I was a bit disappointed with the content of the article.
On the style, the generalization "Only assholes get patents" turned me off quickly.
But on the content itself, there wasn't much in there that related to the title of the article "Defending your app from copies and clones": the author talks about copyrights, and trademarks (not patents, because apparently it's only for assholes), but explains how little use it is: am I supposed to be encouraged or discouraged to follow those paths?
Then the rest seems to be an exercise in despair, convincing you that there's nothing you can do anyways and you're screwed.
I ended the read a bit perplexed, as I wasn't able to squeeze any substantific marrow.
Would have to say for most technology products, there are not much you can do. It is better if you keep your heads down and try to improve different aspects of your apps.
And try best not to compete directly with big players
He was just distilling the real world implications of the 3 paths of IP in a succinct way. I agree with him on all points, and he is accurate in his statements. He's inviting you down from a pedestal you may have put you and your software on: if you create something great, expect to be copied, and expect it to be considered a standard feature when someone else does it. Because really, thats how "features" are born. You can trademark your name for it, which will make it harder to advertise the feature as the same.
I think this piece was intended as an eye opener, and a blunt one, because the world isn't going to soften its blows on you.
Everything in the article was related to the title, but I think your expectations were wrong. Instead of getting a list of things you can do to protect your app from copies, the author goes through the list of possible tools and explains why they don't really work.
However, this also means you can do the same thing to others without much fear so it really comes down to who can learn, adapt and create more quickly.
Not really - It's just that I don't feel like I learned much: I already knew about the challenges he mentions. Whether I like it or not is irrelevant. I was really interested in something "new" or "original" about it. That was my (perhaps misguided) expectation.
> Everything in the article was related to the title, but I think your expectations were wrong. Instead of getting a list of things you can do to protect your app from copies, the author goes through the list of possible tools and explains why they don't really work.
Yes, perhaps I shouldn't have set my expectations from the title "Defending your app from copies and clones", where I thought I was going to learn how to defend my app from copies and clones.
I think that's one reasonable interpretation of the title, although I can see how someone can argue that he never promised he'd give any solutions or original ideas in the title. It could indeed have meant "Defending your app from copies and clones: a lost cause". Those last 3 words I added do make it a better match for the content, don't you think?
Marco rubs me the wrong way. Blanket statements like this are not meaningful.
What the post is saying is: there is nothing you can do. At the end of the day it doesn't matter who does it first, it matters who does it best. Do it best and you'll win.
If patents were usefully enforced and practical to obtain, it would matter who files first because that’s how patents work.
The $20,000 part of patents (and additional legal fees) are a problem because we are clearly at the point economically where tons of people who need stronger protections will never have the resources to obtain those protections. You can’t do that when it’s impossible to sell anything for more than 99 cents.
And we can't expect everyone to pay more for many of these things when there is so much inequality, lack of wage increases, student debt, etc. etc. etc.
Of course I go ahead and pay $50/year for ulysses or $20 for each Things 3 app, but my sister won't pay $3/m to iCloud on her filled up iPhone nor $5/m to Spotify or Apple Music to listen to music.
I think its the other way around: When you are a millionaire you probably don't care much about patents, because you can afford to pay for the license.
You don't have to sue people if they recognize that you invented and patented something. I've licensed patented tech to folks who range from solo devs to large nonprofits, and discussions never veered toward litigation at any point. I think it depends a lot on what you've created and whether your patents are likely to stand up.
You can make money by either licensing the patent or selling it. Large companies are usually prepared to buy out your patents if they think the tech is worth it.
Quite the contrary. The millionaire knows he can take any idea he wants from a young upstart if it's not patented. He can hire more people, spend more money marketing his clone, get the big PR firm working for him, and relegate the true innovator to a historical footnote (at best).
If there's a patent, the force of law can be leveraged by the upstart to protect himself from such incursions, at least theoretically, and it requires expensive, tedious effort for the millionaire to circumvent the patent and/or defend himself against claims that he has infringed. Even if most inventors aren't going to pursue this course, the fact that the possibility exists is a deterrent.
Patents are not necessarily done well, but like many systems that have been co-opted by large corporate interests, they exist precisely to prevent such exploitation.
