The Rise and Fall of the Independent Developer
furbo.org
furbo.org
There are entire categories of software now where people are now conditioned only to accept an open source product. Can you imagine anyone building a new computer language now, commercially? If it was even vaguely successful it would be cloned and forked so quickly it'd make your head spin.
Between the ideologues like Stallman, who effectively think it's immoral to make money off selling the software itself (I know his position is supposedly more nuanced than that, but free software effectively amounts to this if you have to hand out source to all and sundry) and the open-source-friendly companies like IBM and Google - who have every reason to drive the $$$ available off software to zero, it's easy to feel a bit beleaguered.
I think open source is clearly critical to the modern software ecology, but I wish people would think carefully about why they are doing it in each case, rather than assuming their actions are automatically benign because of a sprinkling of 'open' dust.
The point is not that value isn't created - it's that the people creating it may not be thinking about who benefits from it and who loses.
In the end, when value is created in the free software ecosystem, everybody benefits more or less equally.
If I make an open-source heart rate monitor, doesn't that benefit people who want heart rate monitors, hurt people who make for-pay heart rate monitors, and divert my resources away from solving anyone elses problems?
That's got to be good for society, right? I'm sure I'll just go do all that research anyway, because I'm a nice guy. And I certainly won't go off and patent the holy living shit out of each and every aspect of the monitor or anything nasty like that, because that would be bad, and we've already established that cloning products has to be an unalloyed benefit for innovation. Because it's, like, free or open or something.
http://mises.org/books/against.pdf
http://levine.sscnet.ucla.edu/general/intellectual/against.h...
There's always the prospect of reverse engineering, but it's a lot harder to do and there are situations where it seems rather unlikely. In enterprise software, particularly, it seems considerably less likely that a big company will perpetrate hard-core reverse engineering naughtiness when bound by a thicket of NDAs and contracts - why risk it? A quick browse through the opened source code is a rather different beast.
Competing against a $0 product (which is essentially subsidized by corporate salaries) is likely to increase the barrier to entry. That will make bootstrapping harder, favoring existing enterprises. This is clearly not an unalloyed benefit to society.
The existence of a reasonable 'free' product may also deter the entrance of multiple competitors who would have advanced the state of the art. Clearly this is bad for society.
I'm not arguing that this always happens, or that free software is inherently bad. I'm arguing against the dogma that free software is always automatically good.
It's a choice with consequences and the dogma is an excuse not to think about them.
That's not clear at all. If you can't beat a "good enough" free alternative in the market, that's a signal that society would benefit more if you spent your efforts elsewhere.
I'm arguing against the dogma that free software is always automatically good.
As a first approximation, it pretty much is. Would you care to name specific free software that we'd be better off without?
That's not a really useful way to argue. See http://www.paulgraham.com/disagree.html
Your open-source heart rate monitor will arguably directly help more people than the proprietary ones (don't confuse open-source with gratis) because it can serve as a base for other solutions to different problems. What you actually did by developing your open-source equivalent was to employ your resources in a much more efficient way, distributing its returns across a much wider field.
It's true you hurt the very limited group of people who made proprietary monitors, but you benefited a much, much larger group much more than they could ever do.
It's not so simple, everything has a cost.
Yes, I would like to drive the cost of software to zero, for the same reasons that I would like to drive the cost of energy, housing, and food to zero.
and make us all work for a wage providing 'services'.
This is in fact how most developers are paid today. It's a relatively small minority that's writing software for sale to end users.
There are entire categories of software now where people are now conditioned only to accept an open source product. Can you imagine anyone building a new computer language now, commercially? If it was even vaguely successful it would be cloned and forked so quickly it'd make your head spin.
And that's good. The market has figured out that using a proprietary programming language is a very bad idea. And by reducing the costs of development software, the demand for developers is increased.
I'd almost say that most developers are paid to write software and not provide services. Though it really depends who you refer to as 'end users'.
Exactly. I consider that to be more on the "services" side because the ultimate goal is for Bob to get his TPS reports, not to produce a J2EE reporting framework.
A better way of expressing my point would be that most developers aren't in competition with open source; it's much more likely that open source software allows their jobs to exist.
Microsoft, EA, Apple, Valve, Blizzard, Riot Games, all microtransaction-financed games, every iphone developer that sells their game instead of using advertisements for income, they're all developing software that make money off that initial software purchase.
