#211 - Back to the Future: Lessons from My 42-Year Career in Tech - Paula Paul

[00:01:39] Introduction

Henry Suryawirawan: Hi everyone. Welcome back to the Tech Lead Journal podcast. Today, I’m very excited. It’s very rare for me to have a chance to talk to somebody who has like multi years of experience, 40 plus years tech career and software engineering career. So I think today we are going to learn a lot from someone who have been in the industry for that long. And maybe we learn a little bit tips and tricks how to be successful in the tech industry. So welcome Paula Paul to the podcast today.

Paula Paul: Thank you. Thank you, Henry. It’s wonderful to be here.

[00:02:10] Career Turning Point

Henry Suryawirawan: Right. Paula, looking at your profile, I know there are so many things that we can dig into but tell us a little bit more about yourself. Maybe if you can share a little bit of turning points that you think will be great for us to learn from.

Paula Paul: Sure. And I think it is interesting to note that I’ve been a software engineer for many, many years. I started my career in the early 1980s at IBM as a software engineer where compiling code for mainframes was a batch job. So, you know, the transition from waiting an hour to see if your code compiled to everything now is very instant has been a really interesting perspective. And it always leads me to one of my favorite sayings is now the technology is the easy part. We live in an abundance of compute and storage and network reach that wasn’t available at the beginning of my career, which is, you know, in some ways that’s a long time and in other ways, it’s not a long time. So I do expect that technology as an industry is always about change. And the thing that I find very interesting is, the thing that people struggle with the most, is change. So I work a lot with people on what is your relationship to technology, which often is about uncovering resistance to change or fear of change. And that’s much harder than writing the code these days, so…

Henry Suryawirawan: Right. As someone who have been around that long, right, so as you just shared just now. So you started your career by doing these kind of IBM mainframes, you know, the punch cards. I really don’t know what it looks like. And now to the era of, you know, like cloud computing, mobile computing, and now even AI, right?

[00:05:59] How to Approach AI and Rapid Technology Changes

Henry Suryawirawan: So if you can share, maybe looking back with the pace of technology changes, what do you actually see and how do you actually reflect back to your early career? What can we software engineer appreciate a lot these days?

Paula Paul: One of the things that I encourage people to do is really understand the meaning of the words that they’re saying. We have a lot of terms that are very overloaded in technology and AI has become that way now. When people are a little afraid of AI, or is it going to take my job? I always go back to what are we using it for?

I use a note taker, large language model text summarization. So you could say it’s an AI note taker. And I find that when I’m in a call or in a meeting that has been saving me a lot of time. So if we just talk about what we’re using large language models to do, it’s much less. scary and much more approachable. And we’ve been asking computers to generate outcomes for us for a long time. Even before my career, we’ve asked computers to generate outcomes. And I think the terminology around AI sometimes makes it less approachable than it needs to be.

Henry Suryawirawan: Right. So I think that’s pretty good call out, right? So we like to always refer multiple things using the same term and it gets overloaded. And, you know, sometimes people even confuse that to some other meaning, right? So I think thanks for sharing that.

[00:07:27] Long Feedback Loop in Software Development

Henry Suryawirawan: Another thing that I picked from what you shared earlier, right? You mentioned in the very, very early of your career, compiling code takes a lot of hours, right? So the feedback loop seems to be really, really long. But I mean, if you look at it these days, I still think the feedback loop of software development cycle still considered long. So tell us why, even though the technology has changed so much, you know, a lot of compute, a lot of power and all that, still the feedback loop cycle for software development is really long.

Paula Paul: That’s a great question. And it goes to, what do we mean by the feedback loop? So the feedback loop of a compile is really to tell the software engineer, is there a syntax error in the code? It might not necessarily tell you if there’s a logic error. So there’s a feedback loop of syntax and then there’s a longer, I think, there’s always been a longer feedback loop for logic or did you actually solve the right problem? So in my early career, I coded in Assembly language and then a logic language that was proprietary to IBM called PL/S. We had to drop into Assembly language in order to do the I/O, because that made the code portable. Um, so, So even though we were considering portability across different IBM operating systems, and we had a formal estimation cycle, and we typically spent only 25 percent of the product delivery cycle actually writing, compiling, and testing our code. So we did spend a lot more time upfront, really understanding the problem as software engineers.

I worked in the manufacturing area in the early days of CAD or Computer Aided Design adoption. We really had to understand the problem well. And then we spent quite a lot of time in design and architecture before we would write code, doing a lot of work on paper. And then when we felt very confident and could talk about what we were going to build and how we were going to build it, then we would write the code. I had eight people on one of my teams. And because your code is, you know, you punch a compile and maybe you wait an hour, you’re communicating with your teammates much more regularly. So I do think that while a lot of things have changed, some things have not.

And Agile approaches coming along are not so much changing the fact that we can get early feedback on syntax, longer feedback on logic and solution correctness. But we’re breaking the solution into smaller pieces. So I would say that that’s been an improvement, but it’s always been a challenge to understand if you’re really building the right thing and the IDEs and rapid feedback from a instant compile of your code isn’t necessarily telling you you’re building the right thing?

