Live data from Hacker News

What killed the Linux desktop

tirania.org

231–240 of 385 posts

Re: What killed the Linux desktop

#231
post #170

Earlier quoted context omitted.

I bought Samsung Galaxy S3 recently, plugged it in via USB and was not able to browse it because, as far as I understood, MacOSX has no support for "media devices" whatever that is. I had to download some obscure Samsung "Kies" software through which I was able to get to the filesystem and upload some files. Wasted hours.

Oh my This is 100% to blame on Samsung, sorry. Even though my Samsung phone is seen as a USB device (and it works on MAC OS X), but maybe in the newer models they removed this functionality (and called it a feature) People may complain that iPhones need iTunes but then again it's iTunes not the gigantic pile of crap that is Kies

No it has nothing to do with Samsung. In ICE Google changed how the partitions are handled: http://forum.xda-developers.com/showpost.php?p=11714077&...

It was pure engineering tradeoff to improve things in the long term. If anything OEMs would have preferred the old 'just works', sub-optimal, gives-an-excuse-to-obsolete-a-phone, lower support issues, solution. The ball is now in operating systems' court to support this standard in a way it wasn't envisioned to work a few years ago.

Even ubuntu has been slow on this front.

Re: What killed the Linux desktop

#234
post #164
post #128

Earlier quoted context omitted.

> It's silly to pretend that Mac is somehow perfect in these matters. Straw man. Nobody said OS X was perfect. > In my experience it's only been marginally better than Linux, if at all. I used Linux on my desktop for several years and have now used Macs as well for several years. I won't say this is bullshit because I don't think you are lying about your experience, but I think you are extrapolating way too far. > An…

I don't see any material difference between "hardware compatibility just works" and "hardware compatibility is perfect". As for fixing the issue, both of my issues were kernel issues, as are most hardware compatibility issues. On OS X reverting to a previously working kernel would have meant backing up all of my data, reinstalling the OS from the DVDs, installing the combo update to the last working version of the co…

Since you mentioned using Time Machine, why not revert to a good state using that (instead of a re-install)?

Re: What killed the Linux desktop

#235
post #170
post #127

Earlier quoted context omitted.

Hardware? What hardware? Frankly, I havent bought nothing over last 10 years. I have laptop, you see...

I bought Samsung Galaxy S3 recently, plugged it in via USB and was not able to browse it because, as far as I understood, MacOSX has no support for "media devices" whatever that is. I had to download some obscure Samsung "Kies" software through which I was able to get to the filesystem and upload some files. Wasted hours.

This is not an issue with Apple or Samsung, it's an Android decision.

Support for the USB Mass Storage protocol has been deprecated in Android in favor of MTP (Media Transfer Protocol) starting with Ice Cream Sandwich.

The experience in Windows and on Linux is just as poor with this phone. (I love the phone, though.)

Re: What killed the Linux desktop

#236
I more or less disagree. My main frustration with Linux For Personal Use is that I can't buy a piece of hardware that I know won't regress with new versions of a distribution for three plus years or get any service if it does. My reference for the importance of this is a perfectly usable 2008 refurbished Macbook. I upgraded the RAM once recently for a bit more snap, but otherwise have no complaints over the three or so Macintosh releases since then.

Could the UIs and third party application situation be better? Of course. But considering all the garden variety crash bugs, power management bugs, lockup bugs, video driver misbehavior, hit and miss peripheral support, and in general just analysis paralysis about what hardware I should buy, and even then there is a less-certain future with regard to regressions.

Even given Windows's monopoly power in the commodity desktop and laptop markets, its reputation for dealing with sleep and drivers is only so-so compared to Apple Hardware and Software. If Window's monopoly power -- which buys you full attention from hardware manufacturers and their driver divisions -- only gives you mediocre results, what are the odds that a bunch of kernel hackers who receive almost no continual consideration from hardware vendors have a chance? To me, it looks like absolutely not a chance of becoming stable over time. I have completely given up on Linux laptops for this reason: by using desktops with Linux only I have cut out a lot of the problems, but not all of them. It's a kind of medicore that I can bear.

