> Sadly, these groups are small groups with eclectic needs that won't be met or improved by real systems engineering with the kernel and core modules. What do I mean by real systems engineering?
My fourth group isn't small (though perhaps shrinking due to the popularity of tablets), but also has no need for "real systems engineering". My question is this: who does need that stuff? Isn't it really just administrators of high performance servers?
Don't get me wrong, I plan to have a lot of fun and learn a lot from reading up on all the things you mentioned that I either haven't heard of or know of only as buzzwords, but I just don't see how any of it is relevant to most peoples' interaction with computers. Selling "we changed some superficial design elements in the window manager" really does make more sense.
At least my Mac recognizes and uses a second monitor when I plug it in, unlike the bloody Fedora 19 box that I use at work. And don't get me started on that second graphics card that the system can't figure out how to use! And the wifi Just Works(tm), unlike the Ubuntu laptop that I had before that.
Well Linux gets better battery life on my laptop than OS X. (11 hours with ubuntu gnome 14.04).
> my Mac recognizes and uses a second monitor when I plug it in
Not mine. And this is a common problem: http://www.imore.com/os-x-mavericks-problems-drive-me-nuts-h...
Also, I had issues with sound going out and not returning until after a reboot on OS X. This drove me nuts.
Thing is your points only apply to OS X if you run it on Apple hardware. If you get ubuntu certified hardware you will have no problem running ubuntu either. Ubuntu is going to run on more devices with less issues simply because the linux kernel is designed to target more hardware than the OS X kernel.
This one bothers me endlessly. I got my first Macbook in '09 and have had two others since, and every single one of them has had this problem. I just don't get it.
Meanwhile Homebrew is a poor substitute for aptitude and PPAs, multi-monitor support is relatively poor (xmonad excels here), and the Apple keyboard is missing a bunch of keys I use on a regular basis.
Wifi constantly needs manually disabling/enabling, Bluetooth Audio crashes are common, I average one full system crash per month, display settings reset themselves magically and Finder is infuriating. I should be keeping a log of all the issues I have.
This is on a 2014 Macbook Pro.
On Mac things like Alfred (probably will drop this for the new spotlight), better touch tool, some capabilities like ability to swap USB MIDI controllers without restarting OS are cool. Honestly though, performance wise there isn't that much difference between my Mac and Win Machines and being in programs full screen most of the day I notice much difference at all. The really flexible information moving things I've grown accustomed to on Win seem a bit crippled on Mac because of lack of a filesystem.
I'm back on 10.8.5 and not in a rush to upgrade again (because Mavericks crapped itself and I had to restore from Time Machine)
I'm now on Debian 7 with XFCE on an old Think Pad. Life is better here.
Recognizes all monitors and projectors (even the ones Macs fail at) at my work place. The WiFi just works (TM).
You are not using a "Ubuntu laptop". As danford said, just pick a Ubuntu certified machine* and things will just work (TM). And, you will still have more choice than OSX ;)
demallien is right. Nobody on the Linux side has done this kind of aggressive power engineering for consumer-oriented x86 hardware. Google might, for Chromebooks, but it's not clear how portable that will be to the rest of the ecosystem.
"Hours" is a really poor way to gauge these things, because while it may seem practically applicable , we're really comparing apples and oranges.
My T420 before my T420s had a battery run-time of 15 hours. I could fail to mention that it had a slice battery and an ultrabay on top of the primary battery and tell everyone that the particular model and OS ran for 15 hours straight, but that would be disingenuous. If I was a marketer I could run the processor at lower voltages, dim the screen unread-ably, turn WiFi off, etc etc until I got whatever numbers I wanted.
As i'm not a marketer, I also feel as if the claims people make about Apple's hardware being better at power management are also somewhat dishonest.
They are a company that has ultimate say in their hardware platform and software ecosystem. No other computer company locks down production that much in an effort to increase the individual experience of each consumer. That lock-down allows for an amount of control that would be unrealistic to expect from the FLOSS community, as they have little power or say in production and ultimately must code to support a wider range of people rather than a narrow set of known hardware packages that are available.
My point : optimizing narrowly for a single platform is much easier than the opposite; a more fair comparison of battery life would be watts-per-hour usage during a standardized testing/benchmarking suite of some sort, along with the watt hours of battery capacity(although historically handset manufacturers have sidestepped that issue by having the software recognize when it's being benchmarked..). "Hours of use" could be practical if it wasn't arbitrarily inflated or deflated by uncommon usage profile testing and specialty optimizations for the sake of marketing a bigger number.