Software people must understand that software is not very hard or expensive to replicate. You don't get acquired because your software is good. You get acquired because you have a brand, an audience, a trademark, or a patent. There has to be some asset that the prospective acquirer cannot easily recreate by throwing 3-5 coders into a dark closet for a year or two (this costs around $1M / yr, much cheaper than any newsworthy acquisition).
So yeah, millionaires are likely to be very irritated that there's a patent that may force them to compensate the original innovator, or worse, buy their company entirely.
I think he meant it in the context of "Defending your app from copies and clones" (aka "to defend you intellectual property"). If you hold patents just to make them available for ever and free to anyone but patent trolls, that might be a different game.
It's literally the ONLY way to defend against copies and clones. At least, if you're dealing with lots of money and big ideas.
Marco is coming at from the perspective that everyone is a nerd, reads hacker news and knows the difference between "free beer" and whatever the other one is.
Most, nay, all software companies get shit patented. It's the difference between having a legal document protecting your ideas and, well, not.
Yeah, this is just bad advice. I think for most indy developers patents aren't worth the trouble for defending your IP against other indys--which, in the context of the article is probably what he meant. But for defending yourself against major players? Absolutely. And if you want a major player to buy you (instead of buying one of your competitors) it'll be the key differentiator.
> But for defending yourself against major players? Absolutely
The point is that you have no chance to defend yourself. Can an indy developer afford a legal defense of the scale required to "defeat" a major player? If they can will it even be worth their time and energy?
Oh, i tend to disagree on that. He who comes first trains user-interface behaviour expected. If s/he is fast enough, the people will expect a certain behaviour - and be always annoyed if the others do not have it.
So if you are good and stay in the race, and you are first- that can be something that binds customers.
This reminds me of when the game 2048 came out, and became hugely popular. It was mostly a clone of the game “Threes”, released a few months prior, which while popular was less so because it was iOS only whereas 2048 could be played freely in the browser.
The press and public opinion was mostly on the side of the Threes developer, perhaps because they were already fairly well respected in the indie game dev community; see for instance
I mostly agree with Marco’s advice; but if you have your peers’ trust and approval, calling out your copycats seems like it might work in your favor (by bringing attention to your product for people who only know the copy). Or rather, let your hardcore fans call out the copy cats for you, and be brutally candid in your reaction - the devlog format of the Threes developers’ response was perfect for that.
All good points but I would add that you should never make it easy for people to copy valuable things from you if you depend on being unique.
Unfortunately what’s easiest for programmers is pretty easy to copy. For example, any ordinary file in an app bundle is dead simple to copy (and unfortunately the most straightforward APIs for using data from code will assume that files are plainly located in the bundle, which I always felt was a bit of a design oversight).
One fairly low-effort defense against copying is to use your program to reassemble valuable data from parts. For example, instead of having easily-stolen complete files in your bundle, store a variety of bits and pieces that only the program knows how to assemble correctly. You can even create a technically-corrupt/invalid binary file and then have your program restore the bits that you know you removed. Obviously this is an extra pain to do but if you really need to reduce the chance of losing to a competitor then it may be worth the trouble. (Note that this isn’t foolproof either, as clever people can still figure out what a program actually does; it’s just way better than doing nothing.)
> Only assholes get patents. They can be a huge PR mistake, and they’re a fool’s errand: even if you get one ($20,000+ later), you can’t afford to use it against any adversary big enough to matter.
Okay - Google patented their search. It's enabled them to stop the whole industry. Lets not go too far.
For the longest time now I've always held a bit of bias towards closed-source software, but over time I'm slowly realizing that open source is probably the way to go.
If you look at the world as if it was running on a tech tree in a video game, making closed source technology only forces people to retrace your footsteps(re-research the tech, if you will) or worse, find a better alternative (leaving you and your creation in the dust). By open-sourcing, you effectively stake a flag with your name on it in the tech tree and incentivize people to use your creation as a foundation for future techs. Sure, people will be free to use copies of your work at will, but they still have to give credit to you, and won't be trying to do things differently.
And that's not even mentioning that open-source stuff generally has a PR-boost compared to closed-source. If you make something closed-source, you will always have to worry about the appearance of an open-source competitor. I have seen closed-source stuff be completely eclipsed by open-source alternatives once the latter reached a critical level of quality. I mean think about it: Unreal Engine 4 is open source, would anyone start a closed-source project to compete against it? All the best minds will naturally ask you, 'Why not just work on UE4?' Yes you can fork it, but the public will always turn to the original of the fork unless your fork achieves something really different in a good way.
