text
stringlengths
0
2.68k
• Early experiences with stream processing services and understanding trade-offs between delivery guarantees and high throughput
• Development of open source software Benthos for stream processing with delivery guarantees, quality of service, and metrics
• Challenges in implementing delivery guarantees with ZeroMQ and disk buffers
• Experience with acquiring and integrating a company and its data engineering capabilities
• Observations on the difficulty of proving resilience in stream processing and the need for "show and tell" testing
• Acquisition of the company and its data engineering capabilities by Meltwater
• The speaker created a project, Benthos, as a solution to a complex problem in their company
• The project started as a personal project, but eventually became a company-wide solution
• Benthos was pitched as a simple, stateless solution to a complex problem, replacing stateful services
• The speaker's pitch was successful, and the company adopted Benthos
• Benthos has since become a widely-used open-source project, with various companies using it
• The speaker attributes the project's success to its simplicity, flexibility, and composability
• Benthos stream processing made "operationally mundane" according to Ashley Jeffs
• Early users were large orgs with data integration problems
• Benthos popularity grew, but Ashley struggled to sustain himself on donations and sponsorships
• He tried to start a product and pursue enterprises, but found the procurement process too complex
• Ashley now realizes the need for a more sustainable business model
• Hiring out vs self-service approach
• Enterprise readiness requirements (SOC 2 compliance, contracts, etc.)
• Balancing personal lifestyle and business obligations
• Scaling a project (Benthos) and potential loss of control
• Partnership with Redpanda and its benefits
• Integration and ownership of end-to-end pipeline
• Personal and professional expectations (e.g. from children)
• Benthos' early days and collaboration with Ashley Jeffs' company
• Challenges with monetizing open source projects, specifically Benthos
• Ashley Jeffs' consulting company and its role in supporting Benthos
• Acquisition process of Benthos by Redpanda, including formalities and timeline
• Ashley Jeffs' personal experience and emotions during the acquisition process
• Discussion of Ashley Jeffs' rejected proposal to spend a month in Las Vegas and engage in a "shampers binge"
• Ashley Jeffs' initial job at Benthos and his decision to leave the company, citing a desire to focus on enjoyable projects
• The impact of joining Redpanda, allowing Ashley to tap into new customers and revenue streams
• Ashley's concern about being trapped in a project he no longer enjoys, and his decision to walk away if necessary
• The challenges of managing revenue from startups, with Ashley acknowledging it's difficult to secure significant payments
• Discussion of Benthos' status after Ashley's departure, with the project effectively being discontinued
• Plans to split the Benthos repo into two separate repos for the engine and plugins were already in place, but were accelerated due to the acquisition.
• The Benthos engine is still MIT-licensed, while the Redpanda Connect plugins and enterprise features are licensed under Apache v2 and some are enterprise-licensed.
• The acquisition led to a "mixed bag" of reactions from Benthos users, with some unhappy about the licensing change, while others are unaffected or even happy about the increased control over their builds.
• A fork of Benthos, called Bento, was created by WarpStream, which has since been acquired by Confluent.
• The host of a Hacker News post was criticized for changing the license of their open-source project from MIT to a more restrictive license after being acquired by a company.
• The host defends the change, explaining that the original MIT license only guaranteed that the original code would remain open-source, but not that future changes would also be made available for free.
• The conversation touches on the concept of "eventually open-source" licenses, which allow a project to start closed-source and later become open-source.
• The hosts discuss how people often misunderstand open-source licenses and assume that by changing a license, the original code is being taken away.
• The discussion also touches on the concept of a "rug pull," where a project or creator is accused of abandoning their responsibilities or commitments after a change in circumstances.
• The speaker's experience with Benthos, including dealing with criticism and haters
• The impact of the acquisition on the project, including the transition to a paid model and the addition of new features
• The importance of finding a sustainable business model for open-source projects
• The possibility of alternative outcomes, such as VC funding or continued sponsorship, and the speaker's preparation for different scenarios
• The speaker's perspective on the acquisition and the benefits it has brought to the project, including a more stable and sustainable future
• The challenges of maintaining a company like Benthos, including storage and scalability issues
• The difficulty of relying on sponsorships from other companies, including the uncertainty of renewal and the burden of constantly seeking new sponsorships
• The consideration of accepting a sponsorship-based model indefinitely, including the impact on personal life and family responsibilities
• The potential for acquisition by another company, including the pros and cons of such a scenario
• The relationship between Benthos and other companies, including Synadia and NATS, and the potential for shared marketing efforts or collaborations
• Ashley's past involvement with Synadia and the potential influence on his work
• The closure of his GitHub Sponsors page and his feelings about it
• His future plans and goals at Redpanda, including making himself "obsolete"
• The uncertainty of his employment at Redpanda and the possibility of being "expendable"
• The realities of working on an open source project and the need to find balance and enjoyment in one's work
• The desire to work on one's own projects versus having a traditional job
• The challenges and risks of running a business or working on open-source projects
• The "grass is always greener" mentality and the temptation to switch to a perceived easier or more secure option
• The importance of self-motivation and agency in pursuing one's own goals and dreams
• The potential for exploitation in open-source projects or startups, where individuals may feel obligated to continue working without fair compensation or recognition
• The challenges of maintaining open-source projects and the emotional toll on contributors
• The lack of support and resources for open-source maintainers
• The idea of creating a "resource for healthy open source habits" to help maintainers manage their projects
• The concept of a "global daddy pig" or mentor figure to provide guidance and support
• The importance of self-care and taking breaks from open-source work
• A discussion about thanking and appreciating open-source maintainers for their contributions
• A conversation about the difficulties of finding the right balance between work and personal life as an open-source maintainer
**Jerod Santo:** Well, we are here with Ash Jeffs, originally from Benthos, at least when we first heard of you, and now at Redpanda... And now on the Changelog. Welcome to the show.
**Ashley Jeffs:** Hey, thank you for having me.
**Adam Stacoviak:** You've arrived, with that awesome chair.
**Ashley Jeffs:** I'm here, in my commanding chair.
**Jerod Santo:** That is a commanding chair. I feel like maybe we're your subjects at this point.
**Ashley Jeffs:** It gives me more authority than the hosts.
**Jerod Santo:** \[laughs\] Yeah. I feel like you should take charge here and just tell us what we're going to talk about.
**Adam Stacoviak:** Yeah. What's the show about?
**Ashley Jeffs:** The camera angle is as if you're on my lap, which was intentional.
**Jerod Santo:** That's cute. It's diminutive. I do want to give you props, because I've actually used your previous content for future content, which you might not know about... And that is this. When you were on GoTime, our Go programming language podcast, which has been retired and it now lives on spiritually throug...
**Ashley Jeffs:** It was meta. I tried to be clever about it.
**Jerod Santo:** You were trying to hack -- because what we would do is we would take the opinions, put them on Twitter, on a poll, and find out whether or not they're actually unpopular. And so you went meta and you said "If you take a Twitter poll, your opinion doesn't matter. Nobody cares. You should go outside." So...
**Ashley Jeffs:** Yeah... A good social experiment, but we both fumbled.
**Jerod Santo:** Yeah. We both failed to have an unpopular opinion.
**Adam Stacoviak:** It's hard to have an unpopular popular opinion. I find that it's challenging for me.