[00:10:35] Importance of Building the Right Things

Henry Suryawirawan: Right. Building the right thing, it seems like the theme from many, many people I talk to, right? So because these days you can really code really fast, right? So you have a lot of technologies available out there, but seems like building the right thing, you know, building the things, right, also seems still like the problem that we have to solve, right?

And you mentioned that you spend just 25 percent of your time actually just to do the coding, the compiling and all that. And you spend quite a number of time doing the planning and maybe design upfront. Is it still the case that you think some kind of planning are still necessary while people are moving into more like, okay, let’s do Agile. We don’t need upfront planning. We don’t need upfront design and all that. So what’s your take about all these, you know, spending time to actually thinking what problems you’re trying to solve and building things right?

Paula Paul: I am a big fan of having a strategy. I think it’s important to test hypotheses very early, which doesn’t involve code. So if I want to say I’d like to build a new home delivery app, having a hypothesis about that and then being able to reach people that you think might be good customers and testing your hypothesis. I think that’s great.

There’s an old book called The Mythical Man-Month written by a person named Frederick Brooks. And he was somebody who worked at IBM on operating systems even before my time, certainly. And he had a vision and an essay called There’s No Silver Bullet, meaning there’s no kind of brass ring that’s going to make us 10 times faster at doing what we’re doing. He actually said that one of the things he had great hope for was rapid prototyping. So rapid prototyping is awesome, because it’s a way to show people what you’re thinking.

I do think a lot of people are very visual thinkers. And if you want to test a hypothesis with them, you sort of have to show them what you’re thinking. And I would say don’t invest all of the software development time and effort in testing for the finished product. But find a way to demonstrate what you’re thinking before you make those investments.

So that’s… I like that shift in the industry. I did work at IBM. The process I described was really what we would call today a pure waterfall process. And it worked well in the manufacturing area, because we did need things to be right by the time they were put into production. So there’s no kind of crash and burn phase. um, but you know, there’s… There’s different approaches to delivering products and to delivering software. And I’m very much a fan of testing hypothesis very early in the process.

Henry Suryawirawan: Yeah, especially these days where people build a lot of new things, right? New startups, new ideas, new innovation, new disruption. Probably it’s less difficult to actually plan. I think what you mentioned, doing the testing approach, right? Experiment, hypothesis, rollout prototypes, maybe MVP what people call as well. So I think that will be still the go to approach for software development these days.

[00:13:35] The Fear of AI and Technology Changes

Henry Suryawirawan: As someone who may have gone through a lot of, you know, cool trends, fads, and maybe great technology that some people say will change the software engineering, the way software is being delivered, right? So be it, for example, the internet era, the web, you know, mobile, cloud computing, and now AI. Some people are worried these days, right, because some of these technologies seem so powerful that we are scared of our job. So as someone who have gone through all these cycles, up and downs, maybe some learnings that you can share with us, how should we approach this?

Paula Paul: Sure. I think it is just the nature of technology that it frightens people, because it represents change in what we’re doing or how we’re doing it or how we experience things. And the first fear that people have is my livelihood. Is this going to take away my livelihood? And I’m still in the industry, so I am an existence proof. And so I do think even back in my first roles at IBM, it was during the adoption of computer aided design software. And I had to show people who had been working on drafting boards, how to use CAD systems. And their concern was, you know, is this going to take my job? And I would say no, but if you like this kind of work in designing manufacturing parts or whatnot, you will have an opportunity to learn new ways of doing things. Some people like the opportunity to learn new ways of doing things. And some people really like the comfort of the way that they had been doing things.

So that’s the challenge. And I don’t think the technology takes the jobs away. I think the people make choices. So it really is a choice.

And I, I’m very lucky because I’ve always been insatiably curious to learn new things. And if you take it all the way through AI, I really do not think AI is going to destroy jobs. I think it’s opening up tremendous opportunities to learn some new tools. And no matter what, a large language model or other model produces as an outcome, you still need people to evaluate, is that what we really intended or does that thing that was produced have value? So I don’t really see jobs being destroyed. I see opportunities being created and people being given choices, which I think is hard. The hard part is the choice.

Henry Suryawirawan: Yeah, I think it’s very encouraging to hear from you, right? So all of us here could probably learn a thing or two just now what you mentioned, right? So first of all, change is always there, especially with technology. It could be really, really fast, really, really rapid, right? And at the end of the day, it’s the matter of choice. How do you want to go on board with the new technology? Or do you resist and, you know, become obsolete? So I think it’s also coming back to us whether we are curious to learn something that is new. And sometimes, yeah, we have to change our role, right? Change our skill set. But I think that is given in the technology world.

[00:16:46] Timeless Tech Career Advice

Henry Suryawirawan: For people who are in the software development or software engineering career, right? Is there any kind of timeless advice that you want to give for us so that we can focus on the, maybe the important stuff rather than following just the new trends all the time?