So in a way, open-source is the most asshole-way you can do things cause it shuts down any idea of a copy/similar alternative to your work.
UE4 isn’t open source. You get source access with the license but it’s quite restrictive on what you can actually do with it. If you tried to setup a competing engine based on a fork of it you would quickly find yourself being sued.
Godot would be an example of an open source game engine.
Thanks for the clarification. I believe UE4 is still a good example of how the PR-bias towards anything 'open-source'. All it took was a few headlines with "Unreal" and "open-source" and I naturally grouped it with the likes of Blender and Godot.
Right but all that means is that Epic were able to exploit a lack of understanding. Not really that noble or exceptional. There is a PR bias towards open source because it’s seen as a selfless act to provide free software. Conflating that with commercial software such as UE4 actually takes away from that since it dupes the unwary.
Tell that to 321 Media Player, who raked in millions by illegally copying VLC.
Simply attaching a big bad license to your code doesn't prevent 3rd world shovelware farms from copying your work and making tons of money from it for free.
Originating a widely-used open-source technology is a huge liability with virtually no financial benefit. It's a goal an engineer fetishizes and a product creator recoils at. I can't pay my rent with a pat-on-the-back buried in some acknowledgements screen, but hooray for my competitors being able to copy my product in a tenth of the time and effort.
Also, if everything were open-source, then nothing would get brownie points just for being open-source.
But if everything were open-source, anything new that's closed-source would get negative-brownie points cause it's not free.
That's the thing about open-source though, it's a race to the bottom for market share. We saw it with React and Vue.
I also get that open-source is a maintenance liability with issue and requests, but I also see plenty of open-source software that do well "provided as-is".
That's where copyleft (GPL-style) is better than BSD-style. Competitors can copy your product, but you can copy back. You presumably have some first-mover advantage in an existing userbase, so you can improve your product and retain those users + gain more. If you went with a BSD license and the competitor stayed closed you'd be up the creek.
That's how it's supposed to work, but what are you going to do when a start up on the other side of the planet takes your GPL code to build something great but doesn't share their changes?
UE4 is under a proprietary license where you can’t distribute your fork. It’s source is available, but it’s not “open source” in the copyleft/freedom sense that the phrase is usually used in.
As for closed-source competitors: game engines have always provided their source under proprietary licenses to people who paid enough, so the only real difference now is that you can get access to it (but with limited rights) without paying anything. But besides a few big names (Unity, Unreal and CryEngine/Lumberyard) there are quite a few closed-source competitors in use that don’t have generally available source code, although many of these are used only within a publishers development studios (idTech usually got open sourced after some time but until then the code wasn’t available to the public, Game Maker Studio, Frostbite etc)
There is no opinion of him on Apple implementing features from third party apps into it's platform. Would be interesting since he's such a big Apple fan.
"Only assholes get patents"... Everyone would get and enforce patents if it weren't so hard and expensive to obtain them... I think saying don't get software patents is terrible advice...
Of course, don't go and patent a mediocre interaction or technique, but definitely, pursue a patent if you have worked hard on researching and developing something that is key to your business and that could be used by a better established and wealthier competitor to take a market advantage.
Would they? The few times I've had a conversation with people in software on the topic, the conclusion has been they shouldn't exist. At all.
I'd venture to suggest that most software patents are the obvious way of doing something. Even having read some attempts to justify them I see no merits in patents for XOR cursors, online auctions or shopping carts.
> Defending your __________ from copies and clones
Fill in the blank, and the same lessons apply.
Any idea, project, business, app, website, article, podcast, TV show, media production, procedure, product, art, creation, etc are frequently copied and cloned. And he's right, ultimately nobody cares.
Sometimes a copy or clone is better than an original. Sometimes a copy or clone is much worse, but ends up successful because of positioning, luck, marketing, or savvy.
Perhaps it's better to think of the risk of being copied or cloned as part of the risk of business and creation. It just comes with the territory.
I guess Marco really really LOVES assholes. His whole career revolves around Apple/iOS, a company legendary for it's frivolous patents & history of enforcing them vindictively when the competition's execution poses a threat.
