Transcript
Corey: This episode is sponsored in part by LaunchDarkly. Take a look at what it takes to get your code into production. I’m going to just guess that it’s awful because it’s always awful. No one loves their deployment process. What if launching new features didn’t require you to do a full-on code and possibly infrastructure deploy? What if you could test on a small subset of users and then roll it back immediately if results aren’t what you expect? LaunchDarkly does exactly this. To learn more, visit launchdarkly.com and tell them Corey sent you, and watch for the wince.
Jesse: Hello, and welcome to AWS Morning Brief: Fridays From the Field. I’m Jesse DeRose.
Amy: I’m Amy Negrette.
Tim: And I’m Tim Banks.
Jesse: This is the podcast within a podcast where we talk about all the ways we’ve seen AWS used and abused in the wild. Today, we’re going to be talking about AWS, an open-source software. Now, that’s kind of a broad topic, but there have been some specific, recent events I’ll say, over the last year maybe or maybe even less, related to AWS and open-source software that really got us talking, and I wanted to have a deeper conversation with both of you on this topic.
Tim: Well, you should probably start by going over some of the things that you’re mentioning, when you say ‘some of these things,’ what are those things, Jesse?
Jesse: Yeah. So, I think the best place to start is what constitutes open-source software. And specifically, I think, not just what constitutes open-source software, but how does that differ from an open-source company?
Tim: So, open-source software can be anything: Linux kernel, bash, anything like that, any Python functioning module. If you make a piece of software, whatever it is, and you license it with one of the various open-source licenses, or your own open-source license or whatever, it’s something that the community kind of owns. So, when they get big, they have maintainers, everything like that, but at its essence, it’s a piece of software that you can freely download and use, and then you’re free to modify it as you need, and then it’s up to the specifics of the license to whether you’re required to send those modifications back, to include them, or to whatever. But the essence is that it’s a piece of software that’s free for me to use and free for me to modify under it’s license.
Jesse: And one of the other things I want to add to that is, correct me if I’m wrong here, but isn’t a lot of open-source software is very community-owned, so there’s a lot of focus on folks from the community that is using this software giving back not because they need to under the licensing, necessarily, but because they want to continue using this and making it better over time.
Amy: I think one of the issues is that becomes a very opinionated kind of statement where there are a lot of people in the open-source community who feel that if you’re going to use something and make changes to better suit what your needs are, that you should be able to submit those changes back to the community, or back to whoever owns the base of the software. But that said, it’s like the community edition of MySQL before Microsoft bought it, where the assumption was that there’s essentially a candidate of it that anyone can use without the expectation of submitting it back.
Jesse: So, that’s a broad definition of open-source software, but how does open-source software, broadly speaking, differ from an open-source company? I’m thinking specifically there is the open-source software of Elasticsearch, for example, or I should say, previously the open-source software of Elasticsearch that was owned by the open-source company, Elastic. So, what does that relationship look like? How does an open-source company like that differ from the open-source software itself?
Tim: So, there are typically a couple of ways. Usually, a company that is the owner of an open-source product still has some kind of retention of the IP in their various licenses that they can do that with, but essentially—and this is in the words of one of the founders of Elastic—that they’re benevolent dictators over the software. And so they allow folks to contribute, but they don’t have to. And most of those open-source software companies will have a commercial version of that software that has other features that are not available, packages with support or some of the things like that, some kind of value-added thing that you’re going to wind up paying for. The best way to describe—like you said—there’s the company Elastic and then the product Elasticsearch.
I relate back to before: there was Red Hat Linux, which was open-source, and then the company Red Hat. And I remember when they went public and everyone was shocked that a company can make profit off of something they gave away for free. But while the core of the software itself was free, the support was not free, nor was the add-on features that enterprises wanted. And so that tends to be kind of what the business model is, is that you create the software, it’s open-source for a while to get a big user base, and then when it gets adopted by enterprises or people that really would pay for support or for other features, that’s when the license tends to change, or there’s a fork between the open-source version and then the commercial version.
Jesse: And it definitely sounds like there can be benefits to an open-source company essentially charging for not just the open-source software, but these extra benefits like supports and additional features because I know I’ve traced multiple code bugs back to a piece of open-source software that there’s a PR or an issue that has been sitting open for months, if not longer because the community just doesn’t have the time to look into the issue, doesn’t have the time to work on the issue, they are managing it on their own, separate as a side job, separate from their day-to-day work. Whereas if that is a bug that I’m tracing back to a feature in an open-source piece of software, or I should say software that I am paying for through an open-source company, I have a much clearer support path to a resolution to resolving that issue.
Tim: And I think what the end up doing is then you see it more like a traditional core software model, like, you know, a la Oracle, or something like that where you pay for the software essentially, but it comes packaged with these things that you get because of it, and then there’s a support contract on top of it, and then there’s hosting or cloud, whatever it is, on top of that, now, but you would still end up paying for the software and then support as part of the same deal. But as you know, these are for-profit companies. People get paid for them; they are publicly traded; they sell this software; they sell this product, whether it’s the services or the hosting, for profit. That is not open-source software. So, if company X that makes software X, goes under, they are acting like the software would then go under as if the software doesn’t belong to the community.
So, a business that goes after a business is always going to be fair play; I believe they call it capitalism. But when you talk about going after open-source software, you’re looking at what Microsoft was doing in the ’90s and early 2000s, with Linux and other open-source challenges to the Windows and the other paid commercial enterprise software market. When folks started using Linux and servers because it was free, c...