Episode 799 transcript

FLOSS-799

Jonathan: Hey, this week, Laurie LaRusso and Steve Hoffman join me to talk about Percona, the open source database solutions company that after all this has managed to keep open source, we talk about how and why you don't want to miss it. So stay tuned. This is Floss Weekly episode 799. Recorded September 3rd, still open source at Percona.

It is time for Floss Weekly. That's the show about free Libre and open source software. I'm your host, Jonathan Bennett. And today we're talking about Percona. It's an open source database software. It seems like anymore. It's one of the few opens big open source database solutions that still actually open source goodness.

I think we're going to talk about that just a little bit during the show today. I, I don't know much about Percona. I, I kind of have to admit my database experience is sort of just My SQL and SQL light and a tiny, tiny bit of fighting against Microsoft SQL. It's been, it's been something because of because of things happening outside of our control, I am the host and co host today.

So that's going to be just a little bit different, but we've done it that way before. It's, it's not going to be a big deal. But let me go ahead and bring our guests on today. And we've got Lori Lorenzo and Steve. Hoffman and we talked a little bit before the show and Laurie, you are the, the community manager and Steve is sort of the, one of the VPs of, of technical.

Is that, is that about the way that, that lays?

Lori: Yeah. So just, my name's Laurie LaRusso, like Karate Kid LaRusso and yeah, I'm head of community at Percona and yeah. Yeah.

Steve: And I'm, I'm one of the engineering managers or engineering leads here for Percona run the the development side of the house. So, Loray and I are joined at the hip.

Jonathan: I, I understand. I know how that goes. So, let's, let's talk for just a minute about, I guess the technical side is where we'll start. And, what, what is Percona? Like, how is it different than, you know MariaDB? Or any of the other database solutions that are out there? What's, what's the kind of the, the thing that makes the difference about Percona?

Steve: So it's not wildly different than MariaDB's original model. We both took a fork of MySQL and, you know, added our own features to it. What we think, you know, was the, the answer to the needs of the market, you know, from an enterprise standpoint. We've also since done that with MongoDB and we're also working in the Postgres space as well to bring new features on top of an already great base.

So in that sense, not wildly different. I think. You could say we're similar there, but then I think we go a lot further than them. So we also have our Operators product. We also do our, our monitoring and management suite. So maybe we have a wider portfolio, but I think it's fair to say MariaDB model is similar to Percona.

Okay. And

Lori: so this is where, let me just jump in. This is where I think Steve is being humble and that's why you have a community person to back him up. So, you know, I think the, what you're missing is the innovation, right? So we have community editions of MySQL, of Postgres but what we do We make them better, and we're open source.

So, we take, we take the regular edition, we open source it, and we add in our little spice to make it better for users.

Jonathan: So is, is Percona a database? Or is Percona a database company? Or is it both?

Steve: We like to call ourselves a solutions company. We produce database software, and and you can get it in all different formats, but we make the majority of our money on the services that we provide.

But when you take our software and our services and sort of smash them together, you get sort of the overarching Percona solution.

Jonathan: Okay.

Steve: You don't have to use both, but I think that's where you get the best value is our software with our services. We can, we can help you run it better than anybody else.

But if you're using, let's say a community edition. Of either Mongo or Postgres or MySQL. We can still service you but it's just, you don't, you don't get the full value. Okay.

Jonathan: And, and, I, as you, as you work on some of these pieces of software I'm sure, so you make software changes to them, you make changes to the source code, because it's source available without being an open source.

Do any of those changes ever get pushed back up to the original project? Or do they kind of all stay siloed in your, I mean, obviously it's open source, but still, you know, there can be either an effort or not an effort made to push them up.

Steve: Yeah, I mean, we like to contribute everything that we do. Now, we have learned over the years that certain things aren't really necessarily welcome or valuable back upstream.

In many cases, we're actually building enterprise features that already exist in the upstream's enterprise product. So, us contributing that backup there, you know, they don't want that. But we do bug fixes, we do, you know, extensions and enhancements all the time. And again, we do it all, make it as open as humanly possible.

Direct contributions. You know, and hope that we get it included in, in upstream so that it just everybody gets to benefit from the change.

Jonathan: Yeah, I mean, every, every change that you can make upstream is just one, one fewer change for you guys to have to be the stewards of. That makes sense. That's, that's kind of interesting that you're re adding some of the enterprise features that have been not added intentionally.

Is there, is there any animosity from any of the upstream vendors because of that?

Steve: I think on the community side, there's no animosity. I can't even say that I'm aware of it on the enterprise side. I'm sure, you know, if Percona got way too big, people that haven't taken notice certainly would, but we seem to have a great relationship with the MySQL community, certainly the Postgres community.

And Mongo, you know, I mean, I don't think we have each other as adversaries. Business wise, surely we're competitors, but I don't think it's. I've not received any death threats, I don't think any of you on the team have, so

Jonathan: we're good standing there. You've managed to remain friendly competitors, that is handy.

Yeah, exactly. That's fun. Well, but the database world seems to be sort of cutthroat these days. We talked just a little bit before the show about like all of the different licensing changes that have happened and forks because of it. And it seems like, well, no, I know one thing that has happened is you'll have a company that they have a database, like they have their secret sauce in their database.

But instead of making it all secret, they've made it open source. And so they've got this database code that they work on, and then they also offer it as a service. And like, that's great until Amazon comes along and says, it's open source. We can offer this as a service too. And suddenly this, this, this business that does open source, like their revenue stream starts to die, dry up.