The subtle counterpoint would be in companies that hand out stock options,where there is a sort of indirect correlation between the sale of a unit and your wage.
Then you're going about it exactly backwards, the worst possible way.
Wages/prices are driven by perceived value and constrained by price floor. If you want me to work for free then you first have to get energy, housing and food to be free. As long as I have to pay, I have to charge. What kind of stupidity leads one to first kill wages instead of rent seeking?
>This is in fact how most developers are paid today.
So what? The reason many of us are in this field is because it is one of the few left where we have the potential to escape wage slavery. If you want to close that door expect to meet resistance. You're basically telling us all to get back in the mines and stop dreaming about the future.
I work for Atlassian. Our two biggest products are a wiki and a bug tracker. We have dozens of open source competitors. There is also a long list of commercial competitors. Many of our commercial competitor started up in the last few years and they seem to be growing or at least they haven't gone out of business.
You might argue "but then the client can take the source and change it and redistribute it", but how is that any different from real life? The fact that you can charge so much for scarce copies of physical goods is a flaw of physical objects, not a flaw of digital objects. The digital world doesn't support the monetization schemes of the physical. Fighting that seems to be a losing battle (see RIAA, MPAA, et. al)
Honestly I think the bigger threat to the indie developer is big companies dumping products on the market at rock-bottom prices and making up the difference with ads and addons. When everybody is used to paying a dollar for Angry Birds it's very difficult to charge a more reasonable price for other software.
The process of building, say, VisiCalc is a lot riskier and harder than turning Excel into Libre Office Calc. If Bricklin had known he was going to have to compete with free in a matter of a few months after release, he might have done something else entirely.
This may be hard to understand for people that haven't ever designed anything difficult, but frankly, it's _so_ much easier to clone something than build it the first time. Even just knowing that something is possible is a huge leg up.
If you think routinized F/OSS cloning of software is an actual service to innovation, you need your head examined. If nothing else, it makes going into a patent frenzy with every idea far more tempting.
It's hard for me to prove a counterfactual, of course, and show all the wonderful things that have been discouraged by the prospect of immediate shitty open source clones. I fully accept that open source allows us to build innovative things on top of other software layers - it's not necessarily all bad. But it's not necessarily all good either.
You may feel like they are hurting your piggy banks, but you certainly have no ground in going after them.
In any case, I don't think people should worry too much. FOSS developers often work for free, and they probably sucks at doing the hard bit of advertising, documentation, and finding out what users find hard about it.
If you can't compete with hobbyists, you have no business in being in that field.
So sure there might be a FOSS clone of your product, but they would only be copying the features that you've currently got, not the ones which you are developing. Could almost think of it as a lite version.
This is of course only one possible way of thinking about it.
A much more practical angle: there is no MS Office or equivalent for operating systems such as Linux and BSD, so the OS people had to write their own.
To be honest I think the developers of LibreOffice are heroes. Not many people in the open source scene want to be seen working on 'boring clone' projects like an office productivity suite, and would rather innovate. But the fact is, it is a necessary evil for many users, and the availability of OO has helped Linux adoption a lot.
To the grandparent poster, even if F/OSS is copying features, you still have the advantage of being first mover; it just means you have to keep updating and refining. There is also the non-trivial advantage in being paid to do this work. Moreover, if customers find it better to just switch from your product, incurring inevitable transfer costs, there is something wrong already.
Nor are you making a germane response to my post, which is not that LibreOffice is bad and Excel is good, but that if the creator of Visicalc had been in the situation where a few short months after he created his spreadsheet someone was going to clone it and give it away for free, he may not have bothered.
This is just lazy. Stallman's position has literally nothing to do with money.
Also I don't really understand the rest of your post:
Linux and the GNU tools are clones, especially the GNU tools. And free software commoditizing certain classes of software is a good thing since we don't have to reinvent the wheel a billion times. Without free software "clones" it would have been massively difficult for products like OSX to even exist.
There are still "commercial" programming languages being created (off the top of my head: F# and Q) and I'm having a hard time coming up with any commercial language whose success was damaged by free software clones.
Can you give any concrete examples here? LibreOffice is still crap compared to (also kind of crappy) the MS Office Suite. If we all had to pay for IIS and SQL server the web would be much much smaller. Where is the damage being done?
However, the message hasn't seemingly sunk in, as there are many people who seem to be followers of stallman who get up in arms any time someone sells open source software (even if they are distributing the source.)
