As a bioinformatician in industry, if you are and academic who wants to charge for your software, please don’t do this:
Software is free for academic use only. For industry use, please contact our lawyers.
Please do this:
For industry use, software costs $[price] per license. Click here to download a $[more reasonably price] 30 day trial version.
Why?
I will never again* contact an institutional patent office.
What invariably happens is that a lawyer takes weeks/months/never to respond to me and then they ask for $15,000. While I am waiting for that to happen, I have found a similar, open source substitute and coded up some other stuff to get it to do exactly what I want.
I have nothing against people wanting to make a buck. I like making a buck as much as the next person. Also, many industry people have deep pockets to spend on things they really need, and I myself would be happy to support academic research. But what we don’t have is time. If you are in a startup, your company's biggest expense is usually salaries for workers and the space they work in. Those costs don’t go away while you are waiting weeks (or forever) to hear back from institutional attorneys.
Yes, we have all heard stories of a big pharma at the beginning of bioinformatics swooping in like Daddy Warbucks spending big coin on some application. But that was when there was one tool to do each thing, and also when big pharma didn’t have any bioinformaticians and didn’t really know what they were doing yet.
I don’t know if that still happens or why software is always $15,000. It seems to me that there is a price point where people can just buy things without having to call a meeting with six people to discuss whether it’s a good idea, and that price is definitely way lower than $15,000.
If software were cheaper you’d get more customers and there is some optimization formula that I don't think anyone is calculating. I expect most software gets no customers. At least they get none from me. But knowing the price, at least, would be a good start.
There is also something about transparency. Am I getting charged more than my competitor for software that was largely funded by government grants? Who knows! Maybe they're getting it for $10,000.
So, in short, if you want to sell your software, just sell it. With a price and shopping cart. Maybe put it on Amazon.
And that, fellow bioinformaticians, is the end of my rant.
*OK, never say never. I would if it was for something very awesome and specific but I would complain about it all day.
I will say that in majority instances (dare I say close to 100%) people who wrote the software package probably did not have a choice. Given a choice they would not have wanted to do this but they are required to follow institutional rules (to keep their jobs), which leads to situation you describe.
Out of curiosity, do you expect to get prompt support for software that you license? What has been your experience with academic software that you may have licensed in that regard? Supporting software is probably not high on the list of priorities of academics who are focused on other important things. Edit: Looking at your profile it looks like you were in this camp once upon a time.
I have honestly never gotten all the way through the process of purchasing academic software because of the problems above.
I would expect the software to be documented and tested and not really require support. But that is what I expect of open source software and it is largely the case. (Now, this wasn't true 10 years ago).
I still support my own academic software, which runs on an academic server, but that mostly consists of kicking the server periodically.
The one time I licensed something like this it was because the IP office at the institute where I was demanded it. I suspect that's the case 90% of the time. We also usually have no idea what the IP office might actually want to charge (15k seems absurd for most academic software), which is why no price is listed. But I agree that the whole thing is absurd and that almost all of these licenses should just be removed.