How are copyrights or trademarks ANY different? Together, they constitute essential tools in the knowledge/service economy. Demonizing patents but favoring the other two is ignorance on Marco's behalf - no one wants frivolous patents clogging the patent office, courts or legal departments of companies.
Apt advice would be to carefully research recent verdicts and legislation around patents, and only apply for ones that merit the designation (yes - this is very subjective).
Maybe it was just me, but I got the sense the target audience of his piece was individual or small teams. Hence, "even if you get one ($20,000+ later), you can’t afford to use it against any adversary big enough to matter."
These comments citing Apple and Google with their effective use of patents as some sort of counter argument seem to have missed the mark. Those guys are certainly able to afford $20k, and they can certainly afford to take on any adversary big enough to matter.
If one of your potential exit strategies is acquisition, patenting your technology makes sense. Even if he's right that a startup can't afford to defend it's patents, if Amazon/Facebook/Google/Apple acquires you your startup is potentially 10x more valuable to them because they can definitely afford to defend the patent.
> You can publicly call out a copy, but you won’t come out of it looking good...These disputes are best kept private, or not fought at all
Interesting take, considering how Marco has called out both Pocket (formerly Read It Later) and Readability as having copied features that he claims to have created first.
Perhaps his recent comments reflect how well his prior callouts were received?
EDIT: curious to know why this was downvoted. It wasn't meant as a dig at all—I was pointing out that Marco has experience with callouts and might have relevant expertise that many people lack. I would personally be very interested to hear his experience went dealing with those issues.
Don't know how much it will help, but I upvoted you.
I used to like Marco, but after following him a while and noticing inconsistencies of principle like this, it got to me. (That, and when he said in a podcast that he didn't use SQL joins when building Tumblr because he doesn't like them. Trying to think about how he could have possibly accomplished that makes me shudder.) I can give him the benefit of the doubt that perhaps his perspectives have changed with age or experience, though.
All that said, I upvoted the OP too, because as much as I don't like the guy, he's not wrong here.
I stopped following Marco years ago because I didn't like the tone of his writing, but this wasn't anything like what I remembered, or what I was prepared to read based on other comments here. It was clear, well-written, and not incendiary like I expected.
Only assholes go around trying to enforce patents, especially as a business model. But if you don’t get any patents, you’ll be really vulnerable to these assholes
> Only assholes get patents. They can be a huge PR mistake, and they’re a fool’s errand: even if you get one ($20,000+ later), you can’t afford to use it against any adversary big enough to matter.
56 comments
[ 2.8 ms ] story [ 96.3 ms ] threadOn the style, the generalization "Only assholes get patents" turned me off quickly.
But on the content itself, there wasn't much in there that related to the title of the article "Defending your app from copies and clones": the author talks about copyrights, and trademarks (not patents, because apparently it's only for assholes), but explains how little use it is: am I supposed to be encouraged or discouraged to follow those paths?
Then the rest seems to be an exercise in despair, convincing you that there's nothing you can do anyways and you're screwed.
I ended the read a bit perplexed, as I wasn't able to squeeze any substantific marrow.
And try best not to compete directly with big players
I think this piece was intended as an eye opener, and a blunt one, because the world isn't going to soften its blows on you.
Everything in the article was related to the title, but I think your expectations were wrong. Instead of getting a list of things you can do to protect your app from copies, the author goes through the list of possible tools and explains why they don't really work.
However, this also means you can do the same thing to others without much fear so it really comes down to who can learn, adapt and create more quickly.
Not really - It's just that I don't feel like I learned much: I already knew about the challenges he mentions. Whether I like it or not is irrelevant. I was really interested in something "new" or "original" about it. That was my (perhaps misguided) expectation.
> Everything in the article was related to the title, but I think your expectations were wrong. Instead of getting a list of things you can do to protect your app from copies, the author goes through the list of possible tools and explains why they don't really work.
Yes, perhaps I shouldn't have set my expectations from the title "Defending your app from copies and clones", where I thought I was going to learn how to defend my app from copies and clones.
I think that's one reasonable interpretation of the title, although I can see how someone can argue that he never promised he'd give any solutions or original ideas in the title. It could indeed have meant "Defending your app from copies and clones: a lost cause". Those last 3 words I added do make it a better match for the content, don't you think?
Marco rubs me the wrong way. Blanket statements like this are not meaningful.