It is apparently offline now, but there used to be a huge list here: http://ffmpeg.org/shame.html
I picked some at random awhile back and looked into them. Every single "shame" was a commercial company that was complying with the GPL, and was distributing source code to ffmpeg. (Running it as a separate process, not mixing the code.)
This didn't matter to them. They documented constantly harassing these companies, who would try to comply with their demands (most of which were not required by the GPL) and then would give up, because the demands would just escalate.
It was kinda like software terrorism-- harassing people for selling software that included -- sometimes as a small part-- open source code.
I'll give you an example: WordPress is Free and Open source. The PHP code that you write for WordPress must be licensed with GPL. Even with that, dozens and may be more of developers and designers are making mad money selling plug-ins and themes. There are free Open Source plugins and themes, too. There can't be a worse situation than that.
Figure out your market. Know what the clients are ready to pay for. Compete. Make Money.
Freeware isn't open source, but it is free, in some very real sense.
Please quote him and provide us with the reference.
I am not familiar with his making such a statement. The free in free software has nothing to do with price, but has everything to do with freedom.
Richard Stallman, Free Software Free Society, 2nd Edition, Chapter 10, page 65, pdf edition: "Many people believe that the spirit of the GNU Project is that you should not charge money for distributing copies of software, or that you should charge as little as possible - just enough to cover the cost. This is a misunderstanding. Actually, we encourage people who redistribute free software to charge as much as they wish or can. If this seems surprising to you, please read on."
Many conventional software business models are unsustainable if you give away source code. Yes, there are always exceptions (where, typically, you have some closely related bit of software or IP you can charge for - think Pathscale with their libraries or Snort with their ruleset), and you can slave away chasing your product around with services until the End of Time.
However, there are a couple things you just plain can't do with open source - one is to prevent people from making builds of your software for free, and the other is to prevent your competitors from free-riding on your research efforts and duplicating your hard work for a much lower cost than it took you to create it. The only way you can prevent the latter strategy is to patent the living shit out of everything, which I also find objectionable. I don't like the idea of trying to prevent someone who independently invents the same idea after me from competing with me. This sounds like honest competition. He or she got the idea by thinking hard about it - not by looking at my code.
Another horrible weakness of F/OSS business models is that, in many areas, there are ridiculous mazes of interlocking patents, often covering many versions of the same thing. Putting your source code out there is like sticking your head into the patent troll's mouth and begging to be bitten.
"Charge as much as they wish or can" is fine up to a point, but you just plain can't, and any attempt to create a situation where you can charge more for the fruits of your labor than these crippled business models allow is treated as immoral.
An invented 'right to tinker' somehow trumps my (equally arbitrary and invented, in some senses) right to be paid exclusively for innovations that took a huge amount of work to come up with until such time as I choose to release them or someone else figures it out - or something better - too.
The (closed source) company I work for is full of grown-ups, our customers are fairly grown-up too :-) and if we are so benighted as to enter into a mutual contract where we get paid and no-one gets to see the source, it's None of Richard Stallman's Fucking Business, or yours.
I'm not taking away a freedom of his because I never would have invented a goddamn thing if I had to build things under the very limited range of business models possible in the world of F/OSS (note: I'm not hating on those models, just pointing out that they are not appropriate for every type of software).
In some parts of the world, if you farm a little too much, gangs come by in rusty toyotas with AK47's and help themselves (and might kidnap your children for good measure). In our society, we've allowed these men to wear suits and pretend they are contributing members and in exchange, they no longer use guns. This seems like a valid compromise until we as a society outgrow this sort of behavior. (Here's hoping)
There are tools for managing risk. Create an LLC and have professional liability insurance. Sure bad stuff can happen. I had an over-zealous zoning official try to revoke the business license for my house because I was "manufacturing" software (and manufacturing is zoned industrial, donchaknow).
I wish there weren't patent trolls, but at some point you have to admit that you can't spend your life worrying about all of the bad stuff that could happen. Take a few reasonable precautions and then do your thing.
Would you suggest an LLC for each product so that they can only be attacked individually?
If a product is good enough to make money on it's own, it's good enough for its own LLC.
I don't think a dba does this. If any of your products incurs liability, the assets from the entire parent are on the line automatically.