And they immediately go, we've got to make some kind of change to be able to keep making money. And it seems like in a lot of cases, the change they make is we're going to use a business source license of some sort. And. It, it, it, well, it fractures the community. It, it inevitably results in a fork, you know, like Redis and Valky.

We, we now have Valky because Redis did the exact same thing and they, they, in their wisdom, they made this decision and they made the announcement to you. Move away from an open source license with Redis during a, I think, a Kubernetes conference. And so there were a bunch of people that were using Redis all together in the same place when this was announced.

And apparently by the end of the conference, they sort of walked away going, Yeah, we have a gentleman's agreement to make a fork and call it, you know, all get on board with the same fork. But it's just it's got to be it's got to be cutthroat.

Lori: Is it a

Jonathan: cutthroat environment? Have you seen that?

Lori: Let me take your, your Amazon example for one.

And I think that's where you come up with strategic alliances, right? That's where you really work within, I mean, it's a cloud company, right? It's a hyperscaler. So if they can. That's what the marketplace is for. If they can package up what we do and add it to what they're doing, I think that's a win for both, right?

So the way that we treat our core customers, the level of service that we provide is super unique. So to have a place for us to, a platform for us to list out like our version of MySQL or Postgres, I think it's, it's a win for both. And then when you think about how the community kind of rallies around, you know, open source and wanting to keep things open I think For me, the biggest shift I saw was when Terraform went closed and there was a manifesto.

The community was so irate that they came up with a manifesto. And then from the manifesto, like they ended up The, the name of the project now is OpenTofu, but it, it started, I think it was OpenTF and it's underneath the Linux foundation, just like Valky is now underneath the Linux foundation. And I think when you have something that's open, that people use it and adopt it because it's open and they build their businesses off of it because it was open.

And then you close it.

Speaker 4: Yeah.

Lori: Then you have like this big community uprising, specifically if it's something so popular, like Terraform or Redis and to be a company, to work for an open source company like Percona that is there to carry the open source flag and say, okay, so Val Redis closed source, guess what, we're going to be a part of Valkyrie.

Like we're the only ones that offer, you know, like on prem support for that. And in that look at the other companies that have joined Valkyrie. You've got your AWS, you've got your Google, you've got your Oracle. So it's like. All of us working together to have a piece of this pie, if you will, in open source land, I think it's, it's, it's really awesome to be a part of a company that supports open source, that it's that core to who we are.

Jonathan: And so is it fair to say that Percona's business model is not simply hosting databases for companies? There's more to it than that.

Steve: Yeah, perfectly fair to say. What's the, our model is actually you hosting your own database. So we give you the software that lets you. Run it on whatever your infrastructure is.

So if you want to run it on prem, if you want to run it in the cloud, if you want to run it across multiple clouds, we, you know, we actively support and test all of those scenarios that's given the power to the, to the people. Yeah, yeah. And

Lori: I, and I think what we do is we help you level up, right? So we have not only the innovations that are coming out of engineering, but we have the experts in our support and our managed services.

So if you are trying to elevate, if you are trying to do something and you realize you don't have the staff or the support in house, that's where we come in. So having that, that like hand to hold you through a migration or to help you tune your database so that you're getting the best results to take your database and put it on the cloud with our new product, Everest, like just because we're open source doesn't mean, you know, that we are a less than we are actually like a

Jonathan: No, absolutely.

I agree with that. I, I think part of the problem is that so many companies, they, they love the open source model, but they've not really figured out how that model actually ties into their business model in a way that makes sense. And sometimes they end up sabotaging themselves. And it, it sounds like Percona has managed to avoid that.

So that's, that's good. That's a good thing. That's a good, good sign for the future. We just celebrated

Lori: 18 years. Oh, wow. Right, Steve?

Steve: Yeah. Yep. 18th birthday. We became an adult. Percona can now legally Drink and phone? Do what 18 year olds do. Well, just different in every country as we learn. We did a poll internally about turning 18.

It was like, wow, it's amazing. Globally 18 means different so many different things around the world, but that was a lot of fun

Jonathan: What was what was the beginning of the company? What was the what was the first thing that percona did was it was it literally my sql stuff My

Steve: sql

Jonathan: phone support. Yep phone support really Ah,

Steve: so so our our our co founders were actively taking calls from customers to help optimize their databases.

So they were a consulting company. So if you think about Percona, it's a per con performance consultants and dot com was taken. So per con a dot com came to be. So yeah, little, little trivia history on our name, but yes, Peter and Vadim took many, many phone calls. I think, I don't know the exact date, but we have a book.

That they put out at about the 14 year mark, or maybe it was the 13 year mark that said, like up until at the time, eight or so years ago, they were actively taking phone calls from, from customers who were having performance issues. Fortunately we have enough staff, we have, you know, smart enough people that they don't have to anymore, but that's how we got our

Jonathan: start.

Oh, that's, that's great. That's, that's a lot of fun. And, and so what's the, like what's the portfolio now? What, what are the different databases that you guys will. Still take support calls for her.

Steve: Yeah. Yeah. I mean, we're, we're active. Like I said, my SQL is what most people know us for MongoDB, probably second.

But then we, you know, we're very active in the Postgres space and getting more. So every day and then from there, you know, where you choose to run it, we support, like I said on prem bare metal installations, cloud installations. We create operators now, so you can run this in Kubernetes. So in your orchestrated environments.