Paula Paul: I have children. And when I was in college, my father told me just work hard and you’ll get ahead. As I’ve gone through my career, I know that was the correct first half of the equation. And I’ve added more to the end of the equation that I give to my children. I say, work hard at something you love and then find people who appreciate you and that you will enjoy learning from.

So I mentioned earlier that we live in an abundance of technology now. Compared to when I started, we have an abundance of compute, we have abundance of networking reach that is, you know, the internet, and it’s just amazing. Abundance of storage. Our whole lives could be recorded and kept in the cloud, which is it would have been unheard of when I first started. So you have an abundance of opportunity and sometimes now the hardest part is choosing what you’d like to learn more about, because there are so many opportunities.

So my advice is find something that sparks your interest. That you’re like, oh, that really gets me excited. Don’t build a large language model because you think you have to, because everybody tells you it’s the only way forward. Find something that you enjoy. Like I enjoy music. Can I apply some of my skills to music? Can I learn some new skill that I could apply to music? So if you narrow your opportunities to the things that interest you and spark joy, I think you’ll be much more successful and you’ll find it much easier to spend your time working hard at it. Don’t worry if you’re not out building large language models right now, but find a way to connect yourself through something you enjoy.

Henry Suryawirawan: I like the second part of equation that you just mentioned, right? So find the work or the people that really appreciate you, right? So sometimes we work at someplace where, you know, it’s more like a grunt work, not really we get appreciation, right? So I think it’s really important not to just work hard and earn, you know, but also find something that appreciates back on what you’re doing, right, so I think that’s really key. And yeah, choosing the things that matter to you. I think still the hard part, because again, like you said, there’s so many abundance. People maybe talk about different programming languages, different platforms, you know, different whatever that is, right? Kubernetes itself has a lot of ecosystem in itself.

[00:19:34] Navigating Career Decisions

Henry Suryawirawan: So I know that you have spent a lot of your time as well, you know, being a polyglot developers, trying multiple roles as well, not just software development, maybe in the product management, leadership and all that. How can we choose wisely? Because still, I think it’s a challenge and it is even more harder as I see going forward, right? Because there are so many choices, even more.

Paula Paul: Yeah, another really great question. I like to think of things in terms of architecture and strategy for your life and your career. And, you know, the definition of architecture, Grady Booch, is the significant design decisions that you make where the significant design decisions are based on the ability to change your mind, basically. So as you go through life and you have all these options, things you could invest your time in to learn or study. It’s kind of to me like going to a big restaurant where they have a big display case full of all the desserts, and I want them all. So, you know, you can’t, in any one visit have all the desserts. But you could say, ooh, this chocolate cake, I’m going to try the chocolate cake.

What’s the cost of changing my mind? Well, you know, once I pay for it on this order, I’m going to try the chocolate cake. But it’s not going to be my last meal. And your career is like that. Like, if you have an interest in, it might be fun to be a product manager or product owner. Don’t be afraid to get involved in it. And then change your mind and say, no, maybe that’s not exactly right. Let me try being more of a tech lead or the other way around. Or maybe I’m going to try doing test automation.

So I do think many of us are raised to think that your career is a ladder and you just keep going up, up, up, up, up. I like to think of it as more of a canvas, that you have a paintbrush and you’re just connecting the dots across the canvas and it’s a great picture. You get a very rich career that way. I also think that some people are afraid if you’ve been in a hard engineering role to step into product management. What if I forget how to code? I’ve never seen that happen. Never in my entire career. I’ve never seen anybody forget how to code. And the thing that you can tell yourself is you have already demonstrated to yourself that you are capable of learning. So no matter what dessert you try or where your paintbrush goes on the canvas, you can learn or relearn anything along the way if you decide that that really was the thing you’d like to do more.

Henry Suryawirawan: Wow, I think this is a golden advice to me, right? So sometimes we are overthinking in a way, right? I also felt guilty in the past where I think always, you know, your career has to be linear, right? You always have to step up, either your role, your skill set, or whatever that is, right? But sometimes it’s like what you said, right? You make decision, you want to try something, how hard it is to reverse? Some people call it the type one, type two or two way doors, one way door kind of thing. And our career is kind of long, right? So I think our life, maybe we could easily spend like maybe 40 years kind of a career, right? So taking one or two years in between just to learn something and figure it out, whether it is meant for us, it’s not really that long, right? Looking in the bigger picture. So I think, thanks for sharing this for some of us who are still overthinking. So I think the key lesson here is if it’s not that difficult to reverse, just give it a try. So I really love that.

[00:23:03] Every Company is a SaaS Company

Henry Suryawirawan: One of the things that you shared before we had this conversation as well, right? You’re saying that every company used to be a tech company, but everyone is going to change into a SaaS company. So tell us how do you come up with this kind of a thought process and why do you think everyone should be a SaaS company now?

Paula Paul: I think largely, if you’re depending on cloud services now for email or whatnot. So I think it goes back to what the company does to deliver value to its customers. So I work with lots of different companies. One recently was involved in the energy sector as a nonprofit to help organizations take advantage of more solar or natural energy or reduce their energy costs. So the value that that company exists for is to deliver those services and that value to their constituents. You can say, well, what truly custom software is needed to do that? And perhaps not that much now. We have a really sophisticated CRM systems in the cloud that help with reaching your constituents and help with delivering value. So the core of most organizations these days can be represented in many of the large SaaS platforms.