It's not impossible to get good numbers out of a Linux build, it's just fairly difficult for someone unfamiliar with the system (and sometimes impossible without doing the work yourself with certain hardware stacks). There exists a whole subspecies of Linux users that are somewhat akin to hypermillers who turn this kind of dilemma into a game and compete for lowest wattage.
The machine is dual boot as I'm using it to develop software that targets both OS X & Linux. I run a Gentoo with all kernel- and other power saving features enabled, on a 2nd partition.
The time I get out of the battery on OS X vs Gentoo is pretty much the same.
What I do is "surf" (FF, mostly text/reading docs, flash blocked etc.) and code (I do a build roughly every 5-10mins that will probably touch 10 sources max).
Build environment matches except for compiler being used (gcc 4.1 on Linux, latest LLVM gcc frontend from XCode on OS X but I would think this is minor).
Bottom line for me: Linux is on par with OS X in terms of time you get out of a MBP, in my case specifically.
Caveat: I dunno ofc, how true this is, generally, if one ran something like a pre-configured Ubuntu kernel.
IntelliJ however eats some of my battery pwoer, but I still manage 7 hours or full work day.
I'm using Crunchbang on that laptop now and it's mostly better.
That's a lot actually. Current MBAs are around 5 watts with full screen brightness.
For what it's worth my debian wheezy laptop detects second monitor just fine (and I am using awesome-wm) while my 2007 iMac randomly lose internet though it's connected to my routeur via wifi. So... data points.
This is mostly a solved problem in linux these days. Macbooks have consistent and quality hardware. It's no surpised they work well. If you use a mainstream laptop or PC parts, it will 100% work with Linux. Using cheap noname parts or manufacturers with poor software support then you'll have issues.
For example, Macbooks + Fedora 19 work perfectly. So does the 3 Lenovos I tried and a Dell XPS.
You can write a udev rule for that
Yes, please, let me spend time fixing distro BS instead of, you know, working.
I don't remember any time when Wifi worked out-of-the-box in my previous Linux laptop. It worked if I did things manually (wpa-supplicant and stuff), the NM widget? Never worked and made things harder
Another issue: opening and closing the lid doesn't have a consistent result, and sometimes it just keeps the laptop running.
This is for the basic functionality, and sometimes even things that worked before stop working.
Older distros had to be more manually configured, but when that was done it was good to go, today it is even harder to shut down the non-working stuff.
So yeah, it's Mac for me from now on. Recovery from Time Machine needed some "Linux style hacking" once, but apart from that it's a breeze
So, I guess, per your definition, that Linux geeks can school Apple today.
I didn't notice how quickly until I switched back to a Snow Leopard install I maintain for Final Cut Studio (don't get me started on FCP X). All of a sudden the old OS felt...outdated. Totally reliable, sure. But not quite as friendly.
One unexpected change is that Safari has become my favorite browser. Its integration with the OS and iOS, plus the integration of iOS and OS X gave it an edge that Chrome and FF can't match. Judging from what was revealed today, it's looks like that kind of seamless transition between devices is what we can expect a lot more of.
More recently I saw massive power savings by switching browsers. I love Chrome, I love its UI, I love its web devel tools. But even when I only have a few tabs open my Mac's fans just won't stop spinning. :( I kind of swapped back and forth between Safari and Firefox since then, but mostly settled on Firefox. I may give Safari another chance when I install Yosemite.
I haven't felt the urge to check out the release notes for a Linux kernel release in years. Wake me up when they get VM hooks to speed up GC or something. The new developments just aren't that relevant on a client machine.
It's not how I want to operate either, but that's the direction Apple has been heading to for a long time
[1] http://m.techradar.com/news/software/applications/apple-chal...
- silly mixed-case directories with spaces
- case-insensitive filenames
what negative aspects of the filesystem would affect an 'intermediate' user (e.g. a high-level hacker) on a day-to-day basis? Admittedly, those two are very frustrating, but I'm struggling to think of any others.
That and permissions seem to get screwy at times. I had a long running issue where I could not launch JAR files, found others with similar issues and finally figured out, I no longer had permission to my own user folder and the JAR I was launching needed to create a working directory. Ended up solving the problem for more than myself.
