Embedded - 71: Dirty Your Mindscape

Episode Date: October 8, 2014

Intellectual property attorney Judith Szepesi (@Judith_IP) discusses what Elecia (and startups) need to know about patenting. Judith is a founding partner at HIPLegal, LLP. They will soon have a guide to addressing patent trolls (link to be added when available). Ask Patents - a Stack Exchange site to discuss patents (and patent trolls)  Judith and Elecia both recommend the Patent It Yourself book from NOLO Press (always get the latest of this). Even if you seek legal counsel, you'll have a better idea of what should happen through the process. [Note: we got to talking after the show and Judith reminded me that if you do research for other people's patents, you should track that because you have an obligation to tell the patent office about whatever you are aware of that is relevant. -El]

Welcome to Embedded, the show for people who love gadgets. I'm Elysia White. My guest is Judith Sipeshi, a patent attorney here to help us navigate those dark legal waters. Judith, thank you so much for being on the show. Thank you for inviting me. Can you tell me a bit about yourself? Let's see. I am an engineer, an electrical engineer, fourth generation engineer, in fact. And I really enjoy learning about technology. And I realized as I was working on my first job and helping my father with a

patent, that I really enjoyed learning about the technology and talking about it more than I enjoyed actually having my hands in the guts. So I went to law school and decided to become a patent attorney, which I did. I was a patent attorney at a large firm for 18 years. And I recently founded my own small IP law firm called Hip Legal. Founding your own company, even a law firm seems adventurous. It's very exciting. And it gives me even more sympathy for my startup clients that I'm working with. I work a lot with startups and I see some of the issues that I intellectually knew were going to be difficult for them and I'm seeing it firsthand and it gives me a really better feel, I think, for what they're going through.

So if I start my own company, what should i know about filing patents well i think you should figure out why you want to file patents there is some reasons good reasons to file patents and then there's a lot of possible reasons for example if you want to be acquired it is an asset that would be evaluated when you're when somebody's looking at you to acquire your company and depending on what technology area you're in, it can be a sword or a shield, right? A patent can be a sword. You can say to a competitor, you can't do this. Or you can say to somebody who sues you, you have a patent that applies to me, but I have a patent that applies to you. So I can defend myself with this patent. So that's the

other aspect of it. For a lot of startups, they do it mostly as a marketing tool, to be honest. They look at it and they say, I'm going to tell my clients that I'm the only one who does this. This is new, and it's patented. It's so new that the patent office has acknowledged that it is something new and worth protecting. I'm seeing the patent pending as part of marketing materials. Exactly right. I understand that. Okay, so I need to decide why.

And I think for most of the companies I've been in, it's both the sword and the shield, making sure that we're in the space and nobody else can be here right now. I've done the baseball card version, too, where you get them so that when you get in trouble you can trade them so okay so my patent some embedded system, small gadget that I want to build

and I've decided I'm doing something novel because that's a requirement for getting a patent and I'm doing something more than software we'll talk about software patents later but I don't want to start there absolutely what do I do? and I'm doing something more than software. We'll talk about software patents later, but I don't want to start there because... Absolutely. What do I do?

Well, if you're looking to potentially patent something, I think you should start by taking a look around what's already out there. Remember that your idea has to be novel, right? And non-obvious, which means that if you're just putting together two things that are separate now, and all you're doing is inserting them into the same system, that is not patentable. If they-

There are a lot of patents that do exactly that. Yes, there are. Well, so in most cases, what they do is they're synergistic, right? So they work together and they change the outcome. The patent office training has this example. Somebody combined a grater and something that laid the asphalt and grated it flat. And it basically hooked the two things together, literally in one truck, two pieces hooked together. And the patent office said, look, you're hooking two things together. Neither of them changes how it does it or what happens. It's just like there are two trucks driving next to each other.

And that is not patentable. So that's what I tell clients, if you're just putting two things together, and they don't interact or change how they work, as a result of being together, then they are not patentable. So if before I was streaming video from my computer over Ethernet, and now I go over Wi-Fi or Bluetooth, that shouldn't be patentable that's correct doesn't mean that there hasn't been a patent granted for it but it should not be patentable that's correct it should yes yes and so what are the other criteria the the novelty the non-obviousness and then you have to be the first one and then the first but now it's first to file. It used to be it was first to invent. So

if you were the garage inventor and you kept notebooks and you could prove that this idea that now has made $10 million in licensing was yours initially, you were the first to do it in your garage in 1972, you could say, no, no, no, that patent, no, that belongs to me. But now it's not like that right well even back then it wasn't really because you not only had to show that you invented first you had to show that in the entire interim time until you filed your patent you were diligent so you invented something in 1972 put it on the shelf ignored it for 10 years or 20 years right and then somebody came up with it and made a product and you you said, Oh, I have something on my shelf about that, then you would not get the patent. But if you developed it in

1972, and you worked on it, and you developed it to productize it to make it work, and you're working the whole time, then you would be able to say, according to the old system, that the patent should belong to you. Now, it's first to file, which means that it is to some extent a race to the patent office, which means that we now advise startups and other companies as well to consider very carefully when they're going to be disclosing something and make sure they file something as soon as they can. Well, with first to file or with before the first to invent. I could invent something, put it in my notebook, make sure I had good notes on it, and then even write an article. And for a year or something, I could still patent it.