And I do think that custom software, if you separate it into customization of the platform versus truly unique custom work, the companies that are doing specific software products that are offering unique value, that’s definitely not SaaS, but I’ll bet those companies also rely on CRM systems for marketing and whatnot. I can’t really think of a company, a large company that doesn’t rely on some kind of SaaS. And I think where companies are maybe erring on the side of too much custom software now. That they need to leverage their platform whether that’s, maybe it’s an in-house platform, even if it’s a fully custom software product, you have an in-house team that’s doing your DevOps platform or your delivery infrastructure. So I think that people need to really say, okay, how can I get the most out of my platform? And then what’s the value add that I really need to invest in and focus on? That’s truly custom software, custom technology.

Henry Suryawirawan: I think one fundamental things that I learned from my past guests as well in this, era, right. So opening up your capability as an API, I think is also crucial, right? So to become like a SaaS, you know, leveraging a SaaS, you need to have an API, right? Good, well designed API that can collaborate with each other. I think this API economy is also kind of like important for many companies to adopt, right? And especially, I don’t know whether this is going to be the trend going forward, right? We come to this agentic AI kind of trend, where AI agents will talk to each other. So I think without good APIs available for them to talk to each other, I think it’s gonna be really difficult, right? So for people who build strategy for your tech, don’t forget about this API economy aspect and how you think you can deliver your values through the API rather than always all the time through UI, the apps itself and things like that. I think that’s really important.

[00:26:22] The Huge Impact of Open Source

Henry Suryawirawan: So Paula, as someone that are also active in the open source world, I know that you serve in the OpenJS Foundation and maybe a few other open source initiatives out there. What do you see the trends in open source these days?

Paula Paul: Another great question. Even just in the past, maybe six to eight years, if you looked at any given software product, maybe the number of dependencies, maybe 500. And now if you look at any given software product, the dependencies on external software packages, maybe in the thousands, typically. And 90% of that from the Linux foundation, you know, studies are that 90% of any given software product is open source packages. It’s taken a long time for organizations to truly accept that fact. And I’ve worked not that long ago with organizations that say we don’t do open source. And I would just say, well, let’s look at some of your public websites or your internal software. And it’s like you’re relying on it, what does it mean to do it?

And these days now with the very visible kinda supply chain attack, incidents, and so forth, the awareness of open source has become much better. And I think organizations are now shifting into how do we really manage that. So I do see organizations really embracing the formation and staffing of OSPOs, Open Source Program Offices, that help really understand those dependencies, manage them well. You know, my ideal is beyond awareness, then kind of managing it and optimizing it with things like OSPOs. Those organizations really learn how to contribute back to the ecosystem. In highly regulated industries, there are often a lot of barriers to contributing back to the open source ecosystem. And I think because so much value depends on the open source ecosystem that every company should have an easy way to allow their employees to contribute back to the ecosystem. So we’re getting there.

Henry Suryawirawan: So for people who have ever analyzed your source code, right, how many packages and dependencies you are actually relying, I think, yeah, now probably in hundreds or thousands. Not to mention, right, one dependency depends on other dependencies that come from open source, right? And I think if you imagine these days, we can run a web server just using one line or something like that, right? So just imagine how many dependencies that it has to pull in order to make that happen. So I think open source really becoming like a value adding.

[00:28:59] Open Source’s Security Challenges

Henry Suryawirawan: At the same time, you mentioned it could also be a risk for many companies because of the supply chain attack, right? And sometimes we don’t know what software that gets pulled because we just know that our code runs. And we know in some news out there that there are, I don’t know, some hackers out there who mimic a popular open source, try to change the name a little bit, and then it infiltrates some people’s code base. How do you advise us to actually look at this problem? Is it something that we have to be really, really worry about? And what can we do to mitigate that?

Paula Paul: Great question. I do think that if you are building software product, custom software, knowing your dependencies is absolutely critical. And then evaluating, are you just pulling these software libraries into your product, because it does a cool thing, or did you really look at what this package does? Are there alternatives? How well supported is that package? I do tend to say, you know, if a package is supported by a foundation, like one of the Linux Foundation foundations, it has the backing of that foundation. Certainly the OpenJS foundation is heavily involved in security and the security of the JavaScript ecosystem, which everyone depends upon.

So I think looking for packages that are supported by healthy contributor communities and by foundations. If it’s a large package like Node, that really puts a level of comfort behind the package that means that it’s going to be around. It’s not a new concept, but there are organizations like HeroDevs, which is affiliated with the OpenJS foundation that will support packages that are out of the mainstream support. So if you have an old version of Node, but you still need a security patch, you can still get help. And looking for those kinds of things in the packages that you adopt is important.