I still try to figure out the desire for case sensitivity in about anything
Just one example:
Mac: > Full disk encryption and home folder support is, again, hack-ish, and largely built on work other people did.
Windows: > Full disk encryption improved - though home folder encryption is still tragically stuck with NTFS "EFS" support (lackluster, at best.)
Somehow this supposed to show that Windows is doing it better than Mac?
I think you are focusing on features that are unimportant to most users. People hated Vista because it was so slow, clunky and crashed. I remain amazed that between clicking a folder an seeing a file within it can take about a minute on my friends old Vista machine. I'm also pleasantly surprised on my MacBook that it pretty much never freezes and resumes from sleep in about 2 secs with going wrong unlike any windows machine. What counts for users is how the machine works, not whether the underlying implementation is trendy, I think.
I dropped $1200 on a Vista laptop (Inspiron 1520) back in 2006/2007. It wasn't slow (Kubuntu on the same HW was faster though) and I don't recall it ever crashing. Sleep worked reliably although it was more on the order of 30 seconds with a spinning disk. I very much preferred it to XP at the time mainly for the the start menu improvements. I think a lot of the Vista issues - and to a larger extent, complaints about Windows - come down to subpar hardware. It's my suspicion that if you pay less than $1000 for a laptop it's probably not very good except on paper.
I am still using that laptop today - to write this post in fact. About a year ago I replaced the harddrive with an SSD. I can't speak about Vista performance since I installed 8 (now 8.1 Update 1 or whatever nonsense MS calls the latest Windows), but sleep is extremely fast; the instant I open the lid everything is usable.
OS X sort of lacks in default command line tools, but that can be fixed rather easily. I also feel that as Windows 8 compared to OS X loses easily both for Design and Usability, Windows 7 came closer. There are obvious limitations to the approach of all desktop operating systems and for that matter the whole "desktop" and "application" metaphor, compared to systems like the Lisp Machine. Ideally a power user would have access to a fully programmable system, without prepackaged applications and type directed interactions.
* Students looking for a stable and/or sexy device for their university.
* Artists using applications that don't exist on other platforms (or that prioritize Mac platforms).
* Hackers that want a Unix-like system with a BSD userland, or want Anything But Windows on a laptop.
Sadly, these groups are small groups with eclectic needs that won't be met or improved by real systems engineering with the kernel and core modules. What do I mean by real systems engineering?
* Major kernel development
* Novel and/or modern filesystem support
* Fundamental or deeply integrated "platform" features
In Linux land, every major kernel release brings these features. There are tangible improvements to filesystems, to core features that enable new things to be developed on top of them in every Linux release. These are, largely, absent on OS X. The system is too closed, and the result is that things like Time Machine or even security features and ACLs are hacks upon hacks. Full disk encryption and home folder support is, again, hack-ish, and largely built on work other people did. OS snapshots is essentially an "rsync" to another drive with a smart "restore" utility that repairs changes.
The major kernel development that Microsoft undertook with Windows Server 2003 and Vista is still paying dividends. Folder shadow copies became integrated into fully consistent backups with built-in snapshots. Full disk encryption improved - though home folder encryption is still tragically stuck with NTFS "EFS" support (lackluster, at best.) UAC, AppLocker, and integrity levels brought foundational improvements to the security model. Networking stack changes brought DirectAccess, a woefully underused and under-marketed technology. Storage Spaces and ReFS, though years too late, are interesting alternatives to ZFS/BTRFS. Transactional NTFS was woefully underused, but maybe it will return with ReFS. The core improvements Microsoft is making to the NT kernel are still worthwhile, though. Hyper-V is a fantastic technology, and could really blow people's minds when it's baked into the client OS. (For reference, Hyper-V powers the Xbox One's dual app/game personality. It allows isolating the management OS from games running on it, and also keeps the management OS from interfering with game performance with resource limiting. And they both share high performance access to the GPU.) Ah, I could go on. Reading about new stuff in kernel development is a joy.
Of course, I could go on ad nauseum about Linux changes since 3.0x, but http://kernelnewbies.org/ does a better job than I will.
The result is tragic: Microsoft invested in platform features and then didn't sell users on what it could do with Vista. Apple continues to apply lipstick to the OS X pig and sell users on changes to the window manager and built-in applications.