Foojay Podcast #76: DevBcn Report, Part 1 – Learn from the Community

Transcript

Automatically generated from the audio, so it can contain errors in names and technical terms. Suggest a correction.

[0:00] Let’s visit Barcelona and learn from the Defbn visitors.

[0:04] Welcome to the Foojay podcast. All your news about OpenJDK.

[0:10] Early July, the Defbn Conference in Barcelona brought a diverse lineup of speakers with topics across multiple technology domains. Hertiling took the camera and microphone with him to Spain. Together with Nacho Quil and Jonathan Villa, two of the organizers, he spoke with many of the visitors about what they like the most in Java, how AI influences the way they work and what is important to them in the work they are doing. We have more than 20 people who are passionate about the Java community in this podcast and are eager to share their knowledge with you. Let’s start.

[0:45] Okay. I’m Nacho. I’m from Barcelona and I’m the organizer, one of the organizers of the conference. So here I am trying to make everything working.

[0:52] My name is Jakob. I am also from Barcelona right now and I volunteer at WCN.

[0:59] Perfect. So it’s a wonderful conference. can you tell us about the background and what’s been going on over the past years? How it came about?

[1:06] Well it’s a very good question. the thing is the story starts like 10 years before let’s say when I was unemployed. some of my colleagues told us in fact in a barbecue saying oh we have to do something bigger what about doing something like we can meet more people etc and that’s why because I wasn’t employed we started thinking oh we could gather more people instead of small meetups you know to do something bigger because we started to grab attention attention in Barcelona and that’s why we started the first JBC conference in 2015 and from there well you know here we are and nowadays you know looking for Kubernetes topics to the AI topics to Java still Java for the future what is coming from the front end what is the cloud whatever so everything all the topics right now are let’s say they can join let’s say in the conference and eager to know what the community let’s say wants to know let’s say in the future

[2:06] You said it’s 10 years but it’s not the 10th conference

[2:10] It’s 10th conference we are not celebrating it as if it was a 10th conference

[2:14] That’s that’s the issue Where’s the grenade?

[2:15] Where’s the cake?

[2:16] Where’s the party?

[2:16] Where is the cake?

[2:17] The cake.

[2:18] Where’s the cake?

[2:18] That’s a good question. do we need a cake for the 10th?

[2:22] Why not for the ninth or seventh, eight? So, a cake every year.

[2:26] 10 is a big number.

[2:27] Oh, okay. Oh,

[2:29] When my kid turned 10,

[2:31] You will celebrate.

[2:32] Yeah, it was celebration.

[2:32] No, we not celebrate when he has nine.

[2:35] No,

[2:36] That’s a good That’s a good question.

[2:37] Well, well, I will I we can fix that.

[2:39] We are celebrating by having this interview for the Foojay podcast. This is a celebration.

[2:45] Thank you very much. My name is Bachanver. I’m CTO at Open Value and I’m also in the board of the Netherlands Java user group. So what excited me about the Java community is the people like you and me. So at events like this, it’s amazing to meet people who we known for a long time or new people and hear about what they’re working on and all the amazing stuff they’re doing with Java basically.

[3:06] So my name is Ricardo. I have a YouTube channel actually from like basically main Java and also spring boot. So the name is programmandon Java. It’s in the Spanish unfortunately. So I have been just working with Java for 12 years. Well Java, cutling like everything related with GVM

[3:24] And yeah I mean to be honest I find like amazing language.

[3:30] Tell us something about your YouTube channel and what you do there.

[3:33] So I have been starting for four years. So right now we are like around 16,000 people of course like Spanish people like Latin American people and I just like from the basic from what is a hashmap an array list to like architecture things like adapters and ports architecture or like everything related with configuration with spring spring boot. I try to make more like personal now more for example the last video that I uploaded was more about like imposttor syndrome so I try to help people out of course and I try to teach them like with my courses with my videos I also have like a small community in discord that we’re chatting and we just try to make progress together like because some of them like also juniors so they’re trying to find the first job and they share their knowledge they share the experience And to be honest like if I will I would like to share it like in my point of view the level of juniors nowadays is really high. I mean I don’t know like when you started like but when I started like around 12 years ago or something

[4:45] I guess like there was not that much information out there

[4:48] But now they are like wow I get like super surprised about all the information they have. And do you have any opinion on AI and the impact that it can have on developers?

[5:01] This is like kind of a tricky question because in my opinion everybody just talking about like

[5:05] Yeah we’re going to be removed the program will not be a 60s anymore and there has some like subs and services that try to do that and in my opinion that will not happen. It’s really difficult to do that.

[5:19] Great.

[5:20] And what is going to happen is basically we are going to be like super black like I think it’s a tool that make the developer four times better, four times faster and it’s an amazing tool that if it’s right use it’s it’s is super productive for the developer basically. Yeah. Yeah. In my opinion. Yes. My name is Kristoff Newman and I work with the Closure programming language which is built on the JVM and targets the JVM and it’ll gives you a nice dynamic lisp and a pretty high performance lisp that runs on the JVM.

[6:00] And what do you do in the closure community? So I am the developer advocate for Closure and so my job is to help people hear about Closure and learn how to use Closure and be successful when using Closure.

[6:16] Excellent. And what are you doing at this conference?

[6:19] So we’re here at this conference to talk about Closure and then also talk about Daytonic which is a database built using Closure that doesn’t forget anything. So it remembers every change that was ever made. so you could think of every change being its own snapshot. So you can go back a month or two or three in time and query data from that point in time. And so that was built on closure because closure is a has immutability. And so you can’t change data in place. And so you could think of it as like getting snapshots for every change, but it does it in a performant way. So you it gets rid of a lot of concurrency issues with people changing things in place in mutable languages. So closure emphasizes immutability and so datomic like a database that takes that principle too.

[7:09] So final question what is one thing that people typically don’t know about closure that they should know about closure?

[7:17] One thing they typically don’t know about closure is sort of the development experience. So it’s extremely interactive and live. So a closure program you load the runtime and then as as you edit you’re updating the program live and so you can edit the function and immediately run it and edit the function immediately and it can immediately affect the current state. And so you’re not compiling and running compiling and running. you’re working with the whole program right there in the middle of execution in the middle of runtime. So most people don’t understand that about closure because it’s such a different experience than most languages where you have to compile run.

[8:03] My name is Victor. I’m a Java champion with almost 20 years of experience in Java and what I do in the Java ecosystem if I can say it like that I’m doing workshops for companies using mostly Java

[8:19] I’m I’m talking with groups about performance about architecture about the better best way to write the code in a maintainable way that kind of stuff. So I’m basically a freelance trainer so I don’t have a company. Many people ask me who’s your employer, my wife or me. And I’m just going from company to do workshops. That’s what I do every every day basically. Yeah.

[8:42] And what are you excited about or what do you find interesting right now?