[00:31:04] Managing a Healthy JavaScript Ecosystem

Henry Suryawirawan: As someone who is really active in the OpenJS Foundation, and we all know in JavaScript, right, the ecosystem is really, really, thriving. There are so many open source software maybe being written as we speak, right? So tell us the kind of challenge that you have to deal with in order to, you know, build this thriving, healthy community within the OpenJS ecosystem.

Paula Paul: Well, I, I give a lot of credit to Robin Ginn, who’s the executive director and the entire board. It’s really amazing, because it’s open governance of those JavaScript packages. Because if you say, well, if I’ve got a package or a library that really is only maintained by a large vendor, are you really going to get the fixes that you need in priority of your needs, or are you going to get fixes and things that you need in the order of the vendor’s direction for the product? I really enjoy and admire championing open governance, which is something that the OpenJS foundation does. It’s not easy because you don’t have the hierarchy of a large organization dictating terms from the top down. It really is a community. And that community takes care and feeding and a lot of really wonderful and dedicated people to move forward. So I give Robin again a lot of credit for that or all the credit for that. She’s an amazing leader. So I think it does take good leadership and collaboration across a lot of very talented and wonderful individuals. So we’re very fortunate.

Henry Suryawirawan: Yeah, I would also like to express, you know, appreciation, gratitude, kudos to, you know, those people who really, first of all, like contribute the open source project, right? That itself is kind of like voluntary basis, most of them, right? And for people who run this kind of a governance, especially for big ecosystems. You know, the OpenJS, CNCF, and things like that, definitely takes a lot of effort and complexities, especially governing people you don’t work with really, really closely, right? Because people are distributed. So I think kudos and thank you for the hard work that you all contribute to the healthiness of this ecosystem.

[00:33:11] Recent Trend of Open Source Licensing Change

Henry Suryawirawan: Apart from having more open source software being delivered, we also see some, especially big open source companies pulling back from open source, either changing license, make it less permissive and things like that. Some popular examples could be like Terraform with OpenTofu. Maybe Redis as well, now there’s a Valkey. So maybe tell us a little bit what was your view about some of these trends? Are you worried or concerned about open source maybe not going to become profitable? Maybe that’s one aspect that people see, right? Because these companies tend to want to make profit as they go larger. And would it be something that we can still change, you know, to have more open source thriving?

Paula Paul: Yeah, I don’t think open source is going away. I think actually open source is becoming stronger. And if you look at things like OpenTofu as a reaction to, sometimes it’s called the rug pull, the license change. The community immediately reacted and forked what was available. And within a week, I think it was submitted to the Linux Foundation as a new foundation, embraced by the Linux foundation. I was at a conference where it was announced, and we had a very large insurance company come up on stage and say, you know, they were moving from the commercial product to the open source product because they embraced open source. I think that it’s a really good example of the influence of community. And people may not want to be tied to an organization that is driven by, you know, private equity pressure for profit, causing these kinds of licensing changes. I think if you bring a library or a product to open source and you license it under a permissive license, I would rather that be a lifelong commitment. And it sort of is, because the moment that you publish a version under that permissive license, that’s out there. It’s in the ecosystem. People can fork it. I suppose you could delete the repository, but there’s copies of things everywhere.

I don’t really think these licensing changes are very effective. And certainly the Terraform licensing change did not have the desired result. It created a competitor, OpenTofu. I think that should be maybe a cautionary tale for companies thinking that they should do a rug pull like that, that are they really just going to create a competitor?

[00:35:46] Choosing Open Source vs. Commercial Software

Henry Suryawirawan: And for us, many developers who are relying on this software, do you advise us? I know maybe this is a hard advice to give, but do you advise us to always choose, opt for the open source software rather than the commercial version of the software? Or there’s a little bit of a thinking process that you should take. Maybe some advice here.

Paula Paul: Yeah, yeah, no, I have friends, a really wonderful friend of mine, Michaela Riva, when we were working together, he had an idea for a product. And he had, and has a strategy that there would always be a core that would not be open, and that would be part of his value add as a creator but that all of the infrastructure around that product would be open source. So that was his strategy from the beginning. And so that strategy is open. People know that that’s the strategy and then they have a choice. If you as an engineer or developer are using an open source package, I think there is an assumption that I should always be able to keep using that package. I don’t really feel that it’s full of integrity to change the licensing like that when there are so many people depending on it as an open source package. That doesn’t mean that there was, would never be closed source products. I’m really opposed to doing what what’s called the rug pull, is changing the licensing after it’s out there and in the community.

[00:37:18] Challenges of AI Model Training Based on Open Source

Henry Suryawirawan: Maybe there’s another trend these days as well related to open source, right, is the AI actually learning, training their model using these open source software that are publicly available, right? And they are permissive and make it their own, you know, like training the model, giving people suggestions, and building software on top of that. Sometimes the license probably is not evident, right? From the suggestion that AI gives. So what is your take about this? Is this a threat to open source?

Paula Paul: I think it’s certainly a challenge. I do think that a lot of the way that models are trained is not well governed. You know, if I train an AI model on say 10 different open source packages, but those open source packages all have different licenses like MIT, that’s a challenge. There’s nothing stopping you from doing that, but I think that the outcome prompts from that model. I’m a little concerned about that, but it’s also, I fully admit that it’s a very hard problem to solve right now. Say if you generate code from models that are trained on open source, I would really try to understand as best as possible the licensing implications.