Which we're seeing a ton of traction there. We're now releasing you know, if you think of Amazon's RDS is sort of like database as a service, you know, point, click, get a database. We have an on prem equivalent to that in our Everest product. So it still uses Kubernetes and work and operators underneath, but it gives you that that single API that you can call and wire into your CI CD or.

Or, or turn over to your, to your developers and let them point, click and create databases as opposed to having to know all the, you know, behind the scenes commands.

Lori: And you forgot Valky. So Valky is our latest database that we're going to support.

Steve: Yeah. And Valky, we are only, we only support it. We don't actually build our own version of it.

We contribute to it. We're, we're active in that Valky community, but there's not like a Percona fork of Valky. So Valky, we, we're trying to. Support that community not compete with them by any stretch,

Jonathan: right? And so i'm sure you could you could imagine a future where You have a slightly different take on something and you do end up having an in house valky fork.

But I i'm sure trying to avoid that if possible, right? Yeah.

Steve: Yeah, I can't imagine that future I don't really see it. I don't ever see it as good right like I mean it it If we're really community, right like and we mean it then We find a way not to fork now. I can't fault any of the projects that have had to go, you know, go their separate ways.

I know there were certainly valid reasons, some of them personal, some of them technical, some of them vision, but I don't know that that's always the best outcome. So for me, that's like the last thing. If we've hit that, hopefully we've exhausted all options because, you know, to take to take a thriving community and effectively split it in half It's gonna slow, slow innovation down, right?

That's one of the beauties of open source is just how fast we can innovate when, when all of us are focused on the same thing. So again, I'm preaching here. This is just my personal belief, but yeah, I don't, I don't love that, you know, direct competition with each other. If we, if we Percona fork, first and foremost, we want it to be to help innovate and add back to the community.

But in some cases when that, you know, when that, when that takes its turn, like, you know, for us, we do have a fork of MySQL. Oracle Enterprise isn't going to accept our LDAP authentication. They have their own, right? They don't need us, but, but we do feel that it's good to bring to the community. So I think, like I said, we do it for a slightly different purpose.

Jonathan: And so, for example, we've talked about Everest a couple of times. So that's, that's fully open source. If, if somebody wants to just host their own version of Everest and not have to pay you guys anything for the contract, that's out there as an option. And then your, your business models, essentially, this is a complicated thing to do.

We suspect that eventually some of these businesses are going to want to have support for it.

Steve: Yeah. Yeah. I mean, that's been our model is if you've got the manpower, the time and the talent to do it yourself, you Our software is free and open source. If you don't, we can certainly help with that. You know, that's been our model and that's fortunate for us that our business model has been successful enough that we've been Continually growing and continually profitable company since the beginning.

Jonathan: Yeah, well, so I mean that that business model has worked for other companies over the years That's essentially what red hat was founded on and it it grew them to be a billion dollar business so there's there's definitely nothing nothing wrong with that And that is that is proven out to be a a winner a winner in many industries at least yeah, that's very neat.

So I'm, i'm curious in addition to that. Do you ever get someone say You Call you up and say, Hey, it would be great if my sequel did this. It would be great if Postgres had this feature. Can you give us a quote on adding that to the code base? Does that happen?

Steve: That's

Jonathan: only on the days that end in

Steve: Y. Yeah, we get, we get that a lot.

Like every customer has a unique need. You know, that. This is the way our business works. We need this feature, you know, and we have a product team that does the evaluation. Is this solving a problem exclusively for you or is this a problem that many, many people are experiencing? And by creating it, we can we can benefit the broader communities.

We try to stay on that, that, you know, community based side where we're looking for large swaths that would be impacted for the products. And that's not to say that we don't do some, you know, custom stuff. We have some customers that have specific needs. If the ask isn't huge. We put it in there. And there's a, there's a monetization model for that.

I mean, sure. But I

Lori: think that is that is the value of having community live under engineering as well, right? It's this feedback loop. So as many customers as we have, as many conferences as we go to, being able to have the pulse of whatever community we're in, if it's like an all things open, it's just a general open source.

Speaker 5: You

Lori: know conference, but if it's a PG comp, we can really focus our attention to that particular language, that particular particular database. I'm sorry. And, you know, really work within that sort of feedback loop. What are we doing? Well, what are we missing out on? Like here is a talk that we're giving that talks about why we decided to do this, you know, and And I think that's where community wins, right?

Again, Percona being an open source company and community has always been at the forefront of what we do. And then being able to really like zero in on the events that we go to and having like a forum where you get answers, having a very good resources within Percona to give that knowledge out to the community is a win win.

Jonathan: And so that does kind of lead onto another question about that community effort. What, what does it look like as far as in, in these places where you guys are the, the, the source code, like you have your own. Your own repos for this. Do you get a lot of patches and work from the community? Or, or is it pretty much just a a one way street where you guys do the work in the community or you're out there using it?

I

Steve: mean, I can tell you we get a ton of contributions. And, and, you know, community contributions don't always come in the form of long lines of thousands of code changes. Sometimes it's improving our documentation. Sometimes we get a lot of community give back on our forums where they'll come in and answer questions for, you know, for the masses.

Like, how do I set this up? How do I use it? I'm getting this error. So we're very, very active on that front. That's one of the things that shocked me. I don't. come from open source, like, you know, way back when this is something within the last five years for me, but seeing how active our communities are is the exciting part because number one, that the entire burden doesn't fall to us because I think if it did you know, and maybe this is where some companies go wrong is you know, the community is not paying your bills.

So you tend to favor the ones that are but because we have such an active community, it allows us to work together. And, and accept code contributions, accept documentation updates, accept you know, answers on the forums, things like that.