[8:45] Right. Well, as we all know, there is a humongous scope. There’s a lot of stuff you need to learn, a lot of stuff you can spend your time understanding and digesting and going through your brains. And I’m really excited right now about how AI can help us in distilling and synthetizing humongous knowledge to our understanding giving us like bit by bit and that’s what I like AI most for synthetizing common sense wide u knowledge that you for some reason were let’s say left behind or aren’t really fully up to speed with. So this I find best about AI the ability to crunch the knowledge into digestible pieces that you can understand and you can set a context and you can tell I tell me what I need to know next based on my current. So I’m basically if you want to put it like a personal tutor that’s something that really interests me.

[9:46] I’m Justin Riyak. I’m the deputy CTO at DX. That’s getdx.com. And we have a platform for software engineering intelligence that blends qualitative and quantitative metrics to try to paint a picture of developer experience at a company and to try to make predictions about that company’s developer productivity capabilities.

[10:03] Excellent. And you’ve been involved in a number of organizations before this. how does what you’re doing now connect to what you’ve been doing before? So, it’s interesting when I was at Open Logic, what I realized was, you know, giving developers access to secure and highquality open- source really was something that contributed to the developer experience, which took me down the path to working for Gradel, and talking a lot about developer productivity engineering and that philosophy. that would take me for a little while to a company called Cortex, who ships an internal developer portal again to try to build a more comprehensive developer experience and improve productivity. And that would finally take me to where I am right now at DX, which is in my opinion you know the most efficient way to measure productivity in an organization.

[10:46] We have a very comprehensive view of what the developer experience looks like. We have a custom metric frameworks built from Dora space and then the DevX framework called the DX Core 4 as well as the developer experience index or the DXI. And so it’s always been building towards this attempt to measure and thereby improve developer productivity by way of improving developer experience.

[11:12] So what is developer productivity and what is developer experience and how do they relate to each other?

[11:17] Well developer productivity I think is measured by a series of different metrics. You have to be very careful not to overindex on a single metric or try to look at one metric for what productivity means. We need to look across multiple things like PR throughput certainly but also things like what we call the innovation ratio which is the amount of time that a developer can spend on new features versus maintenance and keeping the lights on. Developer experience is a cumulative set of different factors that sort of work together to form what a developer deals with in their day-to-day. So these are things like code maintainability, change confidence, documentation, and a lot of other drivers such as that. So DX has a dashboard or something similar or some tooling that shows all these different metrics.

[12:00] Yeah, we do have a great dashboard that is able to report on a lot of metrics that are gathered from different areas. So for instance, we gather qualitative data through highly effective developer surveys and by that I mean that we see 85 90 95% even more participation rates on these surveys. The surveys are engineered to look at developer experience as a systems problem. In other words, asking developers about how the platforms that they work with, the processes and the culture of the organization contribute to their own productivity. And we gather data qualitatively that way and then we correlate it with system metrics that are gathered from you know APIs and various tools that the developers might be using. We have over 40 connectors that we can gather that data from.

[12:43] So what is a key learning surprising key learning that users of DX gain from it? I think that you know right off the bat you gain a lot of insight into where the actual bottlenecks in developer experience are. and you know from there you find that maybe our velocity looked good according to a single metric but when we start comparing this AC across multiple metrics the we find that we’re hurting in other areas. usually when we take one of these surveys and we gather one of these first snapshots you find multiple areas of improvement in the organization. And of course, we recommend that you try to focus on what you think will be the most impactful and probably least effort to improve. But I think the biggest learning that people people think is that they come in thinking that they’re going to have, you know, already an optimal developer experience and they find right off the bat multiple areas of improvement.

[13:33] How to get started with VX?

[13:35] So, we have a survey template that you can grab from our website that goes through the exact same questions that are captured by our qualitative metrics gathering. each question maps to a different one of the DXI drivers that I mentioned.

[13:47] So it’s a it’s a questionnaire. It’s like a PDF or it’s like a document that you download.

[13:50] There’s a Google form template that you can use as well as a PDF that shows you each of the different questions and it shows you two different ways to calculate your DXI score based on the responses to those questions. We’ve also published benchmarks on our site that can tell you kind of where you are in terms of your competition. So if you have areas where you need to work on or improve, this survey template can help guide you there.

[14:15] So I’m kind of randomly quite interested in the documentation side of things with a long history as a technical writer and so on. So what’s the documentation angle that you mentioned earlier in the story?

[14:27] Well, documentation is obviously a huge part of developer experience, right? how accessible the documentation is, how easy it is for developers to get their hands on that documentation, how upto-date that documentation is, the modality of that documentation, are we talking about a knowledge base, are we talking about commenting, are we talking about both? So, we ask that question to developers whether the documentation that they need to use on a day-to-day basis is accessible and usable. And, you know, lo and behold, we find that that’s actually one of the worst drivers normally when we go through this with organizations. Most organizations aren’t getting it right. I think it’s a combination of it’s not coding, right?

[15:01] And so it’s not necessarily the work that engineers are most excited about, but it creates sort of a vicious circle because engineers absolutely need highquality documentation to do their jobs effectively, but it’s hard to get engineers motivated to create highquality documentation. Of course, AI is helping with this. we’re about to publish a big report on various developer experience drivers that can be improved by integrating agents in the SDLC and talking about different ways to integrate AI agents into usually CI. So you can imagine a workflow where a developer commits some code, an agent notices that, looks at the committed code and then updates the document or maintains the document, sends a draft to the engineer and the engineer just has to say okay that’s that’s pretty good.

[15:45] But that’s that’s kind of developer specific documentation like just describing the services and the APIs and so on which is different again to the kind of front end usage. If you have some product in production where users are going through your online application, your web forums or whatever you’ve created, how is that also incorporated into this? You know, end user guides basically end user documentation of your technology.

[16:12] We really are focused on the developer experience, right? And so for us, we’re talking about the documentation that’s necessary for developers. but that front end and user documentation absolutely is essential for your customers. That’s just not really what we’re measuring. Makes sense.

[16:25] Probably shouldn’t trust AI for that fully yet. So,

[16:28] Right.

[16:29] And Okay. So, you’ve got this Google doc and you’ve worked through this these questions. Now, how do you go transition from that to the DX technology itself?

[16:38] Yeah. So, surveys are hard, right? Getting engineers to take the survey, knowing where to integrate the prompts, knowing how often to remind engineers that the survey is available. so, the best thing to do is to start with this survey template we have like in Google forms. roll it out in your organization, understand where some of the pain points are, and then realize that we have a platform that can automate all of that for you, that can make the surveys more customizable, and then also can tie them to actual things that you should do in response to finding that some of these drivers may not be as healthy as they could be. So we have part of the product called Atlas which will actually recommend investments that you can make over the next 3 months or 6 months before you run another survey and grab more snapshot data to make sure that you’ve actually improved things. So you can think of the platform as taking the survey a step further and building like a whole framework for continuous improvement in the organization based on not just the qualitative data that comes from the survey but also the system

[17:36] Metrics that are coming from the platform that are coming from SCN incident response and all the other all the other elements of the platform.