Because I had the documentation. That is still true under the new regime. The U.S. still has what's called a grace period. So from the initial publication, if you make a publication, you have one year to file in the U.S. still under the new regime. The risk is that somebody else files before you in that interim period who didn't get it from you. They developed it themselves. How do you prove they developed it themselves instead of reading my article? That's a very good question, and we have not yet tested it. But there is something called a derivation procedure, which says they derived their patent from my idea,

and therefore they shouldn't be granted priority over me. And as far as I know, there's been no decided derivation procedure cases yet. Wow, it makes those engineering notebooks go back to being important. A little bit, but most of the time we highly suggest that before you publish, you go and file the patent if you're going to be patenting. Yeah, that makes sense. Most lawyers that I've talked to have said, no, you can't publish that paper on this year's conference. Wait till next year's conference. See, I think that is one of the big differences between being in a small firm dealing with startups and being in a big firm dealing with big companies.

I would never tell my client, don't go to the conference because I know that in order to succeed, they need to. So what we would do is probably take that article, say, let's file a patent application now, a provisional application today, and then you can go and publish it, and you already have your patent date at that point.

Okay, so there is a multiple-step process. And I remember this. I filed a couple of patents at HP and a couple at various startups, and the process has changed a lot, but maybe that's just because each company was so different. The first step is the application. And that helps you get the provisional, right? Or am I skipping steps?

Well, as an engineer or an inventor, the first step is to meet with a patent attorney and do a disclosure meeting. And you explain what you invented, hopefully with pictures and a good explanation. Although I have had inventors come in with literally a napkin from a coffee shop so you get a description from them the patent attorney writes a patent application from that which includes usually flow charts and figures and text and

then you have to go read it which a lot of engineers hate but you do have to give that so hard the language is so difficult and, I think that is a choice. And you will find that younger attorneys, including myself, really prefer writing in English. I think if you can't read it, then you're not doing a good job. If the engineer can't understand what they invented, then I think it's a failure on the part of the attorney. So you prepare an application and then you file it. You can file it either as a utility filing or a provisional filing. The Delta is for a small inventor about $600 in filing fees.

Oh, that's not so bad. No, it's not that bad. For larger companies, it's 800, the Delta. How do they decide if you're a small or large company? The Small Business Administration has a definition, and it is a fairly straightforward one. I believe it's fewer than 500 employees. So most small companies are small.

The caveat is if you license it to a big company, then you have to pay big company rates. That's right. So you have to be careful if you're a small company who licenses. When you're licensing your IP, then you have to be very careful about the patents that are around that IP to make sure you use big company fees. Well, that prevents a big company from using my consulting company as a way of getting cheaper fees.

Exactly right. So the provisional gets filed for 600-ish, 800-ish, less than a thousand. No, 600. The provisional for a small entity costs, I think, $110 right now. Oh, okay. Or 200. Yeah, I think it's 100. And so you file the provisional, and the provisional is basically just a stamp. The patent office does absolutely nothing with it. They put their date stamp on it, and that's it. Then they ignore it. You have a year to file a utility patent that references that provisional patent.

And that's it. If you don't, the provisional patent disappears as it's never been. Yes. It never gets published. It never gets public. Nobody will ever get access to it. And it wasn't public at any time. That's correct. From the patent office perspective anyway. Well, yes. So there's very little risk. That's right. And when you do finally file the utility patent, there is some risk because people can read it. That's right. So for most patents, I think about 85% of currently filed patents, they publish them at 18 months, 18 months after the filing date.

And that is the original filing date. So if you had a provisional for a year and then you file the utility, it's six months after you file your utility patent that it gets published. And at that point, everybody can read it. It's on the internet. You can search for it under USPTO or even on Patents Google. And there's some, I mean, that's neat because it's sort of like open source, except the patents don't end up being like open source. I think originally they were supposed to be, it was supposed to be, I am doing this really cool process and I'm going to make money off it for a little while because I invented it and then everybody can use it. That's right. And that's,

in fact, the exchange of patents. You're supposed to publicly disclose how you do it so that everybody can figure it out and build on it. And then in exchange, you get 20 years of sole use, the right to be the only one who uses it. And after 20 years, it expires. And then the whole world can use it for free. So were patents the original open source? And how did we lose that vision? I'm not sure it would be considered an open source. It actually was built on historically the king would give a grant to somebody that they were the only ones who were permitted to manufacture something. And so that is really what it's built from. And that was really its basis. And one of the things that you see is if you see, for example, trade secrets are wonderful. That's the counter, right? The opposite

of patents is trade secrets. Never tell anyone how you do it. And you can keep it forever secret, which has its advantages, which means that, you know, Coke can still claim that its recipe is secret. That's right. Which is what trade secret gives you. You know, it's been more than 100 years since they changed the recipe, and it's still a secret. Patents only give you 20 years, and you have to tell the world how you did it. So that's the other half. The problem with trade secrets is you can lose them. There's been more than one case where nobody knew how to make it after the fact, right?