Jonathan: Are there, are there any, are there any rough spots? So I'm, I'll give you the background to what I'm thinking here.

I'm part of an open source project and we just recently got sort of called out in our one of our, one of our places where people can ask questions. One of our forums of essentially you guys made this big change and you didn't ask us about it first. And it's, it's been a little bit of a thing because like there's, there's a couple of ways you can look at this.

For one thing, almost every open source project is a meritocracy to some extent. And so, one of the answers that we give is, if you guys would like to see changes, come write source code. Come be a part of Discord, where the, where the developers are talking so that you can give feedback. But it's really, it's really become a little bit of a source of friction.

There's some other things going on, but not important. Of course, it's always more complicated. Yeah, but I'm just curious. Do you guys, have you guys had any of that kind of friction and how have you solved it?

Steve: We, we have our own forms of it. I don't think we've seen that specifically, but with, you know, every, for the products, That we are in control, like PM and Percona monitoring and management.

We don't have an upstream necessarily beholden to, I mean, we do have you know, a copy of Grafana in there that we base it on. But, by and large, we're driving the, the roadmap. And a couple of times we'll deprecate a feature or, you know, we'll put something in that ends up not catching on and so we sort of abandon it or start pulling it out.

And so we definitely hear about it.

Jonathan: But I think I think it's impossible to deprecate a feature without someone coming along and saying, Yeah. That was part of my workflow. You know, it's the XKCD comic about holding down the spacebar, heating up your CPU. That was part of my workflow. Put it back.

Steve: Yeah. We get that, right? Like, but you realize we took that out because there's a better way to accomplish this. Right. And so most of the time we luck out with, you know, sort of a re education on it. We, we didn't Take something away without giving you something back. Here's, here's why. And, you know, it hasn't ended sourly.

I can't say that people are like, oh, thank you for that. But they, they at least walk away with an understanding of like there, there was a rationale behind it. It wasn't. Who can we piss off today, you know?

Lori: Also, I think it's a it's a matter of communication, right? Like it's a matter of giving people warning letting them know what's going on being able to have you know I don't want to say standard talking points but you know those standard talking points as to like this is why we did it and because you you don't want to Detract from the end reason, right?

Which is maybe there's a better way, or when we actually looked at our logs, nobody's using this in terms of like the hand scheme of things. So having a good communication and like Steve said, like having the forum as a place where people can converse, you know being able to kind of contribute back, submit issues that I think is, is key to keeping the community Safeguarded from some detractors.

You know, it's always the, the small group can sometimes make the loudest noise, but when you're just very transparent, like in the whole methodology of being open source, I think, you know, you end up just winning long term.

Jonathan: Yeah. I was, I was going to ask you, Laurie, to, to speak to this next, because sometimes the community people can be on the other side of the fence from engineering.

And if there's not, if there's not decent communication between those two teams, they can be actually working at cross purposes. What, what kind of things do you guys do to make sure that doesn't happen to make sure that the community team, so this is something that just recently happened to us the community team got surprised by a change that the engineering team made and it's like, Oh, that's gotta be a bug.

We're sure to get that fixed. And then the engineering guy is like, that wasn't a bug. That was an intentional change. Like, please tell us about these things. How do you, how do you, how do you avoid that? Yeah.

Lori: Well, I love that my dog just started barking. So I apologize. Hopefully it won't

Jonathan: be

Lori: too much in the background, but I'm very happy to say that at Percona community has moved under engineering.

At other companies that I've been to, community has always fallen under marketing and that's where you get the disconnect, right? Because marketing has marketing objectives and engineering has their objectives and when there isn't open flow communication across the board, Then you have those issues where you're like, wait, I'm presenting the wrong version of something or I didn't see the release notes.

Nobody told me. And it's it's a communication issue. And so I know, don't laugh. It's she's old. She's 14. She doesn't know what's happening. She's just like, why is mommy talking to a computer? So So now that community has moved under engineering, it's so much better because we are on all of the product updates.

We are looking at the roadmap. We are making sure that we are dropping breadcrumbs to innovations that are going to come out in like four or six months, right? We're letting the community know like why we're thinking the way that we're thinking, what our engineers are doing. And then with that, you know, we have, Surveys that are just product led surveys that are just like, we want to know what version you're using.

What is this? Like, how is this working out for you? Like, what can we do? And so keeping a really tight feedback loop like that is where I think community and engineering can just sing. So it's been a night and day from previous companies to be under the engineering org and showing them our value, right?

Like our tech team, our tech evangelists, our developers, our engineers, like they have. Like the history within their own career of touching source code, of building things, of doing testing and doing that kind of stuff. So we're also working to embed them more within engineering itself, like to be the ones that are helping to contribute, that are looking at pull requests, things like that.

So that I think is like, it's, it's new for us. And we're just kind of testing it out, but I'm very excited because the tech teams. are very like my tech evangelist are very excited to get their hands dirty. Nobody wants to be a showpiece like you want to be able to actually demo what's going on. You want to be able to speak from engineering and say, this is why we're doing things.

And that's what we're working on right now. And I think it's going to go really well.

Jonathan: Oh, yeah, that's interesting. Steve, what, what does that look like from, from, as I say, again, your side, your side of the fence, trying to bring your, your evangelists in and let them work with the code. Has that gone well?

Steve: Oh my god, it's better than you can imagine. Typically I would say, you know, we rely on community for the voice of the non paying user. Because we don't really have too many other avenues for that. And if you're not careful, your entire roadmap becomes the customer roadmap. So getting that continuous line of feedback from them, you know, of what does the community want is awesome.