[17:44] My name is Will Fury and I’m the head of engineering at Zenoder

[17:48] And Zenoder is one of the sponsors of Foojay. So, thank you very much. and it’s been great working with your team already and yeah, it’s it’s been wonderful.

[17:56] Yeah. Yeah, we’re delighted to be sponsors.

[17:59] It’s you know, JVM is a part of our stack as well and of course everything around that is super important for us.

[18:04] So, what does encoder do? What is it exactly?

[18:07] So, Zenoder is a an AI coding agent in your IDE. So, not just in your ID, we also have autonomous agents just in pre-release also. But primarily we’re trying to our goal is to you know accelerate coding accelerate the development life cycle for professional software engineers. So we have plugins for both VS code and all of the Jet Brains family. So IntelliJ, PyCharm, Webstone etc. because we realize that you know most of the professional engineers in the world are living in enterprises and an awful lot of them are Java based so based in IntelliJ as well. So we try and support all of the main tools that the developers use today without forcing to switch to something like VS Code.

[18:54] So how does Encoder distinguish itself from other similar technologies and plugins?

[18:58] Yeah, it’s a I mean it’s a crowded space at the moment. There’s a lot of going on again relative to the primary competitor who’s who’s doing the best in the space at the moment of course is cursor and cursor forces you to switch to VS Code. So that’s a again a primary defense between us and cursor will be that and you know in the space in general that’s again one area we really want to make sure we do you ensure the developers don’t have to switch out as they don’t need in addition to that we have strong integrations like tight integrations with the likes of Jira we’ve kind of got 20 plus integrations that mean you know from your developer flow again you can kind of work with the tools you have and the best tools for you and apart from that I said you know we’re targeting that you know that enterprise space. So our agent as well we believe this about code full completion and which is a standard if you like out of the box these days in all of these powerful IDs but our coding agent as well is one of the best out there if not the best out there. again there’s always on these benchmarks

[20:01] Like SW bench and all of this there’s always different implementations and versions coming out but of the main providers out there that have published you know in our competing in our space you know we’ve got we were top and I think we’re still very close to being top not just in the coding one in the street bench verified but also in street bench multimodal we were top for quite a while and the only reason we’re still not there is maybe we haven’t you know we wanted to get up there but now it’s not so important and for us to kind of stay publishing our results there.

[20:32] So from an engineering perspective and without revealing any secrets or whatever what does it entail to create an AI tool for developers?

[20:42] Yeah, it’s it’s an interesting one. It’s again because the whole developer life cycle we’re trying to figure out where is what’s going to be the most likely place that they will start using it because again there’s early adopters and there’s the late adopters and there’s the people in the middle.

[20:58] Mhm. and where will they find the most benefit from it? So again, co- completion that’s why everybody started there. Everybody started on co- completion because it’s just as part of your normal flow.

[21:07] In terms of the agent side of things then you know some people will use it heavier some people won’t but in ter in terms of trying to draw them into it and get initial benefits and get them hooked on it and find the benefit of it. So you know we have specific ones for unit tests. So again some people hate writing tests but we still have to do them. So we’ve got unit test agents where are basically dedicated to help you understand and write a specific unit test for them or we also have what we call zen tester which was recently released which is built on you know doing full end to end tests. So we’ll generate end to end test for your website you just point it to it and it will go and generate it based on what you provide the information you provided. So that’s the kind of hook into it and the idea is that people we want to tackle the full life cycle. So from the initial development also requirements like so we have kind of a requirements tool and we’ll be adding the planning kind of more advanced planning mode very soon as well. So to try and give opportunities to nudge into

[22:06] Different parts of your life cycle without saying you know without forcing you to you know only pick one let’s say but that’s the idea try and give give the right tools get the most out of it for where you find it in the b most beneficial for you

[22:21] And what would you personally say is the most interesting or unique feature of Zenoder

[22:28] Unique features these days are hard right everyone’s moving around so what I say is unique might not be unique tomorrow in everybody’s family. So, our multi-reo support, I don’t believe it is in any of our competitors. I think maybe Windsurf has something similar, but we’ve introduced multi-reo support recently, which is again super relevant for most organizations, even startups have many repositories unless you’re using a monor repo, you’ve got multiple repositories.

[22:56] And what does it do in that context? It allows so if you’re working in one project you can access the context the files in another project. So we basically you can index it and then the agent can access and reference those for example it can reference an internal library that you have

[23:11] Or it can access you know reference if you’re developing a front end and the API is on the back end obviously the open API spec

[23:19] But it can reference that in the other repository itself. So it’s super useful and the idea is again the difference with all of these is context. Context is king when it comes to making the agent useful whether it’s context that you give it context that you have we help structure or you know the context across your organization.

[23:38] Myself Kamesh so I’m a lead developer advocate at Snowflake. I’m based in Bangalore and a long long time Java guy like doing Java for more than 20 years though I don’t do a lot of Java right now but still love to do that but now it’s possible and now it’s all I do is all about data so what do you do with data

[24:00] So what excites you about data in the context of snowflake or anything that you’re working on what is interesting in the data space right now

[24:07] Yeah the super interesting thing right now is like how do you take AI on enterprise data that’s super interesting because there’s a lot of dimensions that it brings in security is one of the main factor like when people use AI the first thing comes to their mind is things safe

[24:23] That is one point you’d always land up and doing stuff and then like some of the things like coming from a strong application background and infrastructure so I always see that there are some things which the good things that we can take from the application world to the data world which still is missing in my opinion so which you start to apply there and the real real exciting stuff right now is like it’s MCP has everyone doing and asic AI so these two things like how do we make orchestrate tools to work together right if you can think back like it take me back to the old so ESB days like of Java like where we say hey how do you orchestrate stuff together to make it work like using messaging and other kind of platforms which really got up right I think I’m seeing a similar thing with AI I right now with Agentic AI doing all orchestration work while they run some datadriven workflow.

[25:17] Mhm.

[25:17] And you’re at this conference as a speaker. What is the specific topic that you’re speaking on?

[25:22] So yes, I’m I’m going to speak tomorrow for this conference on how to engineering a data catalog on Kubernetes. So that’s going to use Apache Polaris which is an open source project that Snowflake donated to Apache Foundation which is going to use iceberg this polaris and the basic idea for me is to how to get developers started using these technologies right from their laptops especially on peretus and you know like apache polaris is a java rest based API in fact quas I have a little legacy from carrying from red hat so quas is something which is close to my heart So you run transit pockets right now and then like it’s rest API trickles your catalog and I’m going to show people like it’s a open source thing that I did it’s available publicly on GitHub where people can take spin it up and then set the things up on their local laptops and start making Apache iceberg and then move to Snowflake whenever they want to.