Lawyers always try to get through the corporate barrier and get at your personal assets, or connect your companies to get at them all at once. (This even happened in the tiny trademark case I was involved in)
Having this done, and done right, its a big impediment to that happening. Don't mix your accounts, pay yourself a salary from each company and keep your papers in order.
Even if you are an indie one man shop you are in fact a business and there are risks. Same thing goes for being a lawyer or accountant or contractor. If you screw up, it could cost you financially in a very big way, even if you didn't intend to do anything wrong.
Even successful businesses can crumble, even billion dollar businesses can get beat up by litigation to the point it isn't even worth continuing.
To say that the sky is falling for indie devs is a bit silly. There are plenty of places you can build, innovate, distribute, and succeed. Sure, the iTunes App Store thing is big today, or maybe Facebook, or maybe Steam, or maybe Wii's downloadable channel, it doesn't matter. Any one of those could be gone in a year or two.
Here's an example, there are people out there who make millions of dollars selling e-books and mediocre/overpriced software on Clickbank right now. Some of those people get sued I'm sure for their products. More people show up. You could take the same ebook and sell it on Kindle or Nook if you got shut down or you could sell your software on the Mac App Store or Chrome Web Store if CB went under.
There are so many platforms to distribute and build software and businesses on now, it really is hard to complain about the death of the indie developer because there are so many new companies getting started every day on all these different platforms that didn't exist even 5 years ago.
As a business owner you can't control all risk, but you certainly can and should plan around them.
If one channel gets shut down, move on to a different one or a different product.
Great devs and great companies aren't built on one hit one time wonders.
So yeah, the situation isn't impossible, because patent trolls need to keep the success rate high enough that new software entrepreneurs are tempted to try.
Also, the risk/reward of lobbying for intervention may be vastly higher than is practical for people who would otherwise be indie developers, so in practice going to work for a large corporation might be the only way to spend your life writing software rather than being a patent reform activist.
But for trivial cases like most software patents, it is just something to prop up the established big companies (and keep lawyers fed) at the expense of emerging players.
That said, it's certainly different from what they're accustomed to due to that cross-border issue, and thus a bit more difficult and there are probably much easier targets within their immediate legal jurisdiction.
Not that it it really matters for me RIGHT NOW, since I only just launched the other day and revenue is a future dream, but it still assuages my fear a little.
[1] Well, he had to move, but the point is he didn't pay the people who sued him a dime. This is what I would want to do; make sure the ones who sued me got nothing.
TROLLCORP: "We're suing your company for ONE MILLION DOLLARS."
CEO (through a lawyer, of course): "Well, oops, we've only got ten thousand in the bank plus a couple of old Macbooks. After that we're just going to declare bankruptcy and you won't see a dime."
TROLLCORP: "Well, okay, frankly, we don't want to see you go bankrupt. Bankruptcy court is no fun at all. We don't need to pay for a tedious legal battle. We just want the money. How about you just give us the ten thousand plus 5% of all future revenue from your products?"
CEO: "If I spend the ten thousand on a lawyer, I can file some motions to delay your lawsuit. Then I can go bankrupt. You'll get nothing then. So, how about one thousand bucks and 0.5% of future revenue?"
TROLLCORP: "Two thousand and 2%."
CEO: "Done."
As with any parasitic transaction, the parasite has no rational interest in killing you. Dead companies don't pay. The danger, of course, is that they won't be reasonable and will accidentally push too hard, in their attempt to convince you to search under more sofa cushions for loose change that they can take. Or that they are, in fact, happy to kill your company because they think it will be an instructive demonstration for the other companies they sue. Or that they will gradually consume your time and suck your blood and your company will eventually die of exhaustion in 2% increments. But, you know, this is why it's good to be a small company. Having nothing means having nothing to lose.
On the bright side, patent trolls are patenting the most obvious patents right now. In 20-30 years all of those patent will expire. I'll be amazed if 20 years from now we will still continue to patent obvious things like one click button transactions. If so, we all deserve to go down the ship of financial failure since it is us, society, that allows such things.
Meanwhile, if we could only find the best ways to make patent trolls' lives more painful... At least it would dissuade the casual patent troll.
It might not be easy, but society is harmed if we make it artificially harder to respond to market demand.
And as I write this I realize that I sound a bit corny. The shame... (no sarcasm intended)
That said, human political systems can and do lead to sustained bad results. In my mind, 'things working themselves up' really means, 'someone finds and implements a solution'.
