In a way bike frames are inherently open source. Since you get a copy of the entire object you can do what you wish with it. Since almost all the ideas are old and unpatentable, even modern intellectual property laws don't apply to much of it. The precision is relatively low, and the rider insensitive to small variations, so it's possbile to reproduce most of the design from measuring the final product. This is less true with composite parts, and suspension design is sufficiently specialised that simply copying the geometry is unlikely to work. Materials do change shape slightly during manufacture, but allowing for that is fairly straightforward for the most part (but for example thermoplastics with a 10% size change on cooling are much harder).
Most of the real design effort goes into making the frame work well with everything you bolt onto it. Getting a description of the design thought process would be more useful, but that's a book-length project at the very least. Tiny things like where exactly you put rack mounting points or at what exact angle the cable stops go can take hours of thought and experimentation. It's not micrometre accuracy, but a millimetre either way can often be crucial.
Then we swing way out of the mainstream and start designing frames for cargo bikes, recumbents or tall bikes. Here the design effort is often more about getting something that works rather than something that's extremely polished, simply because the market is not very mature (meaning it's not a contest for a fixed-size market, it's possible to build your own entirely new market). One you move away from "diamond frame bicycle" the design gets a lot more complex. Simple things like rideability are not widely well understood and some universities offer physics courses in dynamic stability focussed on bicycles. Sample explanation here. Wikipedia has an outline
You can buy detailed plans for a couple of different recumbent trikes, and Ricky Horowitz's Thunderbolt plans are free and public, but still copyrighted. An open source design for a particular type of bike frame would be mostly useful as a collaborative design exercise. I suspect few people would build exactly what was described, so documenting the design process might be a more useful focus than simply describing the results.