[26:19] Hey so my name is Cedric Klyurn I’m an open source engineer and I’m a developer advocate at the best place Red Hat.

[26:26] Excellent. You know, I love working with Linux and Kubernetes and on this talk today, we were talking about Podman and how developers can deploy to Kubernetes easier.

[26:37] So, what excites you most at the moment? What do you find most interesting in what you’re doing and in the community right now?

[26:44] Oh gosh. oh, that’s a big one. That’s a big one. The theme for me this past year was open source AI. because we’ve always had this definition of what is open source software, right? It’s the code, it’s the, you know, you know, the release of, you know, the packages, all of this stuff that defines open source software, but open source AI, which you know, the keynote today is much different, right? Because

[27:09] You don’t understand, well, you can get the architecture and the weights, but you don’t actually have that training data. So, Red Hat’s kind of in a position where we work with the Linux Foundation on their openness for data and AI framework to define in three different phases how open a model is. and I think that’s something really interesting and it’s going to continue to get more important as more people adopt LLM, especially running their own.

[27:36] So, what’s one thing in this area that people don’t know about that they should know about?

[27:42] Oh gosh. well my biggest thing is that you can run your own AI right so this has always kind of been a possibility but it was very complex in the past few years you would use different frameworks like Owen and X and standards like that but now you can essentially in a CLI run a model from you know Meta or Google or any of these other organizations with one single line. So you can use O Lama, Llama C++, Rama Llama, which exists that one’s a real one. but it’s essentially that you know you don’t have to depend on third party providers. You can control your AI journey and I think that’s really cool that a lot of developers are still learning. Hey, I have this possibility and I don’t need a GPU as well to do so.

[28:30] Exciting times.

[28:31] Yeah, it’s exciting and thank you for having me.

[28:33] My name is Brian Famir. I work for Sneak a security company and the things that we do is focusing on how to develop within Java or within everything and then keep it secure and make it easy for developers.

[28:46] And what are you focusing on right now in terms of AI and security?

[28:52] Well, I think the things with AI and security is that people are looking at AI and if you look at talks at conferences, it’s how easy can you implement AI into a system and that’s super cool. However, there are also downsides and there are also things you need to worry about to basically if you give your LLM opportunities to add tools to it and give it autonomous powers to do certain things or connect it to an MCP or whatever like what could possibly go wrong if we give that to an LM. there are new attack factors when you use LLM in an application or AI models in an application and we have to focus on that as well.

[29:30] What are the new attack vectors in the context of AI? Oh, that’s a good question. if you look at new attack factors, actually it’s basically the same as we did before. But for instance, if you look at an LLM, you have a prompt, but as part of that prompt, we want to add, for instance, documentation or things from your database and we use rag retrieval augmented generation for that. If I’ve poisoned the rag, that means I manipulate the input and therefore I can make the LLM go out outside of the boundaries and do different things, act differently and maybe execute different functions. And I think the same happens for instance things like chat memory which is an artificial thing because most LLMs are stateless. So chat memory is basically a conversation and if I can manipulate that conversation as a prehand conversation and give that to the LLM I might get out of boundaries as as well or out of the scope of the LLM or the actual application and therefore let it execute different things that I normally don’t want to do. Now

[30:29] These are the kind of topics that you talk about at conferences at the moment.

[30:33] It’s it’s there are two things I talk about. it’s general security practices for your average developer and also like how does this work now in the age of implementing LLMs. so yes that’s that’s definitely a that’s definitely a thing I’m I’m researching and I’m talking about.

[30:48] So let’s leave with one tip on security and AI. What would be one tip that you’d like to pass on?

[30:54] I think one of the greatest tips is please look at your essentials. look at the essentials like input validation, basic authentication and also apply that to your LLMs as in if you give your LLM powers to execute functions. Make sure that the person who is requesting this is also the admin for instance and not just a random user. also look at if you do if you put a prompt into your LM, why not intercept it and sanitize it and or validate that this is within scope of what you want. So input validation and sanitization, we had to do this all the time in web forms, but not not any different. looking at role-based authentication things we need to do as well. have a proper architecture that thinks about what kind of data is accessible and what not. So

[31:47] Actually the old things but now basically renewed into with the sauce of LM basically. Yeah

[31:53] I’m Andre Citic I developed a post CSS and a few popular JavaScript tools but right now my main focus is a privacy and especially social media fatigue how we can like reshape the social media and our consuming how we consume information.

[32:12] So the topic of privacy is really relevant right now. what are your what are your observations your insights on this topic that you talk about at conferences?

[32:22] Huh? like I personally big fan of the local first and there is a huge local first community right now huge local first movement which they really try to fight against all this problem with privacy but unfortunately this is only one part and the second part just don’t care and we are like in this situation when like mostly people don’t care and only a small part of the people try to warn everybody about the very big danger of what we are going to like I personally think that main problem with privacy that like destroys the our industry because it’s create a very simple but very dangerous way of making money. I’m talking about the data brokers business model and as a result it’s it reshaped our whole industry and we like selling the soul to the devil and like for instance a lot of businesses are like right now we try to scale very fast and this is the only way to do it and it’s really relative how we like get information how get money if like we will go from this free model but like is we try to find investors, try to sell our product, try to sell user data

[33:35] To the model when users pay for ourselves, the whole industry will be better like our favorite tools will not be dead in some random moment etc. So like this my observation.

[33:46] So what are some solutions that we can work towards or what are some strategies to achieve that?

[33:52] I personally think that like the best thing which we can do like tomorrow is just to remove GDPR pop. There’s a lot of misconception about the law around the privacy and a lot of people think that like GDPR pop is mandatory by the law but it’s not. It’s technically a black design pattern to force user to agree to give the data. The law is pretty simple. Don’t track users. It’s a little more complex but still like the core idea. And in reality like it’s very easy to just stop tracking and you don’t need any pop-up. We need pop-up only because we want to continue our best practices. And instead of you know trying to change ourself we punish our users with the pop-ups until they agree to give their our data and like it’s very shady. Second things which like every developer could do just reduce number of the people who have access to the user data and for instance like don’t use Google things any CDN have full access and some of them they’re okay etc. But like if you will reduce the number of the services of CDNs of hostings providers it will be much safer. So just stop to using

[35:06] Google phones just copy both files onto your system. Stop using CDNs for JavaScript. try to like put everything to one cloud and it’s a very good step step but one is try to think about the local force architecture. This is really crazy moment right now. happening in industry like a local first really reshape how we making a software that’s why it’s impossible to refactor your software to the local first but for the new projects it’s very nice option and people like it it’s very not only good for the political reasons but also for the business it has a lot of benefits but one of them is your application becoming much faster and people pay money for much faster application

[35:50] So what is one tip that you would like to leave the viewers of this interview with one tip that they should think about.

[35:56] Remove the GPM popup, please.

