On Thu, 29 Mar 2018 22:37:12 +0200, you wrote: >The one thing that speaks for it is size: it is only ~4 MiB xz-compressed, >whereas a typical font for any single CJK language (which may or may not >have the same limited support as described above for the other 3 languages; >often, there are no characters at all for the unsupported languages) is >20+ MiB xz-compressed. At the time where the decision to ship it on the KDE >Spin was made, the choice was between WQY MicroHei or no CJK support at all. I think looking into size increases is useful, if for no other reason than to ensure mistakes don't get overlooked. But I also think worrying about an artificial size limit, a limit that made sense in the past, is damaging. In the past arguments have been made that we have to keep it small for those with limited and/or expensive bandwidth. While a noble goal, it is unfortunately something that is only paid attention to when creating the live images and then immediately forgotten. I just started up one of my Fedora vm's, which was last updated 3 days ago. In 3 days I now have 50 packages to update/install with a total download size of 171M. So in 3 days I have accumulated downloads totally half the size increase causing all this angst. While certainly not perfect, it roughly gives me 10G of updates to download and install over a 6 month life span of a typical Fedora release. Consult the relevant experts, and based on their recommendations mandate a base set of fonts that provide a quality first experience with Fedora for everyone regardless of where they live and what language they read/write. Making compromises to save 400M on a distribution that will need 20+G over year doesn't seem very wise. _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