And that’s where, again, if you use software that is supported by foundations, one of the functions that a foundation provides is legal. The OpenJS foundation has, you know, very wonderful, experienced legal support that understands licensing and more trademarks and all the things that are important to software. So I would say, don’t be afraid to ask. And especially in these public channels or GitHub comments or whatnot is like, I’m going to generate a product from this, what license should I use? And I think that that’s going to spark some very interesting conversation.

Henry Suryawirawan: For someone who is really new, like I’m a newbie in all this licensing and understanding licenses and things like that, right? Especially with, let’s say, if I use a lot of AI, AI suggests me a lot of things. There are maybe code that are already blended, right? So like multiple open source being blended together. You mentioned about legal, right? How can we actually first identify, okay, if this is a legal issue or should we actually be actively asking somebody to actually check our code? Like what should someone who don’t really have any good understanding about these legal implications?

Paula Paul: It’s interesting because you would say, okay, everything’s on the internet, and I’m a software engineer. The world is mine. And I think in general, things don’t get messy until maybe you have a, your product goes viral. So it doesn’t cost anything to say, you know, what would happen if my product went viral, and then say, well, what software am I depending on? And maybe the simplest question is which of the packages that I depend on have permissive licenses, meaning free to do whatever you would like with this code. Maybe you have to have attribution. You would say, you know, that I use these libraries or that’s listed in my package file. Or my, you know, software bill of materials. If you come across a package or a closed source product that you depend on. That’s the thing that you would want to make a special note of, okay, what would happen if I really got popular, went viral, started to make a lot of income? Who might claim some of that success that I had? And largely, I like to think that most of the open source that I use is permissive license. But if I ever really got serious or went to market with a product, I’d have a serious look at all my dependencies.

Henry Suryawirawan: So I think some people maybe know this term software bill of materials, right? So maybe if you can produce that BOM, right? B-O-M, as part of your CI/CD process, maybe you can really understand what kind of dependencies and maybe they will also tell you what kind of licenses, maybe there’s a detection of license change. And I really believe maybe sometime in the near future, we will see some tools that can, you know, analyze patterns of your code and check it against maybe some of the open source software and see if there’s any kind of resemblance. How much percentage is it the same? And you should be worried about, you know, if let’s say it’s 90 percent plus, right? So maybe you would have some challenges in terms of managing those. So I think thanks for sharing that.

[00:41:46] Recent Challenges with DEI Programs

Henry Suryawirawan: Apart from being active in open source, one thing that you are also active in is actually the women in technology aspect, you know, the diversity and inclusion. So I know what is happening these days in United States can be quite concerning for some people. Like big tech companies pulling back, you know, their DEI efforts. Some even stop it. So maybe as someone who live there, I may not follow all the news about all this DEI, but maybe tell us what you see and what you observe back there.

Paula Paul: Yeah, I think that, you know, with news, there’s signal and noise. And certainly there is a lot of noise around DEI, and it’s just become this buzzword now, that you blame everything on DEI. The fact of the matter in the United States, I did, I spoke at a conference for women in tech not that long ago, and I said that, you know, the encouraging thing in the United States, as far as women’s participation or people who identify as female, more than half of the people getting college degrees identify as female. So that trend has been continuing for some years now.

I, again, because I’ve been in the industry for so long, it’s easier for me to take a long-term view. I do think that education is so important and people who identify as female are pursuing their educations more and more now and more so than men of the same age, which is a different concern. But these things are going to take time.

I’ll go back to, you know, my father told me just work hard and you’ll get ahead and that’s only one half of the equation. I do think that everyone needs a strategy for their career and not be afraid to try things. And then maybe the most important part is don’t get stuck somewhere that doesn’t appreciate you. You don’t enjoy working with the people or you stop getting opportunities to learn. Those things are all self directed.

I wrote a piece recently that, you know, there’s really no such thing as a meritocracy even in companies that claim that it’s a meritocracy. There’s no such thing as a meritocracy, because, you know, to have a meritocracy, you’d have to have agreed skills, you’d have to have agreed definitions of those skills, agreed ways to measure them and agreed ways for people to judge them. And like that’s a very challenging thing, because people are involved and even the standardized tests have bias in the way that questions are written. So I would say, you know, don’t get hung up waiting for some magical meritocracy to reward you and pursue the things that you’re interested in and want to invest your time and effort in learning and then find your tribe, find your people. So those, I kind of stay away from the DEI aspect of that and try to have it as more of your personal strategy for your life and your career.

Henry Suryawirawan: Yeah, so I think that’s really encouraging, right? So still focus on the education aspect, no matter whether you’re a part of the minority, maybe women, right, whatever that is. Maybe this is just a phase of time as well, right? So just don’t give up, right? So invest in the education. I think that’s really cool thing.

