I think having an open source application is useless to the end user. I mean really useless in and of itself. It's not useless. The end user does benefit from

I think having an open source application is useless to the end user. I mean really useless in and of itself. It's not useless. The end user does benefit from it, just not directly. They benefit from the various modifications and enhancements under copyleft that are either integrated upstream or make it to other projects that they may use. They're also unhampered by usage or redistribution restrictions that may potentially get them into legal quagmire from more zealous vendors. What really matters is standarizing the data format and providing tools to convert/export to various other formats. Data formats are software, too. Other than the spec that standardizes them, there is a reference implementation which should preferably be free. In fact, being free is crucially important here: it means any extensions to the format are sent back, instead of a vendor pulling off an "embrace, extend, extinguish" by adding non-standard features to their proprietary implementation which they then lock other software to. Also suppose that X hasn't been in development for over a year and no one is maintaining it. Then the project is orphaned. Someone may or may not step up. The good thing is they can. Also suppose X is very complicated and no one understands how the code works. And yet the code is there, so it's not insurmountable if the utility is high enough to get past the bus factor.