4 ms·
I admit and apologize for salty tone above. It is not very constructive, and I agree, and would like to elaborate. This isn't the first time such thing has hap
by e3bc54b2 5y ago
I admit and apologize for salty tone above. It is not very constructive, and I agree, and would like to elaborate.
This isn't the first time such thing has happened and it won't be last. Windows has becoming more and more painful as Operating System as in operating the machine and has been trying to operate the user instead.
The incidents I mentioned about Winmerge and Kdiff3 did happen with me and I stand by that. Above comment does not quote anyone but convey my understanding of the replies in general, which I now understand to be less than perfect.
But,
> unknown software -> don't know whether it's safe -> warn users it's unknown
You must be able to realise that it is not exactly presented in such manner as to be "unknown". Unknown would simply mean windows doesn't know about it and its upto user's discretion. Microsoft, by virtue of all the telemetry, happen to know that this application is uncommon. Why not say that? Why not let the user decide whether they want to run an uncommon app? What is happening here instead, however, is it gets presented as "dangerous" and blocked by default. This, along with Microsoft's position ends up being (in my humble opinion) an abuse of powerful position. And that, again, in my own humble opinion is unacceptable in being defended.
- ethbr0 5y agoTBH, Windows shouldn't even be in the business of having any opinion on uncommon software. If they want to collect data sets for virus protection, go ahead. If they want to hash installed files from common software to detect divergences for virus protection, go ahead. But having any kind of notification that an install is uncommon? We call it a general purpose computer because we want to use it for general purposes: which is to say, uncommon purposes. This feels like ape'ing Apple without thinking it through.
- gruez 5y ago> But having any kind of notification that an install is uncommon? We call it a general purpose computer because we want to use it for general purposes: which is to say, uncommon purposes. How much % of the windows install base do you think is running random programs that they downloaded off github, and how many are just running the top 1000 programs (eg. chrome/office/winrar)?
- ethbr0 5y agoDo we want to live in a world where someone running something different than everyone else on their computer is treated with suspicion? This seems like a very slippery slope towards the iPhonization of the PC.
- gruez 5y ago> Do we want to live in a world where someone running something different than everyone else on their computer is treated with suspicion? It depends on what you mean by "suspicion". Should users be wary and not blindly run uncommon binaries they got off the internet, on a unsandboxed system? I don't see why not. I'll even say that most developers are not exercising enough caution when downloading random packages off npm.
- ethbr0 5y agoIsn't this already what Windows does? It's warned about unsigned binaries on first run for a while now, no? And I can see some argument for that, even if in practice it feels more like teaching users to blindly ignore warnings. But holding uncommon or un-Microsoft-sourced (that is, signed) to a higher standard feels wrong.
- gruez 5y ago> Isn't this already what Windows does? It's warned about unsigned binaries on first run for a while now, no? Sort of? There's three I know of 1. the generic warning for files you downloaded off the internet 2. the UAC warning when you try to run any program as admin 3. the smartscreen warning for uncommon files. The first two has the "run/open" equally as visible as the "don't run/cancel" button. The last one is the one where the "run" button is hidden. >And I can see some argument for that, even if in practice it feels more like teaching users to blindly ignore warnings. That's exactly the problem. The first two warnings show up for everything, so users are trained to click through. >But holding uncommon or un-Microsoft-sourced (that is, signed) to a higher standard feels wrong. The problem is that without a digital signature, you can't tell whether a binary from a legitimate developer and a malware developer. Hence the need to rely on file hashes and needing to warn users for uncommon binaries.
- UnFleshedOne 5y agoHave you tried running anything on new Apple architecture -- even running your own apps built locally works half of the time, and that's _with_ the full moon out and correct number of goats dully sacrificed...
- gruez 5y ago>You must be able to realise that it is not exactly presented in such manner as to be "unknown". Unknown would simply mean windows doesn't know about it and its upto user's discretion. Microsoft, by virtue of all the telemetry, happen to know that this application is uncommon. Why not say that? Why not let the user decide whether they want to run an uncommon app? But seems pretty close to what they're doing? I don't feel like downloading random shady programs to get the UI to show up, so I'll base my analysis off the screenshots from this page: http://www.rawinfopages.com/tips/2014/10/unblock-programs-blocked-by-windows-smartscreen/ http://www.rawinfopages.com/tips/2014/10/unblock-programs-bl... 1. the prompt says the app is "unrecognized" 2. the prompt says running the app "might put your PC at risk". I'm presuming that's what you meant by "dangerous", but that's not exactly the same thing. 3. The default option is indeed "don't run", and you do have to click on the non-obvious "more info" link for the "run" to show up, but this seems like a reasonable trade-off. Otherwise users might instinctive click "run" and ignore the warning. >This, along with Microsoft's position ends up being (in my humble opinion) an abuse of powerful position. And that, again, in my own humble opinion is unacceptable in being defended. In other words, "I'm so confident that I'm correct and microsoft is so powerful that I think it's unacceptable for people to present arguments to the contrary?