[00:45:05] The Value of Diversity

Henry Suryawirawan: And for people who are in the leadership or, you know, some companies, because the big tech, maybe the United States government as well set precedence, right? Do you think that many people will just follow, right? And what’s your take here for companies who are maybe, you know, in between? Should I follow? Should I not follow? Should I change my program and things like that? Or maybe for leaders as well, they used to be championing DEI and maybe now they are maybe in the wrong side of things. What would be your advice for these leaders out there?

Paula Paul: Well, don’t throw out the value of the signal because of the noise. And to me now, DEI has become like AI. It’s just this overloaded term. And all of the studies throughout the past decades have demonstrated that having different kinds of people on boards, people with different kinds of thought processes and people with different opinions and leadership, those companies produce more revenue. It’s just a fact.

So if you go back to your strategy as company, if you want to optimize your performance as a company, produce the, you know, optimize your revenue, maximize your opportunity and maximize your revenue, having different kinds of people in leadership and different kinds of opinions surrounding you and the ability to collaborate, it’s like a gold mine. So you can walk away from the gold mine because of the noise and the press. Personally, I think that, especially given we have more diverse people getting their education, this too shall pass. That the noise will go away and the value of the signal will remain. But I can appreciate that for someone who is young right now. And I imagine that it is challenging and I know. My own children are in their 20s, so it’s, yeah, it’s just a challenging time for young people.

Henry Suryawirawan: Yeah. So for those young people, right, especially, for example, the women who want to think taking computer science or software engineering role, looking at this trend, maybe they hesitate. I think the key message still here, if you are passionate about it, if you keen to learn about it, right? I think you should still proceed. Maybe in the US, it’s starting, right? But I don’t see any other parts of the world actually following suit. Maybe there’s still hope, right? Don’t be disheartened. So I think there are still many opportunities.

[00:47:34] AI as Learning Tool

Henry Suryawirawan: And besides, some people are scared about AI taking over the job. I see AI actually is very empowering for people to get started into this world, right, software engineering and all that. Because it can really kickstart your learning development really, really fast. And I personally, myself, learn quite a lot of things from AI these days, especially if you want to learn new tech, new languages, uh, solve new problems and things like that.

Paula Paul: Yeah. Oh, I agree. I love that it’s integrated in search and you can just ask a natural language question. Tell me more about this and it will tell you, summarize what it finds on the internet and you can evaluate that and ask different kinds of questions to see if you get a different answer. I think it’s amazing! I love it!

Henry Suryawirawan: Yeah, so I think this curiosity, right? Asking AI multiple times, you know, multiple stages, right? I think it will really give you so much knowledge. Of course, you need to fact check. Sometimes they hallucinate and give you wrong answer. But still, for most common knowledge out there, I think they would give you like a good information that you can learn from, right? So I think that’s really superpower to me. So again, for those of you who think software engineering might be in a gloomy days, um, so I think there’s still hope that it can become like a great career.

[00:48:46] Creating Healthy Relationship with Technology

Henry Suryawirawan: So one other aspect that I see about you in your LinkedIn profile, you mentioned that you are helping people to create healthy relationship with technology. I’m just curious, what aspect do you see that the relationship between maybe human and technology are not healthy?

Paula Paul: That’s another really great question. And I would start with, if you’re working with a leadership team, technology leadership team or an executive team, and it’s not uncommon to find people on those teams that will say, or anyone that might say, oh, I’m not technical. Then I would dig into that statement, because what are they really saying? That is a description of their relationship to technology. They’re saying I’m not technical. Anyone who has a phone or uses a computer or, you know, listens to media or, you know, reads, is interacting with technology. So to say I’m not technical is a very odd statement for me to hear.

And I think what people might really be saying is I’m overwhelmed by how much I’m going to have to learn. Or I know this is going to impact my organization, but I’m not sure what to do about that. Those are very different things than saying I’m not technical. So I think the other statements will open up conversations. And many times what I’ll start with in organizations is, well, how do you make decisions about technology? And that goes to architecture, right? And it’s like cost of change. How can we make a decision that advances value to our business or from our business in a small way so we can learn from that. And then, you know, either change our mind or make another decision to advance it.

And I think those kinds of decision making skills need some work in many organizations that if you’re just making decisions based on, you know, what a vendor tells you or what you read in the noise, you know, at some conference or some trade show, helping people understand how to make healthy decisions about technology is like helping them make healthy decisions about what they eat. Very similar. But people don’t, uh, tend to think of technology that way. So I do try to borrow from these other disciplines like relationship therapy or nutrition therapy. And you can really apply a lot of those things to how people react to make decisions about and adopt technology.

Henry Suryawirawan: Right. When you mentioned about overwhelm, I can really empathize for some people. Like imagine for us, you know, software engineers who are in this world, right? Sometimes we are really also overwhelmed by the amount of technology advancement. So I could empathize people who are not even in this world, seeing the progress, the advancement out there, they could really, really become super overwhelmed, I guess, right?

Paula Paul: Yeah. Yeah.

[00:51:45] Dealing with Tech Anxiety

Henry Suryawirawan: So maybe for some of those who listen to this episode as well, so what’s your tips for them? Because they are not in this world. They see a lot of changes coming. They are like kind of stuck. Some of them resist the change, but maybe is there any tips from you?

