> But still, Emilio, the reason we on purpose fail on missing
> XDG_RUNTIME_DIR is that there is no suitable "replacement directory
> with similar capabilities" to be found in general.
>
> For example, falling back to anything under $HOME will fail for
> nfs-mounted home, I believe.
>
> The requirements for XDG_RUNTIME_DIR are actually quite difficult
> to fill without specific distribution support. Especially the following:
>
> "The directory MUST be on a local file system and not shared with
> any other system. The directory MUST by fully-featured by the
> standards of the operating system. More specifically, on Unix-like
> operating systems AF_UNIX sockets, symbolic links, hard links,
> proper permissions, file locking, sparse files, memory mapping,
> file change notifications, a reliable hard link count must be
> supported, and no restrictions on the file name character set
> should be imposed."
>
> - http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html
>
> Therefore we want the user to explicitly shoot himself in the
> foot if they have to. We actually used to have a fallback for
> XDG_RUNTIME_DIR being unset, and it only caused confusion.
Explanation for current behavior by pq here: http:// lists.freedeskt op.org/ archives/ wayland- devel/2013- March/007902. html
Contents:
> But still, Emilio, the reason we on purpose fail on missing standards. freedesktop. org/basedir- spec/basedir- spec-latest. html
> XDG_RUNTIME_DIR is that there is no suitable "replacement directory
> with similar capabilities" to be found in general.
>
> For example, falling back to anything under $HOME will fail for
> nfs-mounted home, I believe.
>
> The requirements for XDG_RUNTIME_DIR are actually quite difficult
> to fill without specific distribution support. Especially the following:
>
> "The directory MUST be on a local file system and not shared with
> any other system. The directory MUST by fully-featured by the
> standards of the operating system. More specifically, on Unix-like
> operating systems AF_UNIX sockets, symbolic links, hard links,
> proper permissions, file locking, sparse files, memory mapping,
> file change notifications, a reliable hard link count must be
> supported, and no restrictions on the file name character set
> should be imposed."
>
> - http://
>
> Therefore we want the user to explicitly shoot himself in the
> foot if they have to. We actually used to have a fallback for
> XDG_RUNTIME_DIR being unset, and it only caused confusion.