13 KiB
“I wish the industry would not follow this ever increasing hype cycle for new stuff”
While new technologies can lead to innovations, hype often goes with the territory. Gerald Venzl, senior director, product management at Oracle Corporation speaks with Abbinaya Kuzhanthaivel about the risks of following the hype, and shares his thoughts on how to grow a career by contributing to open source projects.
Q. Can you tell us a bit about your journey with open source?
A. I currently work as a database product manager looking after the Oracle Database. As part of my job, I help steer the direction of the product and help pinpoint priorities in the market that need to be addressed. Often people are surprised when I tell them that open source is an important aspect of that. For example, Oracle recently launched a Kubernetes operator for its database, which is fully open source and is available on GitHub. But we have also Docker build files on GitHub, and even some of our Oracle Database drivers are fully open source, such as our Node.js driver and the Python driver.
Q. Is Oracle your first venture into open source?
A. I was accustomed to open source in my personal life before joining Oracle. For me, open source is a natural part of my job. Oracle has a long-standing commitment to open source with Oracle Linux, Java, MySQL and many other projects. Although the Oracle Database core is proprietary, we see the value of open source and having people to contribute and understand the technologies around it.
Q. What do you think about open source as an opportunity?
A. The thing that I like about open source is that it’s essentially a democracy and everybody gets to participate. Not that all participants are core contributors, but it connects people from all over the world. The exposure to projects and other developers is important for growing your skillset and your career.
In Oracle, we are currently focusing a lot on the Kubernetes operator. That’s an exciting project. In my free time, I also write little tools and open source them. I had worked a little on a command line tool called ‘csv2db’, that allows you to load CSV data into a database. I see these projects as a chance to educate myself. You read something, and then you try to implement it and see whether you really understood it, whether it works and so on.
Q. How did this change happen from being a developer to a product manager?
A. My path was probably highly irregular. I am from Austria originally where I worked as a developer for an American company headquartered in New York. I was part of the performance engineering team. I was actually one of those guys who tried to profile the code and those skills eventually brought me to troubleshoot production systems at customer sites. I relocated to New York, and then a couple of years later Oracle asked me to join their pre-sales team as a result of our earlier interactions in running proof-of-concepts and performance tests on their systems together.
I liked the opportunity to learn by interacting with customers directly and discussing requirements. But for me, the turning point was moving back into development and specifically into the technical aspects as product manager.
Q. What are the top two ‘must have’ qualities of your role?
A. I think one must definitely be able to accept ever-changing requirements. But it’s not just a reactive job. You also have to have a curious mindset. If you have a curious mind, are open to new ideas, and can think out-of-the-box, you can establish your priorities and grow into the role.
Q. What would you like to say about the challenges in your current role?
A. Product manager is a very diverse and ever-changing role. One must be highly adaptive because you need to go through steering the product, prioritising the internal roadmap and also accept changes depending on market demands and customer needs. So the key challenge, and also the fun part, is that no day really ever looks the same.
Q. Any important risk you have taken so far which you think might inspire others?
A. I think my shifting into product management was the biggest risk. The future is unpredictable and one must dare to take such risks. For open source projects, it’s probably taking the step and actually wanting to contribute to a project. It can be a bit daunting in the beginning, especially to know where to start contributing for very popular active projects. Not all contributions have to be done in code; often people appreciate help with the documentation or testing. There are many ways to get started — the important thing is that you do get started. If you are passionate about something, you will have to go after it. You may have people not reacting the way you expect at first, but that’s okay. It will essentially help to learn.
Q. What do you consider as top leadership qualities?
A. There are a couple of things that I think are important for leadership. And the first one is that it’s okay to be wrong and acknowledge it. This also encourages people around you to freely share ideas. The other thing would be to take chances and get out of your comfort zone. I have never formally learnt product management. I just was intrigued by it, gave it a try and I thought, let’s see how it goes. It has put me in a good space for growth.
Q. Anything you are not very happy about when it comes to open source?
A. I wish the industry would not follow this ever increasing hype cycle for new stuff. New doesn’t automatically mean great. We may talk just about all the new things out there, but the world still runs on Linux and, remember, we still use the HTTP, TCP/IP and all those technologies today. The fundamental technologies that connect us around the globe have been there for a long time. Something doesn’t have to be new to be great and often new technologies go just as fast as they came.