What the post is saying is: there is nothing you can do. At the end of the day it doesn't matter who does it first, it matters who does it best. Do it best and you'll win.
Very clickbaity title. This could have been a tweet (and it was: https://twitter.com/sandofsky/status/966721199052013568)
The $20,000 part of patents (and additional legal fees) are a problem because we are clearly at the point economically where tons of people who need stronger protections will never have the resources to obtain those protections. You can’t do that when it’s impossible to sell anything for more than 99 cents.
Of course I go ahead and pay $50/year for ulysses or $20 for each Things 3 app, but my sister won't pay $3/m to iCloud on her filled up iPhone nor $5/m to Spotify or Apple Music to listen to music.
"Coincidentally, superior execution is also the most effective way to steal a good idea."
Right. Everyone should give their content to third party if they want to share it.
Does Mastadon float your boat, versus twitter?
If there's a patent, the force of law can be leveraged by the upstart to protect himself from such incursions, at least theoretically, and it requires expensive, tedious effort for the millionaire to circumvent the patent and/or defend himself against claims that he has infringed. Even if most inventors aren't going to pursue this course, the fact that the possibility exists is a deterrent.
Patents are not necessarily done well, but like many systems that have been co-opted by large corporate interests, they exist precisely to prevent such exploitation.
Software people must understand that software is not very hard or expensive to replicate. You don't get acquired because your software is good. You get acquired because you have a brand, an audience, a trademark, or a patent. There has to be some asset that the prospective acquirer cannot easily recreate by throwing 3-5 coders into a dark closet for a year or two (this costs around $1M / yr, much cheaper than any newsworthy acquisition).
So yeah, millionaires are likely to be very irritated that there's a patent that may force them to compensate the original innovator, or worse, buy their company entirely.
Marco is coming at from the perspective that everyone is a nerd, reads hacker news and knows the difference between "free beer" and whatever the other one is.
Most, nay, all software companies get shit patented. It's the difference between having a legal document protecting your ideas and, well, not.
The point is that you have no chance to defend yourself. Can an indy developer afford a legal defense of the scale required to "defeat" a major player? If they can will it even be worth their time and energy?
So if you are good and stay in the race, and you are first- that can be something that binds customers.
Here’s the developer’s statement at the time:
http://asherv.com/threes/threemails/
The press and public opinion was mostly on the side of the Threes developer, perhaps because they were already fairly well respected in the indie game dev community; see for instance
https://www.wired.com/2014/05/threes-game-design
I mostly agree with Marco’s advice; but if you have your peers’ trust and approval, calling out your copycats seems like it might work in your favor (by bringing attention to your product for people who only know the copy). Or rather, let your hardcore fans call out the copy cats for you, and be brutally candid in your reaction - the devlog format of the Threes developers’ response was perfect for that.
I'm going to educate the public to avoid macro.org
Unfortunately what’s easiest for programmers is pretty easy to copy. For example, any ordinary file in an app bundle is dead simple to copy (and unfortunately the most straightforward APIs for using data from code will assume that files are plainly located in the bundle, which I always felt was a bit of a design oversight).
One fairly low-effort defense against copying is to use your program to reassemble valuable data from parts. For example, instead of having easily-stolen complete files in your bundle, store a variety of bits and pieces that only the program knows how to assemble correctly. You can even create a technically-corrupt/invalid binary file and then have your program restore the bits that you know you removed. Obviously this is an extra pain to do but if you really need to reduce the chance of losing to a competitor then it may be worth the trouble. (Note that this isn’t foolproof either, as clever people can still figure out what a program actually does; it’s just way better than doing nothing.)
Okay - Google patented their search. It's enabled them to stop the whole industry. Lets not go too far.
If you look at the world as if it was running on a tech tree in a video game, making closed source technology only forces people to retrace your footsteps(re-research the tech, if you will) or worse, find a better alternative (leaving you and your creation in the dust). By open-sourcing, you effectively stake a flag with your name on it in the tech tree and incentivize people to use your creation as a foundation for future techs. Sure, people will be free to use copies of your work at will, but they still have to give credit to you, and won't be trying to do things differently.
