We Will Not Sabotage Your Code

This is a very low bar. I am genuinely saddened that I need to write it down at all, because until last week it seemed too obvious to say out loud.

Last week the news broke of an open source developer doing something I find reprehensible. The short version: fed up with "vibe coders," he embedded a hidden prompt aimed at AI agents inside his framework, with instructions designed to destroy data on the machines of people who used it.

So let me say it plainly, on behalf of Codename One: we will not sabotage your code. Not now, not as a protest, not ever. We won't do it for free users, we won't do it for paid users, and we won't do it if we disagree with you politically.

Why this matters

From what I have read of the relevant laws, this is likely a crime in some jurisdictions. But beyond that, as an open source maintainer, I find it deeply offensive: it is a violation of trust which I find morally reprehensible.

Trust is the greatest currency open source actually has. We spend years earning it one bug fix, one release, one honest answer on a forum at a time. When a maintainer weaponizes a dependency against the people who chose to rely on it, the damage is not contained to one project. It teaches every developer that the code they pull in might be quietly hostile, and that fear lands on all of us. We already have supply chain attacks to worry about. We do not need maintainers volunteering to become the next one.

To be clear about the principle: a maintainer can do whatever the license allows. If you do not want AI agents touching your work, that is a legitimate position. Write a license that says so it probably can't be defined as "open source" though. That's an honest choice, and it's yours to make. What this developer chose instead was sabotage dressed up as a stand, and the fact that some people find that normal, or even justified, is the part that disheartens me most.

I understand the fear

I do understand where the anger comes from.

Calling AI agents a "disruptor" is a massive understatement. Many of us will lose the jobs we have today. Things may get worse for a while before they get better. That is real, and dismissing it would be dishonest.

But we are engineers. Our entire profession is solving problems and building on top of whatever the new layer turns out to be. Lashing out at the people downstream of us is not engineering. It is the opposite of it.

Why I am optimistic anyway

The potential of AI for open source is groundbreaking, and I do not use that word lightly. Before this current generation of agents, competing with a company like Google was a fantasy. They can put 100+ developers, designers, and technical writers on the exact field we are fighting in. We could only dream of keeping pace.

I now genuinely believe we can surpass them by the end of the year.

Large organizations are going to struggle to adopt AI properly. Process, politics, risk committees, and sheer inertia will slow them down. That difficulty is precisely the opening for the small, agile player working out of the proverbial garage. For the first time, leverage is on our side.

This means reinventing yourself, again

None of this is free. The skills you need to work well with LLMs are radically different from the skills most of us built our careers on, and that adjustment is hard.

My best analogy is my own start. When I began programming there were no IDEs, no code completion, no internet to search. My biggest asset was my memory. I read everything like a sponge, I could recall every book and every subject, and I always knew where to look for an answer. That was my edge.

Then the world changed and that edge mattered less, so I reinvented myself for a newer age. I am doing it again right now, learning to work with these tools instead of against them. And I love this type of programming even more. I can focus on the stuff I enjoy, build huge ambitious tools and actually finish the work with testing and all the parts I used to hate writing.

That is the spirit I want for this community. Not sabotage. Not fear turned into payloads hidden in someone else's build. We solve problems, we build on top, and we earn trust by being the kind of people others can safely rely on.

We will not sabotage your code. We are too busy figuring out how to make it better.

Thank you for being a part of our community.

• Java

Shai Almog

Author

Author, DevRel, Blogger, Open Source Hacker, Java Rockstar, Conference Speaker, Instructor and Entrepreneur.