I want someone to sell me Linux distribution on a laptop that simply will not break over in its kernel-oriented features in five years of upgrades. I want that distribution to stop-ship if it a new version introduces a power management bug to an old laptop, and do whatever it takes to work around some lousy hardware bug or whatever. I want them to do whatever to work with Skype (such as statically linking whatever libraries, etc) and test Google Hangouts to make sure the webcam and microphone works. And it they don't work, they absolutely cannot ship. Until that day, I use Linux -- and I do mean the kernel in most of these cases -- as my personal operating system most of the time in spite of these problems because of my professional and philosophical needs, and not out of preference in any other dimension.

Re: What killed the Linux desktop

#237

Earlier quoted context omitted.

> It's that this is the correct way to do it. Except when you have multiple monitors. How is it the correct way to do it to force me to mouse 4000 pixels to the left or right to get to the menu for an application that isn't on the same screen as the menu? There's never just one correct way to do things. Your arrogance towards other implementations is overwhelming.

I'm sorry that you consider citing the science behind the decision "arrogance", but I suspect its just desperation to engage in ad hominem.

The arrogance comes in the way you deliver your argument and insist that your preference is the only valid choice, regardless of any difference in conditions; it has nothing to do with the science.

I took the same Human Factors class that cites the same decades-old studies that you're pushing as ultimate truth; all of that science was done when multi-monitor setups weren't even supported, let alone common. I have no problem agreeing that a universal menu is optimal for certain sets of conditions, but that doesn't mean it's the one true way for everything.

Re: What killed the Linux desktop

#238

Earlier quoted context omitted.

Maybe collaborative development is incapable of producing something as complex and difficult as a clean, well-designed desktop environment and API? Forking, competing spinoffs, factionalism are all useful things in some ways, but they're death to this kind of project. The Rails community seems to manage.

Well, Rails is a web framework. It's not an end user desktop environment like Gnome, KDE, NeXTStep, Windows or what have you. Collaborative development does rather well with developer tools.

Sure... so why the culture difference?

Re: What killed the Linux desktop

#239
post #97

This is the money quote: > The second dimension to the problem is that no two Linux distributions agreed on which core components the system should use. Linux on the desktop suffered from a lack of coherent, strategic vision, consistency and philosophy . Every engineer I know likes to do things a particular way. They also have a distorted view on the level of customization that people want and need. I like OSX. Out o…

Coming from the perspective of a developer:

I used linux as my only OS sans VMs from roughly 2008 till 2011. I finally gave up because it handled monitor switching, multiple monitors so poorly (using Nvidia GPU). I begrudgingly switched to windows because I owned the hardware already, and it was the only option. Windows sucks, the lack of a usable shell kills it, full stop, nothing else windows does matters. And no, cygwin doesn't cut it.

Fast forward to now, the only computer I manually interact with is a macbook air. The old desktop hardware became my file server(running FreeNAS) despite being many times faster than the air.

Through this experience I realized what I actually cared about...I run four 'apps' 99% of the time: chrome, sublime text(everything non java), iTerm and IntelliJ(java). That is really all I care about, anything that prevents interacting with these applications as quickly as possible fails. The support stuff, git, a database, a message queue, etc. Runs anywhere. I _rarely_ interact with the 'OS' in ui terms. I rarely use finder, I only use the dock to empty the trash (launch everything through Quicksilver). At the end of the day I love OS X not because it has 'awesome' UI paradigms, but because of all the options I see it the least, it may as well not be there.

OS X is the only OS i've ever used that I just don't think or care about. It sleeps, it wakes up, it deals with new screens, it does all the OS shit so I can just use the apps i want to. Thats why I am hooked using it as a dev machine.

Re: What killed the Linux desktop

#240

Earlier quoted context omitted.

Maybe collaborative development is incapable of producing something as complex and difficult as a clean, well-designed desktop environment and API? Forking, competing spinoffs, factionalism are all useful things in some ways, but they're death to this kind of project. The Rails community seems to manage.

Rails has a pretty strong leader in DHH.

But there are strong personalities in the Linux community as well.
Post reply on HN