And that's not even mentioning that open-source stuff generally has a PR-boost compared to closed-source. If you make something closed-source, you will always have to worry about the appearance of an open-source competitor. I have seen closed-source stuff be completely eclipsed by open-source alternatives once the latter reached a critical level of quality. I mean think about it: Unreal Engine 4 is open source, would anyone start a closed-source project to compete against it? All the best minds will naturally ask you, 'Why not just work on UE4?' Yes you can fork it, but the public will always turn to the original of the fork unless your fork achieves something really different in a good way.
So in a way, open-source is the most asshole-way you can do things cause it shuts down any idea of a copy/similar alternative to your work.
Godot would be an example of an open source game engine.
https://www.engadget.com/amp/2012/11/09/silicon-knights-v-ep...
Simply attaching a big bad license to your code doesn't prevent 3rd world shovelware farms from copying your work and making tons of money from it for free.
Also, if everything were open-source, then nothing would get brownie points just for being open-source.
That's the thing about open-source though, it's a race to the bottom for market share. We saw it with React and Vue.
I also get that open-source is a maintenance liability with issue and requests, but I also see plenty of open-source software that do well "provided as-is".
As for closed-source competitors: game engines have always provided their source under proprietary licenses to people who paid enough, so the only real difference now is that you can get access to it (but with limited rights) without paying anything. But besides a few big names (Unity, Unreal and CryEngine/Lumberyard) there are quite a few closed-source competitors in use that don’t have generally available source code, although many of these are used only within a publishers development studios (idTech usually got open sourced after some time but until then the code wasn’t available to the public, Game Maker Studio, Frostbite etc)
Of course, don't go and patent a mediocre interaction or technique, but definitely, pursue a patent if you have worked hard on researching and developing something that is key to your business and that could be used by a better established and wealthier competitor to take a market advantage.
I'd venture to suggest that most software patents are the obvious way of doing something. Even having read some attempts to justify them I see no merits in patents for XOR cursors, online auctions or shopping carts.
Fill in the blank, and the same lessons apply.
Any idea, project, business, app, website, article, podcast, TV show, media production, procedure, product, art, creation, etc are frequently copied and cloned. And he's right, ultimately nobody cares.
Sometimes a copy or clone is better than an original. Sometimes a copy or clone is much worse, but ends up successful because of positioning, luck, marketing, or savvy.
Perhaps it's better to think of the risk of being copied or cloned as part of the risk of business and creation. It just comes with the territory.
How are copyrights or trademarks ANY different? Together, they constitute essential tools in the knowledge/service economy. Demonizing patents but favoring the other two is ignorance on Marco's behalf - no one wants frivolous patents clogging the patent office, courts or legal departments of companies.
Apt advice would be to carefully research recent verdicts and legislation around patents, and only apply for ones that merit the designation (yes - this is very subjective).
These comments citing Apple and Google with their effective use of patents as some sort of counter argument seem to have missed the mark. Those guys are certainly able to afford $20k, and they can certainly afford to take on any adversary big enough to matter.
Interesting take, considering how Marco has called out both Pocket (formerly Read It Later) and Readability as having copied features that he claims to have created first.
"Weiner systematically copied almost every major Instapaper feature over the first few years of Instapaper’s existence", from https://marco.org/2013/02/21/the-first-read-later-service
and
"Over the next few months, they continued adding mostly Instapaper-like features to their service.", from https://marco.org/2011/11/16/readability
Perhaps his recent comments reflect how well his prior callouts were received?
EDIT: curious to know why this was downvoted. It wasn't meant as a dig at all—I was pointing out that Marco has experience with callouts and might have relevant expertise that many people lack. I would personally be very interested to hear his experience went dealing with those issues.
I used to like Marco, but after following him a while and noticing inconsistencies of principle like this, it got to me. (That, and when he said in a podcast that he didn't use SQL joins when building Tumblr because he doesn't like them. Trying to think about how he could have possibly accomplished that makes me shudder.) I can give him the benefit of the doubt that perhaps his perspectives have changed with age or experience, though.
All that said, I upvoted the OP too, because as much as I don't like the guy, he's not wrong here.
> Only assholes get patents. They can be a huge PR mistake, and they’re a fool’s errand: even if you get one ($20,000+ later), you can’t afford to use it against any adversary big enough to matter.
* the patent office seems to have a policy of grant-by-default
* the term is way too long for software patents
Grant 3-5 year patents and I think most of my objections would disappear.