There's things that people made, and nobody now knows how how because they kept the process a secret. And it's also possible that you just lose the secret. Disgruntled employee, maybe you can go after them, but once the secret's out, it's out. That's exactly right. What is the most difficult or expensive part of the process? I think the most expensive part is the upfront writing. So for a utility patent, there's a bunch of requirements on how it has to be written. And you remember the exchanges that you publicly disclose how it's done. So you have to, your description has to enable someone to read it and understand what happened and what you did and what the invention is. So there's a

public disclosure aspect of it. So you have to usually hire somebody to make sure that the format's right, that everything is done right, and you file the patent. So that's probably the most expensive, and that's upfront, unfortunately, for a patent. And that is so that somebody, if I have this right, familiar in the art, can reproduce exactly what you've done. That's correct. With no secrets in between. Mostly, yes. That's a sort of, unfortunately. So the fact is that anything that you're actually claiming has to be disclosed.

But there is lots of things where you're writing where you can exclude certain aspects. So I'm claiming a car, right? And I'm going to go and describe the engine carefully. But I'm not going to tell you about how I did the suspension because it's not what I'm claiming. I'm not claiming the engine, but it only really works if you also have the suspension there. Ah, yes. So a combination of trade secret and patents is the best way to make sure nobody can copy you. It tends to be fairly often that that is a combination that is used, yes. And trade secrets are actually really valuable, especially for false paths. So one of the things is, you know, the patent has to work too. It's a very low threshold requirement. The patent office actually doesn't really look at it. They assume that you're not going to file something that isn't useful. And yet. And yet, yes. But for example, the WD-40 is a

prime example of this. It's called WD-40 because it's the 40th attempt at building it. So they tried 39 other paths that failed, right? So the 39 that failed is a secret, and they're not going to tell anyone what they tested and how they got to where they got. And so that is one of the ways that trade secrets can help you, is all the stuff that you tried that didn't work. You don't have to tell anyone about that, right?

That makes sense. So say I've gotten the patent. It's been two years since I had the idea. I filed the provisional, I filed the utility. And well, actually, after you file the utility, the patent office looks at it and there's a delay there. A long, long delay, depending on what art unit you're in that can be multiple years so yes and the software and hardware device gadget area i think

most of the patents come years after i've left the companies that i i worked on them for so that must be three four years sometimes that's not uncommon, yes. So usually what happens, what I tell my inventors is, they'll send it to the patent office and they'll ignore it for two years. And then they'll go look at it and they'll usually send you a rejection. They'll say, we found something that somebody did before that is the same. It's not new. You can't have the patent. And then the patent attorney, usually working with inventors,

says, no, this is why that is different. Yeah, yeah. So that's an office action exchange with the examiner, and you do that usually between one and four times, and then the patent usually issues. So do I have to do this all with a lawyer? How much of this can I do on my own? As in most things in life, it's your money or your time. So depending

on how much time you're willing to put into it, you can reduce the cost to you for getting that. Can you do it without a lawyer? Yes. Is it going to be as good? Probably not unless you put a lot of energy into it. But I absolutely, you know, do recommend that you look into it. There's a great book by Nolopress called Patent It Yourself. I highly recommend it to my inventors. I think it's a great book. It's like 40 bucks. And it leads you through the process. And there is an e-book, yes.

And it really, yes, I'm familiar with this book. I was familiar with it from about five years ago when I wanted to patent something. And I realized even in my day job that this book was fantastic because I was saving all the lawyer's time because I actually, I didn't write it properly. I never filed that patent for myself, but I got in the habit of, oh, you need this, this, and this. And so my lawyers didn't have to come back and say, well, what about X? And I'm like, oh no, I got you all

covered with all the claims going this way and that way and matrixing. The claim writing is a very difficult process because you do have to figure out how all of the claims can be put together. That's right. And writing a good claim, a good claim will claim one truly novel thing. That's the goal of the claim, right?

And it's a weird structure. It's a single sentence with semicolons between elements. And so writing good claim is hard. And you don't really know if you wrote it perfectly until it's in litigation, which is often 10, 15 years later. So that's one of the big downsides of writing your own is it might look good. And then somebody, a litigator looks at it and says, oh, no, look at the following words that you used. They have meaning. And this is now a problem. So working with someone who understands why you choose the word you choose is worthwhile. And so you recommend to your clients, they read the book in order to help them

not incur a lot of lawyer fees. So, I mean, it just helps knowing what you're doing, what the lawyer's looking for. That's right. That's right. So once they read the book, they know what I'm trying to get at, why I do the things I do. And some of the weird phrases that like patents always use in one embodiment, which is a weird, weird phrase if you think about it, right?

It's such a weird phrase. But it has meaning. it has specific meaning in the patent world. And so having read the book, you get a better understanding of why we do the crazy things we do. And it also, you know, in terms of saving money, the more of the patent you can prepare, the more of the pre-work you can do, the less it will cost you, right? So most lawyers bill by the hour. And so the less of my time you have to pay for, the cheaper it is. And usually it's, you know, probably it takes you twice as long as it would take me to do the equivalent work. So that's kind of where you have to look at it. But for example, a lot of my inventors will do their own figures, initial, you know, flow charts of how the process works. A lot of them

actually have white papers, for example, with architecture diagrams and things of that nature. And at that point, I can use a lot of that data. And then I don't have to recreate it. And that makes my process much faster, which makes everything cheaper. Exactly. And so the most expensive part of getting a patent really is writing it. It's not the fees from the patent office at all. That's right. It's the hourly fees. But you want to do something good because you want it to be defensible or why did you bother in the first place? Exactly right. So fairly significant time investment as well as money investment. So

