Ask HN: Should I open source or not?
I am currently working on two projects: a PDF alternative for the blind and Google Maps/navigation for the blind in my city. I started from scratch with no AI or third-party reliance. I have written everything by hand (with considerable difficulty since I am blind), so what I am asking is: is it worth it to open source?
Embarrassing coding style aside, I am concerned about parasitic big corporations taking the code and putting it behind a paywall, which is not what I want for obvious reasons. Putting my projects under GPL will obviously hold them back a bit, with the cost that brilliant programmers who would have helped would be the casualty.
Another issue I have is that I want zero AI-bros in my code. Not only would I be inviting disorganized, poorly constructed code, I also don't want these kids copy/pasting to add code they have no right to.
What should I do? If it were me, the accessibility benefits of open source far outweigh the potential cost of letting AI-obsessives use it. You can toss it on a public repo with the license of your choice, update it at your leisure and anyone that disagrees will have to fork it. I'm not super fond of AI myself and I definitely see how it can result in low-quality contributions pushed upstream. At the same time - if someone can use FOSS to genuinely improve their life, even with AI-generated modifications, I'd hope they do. Maybe that's just me, though. Yes. Put it under a license that requires attribution (like MIT) and open source it. These kinds of software are better if open so that more people can analyze and implement similar technology to make software more accessible. That’s exactly what I have in mind too. The only issue is avoiding the “freeloader problem” that most big corporations have embraced lately. For example, with my local navigation system, I don’t want Google scraping all my valuable data to feed into Google Maps—only for them to charge me $10,000 per API call when I try to query the same data they took. If you’re a nonprofit or a small, local business, then by all means, use my maps—that’s who I built them for. But big corporations? No. Open Maps is facing this exact problem right now. Every shady AI and LLM is scraping their data and making millions, yet contributing nothing back to the source. I want to avoid all that by making it very clear: if you're a big corp and you want access to our data, you need to pay up. Otherwise, take a hike. They’ll probably take the data anyway without permission, just because they can afford $500/hour lawyers—but at least I’ll have made the effort to stand my ground. I once visited a community with similar questions. "Is it worth to ...?" I will answer the same as I did back then. What does it mean for something to have worth to you? That is not a rethorical question. This is actually a great question. For me, as an individual, the software I create to fulfill an important need for blind people would be worth it. But having said that, I shouldn't just look at it from one perspective—I need to consider it from other angles as well. I have to anticipate how the code will be received once I open source it, and my question was a way to fill the gaps I might have about the consequences of releasing the code under a certain license. I guess I could bypass all these inquiries, lazily upload my code to GitHub, have vultures like Google find it, copy it, and put it into Google Maps. They may or may not care about catering to blind people (the original intended audience), and then my little code would just fade away, as a more refined version would likely emerge elsewhere. I am visually able, so I cannot relate much to the visually impaired cause. The best I can do is try to understand their fight. Maybe tell their story from time to time. Bring their goals to a wider audience. Your intention in assisting technologies sounds noble. However, precisely because of that, I cannot stain your own path with my opinions regarding how to license it. Put it under the AGPL 3 or later. Bad actors won’t touch that license typically That is my plan, but I am afraid I would have the opposite problem, no programmer, the one we want, would bother contributing. A long time ago, I was a math T.A. in the university and the main T.A. of my course was blind. He used a screen reader for everything, and wrote LaTeX and had a paper versions of the courser exercises in Braile+LaTeX (IIRC). > a PDF alternative for the blind I'm not sure what it means. It's a PDF reader? It's another format that is similar PDF but more screen reader friendly? > Google Maps/navigation for the blind in my city Do you know https://www.openstreetmap.org/ The data is open, so it may be useful to build an easier to use client, so you don't have to collect all the info. > I am concerned about parasitic big corporations taking the code and putting it behind a paywall Probably AGPL is better for your intentions. > Not only would I be inviting disorganized, poorly constructed code, There is no obligation as a maintainer to merge all PR. But remember to be nice. >I'm not sure what it means. It's a PDF reader? It's another format that is similar PDF but more screen reader friendly? Yes, a PDF alternative optimized for screen readers. For some reason, PDFs have clunky functionality when it comes to working with my screen reader. PDF URLs are especially egregious—VoiceOver really struggles to read them. So, I started writing a new file format that allows screen reader compatibility from the start. >The data is open, so it may be useful to build an easier to use client, so you don't have to collect all the info. Yes, I know that website—I’ll use it if I need it. So far, I’ve been preoccupied with some other data that neither Open Maps nor Google provides. I’m about 30–35% done. I also have to think about whether to go open or closed source before using their data. Even though what they provide is mostly public domain, nonetheless, I want to stick to my own principles: credit/donate where it’s due, or go without. >There is no obligation as a maintainer to merge all PR. But remember to be nice. Good advice. > I’ve been preoccupied with some other data that neither Open Maps nor Google provides. As an avid OpenStreetMapper I honestly wonder what kind of data is missing in OpenStreetMap. Is that something sighted persons cannot grasp and thus not add to OSM?
