4 ms·

Fwiw, I recently was forced to use a MBP at work after being full-on Linux for ~20 years. I find OSX to be irritating on a number of levels but coming from the

Fwiw, I recently was forced to use a MBP at work after being full-on Linux for ~20 years. I find OSX to be irritating on a number of levels but coming from the opposite direction. > My guess is that Linux developers don’t want to do boring stuff like fixing obscure bugs for certain peripherals, and they don’t want to listen to designers to create a polished unified experience. I think OSX offering a Unix-like experience drew away a lot of open source developers to donate their free time and expertise towards a commercial company like Apple rather than the community. If this had not happened, it's possible the Linux desktop experience may have kept up with the times better. Decade old coreutils is closer to traditional Unix than Ubuntu. Heck, OSX started with tcsh as the default shell. OSX without the magnanimous gift of Homebrew seems pretty garbage.

wahern 8y ago
Nah. Linux will always have poor ergonomics when it comes to device support simply because of the development culture in Linux, and particularly the Linux kernel culture: support every device under the sun in a way that minimally satisfies every stake holder. That means there are always too many controls and knobs. Device support stops at 80% completion (all the esoteric features working, but poorly integrated) because, theoretically, with all the knobs the end-user has the capability to take it the next 20%, and there's always something new around the corner that needs that 80% support. That's the Linux way and will always be the Linux way. There are two other ways: the Microsoft way which is to invest billions sustaining an OS and commercial developer ecosystem that does marginally better in terms of working well out-of-the-box. And the BSD[1], Apple, Chrome[2] way, which is to focus on specific hardware and specific use cases and outright reject everything else. Pick your poison. [1] BSD is a broad brush. NetBSD and OpenBSD are more focused than FreeBSD, which sits somewhere between those two and Linux. NetBSD is more developer focused (always emphasizing the perfect APIs and frameworks) while OpenBSD focuses on functionality, both server-side and client-side. AFAIU, if you buy one of the laptops favored by OpenBSD developers things will work very well out-of-the-box. The downside is that because of lack of manpower if they can't do it well they won't do it at all (e.g. the previous lack of 802.11n Wifi support for many years). [2] Obviously it is possible and viable for a vendor to complete that last 20% that Linux leaves unresolved, at least if they very narrowly focus their efforts. But Google is for the most part the only such success; there have been (and will be) many other attempts that will drown in the endless complexity and churn of the Linux ecosystem.

wahern 8y ago

Nah. Linux will always have poor ergonomics when it comes to device support simply because of the development culture in Linux, and particularly the Linux kernel culture: support every device under the sun in a way that minimally satisfies every stake holder. That means there are always too many controls and knobs. Device support stops at 80% completion (all the esoteric features working, but poorly integrated) because, theoretically, with all the knobs the end-user has the capability to take it the next 20%, and there's always something new around the corner that needs that 80% support. That's the Linux way and will always be the Linux way. There are two other ways: the Microsoft way which is to invest billions sustaining an OS and commercial developer ecosystem that does marginally better in terms of working well out-of-the-box. And the BSD[1], Apple, Chrome[2] way, which is to focus on specific hardware and specific use cases and outright reject everything else. Pick your poison. [1] BSD is a broad brush. NetBSD and OpenBSD are more focused than FreeBSD, which sits somewhere between those two and Linux. NetBSD is more developer focused (always emphasizing the perfect APIs and frameworks) while OpenBSD focuses on functionality, both server-side and client-side. AFAIU, if you buy one of the laptops favored by OpenBSD developers things will work very well out-of-the-box. The downside is that because of lack of manpower if they can't do it well they won't do it at all (e.g. the previous lack of 802.11n Wifi support for many years). [2] Obviously it is possible and viable for a vendor to complete that last 20% that Linux leaves unresolved, at least if they very narrowly focus their efforts. But Google is for the most part the only such success; there have been (and will be) many other attempts that will drown in the endless complexity and churn of the Linux ecosystem.