[35:59] My name is Isabelle. I’m a senior backend developer. I’ve been working with the GBM for I think around six years now, but I have more experience in other languages previously. And I’m also in a program from Jet Brains H called Jet Brains Community Contributor because I create content around cotling for the back end and server side perspective.

[36:22] Perfect. And what are you doing at this conference? What topics do you find interesting?

[36:27] Well, I’ve been in a talk about cognitive load which is a really interesting matter h especially because well I work with microservices and teams that have a lot of microservices. So this is a an interesting topic and a thing to take into account when you h want to create more microservices or decide to go

[36:50] From a monolith to microservices or things like like that.

[36:54] And you mentioned that you’d listen to the Foojay podcast before.

[36:58] Yeah. Yeah. I actually did especially one of the episodes where you talk about how to land speaking gigs at conference

[37:12] And it was really interesting. I actually I’ve been speaking in conference here in Spain for several years and I did a couple of them in Europe and well this year actually I’m going to speak in another conference in October because I check a bit the tips that you gave and use them.

[37:30] Perfect. And what are the topics that you talk about at conferences?

[37:33] Well, it depends a lot but I talk a lot about testing. I talk a lot about a synchronous communication about well good practices for backend developers and how to the last topics was how to use AI for testing but doing it actually right not just de allow the AI to do everything by themsel because you can work with the agent and get a lot of more value

[38:03] So what is the right way to do testing with AI

[38:06] Well it’s kind of a really interesting question because for example I enjoy a lot working with the agent giving like a specific h commands like for example I don’t know create me this repository for testing or build this object model and things like really specific but I I’m the one in charge of the testing flow. I don’t allow the AI to do everything by itself because it kind of create bit of a mess sometimes. What is one tip that you would like to leave the watchers and listeners to this podcast with? One tip around something that you think people should know.

[38:52] Please do testing. Please, please, please do testing. That’s all.

[38:56] Please do testing.

[38:57] Thank you very much.

[38:58] Thank you.

[38:59] We have here Isabella and u tell me what are you doing here? Well, I came here to make connections since I’m a student and I’m going to my second year right now and I came here to like learn about more of the career I want to go to and kind of see like meeting new people in the industry.

[39:17] Okay. So, have you attended many sessions?

[39:21] Well, I tried to attend every one of every session I can here to learn more and I actually learned a lot here.

[39:28] And did you focus on any particular topic? Well, I focus a lot on AI because I’ve been really into AI and now especially it makes me more motivated to kind of be part of that.

[39:40] Interesting. So, could you just highlight some I don’t know ideas or talks that you have attended that have impacted you the most?

[39:48] Well, I think every talk has impacted me in some like way. There’s not really specific one.

[39:54] And apart from the talks, how do you which is your feedback from being in this conference? Apart from the talks,

[40:01] Well, I think it’s been really positive. I really like the people here and there’s a lot there’s a lot of new things I’ve learned and seen and I just like the vibe here. Like it’s it’s really nice to be here.

[40:13] My name is Reuben. I actually not from Barcelona. I come from Andura. I’m a primary software architect, a platform engineer. mostly do freelance work. and I’ve been a volunteer for the VCN for it’s the third year. Third year. Yeah.

[40:31] So you look like you joining the session but it’s a bit let’s say hard work or it’s so easy to be volunteer at the BCN.

[40:38] I would say it’s it’s stressful sometimes but it’s definitely worth it. just the camaraderie in the organization actually getting to see new faces and familiar faces as well. And you have sort of like this vibe where you know effectively it doesn’t feel you know corporate commercial it feels like people actually want to be here sponsors and attendees so it’s it’s really great

[41:02] Okay

[41:02] And have you enjoy any particular session for example anything that you would like to say oh this session I like it very much because I was in that room etc. anything that you would like to

[41:11] Yeah, several. The one of the ones I would like to actually talk about is private privacy first from the development basically local first architectures by Andre. I think it was really amazing to effectively see someone basically push the idea that we as software engineers are also accountable for the software we develop and who we develop it for. And I think it was a really nice take and also it’s a mixture of tech technical talk with non-technical basically widespread implications of software development and also developing a game engine on IntelliJ. I think it was one of the most creative and fun presentations and talks I’ve I’ve I’ve seen.

[41:58] Okay. Thank you very much. Any any particular thing that you would like to say about about the like to repeat something in the next edition or something? Well, I just feel like you know we encourage everyone to come in and u I think it’s definitely has quite a lot of space and ability to grow and I think we’re just scratching the surface. There are plenty of subject matters. We could do more tracks, right? So that’s a very good idea. Thank you for your help. Thank you for your words and I hope to see you next edition as well.

[42:34] I think so. I think so. I am Orasio Gonzalez and I work as developer relations at CleverCloud and European cloud provider.

[42:43] Excellent. And what does that do? What is Clever Cloud? Can you tell us about it?

[42:47] Clever Cloud is a cloud providers by developers for developers. We believe that today putting applications in production. It is way too complex. We are asking developers to know Docker, Kubernetes, Terapform infrastructure and lot of things that have nothing to do with development. So we want to go back to sample things get to prod when your application is ready you test it works you do your pull request it is accepted and every time you merge in the main branch we take the code we build it in the same way that you do in your laptop and we deploy it in micro virtual machines with the scalability you tell us with the add-ons database whatever you want and we give you back the observability to follow the production simply get to product.

[43:37] Perfect. And can you tell us one thing that you’re excited about or interested in or that you find really relevant at the moment?

[43:44] Well, that’s really complex because there are many things. Maybe I am following with interest how the artificial intelligence is changing the developer’s job for good and for bad. I am not telling that it is wonderful and everything is magic or that it is going to destroy the job but it is creating a earthquake and changing the way as we develop applications and I think

[44:17] Today experienced developers can leverage on EA but people who are beginning in the job new developers I am not sure of how artificial intelligence is going to change the job. I worry a bit. I am optimistic because I prefer to be optimistic, but I love to see how the world ecosystem is in a turning point now.

[44:46] Yeah. So now it’s the time for Jonathan not me the other Jonathan in the conference that he’s coming from Bon Bon

[44:55] Bon a conference I sorry sponsor in the conference and well first why Bonnage is here and which is the objective that Bonnage was searching for coming here.

[45:07] Yeah so Bonnage is here because we are promoting our APIs. We do u communication APIs and network APIs and we are launching these networks APIs in Spain. That’s why we are here and yeah it’s basically APIs to prevent fraud. really interesting for banking and everything that requires an extra security layer. So we can we are provide that and we are promoting that here. Yeah.

[45:31] Okay. But I was wondering so you can promote things I don’t know using a newsletter ads Google ads or something but you are here so you are paying for being here among hundreds of developers. Why? Which is the value that you think that the conference is providing that other means are not providing? Yeah, exactly what you said like most of people here are developers and we want developers to know us to have it have us in mind when they are developing something and this provides that just a great you know audience of developers that come here to learn to know about all the companies to even get hired so that’s why we are here because we want the developers to know us and this is just a great place for that.