Q. What are the major risks in going ahead with new technologies?
A. A major risk is forgetting or not wanting to ask the “why” we need the new technology. In our industry we get excited very quickly about something that we then want to work with. Sometimes that means that we oversimplify some business requirements and kind of omit the downsides of a new technology, just so that we can use it for a new project. I agree that new things lead to innovations and there is nothing wrong with that. But I have equally seen just as many projects fail that tried to replace the legacy system because a new buzz technology is out there and looks attractive. I’ve seen a three or four year-long project fail because no one bothered to look at the ‘why’ when replacing the previous system; and although it solved the new requirements, people forgot to ask themselves what the old system did well and they just ended up with those old issues again.
Q. How does one keep away from the hype around new technologies?
A. I would say that, as a developer, don’t just blindly follow the latest and greatest. If someone is telling you new stuff, well it’s great to know. But it’s no surprise to me that Linux runs the world and HTTP runs the Web, because those technologies are really well-designed. Have an open mind and look at what’s new, but think about whether this new technology actually will serve your needs. It’s fine if it doesn’t.
Q. How can a developer find projects to contribute to while keeping away from the hype?
A. There’s nothing wrong with working on a hype project, but you have to make sure that you actually have interest. Think small — don’t expect to become the main contributor in the next two weeks or in a short span.
Remember most of the open source projects really like non-code contributions just as much, such as testing or reporting bugs or lack of information in the documentation. Don’t just go into it to write some code. Initially have some idea about which area you want to contribute to. It should be something that excites you. Go to GitHub and just read through project contributing guidelines. It will tell you what contributions the project needs and how to make them correctly.
Your work may be small, like adding a sentence to the doc or correcting a typo. But it will allow you to get familiar with the process and with the other people involved in these projects. Do not expect to jump into a project and change its core. Most likely, only an approved committee of committers can actually change that part of the code. Build some trust, show that you understand the project and over time your involvement will grow naturally.
Q. Any examples of projects you think were overrated because they had just used a new technology?
A. I have seen a few working in the database space. When you think about it, relational databases have been around for a very long time and are still going strong. To some extent, we have forgotten why relational databases became so popular. The goal was to organise the data in a way so that five years from now somebody who comes in and has no clue what the data looked like, can make sense out of it. For a while there was a general hype that you no longer need any database, whether relational or non-relational, because Hadoop will do everything. And it didn’t. It actually just led to data cleansing issues for many folks. Don’t get me wrong — there are companies that successfully run Hadoop clusters and there is nothing bad about the technology itself. But you have to understand what it is and when to use it.
Q. Oracle has recently introduced the new MongoDB API for its autonomous database. What was the reason behind it and how did it happen?
A. At Oracle, we follow the converged database methodology. This methodology focuses on bringing the algorithms and computation to the data, rather than the other way around. A very good analogy for converged database methodology is the smartphone, where you can do multiple use cases in one device, like taking a picture and sending it to a friend while being on a phone call, for example. In recent years, we have seen a proliferation of vendors pitching their technology to address an often simple use case For example, developers like working with JSON documents and MongoDB allows them to store and retrieve these documents. But it is one thing to store and retrieve these and another to analyse terabytes of them in real-time. We think SQL is a really good language for any kind of analytics and we have the best database for mixed workloads, i.e., allow real-time analytics while transactional workloads are running. Additionally, Oracle Database has been managing JSON documents natively since 2014.
Developers love the MongoDB API as it makes database interaction very natural for them. And we have the best database for analytics and mixed workloads that can also manage JSON documents natively. So we decided to give developers the best of both worlds — the same MongoDB API on top of the world’s leading database.
Q. Will you say MongoDB is a new hype?
A. The JSON format is a very useful hierarchical format and is nothing new — it’s been around since 2001. If you want to use JSON, then go ahead, by all means. Oracle has done a lot of work to introduce JSON operations into the SQL standard, and we see more and more databases supporting these standardised operations. You will find that if you want to work with JSON, retrieve JSON documents, query and manipulate data in JSON documents, you definitely don’t need to have a document store anymore. MongoDB is a cool technology, but so were XML databases in their days. I think it is definitely not needed for data management aspects.
Q. Any hiring plans?
A. We are constantly hiring great talent and all our openings can be accessed on the Oracle careers page. We are looking for people in a variety of different roles, including engineering and product management. We are strong in diversity and have people from all around the world and all ages, including graduate students.
Q. Your message for our readers.
A. Programming is a universal language and it’s great to be a developer and write programs. Don’t be shy, be always ready to try something new and get out of your comfort zone to do things you are passionate about.
作者：Abbinaya Kuzhanthaivel 选题：lkxed 译者：译者ID 校对：校对者ID