Paula Paul: Yeah. And these are natural. It’s anxiety, right? And it’s like, I’m in my company and all of these, I hear all these things in the media. AI and it’s going to take over the world. It might take over my company. So there’s anxiety. And I do think that it’s important to take a breath, look at your own organization. And also many times, people don’t have a good idea of exactly what’s going on in their own organization. So I start with, you know, what’s your inventory. And then things like yoga, you might do a body scan, head to toe. It’s valuable in organizations. It’s like, what is going on in your organization? You don’t have to take months and months. You can, you know, make some inventory. A lot of times, especially in technology, there’s a very sprawling inventory, and it’s like not all tied to core value streams. So there’s a lot of steps that you can do that are just about becoming aware of exactly where you are and what’s your technology stance. No need to be anxious about that.

And then you can start to say, are there places, you know, simple things. Should we have a policy that allows our employees to use a large language model, a note taker in meetings? Should we summarize our company videos through a large language model and make those available to people who can’t join? Little baby steps that are decisions you can make that are easily reversible. And then just start to give you more comfort with the big scary, you know, AI thing. There’s just a lot of anxiety, which is unfortunate. And I think those are the things that are easily addressed.

More strategic things like should we be putting AI into our call center? Those are really great questions. And I would just make a big list of them. And then grow some more robust decision making skills about what should we do a pilot of or what should we experiment with and learn from? So I mean, those things really have not changed as good things to do in general. But when a new fad comes out, like blockchain, this kind of happened during the blockchain era. And then again with AI, it’s like getting back to some basics can help avoid a lot of anxiety.

Henry Suryawirawan: I really love, again, your advice there instead of being anxious about all these change and resist, right? I think always, maybe I don’t know, step your foot on the ground, right? So scan the organization, scan yourself, do inventory checking. And I think I heard this before as well. Before you make any good change, you actually need to be aware first, right? So if we are not aware, maybe we are taking wrong decision, right? So always look back, you know, what is happening around you, what are the opportunities. Because in any kind of risk, you know, any kind of whatever disruption, there’s always opportunity. And if you can spot that, probably that will also give you a big opportunity to actually advance in your career and in your role, right? So I think that’s a very good tips.

[00:55:03] 3 Tech Lead Wisdom

Henry Suryawirawan: So Paula has been a great conversation. As we wrap up our conversation, right? I normally ask my guests to share what I call the three technical leadership wisdom. So maybe if you can think of it just like advice that you want to give to the listeners, maybe you can share your version of the three technical leadership wisdom.

Paula Paul: In technical leadership, there is the mechanic part, right? The mechanics of the technology. And I would say don’t get overwhelmed because we live in an abundance of technology. Anything that you can dream of that you would like to do with compute network, storage, I think is possible now given the right resources, time, this kind of thing. So I would say don’t get overwhelmed just trying to understand everything that’s available, and start to understand the value that your organization delivers to its community or its constituents, customers. And how you can use this kind of vast abundance of technology we have available to make some small steps.

I would also say, if you ever start to feel that it’s not fun anymore, do the body scan. It’s like, what am I really thinking? Why am I really not happy anymore? Sometimes I’ve just taken a break and gone back and written like little Ruby programs or Ruby on Rails and reconnect with the things that you enjoy because that will make you a better leader.

And then maybe most of all is if you are grounded as a technical leader, the best thing you can do is help your team, you know, the people who work for you, report to you and your peers have a better relationship with technology. And to maybe spark some of the joy that you get from it and those other people. And spread the love. So those would be my pieces of advice.

Henry Suryawirawan: All right. So thanks for sharing them. I think I again, I really love the part where you ask us to actually do the body scan, right? So taking awareness, you know, I think that’s always important in any kind of rapid changes, disruptions and things like that. You don’t want to get lost in all this disruption, right? So yeah taking a break is also a great thing, right? Sometimes you don’t always have to force yourself. But taking a break, I think, again, like if you look at the bigger picture, right? This is maybe just a point in time. And you should not be you know disheartened because of this.

So maybe for people who love this conversation, they want to ask you more questions. They want to follow you. Is there a place where they can find you online?

Paula Paul: Sure, there are a couple of on LinkedIn, I’m @ paulapaul so it’s easy to find. And then I do write on Medium. And then if there are people interested in an engagement or consulting engagement, I work through my own company, which is Greyshore. G. R. E. Y. S. H. O. R. E. dot com. So I’m always happy to talk, paid or just in general meeting people and talking about technology. And I really appreciate the opportunity to talk with you today. It’s been fun.

Henry Suryawirawan: And I would also acknowledge your effort, your contribution in the community, right? Be it the OpenJS Foundation, be it the women in tech and all that. And again, I really appreciate your time, you know, having to share some of this wisdom for us. Again, it’s a rare opportunity for me to learn from someone who has been there, you know, in the tech industry since the very long. So again, I appreciate your time and thank you for being here today.

Paula Paul: Thank you.

– End –