And then even more so that they can be so hands on. And again, it takes all different forms. It's not purely contributing lines of code. Sometimes it's using it. Sometimes it's, you know, Hey, I couldn't talk about this at the next conference. This is too clunky. This needs some polish. And this is how I think it should look.

So, you know, the, the, the forms of feedback take all different forms, but they are wildly valuable. You know, you didn't realize just how much you missed it until you start getting it. You're like, wow, where would we be if we had thought of this. So much sooner you know, it, it, like Laurie said, it's been a fairly, not fairly recent change, you know, measured in six months or so, five months, something like that.

But it's been a great change and we're just keep building on it. So I'm excited to see it continue.

Jonathan: So is, is the next iteration of this to, to, to make sure that the marketing people know the source code too?

Steve: We might be asking too much. I don't know if they have an interest in learning the source code. At least the tech evangelists have an interest in it. And I'm not faulting the marketing folks. You know, don't ask me to design anything that looks catchy. I not a creative,

Jonathan: I get that. I just, I know I've heard stories of, you know, marketing people coming and asking just the most off the wall questions.

Like, could it be possible for us to do this? And, you know, engineering goes, I guess it's technically possible. And then a couple of months later, you see the latest marketing thing, this new feature coming soon. Yes.

Steve: Yes. That, that happens everywhere. That's every job I've ever been at that happens. That has nothing to do with open source though.

That's just human nature.

Jonathan: So it's, it's the marketing guys just trying to grab onto something new and flashy to put in. Yeah. That's great. Okay. So let's see. What, what does I'm curious, what, what does this look like then? So if you get somebody from the community that says, Hey, I've got this great idea, but they're not a paying customer, but you look at it and go like, that's legitimately could be useful for people.

Like what does the decision matrix look like for this to, to, you know, are we going to put some resources into making that happen?

Steve: Yeah, I mean, so I'm going to speak for my my product counterparts. You know, but but one of their big value ads is really assessing the overall market need. And it's not just, you know, take your idea at face value.

Is there enough people that might buy this? Well, that makes the decision, but it's Is there enough of an idea in your suggestion that might be able to expand it or might be able to take it in a different direction, maybe sometimes even narrow it. And that helps us get to a, Hey, we're onto something here.

There is a legitimate market need. And again, we do partner with the community team to make sure that we do vet those ideas before we just run right out and build it. Cause Lord knows that's expensive, but Hey, we're thinking of this thing. We got the feedback. It seems to be valuable. Let's talk about it at the next Pick your conference, you know and see what the feedback is if it's positive if you know if we hint that hey We're thinking about doing this and the feedback is oh my god.

Yeah, that'd be awesome. Then it just starts moving up and up the list You know, I think the biggest thing is Giving that community a voice and then listening to it. Not just, you know, here's a form you can fill out and, you know, black hole it. Right. Right. I

Lori: think one of the things, sorry, that I think that we're missing from this conversation that I'm happy to bring in is not just Percona related community, but community in general.

So Steve is my Guinea pig and I'm so happy to have to be on this podcast with you and one of the things, one of the communities that we are a part of is finos. And so FinOS is the financial arm of the Linux Foundation. And one of the things I was sitting on a call that they were looking for is contributions to one of their projects.

It's the something something control. I can't remember. It's three C's. It's CCC. I'm terrible. But it basically, they need help with, like, Expertise. And so Steve, I was like, I've, I've all told him that he needed to help me. And now we're contributing our sort of reference architecture for for my SQL in terms of like having a hardened, like security features for my SQL things you should look out for.

And so it's not just talking to ourselves or talking to our own community. It's how can we impact the broader community with our, our expertise and, and There's another project that we're working on it's called Gwok. It's in the OpenSSF, so it's a security project. Gwok is being built in Postgres.

One of my contacts was like, Hey, you work for Percona now, we need help with Postgres. We wanted to keep this open source project completely open. Can you help us? And it's like, Yes, of course we can. So we are now looking at their database as they're building out this security project. So it's, it's not just working within our own sort of Percona mindset.

It's what can we do at Percona to really elevate the community in general. So it's, it's a, it's a really exciting time to be at Percona because we are sort of flexing our, our skill set in other communities that maybe we haven't approached because You know, it's, it's not just the database, right? It's like, how does the database make your project better?

And so I think, you know, the next thing you think about is, is AI, right? And so there's all of these tools that are coming out and for us, it's, we don't want to be the next greatest AI tool, but like, Hey, what are your, what are you building out of you're building out of your database? Like, do you want to make sure that you are fine tuning your database that you're getting out of it, what you need?

So. Again, it's sort of like we are vastly more than just talking to ourselves, which I think is, is what sets us apart.

Jonathan: That's, that's an interesting point because, you know, Steve and I, we've been talking a lot about the, the, the community upwards to your upstream vendors and the community down.

Downwards to your users and Lori comes, it just brings along this idea of what about talking about the community outwards to sort of our peers and the other people doing interesting things. And that's something that we we don't think about quite as much. And that is, that is really interesting. And so what is the, what's kind of the motivating factor to want to reach out and work in some of those, some of those kind of vertical directions, excuse me, horizontal directions.

Lori: Well, I think again, it's like everybody has to store their projects somewhere. And if we can lend a hand in making your project successful, I mean, that is the value of open source, right? It's people coming together to build something, to innovate on it and to make it better, to solve whatever problem.