Meanwhile actual indie developers are as we speak, prevented from working on adding value to the economy because they are forced to spend their time and money dealing with trolls.
Independents are suddenly able to take a little bit of risk, take on some money (then: banks, now: angels and VCs), and work by themselves or in small teams to produce something that lots of people need (then: oil, now: good software).
Some of the biggest fortunes in history were made back then (1900 - 1930s), just as some of the biggest fortunes are now being made in this industry.
The digger I deep (no pun intended), the more similar these two stages of history look alike. I'm sure there are other eras that also share these common threads.
One of us is misremembering. I seem to recall an interview with Gates and/or Ballmer where they talked about the realization that the cost of 5 or 6 floppies at wholesale was insignificant compared to the revenue from the software, that it was a no-brainer cost-wise. CDs were the first time when cost became an issue, iirc (again) because you had to pay for one or more masters to print the production copies. I'd feel better if his history was right.
This time the retail channel itself is very cheap, but the ancillary costs, both financially and emotionally, are very high.
And, of course, only large companies and publishers can bear these costs. My fear is that It’s only a matter of time before developers find the risks and expenses prohibitive and retreat to the safety of a larger organization. We’ll be going back to square one.
I can't even address the "emotional costs" he talks about, but there is a precedent for dealing with risks of this kind. For consulting engineers, it's called "errors and omissions insurance." It strikes me that if there isn't something like that for developers, one plausible reason is that this kind of liability hasn't actually been a problem. Is there errors and omissions insurance for developers?
From our experience, it’s entirely possible that all the revenue for a product can be eaten up by legal fees.
The author uses the phrase "from our experience," but I wonder what that experience actually is. At this point, it's not even anecdotal, because he doesn't supply the anecdote! We can't even begin to ask if his experience generalizes. For example, did he get good counsel? Was he actually infringing? Should he have settled? Did he avail himself of the FSF or ESF if applicable? Should his product have been more profitable?
I think I'm either blessed or cursed with too much curiosity to take this guy at his word.
Sure, the cost of delivering a single copy of software to a given customer was relatively inexpensive. But Craig is talking about how, in the absence of the internet or other easy customer acquisition and distribution methods, the way you used to get your software to customers was by making thousands of physical copies, boxing them up in a nice shiny box with manuals, and have computer software and electronics stores display your software on their physical store shelves. And that's quite a bite of money to have tied up in inventory even though a single piece of media itself was relatively inexpensive.
Even magnetic tape reels were not that expensive..I seem to remember that RMS earning half of his expenses just from revenues of supply code by magnetic tape alone.
Craig is a very well-known developer and author in the Mac and iPhone developer community and works at Iconfactory. They're one of the first companies that was addected when the Lodsys patent issue came to light. Even if you didn't know that, the very previous entry on his blog is an open letter to Apple about the problems his company is having with the patent issues. It's a little disingenuous to call someone out about omissions in a blog entry while taking zero effort to familiarize yourself with any of their work.
I know we're supposed to be civil and not say anything to anyone on here that we wouldn't say to their faces, but, well, but nothing. I'd say the following to your face.
Fuck you. It's out of line to call someone calculating and not straightforward when you have no idea what their actual motivations are.
Asshole.
Karma be damned.
Yes. I got a quote through Creechurch for E&O (along with liability) to insure my software dev consulting, but the cost is prohibitive for me right now. Maybe I'll do it when things pick up.
There was a period earlier in the timeline(the late 70s and the start of the 80s) where one might succeed in breaking into retail without a large spend, but between then and the rise of cheap web apps and app stores, the market for independents was in shareware mail-order and downloadables, which have much higher purchase friction, and thus lower conversion rates.
It seems like something that large corporations as well as independents could get behind.
Denying patent protection to anyone who doesn't already have a finished product out on the market opens a whole new can of worms.
Facebook -> RenRen
Twitter -> Weibo
YouTube -> Tudou
Google -> Baidu
Flickr -> YuPoo
See http://www.randomwire.com/chinese-web-20-clones
Yeah, Trolls, go ahead and sue them.
Little has changed with the process of software development since the 1980’s
I understand the point: that things are the more the same than ever, that this guy is a salty old sonofabitch who's forgotten more than i've learned, but seriously. Little has changed with the process of software development since the 1980s other than the tools? The entire way we think about why we're building software and what we're trying to get out of has changed and is still changing.