seeking patent protection, much of their interesting/valuable IP may barred from protection.

2) While expensive, patents are not that expensive considering the other costs of operating a startup. Passing on patents in the beginning means you will lose the right to protect your early inventions. Often this results in the loss of IP protection for your most important innovations -- the secret sauce that the company is based on. Trying to catch-up later is harder and more expensive because you have to patent around the edges of your secret-sauce because it is too late to patent the secret-sauce itself.

3) Plenty of so-called software patents are issued in the UK, EU, and Asia. You can search the EPO patent database for patents assigned to Google, et al. to find plenty.

4) PCTs are placeholders that defer (for 18-30 months) having to decide if you want to file in a covered country. They are cheap so we often recommend them to clients that intend to sell into UK/EU/JP. If you file in the US first and then use the same US application for the PCT there should be no legal fees for filing a PCT, just filing and administrative fees.

Note, non-US countries do not have a 1-year public disclosure grace-period. So if your inventions are disclosed to the public before filing your US patent application, forget about PCT applications.

5) Patent Attorneys. They are like any profession. At least half are average or worse. I think finding a good patent attorney is the hardest part of the patent journey.

A good patent attorney will have relevant industry experience and they will focus on specific/limited areas of technology.

Patent attorneys that are good at software or computer-based inventions will tend to work exclusively with those types of inventions. This is because there is a high demand and patents for software/computer inventions command high fees. Gadget patents typically have lower fees that really are not worth the time for a computer/software oriented patent attorney to handle.

Also, even big law firms with name-brands can have terrible patent attorneys. So the best way to find one is word of mouth from other CEOs/CTOs or your VCs.

6) Startups filing for patents in the US should be doing Track 1 (Prioritized Patent Examination). It guarantees that your that your patent will be examined within 12 months. It only costs $2070 more than normal filing fees for companies with less than 500 employees.

If your US patent attorney has not recommended Track 1 filing, you probably need a different patent attorney. Track 1 is a 'new' option, but it has been around for several years now. There is no reason not be filing Track 1 if you are a startup. If your attorneys argue against it, consider seeking new attorneys.

In three years you could have several issued patents instead of still waiting for your first response from the USPTO.

7) A good patent attorney only needs 1-2 hours of inventor time to draft a patent. If your patent attorneys are meeting multiple times with the inventors, especially to work on claims, you should look into finding better patent attorneys.

8) All good patent attorneys charge high fees, some bad/mediocre patent attorneys charge high fees. However, all low fee work is done by bad patent attorneys (or they are sub-contracting out to bad patent attorneys).

There is too much demand for good patent attorneys for them to charge low fees. Unfortunately, plenty of bad patent attorneys charge high fees too, especially if they are at bigger firms.

This article is software centric.

Good luck getting VC in hardware without patents.

If apple can risk it all by going into the music industry then I think your small startup can take the risk.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact