What's going on everybody? Oh wow, that's excellent. My voice cracks up in a very opening sentence. I'm fifty three, going in my third round of puberty here, so welcome back to the show Adventures and DevOps. I'm your host Will Button joining me in the studio or in parade. Welcome Wren, how are you? Man? Hey? Thanks a lot. You know what's interesting. I do a lot of public speaking, and one trick I've learned is that you're supposed to sort of say like hello to yourself, maybe
quietly before you actually start talking. But we were just coming a long conversation, so I honestly don't know what happened there. I would that's that would That's one of those things that I think would take practice, saying hello to yourself, because then I'm gonna I'm gonna start overthinking it, like who am I saying hello to? Like I mean, it's be you know, create
any sort of existential crisis here. But it's like, you know, I mumbled to myself sometimes when I'm dealing with a difficult problem, so that sort of comes a little bit natural to me. Anyway, the fact that I'd like to share today is something important to remember. I think in the world where we've started to convert to more to DevOps, which likely means agile working that means no sprands, no scrum or anything like that, we lose a
lot of ceremonies that come by default. And with a big release that happens once a year or once every other year, we get we were reminded to stop working on that thing potentially celebrate our successes. So it's still important to do those things. As a matter of fact, if you release every day,
you almost should have little celebratory dances every day. But I usually recommend making sure that your team does something once who are, once a week or once every couple of weeks to actually sit down and be like, we're actually a good team, we actually got something accomplished, because I think it's really important for the health of the team and individual team members. That's that's a
really great point, man. That is I like that. I can't even express in words how much I like that, because we spend so much time, you know, feeling like we're under the gun, and you get something done and there's no time to celebrate. You're behind on other stuff already. So I think that's a great tip. I'm terrible at doing it, but I I think it's important, and you know, I think teams love this too. So in previous engagements we would have like a beer every Friday or
something like that where we actually talked about what we were doing. I think it's super important. Yeah, for sure. Well, onto today's topic. We're talking about awesome DevOps and assuming that if you're listening to this show you're interested in DevOps could be a bold assumption on my part, but I'm gonna roll with it. And you know, like whether you're just getting started or whether you've been doing it for a long time, there's a continuous learning curve.
So our guest today, Maria Stanka, has created a repo and has collected a bunch of resources and how to like break down and categorize the different components and aspects of DevOps and it's really it's really a cool resource to go through and just kind of reflect on, like, oh, I need to work on this, so I don't need to work on that, or even like from a beginner's perspective of where do I even start. So with that long ramble out of the way, Marius, welcome to the show. Thanks
thanks for Yeah. So what is your background in DevOps? How did you get started? How long you've been doing it? Oh? I think I started a freelance project, was a small project, and I think that was that time. Is that popular? And I started to work let's as the boks here, and then I realized how actually, how cool is that? And then I'm doing from that time only the bops. I think I have more than ten years Yeah, the boks right, an, So what was
the inspiration behind creating the awesome DevOps repo? You know, I think all of us, or most of most of us, we open the links all the time the browser. We have a bunch of tubs in the browser, and you keep that I'll read this later, you know. And then I
realized why I should not put this, let's in that document? From that time I was working actually in the corporation, and in that time I spent a little bit of time to take all the tabs and then putting that in this let's say document, and then why not a website written around this? And then I gave this to some of the colleagues, what do you think about this? And they they said, hey, it's a very good idea.
Actually one of them that pointed, hey, you need to put this in this section something like that, And then I said, okay, let's go into this in the in the in the in the open source space, so then we can we can share this with a lot of people. There. You've got quite a lot of things. Yeah, there wasn't only my contribution. Actually, most of I mean more than tell people actually contribute to this, and they're keeping let's say people to let's say let's say fok or
stars that's still keeping let's say beverage is five per day. But yeah, it's it's grewing. Actually, yeah, it's green. Do you maybe this is a trick question, but like do you know everything that's like in the repository, Like are there things that have been added here? Like I don't know what that tool is that someone just threw in there. That's a good question. Actually, I'm not I don't accept any any actually any resource in
that document. Actually I'm thinking not all of them, but some of them might experiment also because I don't want to put any resource which is not related. Most of them, by the way, or not most of them, but some of them. They tried to put only four let's say promote something like that, not accept. Actually, if you let's say, give a free account, or if you give at least away, you can experiment, or you can play, or you can use let's say the project, then
yeah, it's one of criteria I can accept. But yeah, I don't know all of them, to be honest, but at least I true. I mean I I I checked actually all of them. Yeah, I checked to see if it's related to the develops or it's related to the sorry. Yeah, so yeah, yeah, And I think that's a I think that's a really good point for for everyone to acknowledge, especially if you're new to DevOps, as you're never going to have one hundred percent competency across DevOps.
You know, there's you're going to have things that you are good at, and then your company or your job is going to shift and you're going to lose those skills, but at the same time pick up new ones, and then you'll circle back around, yeah, and get mad at yourself because you're like, man, I used to know terrorform a lot better than this, but the skills have died off, and you've got to not only like brush up on the skills that you used to have, but also terraform has added
a bunch of new stuff since the last time you used it or whatever product you know. I'm just picking on Terraform. But the thing is like it's continuous learning. Yeah, for sure, I like that. It's a cool
hack to actually find out about those things. Right Like if you follow one of these repositories, every time there's a new change or discussion, it's like people are coming to you and telling you about that those things that are in the industry that someone is using potentially and get extra experience and know where to look next. Yeah, so go ahead, and I so a few of the let's say, people I know they want to start also in the bops. I mean, what's the BOBS doing? How I can get started?
Just give dis repeal. Then they go to the road map for example, and then see what actually is the BOBS I think it's yeah, it's I think it's helping you. It's helping them people who want to join in the what to see? What actually is the BOB set oriented towards those beginner engineers that inexperienced or would there still be value for those that have three, five
or twenty years of experience. I think it's more available to people who has more experience actually, because there are some other tools or at least the tools for are in the its on the sections. Maybe they want to I want to to see what are I know? How for example the key station is working, for example, but I want to see what are the tools? Who? Actually it's based on my use case or it's based on it's based
on my cover my needs. Actually, then I go there and see all the tools, and most of the tools are quite complex and you need to know at least you need to have that knowledge tool let's say, pick up quickly so you can take and fit in in your work. Yeah. I think mostly it's most for the for the people who has experienced Yeah. So for your own personal use how often do you find yourself referring to your own
website for resources? U can you put? Yeah? How and do you do you come back to your own to the to the awesome develops repo to look up information for yourself? To be honest, I think keep the tabs open. Uh, but oh, but I'm going actually to check the c I C. Actually I have a pipeline who actually check every week the link for each of the repole, if it's still available, if it's responding or has actually TPS also, and I check then that time actually I checked this,
this, this, these links on the ripple. Yeah, I don't use let's say this very often, but yeah, but it's okay to be as a record. Yeah, I mean I can imagine that there's some turn in the industry, but you've had to automate this. So that mean that like a lot of legs get broken, like those products are no longer maintained in some way. Do you have any like stats or just be like one in five just don't exist after some time or something like that. Yeah,
last time it's one which is Vulture cloud. It's not responding. But I saw I saw some of them. They changed a link and maybe there is a security issue something like that, or maybe the link really relates to something which is bad request, let's say. But yeah, last time, just one. I had also ten which are not responding one time and then I get rid of them, yeah, or I change also the link in case there is a ripot, I put the rep link that I'm using one of
the requests. Actually, when you put a link, it's okay to put the GitHub because githubits remain there. If you put a let's say a customer domain, sometimes it is broken because it depends on where you store this. But yeah, I had also ten no more than ten, well this time just one which is not responding. Maybe they put a limit. I don't know, I don't know. Yeah, and more than two yeah, more
than two hundred responding. Yeah, oh wow. Right on. So when I look at the top level organization of this you have You've got like cloud platforms and then file systems, automation, orchestration, database is like all these upper level categories. How how did you decide that those are the top level categories? Uh yeah, interesting question. Actually, I think based on so when I look to the tool or platform, whatever, I just see what
actually for what is made. Actually, for example, let's take orchestration. For example, Let's say we know, uh this, this orchestration is doing something like you can orchestrate let's say the work law or let's say process or whatever. Just I just put Let's say I can find this set of tools which which are doing similar things, for example, but they are for different needs, but they are doing different things. And then I I put a
tag based on this, this these are doing the same thing. Then if it's let's say more than two, three, four, I just put a tag, just put this in this section. This is from them. Yeah. Do you find that there's like a lot of tools that all on the multiple categories, Like it's not so easy to quickly identify that, oh, this is a data work it's like a database as a service or some sort
of data platform or something like that. Indeed, Yeah, yeah, yeah, I found for example, rg if you take Argo, it's not just for the city. You can use for other things which has different components, and it's hard to put only for the city you need to put Also, there are category categories yeah, for the pipelines or data pipelines or something like that. But it's yeah, there there, I mean, maybe I'm I'm actually curious about that. So rg CD primarily, I mean it's in the
in the name that it's the ICD. But you're saying that there's you know, they focus on a bunch of different areas, you know, talking me about that. For example, they have Argo workflow. You can you can run multiple I mean workflow which you can run let's say batch jobs. For example. They have worlds Argo events. You can do this let's say for different events. You can trigger different let's say workflows. They have also also with ugocity you can do not only the deployment, you can use for even
door key station. If you want to integrate this in the products. Let's say I want to deploy restar creation. Let's say let's this are creation in the cloud for example, and then you want to deploy let's say every time new new triggers coming in the argosity you can do. This doesn't have to be let's say for the continuous deployment or continuous delivery. So yeah, it's for different use case. Yeah, even if it's built for the c c D, you can do or other things. I think that's one of the
benefits of having curated resources like this is you're able to to me. It feels like you can get that information much more quickly from a curated list like this than you can from the company's own website about their product. You know, because like when you go to when you go to like someone's website, not to pick on anyone, but there's like all the marketing and the fluff
and the pipeline to get you to cough up your email address. They can add you to their email list, you know, and it's hard to just say, like, like, what do you actually do for a living? I'm here on your page, what do you actually do? And then when you figure that out, you know, like what else can you do? And so I think that's where curated lists like this are really really helpful, just because it strips down to here's what you're going to do. Yeah,
I mean, I think there's a ton of value there. Realistic by its just we have one that we maintain for SoC two related resources and we get requests to update it similar and realistically we have to go to the provider and be like, you know, what are you charging for? How much is
it? What do you get out of there? And some of them just can't even answer us straight, like they want their link to there, and I'm like, you told us that you charge seven thousand dollars per user set for some small companies, but that's not what your pricing page that, So like where did you get that from? Where did you pull it? Like, well, oh we have we're updating that, or the feature set doesn't
match what they're looking at. And so it's like a real challenge to even like you're actually providing more than just the list curation, but also the definitions of those things and what's important there for people that are coming and looking for that tool set and it's not marketing nonsense. How big of a struggle is that for you? I know you mentioned it earlier talking about checking to make sure that the links are not only accessible, but that there's a free version
of it. Do you find that there's a lot of a lot of places where people try to slip in ways to get you into their sales funnel using your site. I saw they saw a link, they open a fulle request in that request has was the link and the end of the link it was a code I think mostly Yeah, some of the people, Yeah, they try also to promote and based on this code, it's referring, hey this this, this request came from these people something like that. I know.
Yeah, not all Yeah, it's not happened all the time, but yeah, it's yeah, that's like a yeah, that's a whole business model. There is go around and plug in like your affiliate links exactly, yeah, exactly, affiliate links exactly. Yeah, I get I get it right.
You want to know whether or not this link is getting clicked on, and you know where these users are coming from, you know, if they're actually going as this tool, because maybe maybe you're like, oh, these users, we're changing our content where this link goes, and right now it's pointing to someplace, and we want to know, hey, are we're actually getting links here so that we can make sure that the content is oriented better for the audience that are coming in. I don't know, like, I mean,
there is something with tracking which is for sure problematic, but I do think that and this is sort of the problem with the tech industry in general, the overlap between use for a good purpose and use for a nefarious purpose, Like it's the same tool, but you know your intent and what you do with it at the end of the day is it can be much different, and there's no way to distinguish upfront, like what they're going to actually
do with that information. Yeah, that's that's one of the that's that's a whole nother podcast episode there. I'm going to stop talking. I mean, I know it's for sure true. You know, Marius, I didn't. I did notice that you you worked at quite a few number of different companies,
and so that experience must have really built up over time. Did you notice anything significantly different in the sort of role that you were doing for different companies at different stages or uh you know where they were at or how they their approaches to DevOps. Yeah, no, I think the most challenging on the website. Actually they don't know how to let's say, structure the teams,
how is the communication actually and actually what the responsibility means. Actually for the devlops, this is the most most problematic in most of the companies. They want the wops, but they don't know how to supplied actually the wos and but in the end, yeah, the boves would be the same. But yeah, most of the companies you join, yeah, they think to let's say the books is different things than the other one. Yeah, some
of them they think it's let's say, new new system. That mean some of them it's sorry, yeah, it's yeah, yeah, I know, but yeah, this is what what like? What does that look like from your perspective? You know, it's not it's not I mean I agree with you for sure that it's not just happened or as ay, those are other areas for other roles, but you know your perspective here, what what does
that really amount to for me? Yeah, it's let's say, I think for me, it's just bringing the barrier barrier between let's say the application and the infrastructure. Actually, for example, I wanted to puy let's the infrastructure, so I should be able to play the infrastructure, but also I should be able to deplayd application. So for me, the debopsites simplify the whole warkflow let's say, from developed application to the prod application and including all the
tools set actually on the back end. That's me Actually. Also another example, let's say even the let's say the the person who's building the product doesn't know about the infrastructure, but at least that person should have ability to grab the tool set and then just click or just type of command and then deploy something that this one actually means for me at the box, Yeah, not making for the let's say, getting the application from the person who developed and
then just deployed by your No, you should give the yeah, the let's say, the instructions and then hey, just to buy yourself to simplify, Yeah, simplify the the workflow. Yeah, good application life. That means yeah, I know, it's combined little bit, I mean and it's using the same process like platform engineering, it'll be but at least, yeah, increase the productivity, increase, let's say, but the experience. Let's say, so you can you can have these processes to use it. Yeah.
Yeah. I think for me, the way I've been thinking about and this is sort of a new approach for me that's developed over the last couple of years, is it's about putting in guardrails and creating a set of guardrails so that the application engineers can build, deploy, and ship as fast as they want to with wherever they're at and their expertise. And then I'm just going to put guardrails in the way so that they can't accidentally do something that they
don't want to do. Yeah, like including your API key in the U r L string or your link. Yeah, you can have. I struggle that it is. Your job is never complete. You know. I want to interrupt you there real quick on that very point, because I get asked by tons of people every week is DevOps still a valid career path or is
it a dead end? And and I think that point right there highlights the fact that we will never be done with this Warren because every every every bad path that we close just opens up the opportunity for someone to forge a new path that we have to close. I mean, I don't know if I would to be so cynical. There's an aspect here for me that's about automating your job away and so whatever it is that you're doing, make that go
away. That either you go into the organization and you make it so it's no longer a business problem or a tech problem or whatever, or you're building some tool around it so it's not education, it's there is a problem that needs to be solved and making it so that something works super reliably, like that CLI command et cetera. You don't own it, some other team owns it realistically, but you're improving it to such a point where it's automated.
And then there's still and then you're out of a job and you look at the organization from the outside end. You're like, hey, look this other thing is broken in the organization. Like let me be the person who bangs are faced against the wall over and over again on that problem. Now you know it's in the it's in the database space, or it's in the security space. Because of what you're doing or there's a new tool out that's nice
and shiny that everyone wants to use or not. I think the bigger problem that I see a lot is the organization has divested ownership in innovation to a different team, and that team is divesting ownership in innovation because they're just trying to automate whatever is in place. So there still has to be this aspect where you go out and learn about new tools through get up repositories out there on the internet and find whether or not that's the best way to be going
about it. Someone still has to be doing that and bring them in. In the past, that role in my teams has been called a full stack
hipster. Ye. I mean, it's just the sort of person that ends up there and they can do a great job pulling this stuff in and they go through every RSS feed that they can find, every email from every source and think about these things and I think realistically injecting new ideas in because the worst thing that you could do would be to create those guardrails that you said, well, and then those guardrails, you know, pointing you off a
cliff ten years in the future that you know, it's very difficult for even people to see realistically. And one of the most challenging also in the biopside, it's how to combine all the tools together because I saw, yeah, different let's say teams they put like this, the other ones like this, all different companies has different It's hard. Yeah, hit somethingdeed. This is actually most for me, think this is most challenging is to grab these tools
and put sort together. Actually, so that's a problematic, yeah for sure, because you get like you work with really talented engineers and they take some different tools and build them together to create something. And you personally may have experience with all those tools but never thought to use them this way. And so when you first into the room, you're like, oh my god,
should I be like impressed or scared? Nah? Yeah, there is for sure, you know there is something and maybe I don't understand there's genius going on there. I feel like something about both approaches isn't isn't the best? Like on one side, it's wow, they're using that tool for that like that, and on the other side, it's wow, they built that themselves,
like whatever it is. And so I think there's some old joke here about there never was the case that some engineer came into a team after being recently hired and said, Wow, that last person's code in this domain, it's absolutely fantastic. I don't want to touch it at all. One of my favorite examples of that is I had this job a long time ago and we we had our application was a web based application using ASP dot Net one
point zero. And I know, right, but it gets better because the guys who built this, rather than just writing the ASP code the dot Net code to render the website, they decided that they would store the dot net code in a SEQL database, and so the web server would query the SQL database and it would get the dot net code in the response, and then it would render that as the website. So if you wanted to like change a frame or load a different image on the website, you actually had to
update a stored procedure in the database to do that. And I was like, Okay, that's unique, that's novel. Well, we're not going to do that anymore. Just wait till get to the point where they're like,
well, you know, what we found was that was inefficient. So we ended up doing was creating our own tool that allowed you to edit that database in less sort of like a like a I see I you know, I saw the one where previous job where people got on the Microsoft train very quickly, and so they're like, oh, silver Light, that's a thing, and you know, for enterprise applications, we're building stuff for silver Light.
And it's very it's very sad when you're trusting in a large organization to maintain a framework very long, and it gets I think for that one, it actually got discontinued in Google style before it was like officially out of the pre release, and so you're left with that sort of thing, and I don't
know what it is. It's maybe it's like on the both sides of the spectrum, right, small companies they have a couple of people who just don't want to use a lot of tools so they do it themselves, and also large companies that have lots of people who are like, we also need to do it ourselves. It's like there's got to be somewhere where someone's saying, you know what, I know the perfect tool for the job. If only
I knew it, the tool was Yeah. I think I think what you're getting at there is something that I've seen for decades in this career is there's this overwhelming desire from corporations to create a tool to let non technical people perform the technical roles. And I get that, because finding tech people is really really hard and expensive. I get it. It's just I don't think it's
been successful and still today. Right. That's why we're using the GPTs to generate source code, right, so non technical people contribute as well to the bugs and our production platforms that one's actually working. Because I've got my GPT bought making poor requests. As we're recording this podcast, it looks like I'm
still at work. I mean you should. You should let us, I mean all of us here, I think, uh listening want to know, well, how exactly you did that, because I could definitely get some extra money on the side doing or three extra DevOps jobs. The key here is my code is so shitty to begin with that they're not really going to see
anything except improvement by letting chat EPT do it. Oh no, I mean maybe you should train it on some good repositories out there that have a list of good tools, and you know, every once in a while throwing integration with one of those just to make it look like, oh, you know what you're doing. Oh, that's a Marius. Have you tried this? Have you pointed chat GPT to your repo and said, hey, I'm wanting a tool that does this? No, I did them all right, I'm
going to check back next week with you. You got to imagine that, especially a company like open Ai, have used Marius's repository to actually funnel aims of products and companies and tools that can be utilized in this space. I don't know how much. I mean, we know that getthub was for sure a source of this. I mean get up as their own solution here that they have to be utilizing. So realistically, it's invaluable not just to human
onlookers, but also our future robot overlords. Yeah, fair point, And we want to be in their good graces as much as possible. That's a roco's basil, right. But that's why I always tell Siri thank you every time I ask for a question. I think that's important to train and to train them that that's the appropriate human interaction. So how much to put together and maintain a resource like this. How much time and effort do you spend on this? I think a couple of hours per week, it's right.
I yeah, so that's not huge, but that's still Look, it's one ongoing commitment. Yeah. I think there's a couple of things that strike me from that is one that you can produce really quality resources without a lot of effort. And two, if I were to just put in that same amount of effort, I could probably close down about two hundred browser tabs already thinking about automated like, how can I create this tool that a lot of closes browser tabs and just moves the links into it like that? I just saw
that's where his thoughts were going. Yeah, I think. I think it's also the experience, to be honest, working in many tools, you have kind of idea that's actually that tool for example, or jump pretty quickly on that tool, and then I just read a little bit about that. Just let's say if in case you have options to play that, they can give you a free account for example, you can play with that. So it's let's see. Maybe also because of these things are quick, you know,
yeah, yeah, so that's go ahead. One. Have you found anything particularly interesting in your you know, journeys through creating getting all these products and you're like, wow, like I didn't know there was this tool that did this thing. It would have helped me so many times in the past. Yeah, yeah, I didn't know. There is a I don't know if you know getub, get lab CEI and if you are if you are to run, let's say, uh, if you want to run, let's say
runners. I mean you see, I mean pipelines. You need to have runners to run for sure. And I did not know that is I think it's was quite a pretty with this tool. And this tool actually simplify the management of the runners of the Good Club actually, and yeah, I actually experiment to that, give me account actually and experiment to that and looks like
it's very very interesting. Actually tool you can spin up runs quickly actually just connected to the gidlub and then so yeah, there's some which is which are very good actually, yeah, and simplify the works that's definitely unsolved bootstrapping problem and develops maybe someone hasn't Yeah, how do you how do you have the runners already working before you show up on the first day, Because as far as I can tell, there's a manual deployment that's got to happen where you
click some buttons somewhere and be like, Okay, finally I can let the you know, machines do their work now without uh, manually playing with them over and over again. Yeah, that one is still in review because I put a tag reviewed so the people can know what is the status of the poll request and that on it's I think it's a review if I'm not wrong. But anyway, I'm playing with with that tool. So every prequest which I have, let's say there's a new tool, I just play with that
tool. I will try to play with that tool in case they give me this dis option or I just read the documentation if it's quite valid for me, and then yeah, just based on this, I can I can march actually that the changes. Yeah, but what's new tools? Yeah, you need to spend a little bit time. Yeah, yeah, because this Yeah yeah, I mean it's not let's say, it doesn't take too much. But maybe because of the experience, I don't know. Yeah, I'm in
other spaces with especially physical products. I know companies will give you, you know, send you a copy of whatever it is so you can try it out and post a hopefully glowing review on your stream or YouTube or whatever. I actually hadn't considered the possibility of that for for software tools. You know, I guess maybe here's my pitch right now. You know, come companies, you know, let us you know know, you know, give us a free version of whatever you've got, and you know we'll we'll test it
out. I think it's a quite a great opportunity for them to get users. Not just put a random promotional link that people can click on, but actually, uh, you know you're willing to bring it in, talk about it, actually test it out and use it and put it give the feedback. I get that from my YouTube channel, Like companies will approach me and say, hey, will you try out our product and do a video on your thoughts about it? Yeah, that work out so well so far.
Not really. Now, maybe that's not the right answer. I should say I haven't done many of them because for me, it's I know that me personally, like whenever I first look at a new tool, I'm like, oh man, this is cool, and I get I artificially inflate my excitement
about it. But then whenever I come back the second time, I look at it from a more pragmatic point of view and say, well, is this really solving a problem that I have, or is this me playing with a tool pretending to solve a problem and that time and effort you know, for me, I just know me in my personality, I have to go
through that process before I can get a full accurate picture of it. And the time and effort it takes to do that in a YouTube video just hasn't been a good fit for me. So I I haven't done Actually many of those. I've been approached many times, but I haven't done it. There's a real challenge there, a time investment to really understand how a tool works. I mean, I'm on the opposite side of the spectrum. I see
something and I'm like, I don't want that. I mean, maybe it's great, but I look at it, my first reaction is like, that's just extra overhead or extra effort to pull on yet additional tool. Like we have to remember that every time we had a tool, that's extra cognitive overhead for everyone involved to have to think about and deal with, which is a real concern, and the only way to reduce that overhead is by using less tools in some way. Unfortunately, there's not a lot of alternatives there.
But I didn't the decision on buy versus buy versus by is also incredibly challenging
to know whether or not a tool be the right one. One of the things that keeps coming back to me I wasn't a methodology that I had a lot of experience with early on in myine career is this idea of one way versus two way doors, where if you're trying to pick a tool that you want or even make a decision, if it's an easy, reversible decision, or the impact of that decision is low, then it doesn't matter which way you go, and you might as well just flip a coin and pick one
and go from there, because you can regret that decision and just make an easy change. But if there's a huge impact of the organization or you'll be bought in for a long period of time, then you'll be forced to pay
that cost if you do make a mistake. But even then it's good to think about how to separate it out into multiple pieces, like can you make a decision at a smaller level instead of buying the whole tool and getting a three year license deeal for that, and then that becomes the justification for continued usage. Go with a one year license and then revisit after a couple of years of renewing it. I think I'd heard that before, but it's never really stuck. But hearing you say it, I'm like, oh, yeah,
that makes a lot of sense. Sometimes these things work on other people, right, Like they get stuck in a decision and you got to figure out what the right words are to say to help them unstick themselves. I think another one that comes up very often is one tool to solve all of
your problems. I think you may have heard of the single pane of glass, which I'm not a huge fan of, but I think it goes around in a lot of the marketing team, sales teams, business analytics that realistically, as a user, like I want to use my ID, right, Like I want to go in there, and that's the only thing I need to worry about, Like I like get the command line. I'm sure someone's
like, well, how can you like the command line? But then we have a venuser right here, so but you know, like that's mine, right, Like I don't want to have to leave that tool unless I really need to, unless I'm testing a UI, and I obviously don't do that very frequently, but the same is true in other areas. So definitely consolidating
that can be helpful. But at the same time, if you're overutilizing a tool to do something that isn't necessarily meant to do or creates extra channel es, it is better to have too in other situations for sure, Marius, what's your favorite what's your favorite tool that you've learned about as a result of
curating this website. Ah, it's sun stuck actually really right, an not only you can do configuration management just putting a file on the selver, just dorink a lot of bunch, I mean, doing a lot of things. Even if you have let's say, to pray some reasons in the cloud, you can do that or let's say, type in a command in see a life from one thousand servers for example. At the same time, so I think for me, it's interesting how actually it's capable, Yeah, this tool
and you can do a lot of things there. This one looks interesting, I mean most it's is the one which I like most. Actually, yeah, right, I actually haven't used to myself. That's like a advanced version of like puppet chef Antable or something like that, And it does configuration management. Yes, exactly. It's similar with tensible, but the difference is UH has a master. Actually it's working like a massive and slave. But with
the new version intends also what Exible is doing with SSA CHA. But yeah, it's I use this tool, to be honest, to do a lot of things, for example, grabbing all the information from those CEREB it's in the real time actually, because you have a slave for each of the servers. That's I had more than five hundred top five thousands, and it's hard to let's say, to keep eye on the services when it's failing. When
you let's say, execute the sole stack for example. Uh, and I use this tool to grab the information in real time in the graph to see when it's failing. What is the server it's failing. What is the line it's failing of the states states it's state, it's the file Actually it's sol stack it's executing. So it's you can do a lot of things. Yeah, that's cool. Yeah, I have a tool which gives you at least
expassibility. You can, let's say, do more things. Even let's say I mean you cannot do that in that time, for example, but you have a gate, you have a way you can extend for other things. I think it's the best tool, is the best to the Yeah, for example, Kubernets, all of us, most Kubernets. Kubernet is not only
just run the application, right, you can run jobs. You can run let's say, you can extend for other capabilities with the CRD for example, you can keep the data by the way, yeah, as a as the objects at c D for example. Yeah, you can do a lot of things. Yeah, it's yeah for sure. So what's the future of awesome DevOps look like? What kind of things have you got planned for it?
Yeah? Yeah, the future of develops it's right. Now, First of all, you need to put price for each of the tool because yeah, because most of them they're trying to put commercial let's say link and it's good to have at least from the single view. The view you just enter the website and then see the the list. You just it's a it takes I mean, the decision. It's it's quicker if it's free, for example, right, so you will see, uh, what is the price for that
tool? For example? The other one that I don't know if it's included in these as the books, but it's okay to be I think it's to put a framework in place, actually a framework which brings the organizations how they can adapt the bobs quickly and what are the steps, what are the tool but the tools as an example, not let's say you need to use these tools, but at least they give this this This framework should give the let's say the management for example, to grab the bobs quicker in the com money
for example, how the teams should be structured, how they should communicate, what are the tools, what are the workflows or process or what are the things they need to automate or whatever it's in the devil. It's it's better to have these including should it should be something complete, you know, take this and then do do the devils into organization. Hm. I see we were smiling a little bit. I mean there is something about that that scares
me not not having it. But like if I look at the Agile manifesto and then I look to see how the word agile is being used today or used incorrectly today, Uh, you know, I feel like everyone's sort of done the wrong thing with develops as well. So it's sort of an unfortunate situation that companies adapt what they're doing they don't understand a word or a mindset, and then map into what they're already doing or what they already understand.
I think that's sort of just humanity, right, It's there's not a lot that can be down there. So I think having a roadmap for how to actually implement that in your organization, I think there is value there. I can just see, you know, there'll be eventually a safe version of oh yeah, and in a last version, and I don't know how I feel about that. At least they can give the basic information. I don't know how they can start because I saw people I don't know what is the starting
point actually to adopt this or something like that. Or they can give you high over let's say, I don't know, they can easy pick up maybe easy, but yeah, at least they have starting point there. Yeah, I think it's I think one of the cool aspects of that approach is asking, like creating the framework from a why are you here perspective, like, Okay, you're interested in DevOps, but why what's the problem you're trying to
solve? Because the answer to that might change the path that they're pursuing and lead them to results quicker. He just wanted his end to talk about selling things and education and like where is your customer at right? You know, when they end up on a landing page, you know what are they looking for? Are they looking for? What is? Like I heard some cool kids saying the term DevOps and I don't know what that means. I want
someone to tell me what it is. And the difference of like, wow, you know, it's perfectly aligned with what we're currently trying to do. And we're innovators in the space and we want to know more. We want to know toolfless and other things that we can be doing. And this is a huge, a massive spectrum from one side to the other, and I think capturing that is quite a challenge, right, Like even individual products go
and have a very narrow mindset of what their audience actually looks. Like Mary said that, you know more of the expert practitioners who probably are looking for ways to convince their organization to implement develops more effectively, rather than the I don't know what develops is, but I'm sure we're doing it because we're a great company. By you also, I mean a couple of years, Uh, someone asked me in the interview, actually, what is the the bops?
I explain a little bit blah blah blah. And he asked me, you are contributed to source? Yeah, I have I contribute, but mostly let's say some of the bobs people they just give this as some of the bobs. And he told me why he didn't give this when I ask you what is the wolves? I think and I think it's a good idea. Actually, if you want to see what is that the wolves? Just keep the link. I don't want to answer, I just give this link.
I feel like this is this challenge the other way around though, because people don't like reading things right, so like I feel like you have to make a custom message for each person otherwise like oh you know what, that answer is one hundred percent defined at this, like just you want to go here, like trust me, just read this. And they still don't want to do that. Yeah, what's funny? What's funny? Right on? Well, yeah, I think we've covered it, Like that's this is a super
cool resource. Like I keep scrolling. I've been scrolling through it and looking at it as we've been doing this entire episode, and it's very very thorough it's easy to navigate. It's really really well done. So congratulations on that because you've been doing this. I was looking at the getubstat you've been doing this for like five years now. Yeah, it's a long time. Actually,
yeah, that's commitment right there. That's impressive. And I think one of the cool things about that is that's five years of history and changing tools, adding new tools, removing older tools, with consistent updates. So I think that alone just gives us a lot of value as far as being a relevant resource. And not only this, but it's help you. Actually, it's helped you so to see if that tool is still available in the market.
You know, I saw some of the links. Oh the tool does not exist anymore, or it's in the high mode or something like that. So it's it's it's nice. It's nice to help. Also this updated, Yeah yeah, yeah, yeah, awesome. Well should we do some picks for sure? Right on, Warren, what'd you bring for us this week? Yeah? So I was reminded recently how the DEVOVS movement may have been started in the agile mindset, and actually before that was lean, which comes
from manufacturing. So many people don't know about this Lean is all about eliminating waste in organizations, most specifically manufacturing. And there's a book that really categorizes this, primarily called the Toyota Way, and it contains a bunch of principles they say management principles, but realistically they're how to achieve more effective flow in
your development team. You can have lean software engineering. And I learned about this before agile realistically, and it's really affected my mindset and put me in a direction of it's not just about doing things quickly, but really understanding the
overhead in any number of aspects. The one that I find most interesting is a product that's been delivered has a negative impact on your organization because until you start getting money in or delivering that delivers value to your customers, it's sitting on the shelf and it requires maintenance and upkeep in order to make sure that it's still effective. And so if a product isn't actually returning enough value,
the best thing you can do for your organization is delete it. And I feel like that's a mindset that is lost on a lot of people, and being in the devov space, it's super critical for us to help remind others that there is it's not just free, Like you don't just write something once and throw it out there and then you don't have to do anything after that.
Yeah, that's a very very valid point. Working a lot with early stage startups that, like I do, I advocate a lot like you have to define the success criteria, like this product, when we launch it within six months has to be meeting these objectives otherwise we're deleting it for that very reason. That's that takes some real discipline to actually achieve the like congrats,
Like that's not easily done. Yeah, it's it's not easy by any means, because when you even if when you define those criteria, like when you get to the measurement day, like you're emotionally attached to this thing, and so then you try to like massage the criteria and say like, wow, we didn't really hit that, but we do have this and and you have to be It works best when you're just brutally honest with everyone and are in a group of people where you can call each other out on your bullshit.
I like, you know, I think what helps me here is I use the mindset. If we didn't have this anymore, you know, we're at the day, Like, let's let's take it. Put everything on the table, Like, we have no engineers, we have no product, we have no software done at all. Would we today hire a group of engineers to go and rebuild this thing by buying it off the shelf and buying them all?
And often you're like, wait, no, actually I don't want to do that, you know, knowing what we know now and then that's a good indicator for what you should do and how to separate the current state from whatever mindset or unfortunate attachment you have to that. Right on, Marius, what'd you bring for a pick? No, fair enough, I'll give you a minute to think about it. I'll go next. So my pick, if you have been listening to the show recently, this one's not going to
surprise you on once again picking platform con. We've got the list of speakers are out on the website now, and so at the end of the conference, I'm going to be doing a live Q and a session with some of those speakers, asking them questions. So I'm treating this as a way to like get to like the humans behind the talks, you know, and learn like the people who are doing this and what what they're into and what drives
and motivates them. So if they're specific questions you want me to ask them, hit me up on XAKA, Twitter or email or whatever place you see me hanging out online and let me know what your questions are. And platform Con is coming up in June. It's five days this week and five full days live streamed, one hundred percent free, tons of great speakers. Check
it out platform con dot com. Yeah, I mean, don't miss your opportunity to bug Will on this, honestly, because I find a lot of the value at conferences is sort of the human interaction, the questions that come up afterwards. You can go lots of places and watch arbitrary content on YouTube and whatnot, so you know, don't if you see something there interesting, definitely forward the question. Yeah, for sure. It's like so many of
the talks. You know, whenever we give talks, we are in a bubble and we talk about what we think is going to be interesting, and there's there's all kinds of questions, and sometimes there's just like questions like you see someone giving a talk and you're like, Wow, that person is doing something really cool. What could I learn from them? And this is your chance to ask those questions? All right, back to you, Marius,
Which guy can be anything? Can be anything? Yeah, content, you know, something great personal life or something technical if you prefer that, Yeah, I think going to some technical Actually, uh, I didn't experiment anything except the technical part because yeah, it's there is no type or other things. But yeah, but if you let's say you want to learn something, uh, you need to experiment. And by experiment you can none. You can let's say you can grab that, let's say piece very quickly. You
can even speak about that. You can let's say you can build. Actually, so I think it's not you don't need let's say, well, it's not enough to read. It's it's better to experiment. Actually, and I think for all the people who are listening, if you want to, for the new people who actually want to, let's jump in specific area, it's not enough to read just experimentals. Yeah, one hundred percent agree, absolutely truth. All right, excellent, Well Marius, thank you for joining us.
For everyone listening. Check out Awesome Develops, Awesome, dash Develops, dot x y z. Are you active on any social platforms? Hang out online anywhere? Marius GitHub blinked in but GitHub mostly Yeah right, awesome. So if you want to reach out to Marius, if you can find him on GitHub and my website mons not x Y you see. Oh right, huh, there you go. I like it. The simple domain names,
they're getting hard to come by. Yeah, it's easy to remember. Yeah, but you know, one day, especially for you, Will, you may find your last name as t l D. But you know what's going to happen. Somebody's going to like jump on Will first, and then I'll miss out again. Yeah, you have a you have a previous traumatic experience. Yeah, dude, it's me forever to get Will button dot com. I can imagine. Yeah, yeah it did, but that's story for another
time. Marius, thank you for coming on the show. I really enjoyed the conversation. Lauren, thanks for joining me again. Appreciate having you here and for everyone listening. Thanks for listening, and we will see you all next week