| ▲ | kelnos a day ago | |
AFAICT neither has any default for `undodir`; the user must set it explicitly. I think a user could be forgiven for assuming that it was fine to use the same value when trying out a migration from vim to neovim, and that neovim wouldn't actively delete vim's state data. Sure, maybe they should have backed it up or used a different directory, but I think it's reasonable to expect that many people wouldn't. If a software developer is only going to consider the scenario where their users are thinking/behaving exactly as they are, then they aren't particularly good at what they're doing. | ||
| ▲ | em-bee 19 hours ago | parent [-] | |
vim: Undo files are normally saved in the same directory as the file neovim: by default they are saved to the dedicated directory in the application data folder I think a user could be forgiven for assuming that it was fine to use the same value when trying out a migration from vim to neovim neovim would also overwrite its own old version of the file. the point is that the expectation that neovim should watch out for files that are used by other applications is in my opinion not reasonable. it won't even watch out for its own old undo files. users may have the expectation that neovim and vim are compatible, but that expectation is simply not entirely fulfilled. you can't share config files unless you have some minimal ones that just happen to work. so why would you be able to share other files? the question is, how are these expectations formed, and what is reasonable to expect? when it started out neovim was very compatible with vim. but from my experience with FOSS and based on the stated goal of neovim at least to me it was clear that compatibility was not going to last. not everyone shares that experience, and thus their expectations are different. neovim should perhaps make that more clear and for one be aware that people expect compatibility and warn users accordingly. an argument could be made that if a user changes the undodir value, there is a higher chance that they may try to use the same for vim. but then vim devs would need to do that too. but does not considering that make devs not good at what they are doing? i don't think so. for one, if someone thinks the devs are not good enough they should stop using neovim (as some people here in the thread announced they are doing. their choice). personally i think this is average for FOSS and good enough. we can't expect everyone to be perfect. the correct response to an issue besides reporting it is not to complain but to contribute a fix. | ||