you know, just looking at the patent office fees over the lifetime of the patent, it'll probably cost you $6,000. So that's not a lot comparatively, but if you're going to be, and probably hours and hours of your time. So don't waste it, right? If you don't think it's worthwhile, don't file it. If only more people took that approach. Well, you know, one of the things that I think is very funny is I've convinced more people not to file patents than you'd believe. Often people will come to me and say, I have an idea, I want to file a patent. And I look at it and I look at what's already out there

and I say, you know, you're not going to get anything that's worth it. You probably don't want to spend that money. So if I did, and I've waited my four years and i i have had a product shipping and i've been hoping that it's covered by patent i'd finally i get my patents now it's a matter of license and prosecute what are the next steps well that depends uh for most small companies you say oh good you make a pretty plaque put it on your wall and ignore it for the next 20 years. Usually you put them in the lobby so other people can see them as they come in.

That's exactly right. That's exactly right. So very, very small number of patents actually ever get enforced in any way. A small number get licensed. Enforced means you sue on them, right? Do you have percents? I believe it was, it's been a little while since I looked at it, but it was about three to five percent get ever get litigated and probably around 10 percent get licensed ever. So 85 percent of patents, that money went down the drain and made a little gurgling sound?

Well, no, because remember when we started, this is in part about marketing and in part about shields and swords, right? So when someone comes to you and says, hey, you can't do this product because I have a competing product and I have a patent. You can say, ah, but I have a patent on my product and you couldn't do yours. Let's cross-license. And that often wouldn't happen. Or you never really cross-license.

You just send them back a letter saying, you know, we think that's a lovely idea. Here's our patent list. If you feel the need to talk to us, let's talk again. And they kind of go, oh, never mind. That is a value of a patent, even though it never gets into that litigation thing, right? That makes a lot of sense. How do I go about licensing? If you're trying to license your patent to someone else? Yes. There is a couple of different paths. Most startups only license with the technology. So what they do is they're selling their technology as a license, right? I'm not going to build this for you. I'm going to license the IP to you, which includes patents. Right. So if I build a little board that does something fantastic, I will let you build a little board that does something fantastic and that will be licensed. Exactly right. And you'll pay me some amount. How do I decide how much it's worth?

That's a very difficult question. And there's people who actually specialize in royalty calculations. It really depends on the value of the end object and the value that you are providing with your IP, right? Yeah, if my IP is 10% of their product, and their product costs $100, I'm going to at least start out asking for $10. Exactly right. When they say, wait, no, it only costs us $15 to build, and other people are taking all these chunks, and we only have $4 in profit left over, then we negotiate.

That's exactly right. That's exactly right. So what about prosecution? I mean, that can be hideously expensive. So prosecution is the process between when you file the patent and when you get the patent, right? And so that's what is referred to as prosecution. Oh, okay, that wasn't what I meant. That's the process before you get it. Okay, so you mean enforcement, patent enforcement or litigation. Enforcement, yes.

That can be hideously expensive, incredibly, incredibly expensive. And it is something that I would advise any startup not to do if they can avoid it. And the reason for that is because it will take up all your mental space and it will take up all your money. And realistically, if you're a startup and you want to be successful, that's not where you should be spending your energy. On the other side of that coin, defending against an enforcement suit can also be very expensive and terrifying. How do you prepare clients for that? Do you work in that area? I don't personally litigate. I have a partner who's a litigator, former litigator, and I have a partner who's formerly in-house. So I get the full spectrum of opinions, but I don't personally do litigation. But the biggest thing you do is, you know, you try to

get all your ducks in a row and you try to get it out of your hair fast. And this is one of the things that, for example, trolls use to their advantage. They say, you can give us $20,000 and we'll go away, or you can spend $100,000 on lawyers and we might lose, but you will be spending a lot more money and time, very expensive money and very expensive time, on defending. Patent trolls, yes. Yes, that is in my list of questions. I never know how to approach that. I know that if you call somebody a patent troll and they don't believe they are, it's a grave insult. On the other hand, when I look at things like this podcasting patent, where they go after small podcasts and demand $10,000, I'm like, yeah, I'm not making any money off this.

You can have all the buying profits, all of them. That's right. The problem is that when you defend yourself, you have to pay a lawyer, right? And that those lawyers are a lot more expensive than I am per hour. So it's very, very expensive. And it is very unfortunate. One of the downsides of the patent troll business model is that they don't go after the podcast application maker, they go after the individual podcasters. There's a Wi Fi, a company that has some patents on Wi-Fi provision, providing public Wi-Fi. They're going after hotels,

individual like small hotels, because that's a lot cheaper than going after Cisco or Google, who made that equipment, who actually provides the service, right? Because they have the money. That doesn't seem fair. No, it doesn't. And it doesn't. And the patent office doesn't have any guidance about going after the person who is enabling the patent infringement versus the person actually infringing somewhat unknowingly? Well, this is actually not in the patent office at all. The patent office does not look at enforcement in any way. So this is all on the courts and the legal system, the litigation part of the legal system that this goes to. And unfortunately, there really isn't anything

that would keep someone from suing you. It's one of the things that people ask me sometimes, can I be sued? And the answer is, regardless of what the question is, the answer is yes. You can be sued for anything in the world. And they might not win, but they can certainly file a lawsuit.