[46:18] Okay, thank you. Thank you. were you in the party yesterday?

[46:22] No, unfortunately I wasn’t.

[46:24] Oh, so bad. So, you missed the karaoke?

[46:28] Yes.

[46:29] But you are coming next year as a sponsor again and you won’t miss the karaoke party thing.

[46:36] Yeah, that’s mandatory. Next year if we want to come back. He told us that yeah, we need to be there in the karaoke. We cannot miss that. That’s a requirement next year. So finally any advice suggestion to that you can give to companies that are not here and that they are losing all the value being in a conference like this one.

[46:57] I mean it’s a it’s a great opportunity to for people to know you for you to get exposed to you know not only locally but globally as well. especially well if you have you know if you’re interested in hiring people in Barcelona this is a great space for that as well just to get exposure and there’s a lot of comp good companies here in Barcelona as well that you know so I mean it’s you cannot miss it next year it’s really great the audience is great it’s mostly developers with a great level great workshops great talks so there’s there’s nothing wrong about it it’s just everything is great

[47:36] My name is Ru Mahano I’m the co of virtual solutions and I’m the principal engineer on the box lang language.

[47:42] Yeah. And I’m Christo Alles. I’m also part of the box llang team and we are basically spreading the word about our new baby which is Boxlang.

[47:51] So what is Boxlang?

[47:53] Oh rough question. All right. So Boxlan is a new dynamic JVM language

[47:58] Compiles to Java byte code is dynamic. It’s modular and it’s meant to be productive. Mhm.

[48:03] So it’s I think the last dynamic language was more than 12 14 years ago.

[48:08] So we really wanted to revive the dynamic space in the JVM and we’ve been tinkering with this idea for a long long time. we didn’t just wake up one day and say hey I want to build a new language. we’re a little crazy but not that crazy. but mostly it was because we’ve been building languages building libraries and frameworks for over 19 years. And so I launched the first initial MBC framework for the coof fusion language in 2006. It’s become kind of the standard in cool fusion and you know all these 19 years we’ve been building all of these libraries and ecosystem and tooling and this was the last part. This was like okay we need to push the envelope with all the stuff that we’re building and we weren’t seeing anything that would allow us to do that right. So we kind of took this decision of say what would a new language be like right in 2023 at the time when we started designing it right and for the future right how would it evolve and that’s how we started basically is we started from scratch we didn’t fork anything we didn’t start from anything we really did not want

[49:10] Any baggage at all from other languages we wanted to make sure that we started fresh with what it would look like for the next 20 30 years right and that was the real challenge right how do you start from zero but it was the best decision we were able to basically go from nothing to a stable release in 20 months.

[49:27] We did over 30 something beta releases and alpha releases, release candidates, and migrating all of our internal apps into box as well. so I can talk a little bit more of the intricacies of Boxlang, but that’s a little bit of a history compacted.

[49:43] Perfect. So why is Boxlang called Boxlang?

[49:47] Well, basically our we believe in open source. We have been working in the open source environment for almost 20 years and all our open source products are under the last name of Box. We are the creators and also the owners of callbox and the command box and a lot of well-known products that use box as a as last name. So for us it was the natural evolution of box language.

[50:17] Mhm. So what are some reasons why somebody would want to use box lang like very specific reasons why I want to start using box lang today.

[50:25] Sure. Well you know languages have evolved through time but they all have become semantically the same right. So semantically the language is very similar to other languages. This is normal right but our niche is productivity right we come from the productivity space in co fusion which was released in 2000. It was a great productivity platform and we wanted to continue with that ethos of productivity right so you will see a lot of framework capabilities into the language itself and that’s what we wanted and when we first designed box line we designed on two aspects multiper and multi-runtime we wanted to make sure that we could run multiple languages in the runtime and we wanted to also basically execute everywhere right so we started first with a chip the lowest common denominator where the JVM could and bite code can run. That’s where we started and we developed the entire language first at the core level and the entire language is about 7 megabytes. It is extremely tiny and right now it’s still big because we still have a lot of debugging aspects around it that are going to disappear eventually. So it’ll

[51:28] Probably go even down,

[51:30] Right? and fast, right? We can even start, you know, repples and servers and everything extremely fast, probably faster than anything in the JVM. And so that was our first kind of focus. And then we did multiper right we have a history of 19 years of the coefficient language that we wanted to bring on day zero and we did right we designed the parser to be able to parse different languages and all transpile into our blast which is our aspect syntax tree for boxite right so everything basically has a translation layer that goes into the blast and then the blast goes to bite code right so that was a big focus of ours to bring basically three ecosystem from day zero java confusion and box sign from day zero So this was very unique, very niche and the entire language is also event- driven.

[52:15] Okay.

[52:16] Right. And this was also important because we saw the huge benefits of this in our frameworks is that everything that works through interceptors and observer patterns was so easily decoupled, right? That we could grow it. So that’s why our core is extremely tiny and then basically there are events being emitted everywhere when a class compiles, when a class is loaded, when a class is constructed, when a function prefunction, post function. So you can listen to all of these things to basically do tooling and extend the language into more framework capabilities. So that is extremely unique and

[52:47] So what does a hello world look like? I’m setting up my very first box application the simplest one. How does it look?

[52:55] Sure. You create a file called hello.bxs bxs which is scripting and you put print line hello and you’re done.

[53:02] That’s it.

[53:03] There there’s no encapsulation class around it. no methods. yeah. Well, that’s that is the method. Yeah. Yeah.

[53:11] That’s the simplest. Print line and

[53:13] You’re done.

[53:14] Okay. And what would be the next logical thing after that?

[53:16] Okay. We can do a class. So, we can do class and then create a function called main

[53:20] And then say print line hello.

[53:22] Mhm.

[53:22] And that’s it.

[53:23] By convention. So, we do a lot of convention by configuration. Cobbox was the first conventions by configuration framework in the CFML space. And so, we have a lot of conventions into the language itself. So by convention it will find the main function it will execute it

[53:38] And that’s it.

[53:39] So is it integration with Java for example that you could call from the one to the other?

[53:44] Yes. So

[53:44] And in which context would you do that?

[53:46] So we really wanted to make sure that this worked very well right because we’ve always wanted this in our tooling. So you can actually use Java and Box natively or you can use Java in boxing natively. So you can go either way right. So if I am a building a boxing application I can import Java classes just as normal use gradal use Maven right by convention you could put it into a lib folder and everything will become alive in the class loader hierarchy that we have in box line and then you just say import like normal Java

[54:14] We’ll take care of the reflection not not a reflection sorry the invoke dynamic because there’s no reflection in box line we’ll do everything invoke dynamic for you right we’ll make sure that we can address all those Java classes right and do all the casting for you right and then from the Java side you can import Moxlang as a library, right? they’re deployed in Maven, so you can just bring in whatever you need from Boxlang. And let’s say you need an enterprise cache engine. Well, Boxon has an enterprise cache engine. So you can basically import that into your project in Maven or Gradle and then basically just leverage the Boxon cache.

