231–240 of 348 posts
The article says that it's the same format as Chocolatey (an existing Windows package management tool) but actually it's a fork of NuGet in Windows 10 and Chocolatey is also a compatible fork of NuGet. Microsoft already owns and maintains NuGet.
Earlier quoted context omitted.
>Android doesn't have that Android has Nexus and Google Play Edition phones.
The Nexus comes with some HP printer crapware that can't be removed without rooting (My Nexus 5 did at least)
Earlier quoted context omitted.
So, what exactly is your definition of a desktop system? If we're going to have this discussion we should get the agreement or disagreement over that definition out of the way. To me, it is a desktop computer. I'm inclined to include laptops in that definition as well. Basically, it's the kind of computer I'd use to get my computer work done at a desk, with as little regard to what exactly that work entails as possib…
So, if I have a keyboard on my tablet that is running Android? I think the point here is that "desktop" as a location has no real meaning - it's the use case that defines it rely. By desktop system one really means a day-to-day used system that is used for word-processing, web browsing and other standard domestic/work uses (photo editing, spreadsheets, games and such). That's the stuff a desktop system has always don…
What exactly is a logical difference in this case, and how is it relevant to the discussion? Is my decision illogical if I consider the ergonomic implications of using a particular hardware/software platform?
Aside from ergonomics, I'm not sure what is so controversial about assuming that the different user interfaces will (and quite obviously do) cater to different use cases. What I can realistically do with a keyboard and a 24" monitor is certainly different from what I can do with my mobile phone. What I can realistically do with Debian is certainly different from what I can do with iOS or Android. These differences alone convince me that the distinction between desktop computer and phone/tablet/crablet is still meaningful, regardless of the literal meaning of "desktop computer". Manufacturers, retailers and consumers generally understand this, no matter how much anyone pretends that a tablet is equivalent to a desktop computer.
> "year of Linux on the desktop" People have been using [insert any relatively obscure OS] as a desktop OS for plenty of years. The fact remains that Linux doesn't hold a significant share of the desktop market, which is what the idea of "year of Linux on the desktop" has always been about.
Earlier quoted context omitted.
Nice... For those who've never heard of him: he's also the main developer behind Coapp http://coapp.org/
Do you have a positive experience with CoApp? Mine has been overwhelmingly negative: it does not appear to work, does not appear to be maintained and does not appear to be supported. I've been trying to get somebody to provide even a little bit of support for Native Nuget packages via CoApp for months : http://stackoverflow.com/q/24331880/729881
Obviously it was not ready yet, but I wasn't really let down: I seldom use Windows, so I don't have any real need for it, and my perception was that with some more months/years it would get there.
But CoApp and OneGet are worthwhile endeavors, so I'm cautiously optimistic that they'll eventually build something that can make software installation on Windows less painful
PS: uh, I just realized that you're a Microsoft developer as well (just like Garret Serack)... is it so difficult to get hold of him even inside the same company? (I don't know... maybe you're actually on different sides of the ocean, and this would make the matter quite a bit more complicated)
> and who knows, that might just trigger some kind of revolution in Windows app management Free trials of un-tar? Installing apps that require subscriptions to the cloud? Dev libraries that require enterprise support packages? Good package management on Linux is owed largely to the tireless voices such as Stallman who understand the core issues here. Yes, tying together install scripts and maintaining repositories re…
> Good package management on Linux is owed largely to the tireless voices such as Stallman I thought it was due to the people designing the package managers. The people who run the repos Stallman-like end up shipping IceWeasel and Chromium and telling long boring stories to someone who just wanted Firefox and Chrome.
Important things like good and efficient dependency management and library reuse are far easier to achieve when all the relevant sources are available to packagers.
IceWeasel and Chromium visually are Firefox and Chrome with different icons and you'll find them if you search their "real" names in the package manager. Not so hard, right? You can find about why they exist if you are interested, it's not required.
Earlier quoted context omitted.
The Nexus comes with some HP printer crapware that can't be removed without rooting (My Nexus 5 did at least)
I was confused by that, too. But it's hardly on the level of Windows PCs.
Earlier quoted context omitted.
http://www.android-x86.org
I'll give you points for tenacity, but Android was never intended to be used on desktop computers. The link you give is just a hobby project, nothing more. And since we're being pedantic here, let's talk about definitions. A mobile operating system is one that has been designed primarily for devices intended to be carried on one's person, such as a phone or tablet. A desktop operating system is one that has been desi…
And one of Android's original use models was a non touch screen connected to a keyboard (mainly for Blackberry style smartphones before they were sure that slates would take off, but still...).
Earlier quoted context omitted.
More like C: \> instpkg32.exe /DefaultSettings:ALL_DEFAULTS C: \> Cryptically named command with verbose arguments, that then goes on to fire a GUI window that requires manual intervention: that's the Microsoft way.
Good job at not reading the article! Had you read it, you would've learned that the commands work like this: Find-Package Firefox Install-Package Firefox I admit it's not as easy as on e.g. Debian: sudo apt-get remove iceweasel sudo echo deb http://packages.linuxmint.com debian import >>/etc/apt/sources.list sudo gpg --keyserver pgp.mit.edu --recv-keys 3EE67F3D0FF405B2 sudo gpg --export 3EE67F3D0FF405B2 > 3EE67F3D0FF…
Earlier quoted context omitted.
I have a really hard time believing that.
For the purposes of this reply, I'm talking about packages as distinct from Windows itself. A reboot is required if something you are trying to install is already in uss. Windows locks executable files and dells while they are being used, so it's not possible for an installer to overwrite them. When an installer detects this it places the new file in a temp store, and windows empties this store on startup, Thus if yo…
Exactly, and Linux doesn't do that, hence my comment. The Windows model is flawed--why do you need to lock a binary on disk when a copy of it is running in main memory? Linux just lets the installer overwrite the files on disk, so there's no need to restart the whole OS, just the program whose files were updated.
I assume Windows 10 will not change this behavior? A command-line package manager on Windows would be cool, but its utility will be limited if you still need to restart the whole OS just to upgrade a program that's currently running.