Costs just a filing fee. And yes, that is a very big problem. Okay, so how do I not get sued? I'm not sure. There's been more NPEs, non-practicing entities, which is the politer way of referring to trolls. There's been a lot of NPEs that are going after startups now because of the same reason, because the Googles and the Apples are saying, we're not paying you. We're going to fight you.

And we have a war chest and we're not, you're not going to, you know, beat us into submission by just making us throw more money at it. But if you go up against a startup that has just received its first 50K investment, you can say, hey, I'll take a quarter of that to go away, or you can spend all of it on lawyers. Pick. Yeah, and a lot of these non-practicing entities, these trolls, come out right after you get a funding round. That's correct.

I've heard they even come out after you have a funding round. That's correct. I've heard they even come out after you have a successful Kickstarter. And all of that's very public information. So it's not hard for them to know that you've just raised a bucket of cash and to want to dip their fingers in. And if enough of them come after you, then there's your whole bucket and you have nothing. Yes, and it is a real risk.

I mean, an unavoidable one right now, right? I don't think, Congress is really trying hard to figure out something they can do that will actually make this problem less of a problem. But nobody has come up with a good definition of how to make this problem go away. The biggest thing that they've done, and the Federal Circuit is leading this, is they're saying, if you bring a lawsuit that is baseless, that you should never have been allowed to bring, then we'll make you pay the other side's lawyer fees. That's a hard thing to prove.

Entirely baseless? Very hard to prove. But in the U.S. system, otherwise, even if you win, even if you prove that they were absolutely wrong in their assertion, you still spent all that money on lawyers and that's never coming back. Oh, man. That's just annoying. And there's a reason I don't do litigation.

So what can I do as an engineer to fight patent trolls? I mean, it's a huge problem and I hate them because I feel like they do limit my scope. Do you have any advice? Well, actually, we're working on putting together a guidance kit for how to deal with trolls right now. My firm is doing that. So that should be going up on our resources page, hopefully in the next week or so. But the problem is that the best thing you can do is to band together with other small entities that are being attacked, which is

what a number of companies have been doing. They put together what's called a defensive pool. And they say, basically, we are all being sued, we're all being threatened for 10k. Let's put half of that or some percentage of that into a defense pool and hit back, right, go back to the patent office and say, hey, look at this patent, it should never have been granted. Let us show you why. Go look at it and pull it back. Pull back the patent. It's called a re-examination. And there is a patent stack exchange.com i don't know if you're familiar with it but some of my listeners might be and it's uh it's a place you can go to and look at patents and then say this is not reasonable and here's why and i'm an expert in this art and so if you want to go look at patents that never should have been granted if you want to go find whatever the next xor or cat laser patent is you can if if you want to give 15 minutes of

your time to read a patent and say no this is dumb and here's why you can absolutely and i actually am familiar with the sac exchange i have an account there but uh still i'm an engineer um but the problem with that is that most people, when they read a patent, and you'll find this especially on Slashdot and a bunch of other places when they complain about patents, they're talking about the abstract, not the claims. So you have to make sure you look at the claims and what exactly they say and what it means if you want to attack a patent. And that goes back to understanding how to write a patent yourself.

That's exactly right. Which is, and we talked about that book, which I have and will be in the show notes. And you mentioned you are an engineer. You studied electrical engineer. How is that important? Well, most of the inventors, I work in high tech, right? So hip legal, my law firm is high tech intellectual property. And so you work here in Silicon Valley, and I work here in Silicon Valley. And my clients

are mostly in software and in telecom, mobile devices, small devices. So that is what their language is. So if somebody comes to me with an invention, and they are talking about the bus, and they're talking about the, you know, the... When they talk about the bus, they probably don't mean the school bus. Exactly right. Exactly right. So I understand what they're talking about. And I do make an effort to keep up with the technology as well. So, you know, when they're talking about the cloud, I know that they're not talking about brain, which isn't going to happen. So there's a lot of language and a lot of understanding of the underlying technology. I don't want an inventor to have to come in and explain to me how the basic state of the art works.

I should know that already. That is very helpful. I've had some lawyers who didn't exactly follow the state of the art. And it does make it a lot harder to write the claims when you're trying to explain what the gadget does. Exactly. And how it's different from the prior art, right? If I don't understand what's already out there, how can I clearly differentiate it, whatever you invented from what's out there?

Did you ever practice as an electrical engineer? For a very, very short time. And then I got into software. Silicon Valley. Yes. And so do you write software now? Sadly, no. Not in a long time.

But having done an electrical engineering degree at a really good university, you went pretty quickly to law school. That's right. How hard was law school compared to electrical engineering? Not that hard. It's interesting. So law school is really interesting because it's really targeted at people who are good writers who have to learn logic. And as an engineer going in, the logic was really easy. The writing was really hard. So I had the flip side of most of their problems. I had to learn how to write an argument both sides.

In engineering, there's a right answer. In law, there often is not. I remember my husband, I think, got me an LSAT book for Christmas one year, not because I wanted to be a lawyer or take the exam, but because it's full of really great puzzles. Exactly right, yes. And so I thought that that part would be easier for most engineers. Absolutely. So that's good to hear.