So for guac, it's an S bomb problem and where Percona might not really be interested in solving like S bombs, you know, like your supply building material, if we can say, Hey, in building Gwok, if you do this to your database, it makes it cleaner. It makes it better than at the end of the day, your project will be better.

Your project will be more successful. That I think again, is, is the value of going, To these other communities and showing what you do and how you can do it better. And when you think about foundations, like look, who's involved in foundations, you've got your Googles, your Amazons, your Azure's, you've got your Facebooks, you've got your Fidelity's, you've got Citibank, you've got all of these massive companies, you know, like what is it?

90 something percent of enterprise companies use open source components. So why wouldn't we want to share the love across the board with foundations that we're a part of?

Jonathan: Yeah, absolutely. I'm curious. Do you find that you're kind of your community? People have to sometimes do some translation work for your your engineering team.

And so what comes to mind with this is I do. I do a little bit of small business it for very small businesses around town, and sometimes they'll come to me with their ideas. The, the strangest asks, like, I need a computer with, and, and they'll say something that they saw from a magazine or on an ad somewhere, like, I need a computer that has one of those new NVIDIA workstation graphics cards.

You're a sprinkler system. You, you put. You, you put lawn sprinklers in what, and then, you know, so then there's this, okay, so what are you trying to accomplish? And then they'll tell me, and it's like, Oh, you don't need one of those really expensive GPUs. You just need, you need a new desktop with a new processor and a GPU in it that will run something with CUDA.

And then you can use your blueprint software. And so there's this, this kind of. Translation issue where they, they, they know what they want, but they're trying to tell you like the details of it and they don't understand the details. And I have to imagine that in the database world, working with, working with customers and even some on this, working with other foundations that this sort of thing happens to, we need a database and I don't, I don't even know databases well enough to tell you the sort of off the wall things.

I'm sure you guys get it though. People ask for off the wall things and it's like, well, let's help you actually come up with what this, what you really need.

Steve: If we're lucky, we can hear the, you know, what they're truly asking for. That's if we're lucky. We do rely a lot on, you know, the community team, the product team, our customer success team, some of our service, you know, services engineers.

Like, they are masters at hearing the problem under the problem that's being described. Right, right. If we're unlucky, we just start building based on, you know, I need this NVIDIA thing, you know. Yeah. And that's happened, well, once or twice, more than I care to admit. I'm sure. You know, I think then, then we circle back, like, okay, how do we miss this?

How do we get this out next time so that we're not just, you know, sort of chasing fantasies, but actually, like, getting to the core of the problem. Mm hmm. Because there's nothing worse when you build the exact thing that you were asked to build, turn it over, and they're like, it didn't fix my problem.

Like, what, what was the problem? Like, maybe we should have started there instead of just building the solution. But yes, yes. Yeah, I

Lori: love it. Because it all comes down to communication, right? Like, there have been many times and like, Steve can tell you where I've dragged. Somebody from engineering on a call with me, right?

Talking about community stuff and people asking for things that I have no idea, like I'm a marketing person, right? Like I have, I am not technical. I can try and translate to a point, but I need to pull someone in to be like, what, what, what exactly are they saying? Like, and then, you know, and then also like being able to take a step back to be able to stop a conversation and say, okay, hold on, like, Okay.

Let's talk about part a you've just went from A to Z. Like, let's just figure out what a is. And I think when you have a good team involved, right? When you have someone that can maybe take the marketing fluff and drill it down to the actual ask. And then you have the engineer that can take the actual technical side of things and and put it together.

You're gonna win. But when to Steve's point, when you just do one or the other, you could build something that's not necessary. And, you know, nobody wants that.

Jonathan: Yeah. Do you, do you get this with pull requests as well? I, I'm sure you must from time to time.

Steve: Usually you get the, yes, it is very, very specific to their problem.

Not, doesn't take into consideration how anybody else uses the feature. You know, but believe it or not, it's, it's less. Frequent than you might think more often than not, you get a, you know, very flexible contribution that's like, you know, this solves only my problem, but it makes extra effort not to step on anything else as a result.

So those, you know, those are the ones that we appreciate. But even the ones that are purely self serving will try to kick it back and say, Hey, you know, is there a way to do this a little bit more friendly to the other people that actually do want this to happen? And, you know, can we improve the. The error messages or the return codes or whatever the case may be.

So we try to, you don't want to tell somebody, you know, thanks but no, don't ever come back. So it's always to keep the conversation going. I mean, it takes an awful lot from someone to be willing to contribute code and databases. They're not easy. So if you find someone giving you database code, there's someone you want a relationship with.

So we do our best to keep that going.

Jonathan: Yes. Yes. It seems like databases and compilers are two of the really hairy problems in, in modern code. Not

Steve: for the faint of heart.

Jonathan: Yes, absolutely. Okay. Probably a question for Lori then. Have, have you guys had any real like pain points beyond what we've talked about with trying to wrangle the community?

I know sometimes there's things that come up that are just not pleasant to deal with.

Lori: So not since I have been at Percona, but I used to be the CDF, Continuous Delivery Foundation outreach chair. And in that there, it was kind of there was a wonky time, like they lost their executive director.

They got a new one. There was, you know, community didn't understand what was happening within the organization and the way that we, like, and there's nine or there's eight projects under the CDF and the, and the projects were wondering what was going on. And so what I did was I held feedback sessions.

Where I just let them tell me everything, the good, the bad, the ugly. There was lots of ugly. And it, it all stemmed from miscommunication or lack of communication and not understanding like what was going on. And, and from that, you know, creating an actual action plan moving forward, having everybody buy into it because they saw.