Have you seen https://wiki.openstreetmap.org/wiki/OSM_for_the_blind ? Is something missing there? > I also have to think about whether to go open or closed source before using their data. No, you don’t. If you use the map data, you have to attribute it, but your code doesn’t need to be open. See https://osmfoundation.org/wiki/Licence/Attribution_Guideline... for the guidelines on attribution. >As an avid OpenStreetMapper I honestly wonder what kind of data is missing in OpenStreetMap. Is that something sighted persons cannot grasp and thus not add to OSM? Have you seen. Thank you for providing that link — it was helpful. The information I want to display is text-based rather than graphic-based. My plan is to have two systems working in tandem: physical hardware placed in areas of interest like bus stops, intersections, etc., and a user interface that queries this data. The UI will be entirely text-based, with an absolute minimum of graphics. In the future, if this project succeeds, I aim to launch my own GPS satellite to bypass Google’s predatory API calls. Yes — but in Phase 1, the map will be text-only to ensure accessibility. What I envision is a sort of Wikipedia for my entire city, where every landmark and point of interest is cataloged and annotated in a rolling fashion, allowing others to edit, expand, and improve. Kind of like Google Reviews — except not owned by an advertising company. > I aim to launch my own GPS satellite to bypass Google’s predatory API calls. That might be a bit more costly than calling any API :)
And launching own GPS satellites has no connection with Google's API as far as I know. Are you interested in location data only?
Then you can actually query https://nominatim.org/, which does the (reverese) geocoding. > the map will be text-only to ensure accessibility. I wonder how that's done, something like https://wiki.openstreetmap.org/wiki/Mapscii ? Pretty sure no one's going to give a ____ either way. No. Do not open source it. Respect yourself and the work you have accomplished. If you have no interest of making a business from your code, you can give away the finished software for free. Then you can postpone the decision if you want to make it a source of income later in the future. There are no benefits at all for you by making the code open source. There are no benefits for users either. The only people who benefit are huge corporations and a few open source fanatics who wouldn't even give you a glass of water if you asked for it. If you want to benefit other people with your code and make a living from it, sell the software for an honest price. That way you attract quality users and are fairly compensated for your work. Accessibility tools often thrive as open source precisely because they can be adapted to diverse needs, audited for security, and maintained by communities when individual developers move on. I see software thrive much better when it's sold for a fair price by a solo developer or small shop. Boutique software is great for users and great for the programmers. And if the OP has no monetary interest, he/she could give away the software for free while keeping the source code for themselves. You are getting downvoted but I want to reinforce this: open source is a tough free work, it is easy to get oneself burnout for nothing and nobody will care, not even the people benefiting the most from your code. Yes, as I read HN and Reddit, I’m seeing all the big corporations acting like vultures. The problem is, as these corporations monopolize the market—either by buying out or rendering smaller independent companies and developers obsolete—and as most of the brilliant programmers end up working for the same 3–4 tech giants, I’d be forced to either: Allow these vultures to steal my code, lock it behind a paywall, and then compete with me using my own work; or
Make my code GPLv3 to effectively block them and their subsidiaries from touching it. In that case, I and other independent developers would retain control and be the ones improving the code—unless, of course, the AI-bros are allowed to steal with impunity, which would make this whole conversation pointless. If you make the code publicly accessible you have no guarantees that people and corporations won't take that code. How about inviting the right people to work with you on the closed source code if they show any interest to help?