I always hear such horror stories about law school, and I've always kind of suspected that the arguments wouldn't be the hard part, that the people would be the hard part. There's two hard parts for most of us. One is that you do have to do a lot of reading. I'm a reader, so it wasn't very difficult for me. And the other one is that you have to write a lot and you always have to write both sides of the story, right? So in law school, whenever you make an argument, you say, well, one side says this, the other side says this, this is why they're right or then this side.

In engineering, it's this is the answer and this is why this is the real right answer. Yes, yes. Perspectives are not always what we're good at when we have a problem. Exactly. But I think good engineers can see both sides of all the things, except for agile, which I just think is wrong. But that's...

Okay, moving along. Do you have any advice for how engineers can find patent attorneys to trust? What questions should we ask as new startups to find somebody who won't take all of our money and then walk off leaving us a patent that was worse than we could have written ourselves? So I think the first and biggest thing to realize with any hiring any lawyer is you're hiring an individual or a team, not a law firm. So don't go to Wilson's and Senior because they have a good name. Go to the particular person you want to work with. Because, you know, every larger law firm has a lot of people. And what you don't want is you get the pitch from the senior partner who is fabulous and brilliant

and understands. And then the second year associate who doesn't have any idea what's going on is writing your patent. So you want to hire the person who writes your patent. And my take on it is, you know, you're going to be wanting to look at other people's patents anyway. Take a look at who wrote those. If you find one you really like, you think it's cleanly written, it includes everything that needs to be included, nothing extraneous, and it's written in actual English, not in legalese, go call the lawyer. That makes a lot of sense. I've been doing judging for Hackaday and looking at all these fantastic projects. And I kind of wish they'd read each other's things

because some of them had really good ideas on how to present the information and others had really good ideas on how to do the video. And if they had looked more at each other's, it would have been fantastic. But this idea of reading other patents is a great one. I think it's just a useful way of progressing through the art. Books are easier. Books, well, books are easier and harder. They're longer usually. Although there are some patents that could be books.

The other thing that if you're actually looking for what's out there in reality these days, you should not be looking at patents. You should be looking at your competitors' products. You probably know who's out there, who's going to be your competitor. You should be looking at their press releases. You should be looking at their white papers. Because as we started with, patents take at least 18 months to publish. But doesn't that dirty your mind escape?

I don't know how to say that properly. That really wasn't it. So looking at patent claims might be problematic in terms of trying to copy them, but looking at what's already out there, you're going to be doing that anyway. I mean, I don't know any small startup who doesn't know what their competitors are doing too.

You have to do the market research or you're never going to succeed. Exactly right. So you're doing the research already. Just look at the tech as well and think about it from a patent perspective, not just from a competitive perspective. And let's go back to dirty your mind escape because that you have to invent in a vacuum and that you can't get information from outside or you have to give them attribution and then that weakens your patent. Is that just bogus? Yeah, that's pretty much bogus. There really is not a lot of risk from being aware of what's already out there.

The reality is that the patent office expects you to know everything. When the patent office looks at your patent, or when the court looks at your patent, when it was after it issues, they said you were hypothetically speaking aware of everything out there already. So it should be novel over everything in the world. Good luck. Novelty and clairvoyance, if only I had that. So the problem is that that knowledge is attributed to you anyway when they're looking at your patent. So it doesn't hurt you any to go look at their stuff.

It will, from a copyright perspective, it can be problematic. If you read a lot of code, you'll find yourself using snippets without thinking about it. That can be a problem. Because copyright's about copying. I read code. So from a copyright perspective, copying someone else's thing is a problem. From a trademark perspective, it's an even bigger problem.

Remember, trademark is the mark you put on it, like the name, company name, the logo. If you copy someone else's there, that's a real problem. From a patent perspective, much less so. And what do software engineers need to know about patents? I mean, should we just take away patents from software engineers and only let them have copyright? I think that's an interesting question and a very difficult one.

I'm of the opinion that the dichotomy between software and hardware is fairly artificial. You know, I programmed FPGAs, and I can probably implement something in hardware if you can implement it in software and vice versa. So trying to define something that makes software not patentable while leaving hardware alone is going to be very difficult. And I'm not sure it's possible, and I'm not sure it's logical.

Back in the 1960s, when software was officially not patentable, IBM got a bunch of software patents by couching it in terms of hardware. So what you're going to be doing by making these rules is saying you can only get a patent if your patent lawyer is smart enough, which, you know, is nice for me, but not necessarily useful in the larger world. Okay, but there are some pretty bogus software patents out there. Oh, goodness, yes. A lot of them. Why?

I mean, the patent office just started rubber stamping them? There is two big problems. Probably the easiest to define is that it was fairly new, right? Software was fairly new when it was initially allowed patents. And so the patent office didn't have a pool. The way that examiners look at whether something is patentable is mostly they look at older patents. If there weren't a lot of older patents, stuff that in software existed, but the patent office wasn't aware of, was a huge scope of stuff that got patentable. Because they should have been aware of it, but weren't. So one of the things they're trying to do is they're trying to set up a pool of software

descriptions and explanations that they're giving to the patent office. So they put together this giant database of stuff that's already out there and say, none of this is patentable because it's already out there. The patent office, the biggest issue is that the patent office is ignorant of what's out there because most of the patent examiners, although they have technical degrees, don't really understand the software world. Well, I mean, I'd like to believe I'm an expert in my little corner of the software world. And yet, it changes so fast. Exactly right. And if you talk about all of software, it changes blindingly fast.