Bits and pieces of the conversations that they had within the plan itself, and then launching the plan and holding yourself accountable to the things that you said that you were going to do. And the change from when I started in June of 2022 to, we had CD con in, um, May of 23 was, or April of 23, whatever it was, was phenomenal.

Because everybody felt like they were a piece of, of the CDF, whether they were a project, whether they were a contributor, whether they were someone that just happened to use one of our projects, a user, someone that adopted it, it was just, it was amazing. And I think that is where community can really shine is when you do take a step back and say, okay, this isn't working.

How can we make this work? How can we fix this? And it, if it is sitting there listening, I had very many calls and The CDF is a volunteer position, so sitting there knowing that people are going to rag on you for an hour at a time, telling you how you're not good, like you're not good enough, you're not doing what you said it takes a toll, but it also I think is inspiring, because it makes you want to do the things that the mission statement says.

Excuse me. The mission statement says you chose to be at this foundation for a reason, right? Like you chose to use this project for a reason like we need to be beholden to like why we made that decision. Like we need to hold true to our core values as to why we're doing this in the first place. So it's, it hasn't happened at Percona.

I mean, I'm new, but I, like, I think again, the way that. Coming in from an outsider looking in, right. From a marketing person, that's always been community to a community person inside of engineering. It's really awesome to see the innovation from our engineers, the feedback from the community, the amount of push and pull that is all positive, you know, it's, it's really cool.

So happily haven't had to deal with anything like that, but again, coming from the CDF, it was just a matter of, you know, letting people's voices be heard and then creating an action plan and then being accountable for what you said you're going to do.

Jonathan: So I want, I want to let Steve answer that if he wants to, and then I have a followup for Lori, because this is fascinating.

Steve: No, no, I actually, I don't, I can't recall an instance like that where, where where there was just an up, upheaval. I think we, you know, I think we're very cognizant of how you treat the community. You asked earlier about, you know, how do you, it, you can't just be. In your own little community, you can't expect everybody to come contribute to you.

So we have to You know, that's why we share our expertise outward So I think because we're very very aware of that. Maybe that's helped us dodge some bullets but I can't I can't remember my five years here any ever being sort of thrown at us I don't think it's luck. I think it's you know, we're very intentional about how we get involved in the various communities either the ones that we depend on our own or the ones that we depend on

Jonathan: Yeah, makes sense.

Okay. So, Laurie, let me follow up on that with, with this question. When you have those conversations with community and you do come to the conclusion that, Oh, there's a In the organization, the broad organization, there is something that we've done here that maybe it needs a course correction. How do you get buy in from those in the organization to convince them that there needs to be a course correction?

Like that seems like that could be the hardest part here. How does that work? Do you have any secrets for us?

Lori: So this is actually funny. So when I first started with the CDF, I had a nail in my tire and I had a meeting with the core team from the Linux Foundation that worked on the CDF, and I didn't know that the new executive director was going to be on the call.

So I'm at a tire place. With my hoodie on, my headphones, like looking very much like hacker crazy, and not realizing who was on the call with me, I just kind of stated everything. I was like, this is happening, this is happening, this is happening. And this is how I see a pathway forward. And without realizing that the executive director was on, he's like, yes, like, yes.

And so I think it's being able to justify the actions that you want to take having an actual plan and a strategy. And it wasn't just like, I want to listen to a bunch of feedback. It was, I want to listen to a bunch of feedback from that feedback. I want to create our plan of action from that plan of action.

I want to present it to everybody to get buy in from that buy in. Then we're going to go. And it's having the overall strategy when you, and then presenting it. Right. Because if he had said, no, like, I don't want you to do that. I would have said, okay, what is your suggestion? And then I'm going to go back to what mine is, or maybe I will alter it and include his pieces.

I tend to steamroll. I don't, I can't, I can't help it. But again, it's all about like opening the communication and, but also having the, the. Like the information to back up what you're saying. Like I was a part of the, that organization for a year. I was one of the people complaining, which is why I stood up and be like, ran for outreach chair, right?

Cause I was like, I cannot stand to be in another one of these meetings. Again, this has been terrible. Like we, like, we want to do something and nothing is happening. I guess I will stand up and I will, I will do more. And I think that's also what makes community great. Right. Is that everybody comes to the table and it's up to you to decide, like, How much you want to put in, like, do you want to be a chair?

Do you want to volunteer? Do you want to be on a program committee? Do you want to initiate, like, do you want to go to GitHub and say, here's an issue, like, do you want to give feedback? Right? Like. Community is only as strong as you make it. And it's up to the individuals to kind of take that leap. Like you can use a project all you want and never say anything.

And you can be, be upset that things aren't working the way you want it to, or you can have ideas, but if you don't share them, if you don't bring yourself to the table, then they're not going to be heard. So I think again, for me, it always comes down to like open communication, feedback like listening to others and then Using that to strategize on how you move forward.

Jonathan: Yeah. Excellent. All right. We have been going for just over 50 minutes. I'm curious. Is there something that we haven't talked about that it's just a burning that you guys really want to talk about? And this is sort of a different question. Difficult question, because you'd have to go through all the things I wanted to talk about and which of those have we checked off, but is there anything that we have not gotten to that you really wanted to?

I

Steve: mean, there's a couple of items in the road map. Maybe I could assign a few JIRA tickets to you if you wouldn't mind just throwing some code down for me.