[54:47] Okay. So what are some aside from that, what are some use cases why a Java developer would do this?

[54:51] So productivity, that’s the first one, right? we’re looking for people to be very very productive. We’re also professional open source. So that means that we’ll support you along the way. There’s a professional open source version which is free and open source, but you can also get plans that you can work with us and we can work with you, right? And we built tons of modules. We already have like 60 modules that we built that are curated by the team and certified by the team. So you don’t have to go scouring the internet for, you know, libraries that you want to work with, right? So there’s things like PDF manipulation that is already part of the team that we’ve developed, right? Image manipulation, FTP connections, SFTP connections, right?

[55:27] Disk abstractions, right? The password securities. So, we’ve we’ve really focused on all these 19 years of all the tools that we’ve built and we want to build them as modules. So, basically the community can consume them and they’re just part of the language.

[55:40] Makes sense.

[55:41] You just use those that you need. I mean, you you’re not forced to take all or nothing.

[55:48] Yeah. You’re basically building them in modules. Makes

[55:51] Sense. So, so what does the documentation and the community and forums and mailing lesson things like that look like?

[55:57] Sure.

[55:58] So, documentation is a sweet deer thing of mine. that’s actually how Cobbox became very famous is because I decided from day one that if it was going to be professional, open source, it had to be documented. So, I’m a weirdo in that space where they call me the documentation weirdo. But it’s so important. So, our book is already written. It was actually written before the code was were done. Nice. So we believe a lot in documentationdriven design and we basically have an entire book at box9.orgbooks.com which is available. It is a live book. It is available and we’ll basically create you know physical printed books out of that but we’re very finicky and very nitpicky about it. actually every single function has to be documented by the developer or by AI, right? But we’re big believers in documentation and we’ve created a lot of tooling just so developers can find what they need. So that’s super important for us. And then as far as community goes, we have all the community I would say avenues that we’ve built for the 19 years that are available and they basically just have a new section which

[56:59] Is boxing,

[57:00] Right? So all the developers that have been using our products, you know, in companies like NASA, Air Force, Navy, you know, a lot of universities in the US, European Commission, General Electric, right? So all these companies that have worked with some capacity with our products now are just part of the box family basically.

[57:18] I think it’s also important to mention that we have an e-learning platform

[57:23] Which is called box academy with all those videos. I mean a lot of people learns easily without with videos and not reading nowadays especially younger people. And there’s a lot of material about how to get started using box lang with in that platform. So that would be important for anybody that wants to join the movement.

[57:49] Perfect. Perfect. So final question, where do you see box going next? You know, what’s on the road map? What are the features that are coming?

[57:57] The things that we believe I mean we have been part of a big community for a lot of years. So we believe that those new features and the road map should be driven by the community. So we are very open and our aim is to respond really fast to community requests. So we have a plan obviously we have a plan which is in our website but we are very flexible and we are very open to new suggestions.

[58:28] That’s perfect.

[58:29] Very important for us.

[58:30] Thank you. Thank you.

[58:31] A little sneak peek into what we’re building. So, we’re actually going to be releasing our new desktop runtime pretty soon, hopefully this month. So, we’re working on a JavaFX capabilities for box lines to do native Java and box line desktop applications. So, we built an entire framework around it to make it extremely easy for developers to create desktop apps because we also have a scripting lang templating language. So, that will allow you to build the UIs even reactively with box line.

[58:58] So, that’s in the horizon for this month. and also Azure functions that have been a huge community request. so we had AWS Lambda core in the runtime and now we have the Azure functions runtime that we’re working on. So those are things that are in the small horizon right now in short term

[59:16] And the pipeline. Yeah.

[59:16] And the pipeline and then I would say after that we’re also working on a groovy parser. So we did a lot of Groovy loved groovy but obviously has dwindled throughout the years and we want to do that the same as we did with CFML. So we’re building the Groovy parser and the Groovy compatibility module. So basically you can basically have your Groovy apps or Gra apps and bring them into box line without touching source code and run in box.

[59:42] Yeah.

[59:42] I’m Miguel. I work as a sales engineer in Dinatries for like Iberia and like I mostly work on like deals where we try to like bring value through like an observability platform where they can like let’s say it’s like the evolution of monitoring traditional monitoring we bring observability into into our customers like IT deployments. So that’s basically what we do. And what do they typically find surprisingly amazing about Dino Trace? So actually I think that most of the companies right now are still stuck in this monitoring mindset where they just use like metrics to get alerted like Prometheus open source things where they just see the data and then they act on what we want want is to bring up that opportunity that information in order to automate those like remediations or actions or even information from like if you have a really high response time on your application. How do you deal with that? how do you avoid these big war rooms in where people get together to discuss about like incidents. So we try to bring value into into that use case where they can

[1:00:52] Like have a whole observability real time of their business through an observability platform. It’s quite quite interesting. I mean it they’re really happy. So we’re trying to we’re here talking more with developers which are the end are the owners of the applications the ones who make it so they can like get value out of seeing what they do in real time like my code is my code is responding like way too slow because like a database is not answering back what I want so how do I get alerted on that and like being able to automate all that process through through like terraform integration obserability as a code all those features are what we’re trying to tell the people here in is in this event.

[1:01:31] Mhm.

[1:01:32] And AI is of course the big topic of the moment. So how does that interact with what you’re doing?

[1:01:37] So quite quite a nice question, right? Like the thing that we’re all talking about lately is a gentic AI. So being able to provide a center of like communication between the different applications. Detrace plays a critical part in this in this model because it gives you the information on where to act. So imagine that you have an incident or a problem or your application has gone down through the agentic AI that agentic AI is going to be able not only to grab information from the observability platform ders in this case. So that AI is going to be able to coordinate the different resources within your environment in order to solve that issue. So you need for example to restart the database then you have to inform the appropriate teams and then you have to create and maybe close a ticket on your J environment. So all those actions thanks to aentic AI where you connect your different endpoints within your company are going to act like independently which is pretty awesome actually saving a lot of time.

[1:02:38] Perfect. And final question what are you excited about or interested in or enthusiastic about right now in the technology space?

[1:02:45] Well I mean what we talk about is quite much the most amazing things because it changes the whole world the whole thing like the whole way we work like all this AI. I mean the other day I was reading that the most trendy programming language right now it’s English because it’s that’s true because like it it’s a huge mindset like change of mindset that we are facing right now and doubly I mean there’s it’s going to impact not only observability but many other like scenarios in the IT world. So I’m looking forward to see how is this aentic AI going to is going to arrive in these customers like in not more than one year from now because this is everything going so fast. So I think that’s definitely the main thing that we should take care of like right now. Okay. So now we have Sylvia from Dow Jones and she’s going to talk about well which is her experience with the BCN. Tell me why are you here?