I mean, much faster than chip sizes and all of those things. New stuff is being written every day that's sort of miraculous. I mean, now they have spectrometers that are the size of your finger and radar arrays that fit in like a coffee can. It just, how can they possibly have a wide enough scope to ever succeed in that endeavor? And that is one of the inherent problems of a patent system. But what they try to do is they basically give to examiners a small slice.

They say, you are only working on mobile devices that are primarily for texting. And that is all they're working on. So the patent office classifies. That is quite the're working on. So the patent office classifies. I know. The patent office classifies your invention. When they get the patent, they actually put it into a class. And then they assign it to an examiner. And that examiner is supposed to be working only in that little class.

So you'll find, for example, I have a lot of patents in the mobile space right now. And so I'll find the same examiners over and over. They're seeing the same kind of things over and over as I'm patrolling around. It's kind of good for you. You start figuring out what they're going to ask about. That's exactly right. One of the good and bad things is I will also find out that they keep using the same references, which even if they're not the most relevant references in the world, those are their favorites. They know it very well. I have a hammer. It must be very well. I have a hammer.

It must be a nail. I have a hammer. Yes, yes. Most examiners I deal with are very, very nice people and smart people, and they work hard. So the problem is that they're supposed to produce a lot of work. And so it's always easier to reach back for the same tools you know. They're not clairvoyant either. That is the other problem, yes.

Clairvoyance would help all kinds of things. It really would. So, okay, software engineers and all engineers, electrical engineers, what should they know about patents? Assuming they aren't the architect who's writing the patents, what should they know in order to be an effective engineer? I think they don't need to do anything about patents. If you want to be a line engineer in a big company and you're not inventing stuff, you're implementing other people's ideas, you don't need to think about patents at all. If you want to have a startup, if you are looking

to architect a new product, if you are looking to add new features that you think are inventive, that's when you need to think about patents. So for a lot of people, this doesn't really matter. True. Good. Open source. Open source is a growing field, and I think you work in that field. But how do patents affect open source?

So the scope of my work is a little broader than patents. I work a lot with startups, and in most cases, I'm the only IP attorney they work with, only intellectual property person that they ever talk to. And so I talk to them about employment agreements, how to make sure that your employees assign their rights to you. I talk to them about open source and how do you use it without tainting your own product. I talk to them about trademarks, which really, in the end, are the most important thing if you're going to be customer facing.

Yes. Trademarks. Absolutely. How you name your product can kill you or make you successful. That is the only thing that your customers mostly are going to be aware of. They're not going to know about your patents, your copyrights, your trade secrets. They're going to know about your trademark because that's the name splashed on everything. Right?

Well, then how come there are like six products out right now called Aria? Are there? I did not know that. Between booking hotels and looking at various things, I just keep hitting them. They're not supposed to be. So I try to do all of that scope. And so open source doesn't relate to patents directly,

but it relates a lot to what my clients are doing. Because these days, I don't know anybody who is doing any software work and really any hardware work without using some open source. Yeah, it is hard to avoid that. Even if it's just the code you get from the processor vendor, that code does have some liability to it in that you didn't write it. You don't own it. Are you sure you have permission?

That's right. And that's one of the things that we actually also do, which is licenses. So if you're going to be licensing your code in, making sure that that license says, and we own this, we promise. And if we don't, we are the ones who are liable is really important. Ah, yes. So what do you tell them about open source? Do you ever have to say, you can't use that, you have to rewrite it? Sometimes. Most open source, unless you're using something like the Linux kernel, which has so many owners, you're not going to be able to license it. But most smaller open source segments can be licensed under other terms. So I have had more than one situation where I was helping a client, most of them in

throes of acquisition, to clean up their record. And we contacted the person who wrote that open source segment and said, we'd like a commercial license, please. Yes, they still own the copyright. And they may have open sourced it through GPL or whatever they chose. But in the end, they own the copyright, they can make it another thing. That's exactly right. So they can license it to you under whatever license you can negotiate with them. And some of them are really fun. I had one guy who wrote a software segment that one of my clients wanted to use. And he said, if you buy my O'Reilly book for every engineer in your company, you buy a copy of my book, I will license it to you. So that's what we did.

When the show ends, I'm going to ask who that was because I bet I know him. I don't remember his name. It's been a long time. But yeah, okay. I can see that. I mean, because when you write open source code, you're doing it as a sense of community. You're putting it out there. And if somebody offers then to pay you for it three years later, you're like, yeah, okay, a little money would be nice. But how can I charge a huge amount for something I did for free a long time ago? I certainly have hit that a couple of times.

Exactly. So in most cases, I don't think I've ever said you cannot use this at all. I said you can't use it with this license. You need to go figure out how to license it under a different license. And talking to the open source owners, copyright owners, can be really amusing because some of them are interesting people. A lot of them are really fun people. One of the things I noticed written up about you was that you do interact well with engineers.

Do you ever talk to other lawyers about better ways to interact with the sort of nervous engineers that come in to talk to you about an idea they have they're super excited about, but not super excited about talking to a lawyer about? Yeah, I used to do a lot of training at my old firm, because I do think that that's an important part of doing things. So I think the biggest thing to realize is that this may be everyday process for you, but it's not for them. And so you have to understand where they're coming from, right? They're going to be spending a bunch of their time and money on something they don't understand. And that's scary.