Jonathan: I can write code, but man, databases is just outside of my expertise. That is actually a good point, though. What are you guys looking forward to for the future?

What is coming down the pike that you're excited about?

Steve: Oh, jeez, all kinds of stuff. I mean, I think the ones that I'm most excited, obviously Everest is huge for us. You know, bringing. database as a service, giving you control, giving you sort of the best of both worlds, the AWS experience, but also control of the data and the systems underneath it.

And then one of the other big exciting ones is our Postgres team is working on transparent data encryption, which is something that we're really excited about. I know there's been multiple There are multiple solutions out there, but we, we think we're looking at a slightly different approach to the same problem.

And hopefully you know, the, the early returns are that the community does see value in it. So I think that's always great for us to keep, to keep on going when we hear things like that.

Jonathan: Yeah. Transparent data encryption being like the, the database itself automatically does encryption and then it's encrypted at rest.

Steve: It's actually stored encrypted in the database. So it doesn't matter. I mean, It's, it's actually double encryption if you encrypt at rest, meaning the disk itself is encrypted, but this is encrypted inside the database. So even at runtime you don't have to worry about your data being exposed unless you are the application that was intending to see the data.

Very cool. So it's being encrypted in real time as it's being read and written to the database. Excellent.

Jonathan: Blory, did you have anything you wanted to plug for the future? Anything you're super excited about that's coming down the pike for Percona?

Lori: Well, as a community person, I'm going to jump on what Steve just said.

Right. And like, we're going to a bunch of conferences, conference season is back. So I am very happy to be able to be to be at events representing Percona, so we'll be at open source summit Europe. We're staying for Valky day. We'll be at all things open. We'll be at PG conf in New York.

We'll also be at. OSFF, which is Phenosis Conference in New York. We're going to be at KubeCon. Like, so to be able to take everything that we just talked about today and then put it into practice, like, in person is what I'm super jazzed about. So, like, to see what the reaction from the PG Postgres community is when we sort of have our presentations about TDE, to get their feedback in person, to launch Everest at at Open Source Summit Europe, right?

To see, like What questions people are asking to see how we can then react is that as a community? Like, what do we have? Do we need documentation? Do we need this? Do we need that? Like, this is like, this is where I get super jazzed because it helps set the stage for what we do in the spring. Right. So with community, everything you do is like a couple months in advance.

Right. So being in person at like. Six conferences in like three months or something crazy. And then be able to take all that knowledge back to the engineering team and then package ourselves up for the spring round, right? KubeCon in the spring, open source summit in the spring, all of those other things is like, it's the fun time.

So I'm well rested from the summer and now I'm ready to get back on the road.

Jonathan: Excellent. Busy, busy, busy. All right. So I have to ask you guys each before I let you go. These are our required two questions. We'll start with Lori. What's your favorite text editor in the scripting language?

Lori: I'm sorry. Did you see the face I

Jonathan: just made? What? I'm

Lori: going to pass, like.

Jonathan: So is it Microsoft Word and English? We get that sometimes. I'm not going to lie.

Lori: Then I will, then I will, I will say that. Like, you just, I was not expecting this question. Catherine did not prep me for like the standard questions.

I've got nothing. I got a little egg on my face. Just say Vim. Vim. Vim. 100%. No,

Jonathan: no, that's fine. And that's, that's something most of the people have an answer for. Everyone's every once in a while, somebody doesn't, I think the really fun one is when someone's like been in management for a while too. And they're like, let me think back to what I used to use back when I actually wrote code.

So that's, that's fine. There, there is absolutely nothing wrong with that. But Steve, same two questions. I imagine you'll, you'll have a little bit different answer.

Steve: Well, I've definitely Vim is my favorite editor. This copilot thing, I'm, I'm digging. So I use VS code a lot more for, you know, like an IDE, but I still, my old go to quick and dirty, just Vim file.

And I could be a thousand times faster navigating the file there than in anything else.

Jonathan: Yeah, absolutely. And scripting language.

Steve: Bash is my go to. You know, Python, if it's going to get longer and more involved, but like bash, if I could type it on the command line, it's just,

Jonathan: it's just easier for me.

That's great. I'm waiting for someone to tell me one of the esoteric extensions on something like bash or Python. So somebody that uses Amber. Instead of bash as their main scripting language or I was I was recently introduced to Python which is Python with braces and some fun stuff out there like that

Steve: I work with a guy who's all about z shell everything was a shell.

That's great. Like I'm lost in that

Jonathan: Fun. All right. Thank you both so much for being here. It was a blast Really really appreciate it and that's Laurie LaRusso and Steve Hoffman. Thank you guys both both for being here

Lori: Thanks for having us.

Jonathan: Yeah, absolutely. Good stuff. All right. Well, that is the show today.

And as I'm sure you noticed databases are not my area of expertise, but we we we sorta, we sorta hung in there. We faked it for part of the way, but had a really good conversation about community as well as some technical things and a bit of database stuff and was It was really, really a blast.

We've got something really cool coming up next week. We are talking with Andreas about Ladybird, which is a browser, but it's not one of the big browsers. It's a new browser, sort of written from the ground up, which is really interesting. So make sure and tune back in next week for that. Do you want to follow me?

There is, of course, Hackaday. I've got my security column goes live every Friday. Friday, and that's also the home of Floss Weekly, as you probably know. There's also the Untitled Linux Show over at twit. tv still, and make sure and tune in for that. We sure appreciate everybody that's here, both to get us live and on the download, and we will see you next week on Floss Weekly.