[1:03:43] We’re here because we heard that this event would have like a great visibility and impact in the Barcelona community. we are Dian. So Dian provides journalism, data intelligence and we have been around for many years but in Barcelona we are not that well known. So we essentially wanted to have a little bit more of visibility in the technology community because we’re expanding the technology hub and it’s so great to connect with the local talent in tech in Barcelona.

[1:04:11] Okay. So what would be your advice for those companies that not that are not that lucky and they are not here? So how is your experience saying yeah I didn’t know that this would be so great and it is.

[1:04:26] I would advise them that it’s it’s an amazing opportunity to connect really to talk to people that is great not just local because we have we talk to people from Germany from people from different countries in Europe that came just for this and it’s also super nice to connect with other technology companies that we didn’t have relationship with. so you create a lot of synergies in a super nice environment. People is super open to come and come to your son and get to know your company what they’re doing. So this has been a super nice experience so far for us

[1:04:58] And I’ I’ve heard that you also have a speaker talking in the conference right so can you talk a bit about it?

[1:05:06] Yes we were lucky enough to have one of the speakers in data science being one of the key speakers. he’s going to talk about data science, what we do at Dawian. So it’s a nice opportunity as well again to get people know what we’re doing at Dawian in technology and also a great opportunity for Gabriel for our speaker to yeah again to connect with the community and share his knowledge.

[1:05:30] Okay, sounds sounds amazing. So just to finish I don’t know can you give a product company pitch for people that well want to know more about Dow Jones?

[1:05:40] Of course. Well, at Dian, we provide trusted information, trusted journalism and data intelligence for companies. Probably people knows us for one of the most popular products which is the Wall Street Journal, the newspapers. But we have other like Baron, Market Watch, we have Factiva, which is a giant database of news that with billions of articles that companies use to do research and make better decisions in the end. So yeah, we have 5,000 employees worldwide. We have offices in many countries. and we have a tech hobby in Barcelona that is expanding. So that’s it.

[1:06:15] I am Rio Sam. I am originally from India, now settled in Netherlands. I’m here in Barcelona for the conference now. And I work as a Java chapter lead in ABRO Bank. yeah, I so I work partially as a Java developer and a manager for a team of eight Java developers. So that’s what I do.

[1:06:32] And you’re also an international speaker and you’re here to deliver a talk. And what’s your talk about? Yeah,

[1:06:38] I’m here talking talking about framework agnostic development with Java. So, it’s about how you can develop in plain Java without a lot of framework. So, we need to reduce the it’s more about how to reduce the framework usage and yeah how to deal with that.

[1:06:53] It’s kind of sounds like NMB the kind of lean lean coding type approach without frameworks without stuff just pure Java.

[1:07:02] Yeah, not exactly. We need framework for some parts. Not exactly. It’s not vanilla Java style. It’s not advocating to only use Java. Use use framework but use it wisely. Limited usage. You need don’t need frameworks everywhere. So your core business you don’t need framework frameworks there. You can use pure Java Jakata there and keep frameworks to a minimal effort so that if you want to switch you can easily switch only that deployment module of framework. So it’s that’s approach I try to do because completingly ignoring a framework will be stupid. Yeah.

[1:07:36] So what are some concrete examples or use cases that you talk about?

[1:07:40] Yeah. for example if you want to connect with database it’s always easy to connect with a string JPA or something instead of doing that use Jakata persistence and keep spring out of that module and only use as a deployment module where you connect your beans. That’s it. So that’s that’s one example simply you can do. So your database module will be completely free with spring spring or Quarkus whatever it is. Also if you change your core business logic to a module where there is no spring or spring needed you just need to initiate it with new that’s also work. So you can completely remove all the frameworks from your core business module which only do your like for example your logic what your application is doing. Yeah that’s

[1:08:27] Perfect.

[1:08:28] So aside from this topic what is interesting to you at the moment? What are the topics that you’re thinking about and what makes you enthusiastic about the developer ecosystem? I always am enthusiastic about culture about how people work outside of developer comfort. So it’s it’s easy to technical things but then we also need to understand a little bit about culture, how people work and how people are wired. Actually all are all of us are different and we all are capable of doing things but we it’s always difficult to work together. So that’s always my favorite topic to work on because world is populated with lot of cultures. So it’s it’s a never ending subject.

[1:09:09] So in your Netherlands content context that you’re in, what’s an example from there versus a different culture that is very specific.

[1:09:16] Yeah. something I found here itself in Dutch Dutch timing and Spanish timing is totally different here. So 6:30 in Netherlands I try to be there at yeah 6:30 exactly or maybe 1 minute earlier. Here it even if I want to be there 6:30 it will be 6:45 or 650. I experienced it here in two days. like last two days. Hello.

[1:09:38] Hello.

[1:09:39] Well we I’m Nacho.

[1:09:41] I’m Jonathan.

[1:09:43] We are the core organizers and founders of the WBCN. hello everybody. what we can say?

[1:09:50] Well this is the second day of the conference. We are tired. I don’t have boys. Maybe it’s because it’s cold in Barcelona or maybe it’s because we had a karaoke yesterday. Who knows? probably the party had some let’s say reason of

[1:10:06] Exactly exactly

[1:10:08] Yeah we are we had let’s say many sessions already the worst happened two days ago and what else there’s a lot of content that happen is still happening here today a lot of AI a lot of Java what else you’ve saw Jonathan

[1:10:22] Yeah Kubernetes Python leadership

[1:10:26] Y

[1:10:26] JBM stuff who knows maybe related to langford chain you know

[1:10:31] But who knows yeah it’s been it’s been full today is very very a lot of a lot of passion a lot of energy a lot of yeah conversations between all the attendees and speakers

[1:10:46] And all of the sponsors I think they are super let’s say eager to come here and to see how the things are working to connect with others to see how the community the developer community in Barcelona is let’s say still growing and as well we have nonprofit organizations with helping other let’s say people to work with the technology etc. So what we can say more we invite the people know to come to Barcelona next year to the bien.

[1:11:13] Yeah. So July or maybe June Barcelona

[1:11:17] Summer pa sangria beach technology people well so come come again.

[1:11:25] It looks like a good plan.

[1:11:26] Yeah looks like a good plan. We will be here.

[1:11:28] Yeah. See you. We hope to see you tomorrow. What? Tomorrow? Next year.

[1:11:32] Yeah. That’s it for defbcn part one. The next part is a bit special. Some of the interviews were done in Spanish and these are combined in part two next week. Thanks for listening. Please subscribe to the Foojay podcast in your app or on YouTube. Keep an eye on Foojay.io for future articles and podcasts about development and everything related to the Java world. See you. Be the friends of OpenJDK.