Especially for many of the engineers who are used to working on things they understand very well. Exactly. You get to be an architect, you get to be the person who starts writing the patents, and now you want me to do something completely foreign? That's kind of odd. Yeah, and it's very expensive too, so it's very difficult for a small startup. So it's a big decision. And making sure that they understand that you are not adversarial,

that you are trying to work with them, and trying to make sure that you look out for their interest is what you're supposed to do as a good lawyer well you mentioned you were at a firm where you did training but fairly recently you co-founded a new firm a hip legal that's right and you said hi tech intellectual that's right becauseTIP legal doesn't sound as good. Exactly. I just like TIP legal. I thought it sounded good. What is it? I mean, how is it special? So we just founded this six months ago. And so there's two big things that I think make it fairly unique. One is that it's three women partners who founded it. We all have, I think, 15 plus years experience. And so one of them was a litigator for a long time. She does a lot of transactional work now, but she litigated patents

and trademarks and trade secrets and copyrights. What is transactional work? Transactional work is when you're not in court. It's work that's not adversarial. So you're not fighting with someone, you're working on licenses, you're working on patents. It's all called transactional work. And one of them was in-house for over 10 years. And so really has a good perspective for what the business end is really looking for, what the folks who are on the inside are looking for. And I've done a lot, 15 plus years of patent prosecution and patent strategy for my clients.

And so between the three of us, we really cover the IP space really strongly. And that's fairly unique. And in patent law, especially, women founders are a very, very, very small number. And does that help you get more women-founded companies, rare as those are? We're trying. We have not yet had a lot of luck with that, but we're working on that. I think all of us are really supportive of women and trying to make sure that they have a chance to found, as you probably know, it's very, very difficult to get funding as

a woman. And it's very difficult to be a lead person or startup. And so we're very supportive of that. And Annie Rogowski, who is one of my partners, founded the Women's Club of Silicon Valley, which is also trying to do exactly that, which is enable women to be as successful as they can be and to encourage them to not feel alone when they're doing these things that nobody else is doing or a very small goal and i wish you the best of luck with it uh let's see i have a few more questions if you have a few more minutes absolutely you have a blog judithip.com it sort of came i have a very active twitter account i really like to read twitter and i really like to post on it's It's Judith IP and Judith underscore IP to be precise. And so the blog happened because I kept posting longer things that kind of slid into multiple tweets and I hated it. So I decided I was going to put anything that didn't fit into a single tweet on a blog. So it comes and goes. I think it's not been very active very lately, but

whenever I have something longer that just doesn't fit into Twitter, that's where it goes. One of the things I like best about it was that you put up the blogs that you read, and there were some general legal and some patent ones, but also high-tech patent areas that I think probably are very educational for me. How do you find these? I spend a lot of my time keeping up in technology and law because not as fast as software, but the law moves pretty quickly too with court decisions coming down

that change how you're supposed to write and change the best way that you can work with your clients. So I really work pretty hard to keep up with all that. And so I've accumulated them over the many years, I think is maybe the best way to put it. And I think it is as important to stay current in software and electrical as it is with law. So that all makes a lot of sense. Well, I think that covers it for my questions. Judith, any last thoughts you'd like to leave us with? I think the biggest thing that I want to make sure that people who listen think about is getting a patent's only half of the process.

Figuring out why you want to patent and what you want to patent is important. And having a strategy that makes sense is important. So one of the things that made me want to go out on my own is I really like working with small clients and really understanding their business needs and what they're looking for. And as I said, I've talked to people out of patents more often than I'd like to think about when I look at my bottom line. A billion hours.

Exactly. It's like, oh, I could have had work. But the thing is, I want them to really get what they need, get what they need to be successful from me. And so having someone who looks at the strategy, not just the patent, this individual patent, is really useful. And I highly recommend you work with someone who understands why you're doing it and what you're doing exactly. And not just says, okay, here's a disclosure of a patent. I'm going to go write you a patent. Period. I'm not going to look at why. I'm not going to look

at who your competitors are. I'm not going to look at how it fits into your portfolio overall. Right? All of those things are things that somebody who's working with you should bring to the table. Yes. Okay. So if I go to someone and say I want to write a patent and they don't ask me why, then I will move along. That's right. If they don't ask you any questions beyond the technology,

you should probably move along. Well, I've learned a lot. Thank you so much for being here. Thank you for inviting me. I really enjoyed it. My guest has been Judith Sipeshi, intellectual property attorney and founding partner at Hip Legal. You can find her and them online at http://hiplegal.com. As always, thank you also to Christopher White for producing.

He'll feel better next time and join me on his squeaky new microphone. And I mean squeaky new, not squeaky microphone. I should have said shiny. Anyway, thank you all for listening. We occasionally ask for iTunes reviews, especially if you like the show. Okay, we only ask for reviews if you like the show. Please give us some lovely stars. It helps get new people,

and that helps us get more guests, and all of the eggs and the chickens continue. If you'd like to talk to us directly, email us show at embedded.fm, or hit the contact link on embedded.fm, or tweet at embedded.fm, all the ways. Final thought for this week is not a lawyer joke. That would be funny, but rude. Instead, I think from Stephen Hawking, who is probably somebody who reads and keeps up on his technology. We think we have solved the mystery of creation. Maybe we should patent the universe and charge everyone royalties for their existence.