Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
nodamage
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
39 ms
·
391.
▲
by
nodamage
6y ago
I'm not exactly sure how you think these examples justify the author's claim, can you elaborate? Cause it seems like what you're doing is the equivalent of pointing to an example of someone killing another person and then say
392.
▲
by
nodamage
6y ago
> Your security and privacy aren’t really protected inside these walls because the main point of these security measures is to enforce control, security against attackers and protecting your privacy is mostly marketing spin. The author
393.
▲
by
nodamage
6y ago
Are you serious? At this point I can't tell if you are trolling. This is a direct quote from the case I linked: Psystar alleges that this conduct has caused harmful and anti-competitive effects in the marketplace (Compl. ¶¶ 68-77). Ps
394.
▲
by
nodamage
6y ago
I'm not sure that group values the lack of side-loading specifically. I think what they do value is the additional security and privacy benefits provided by a platform that strictly controls the distribution of its apps. To the exten
395.
▲
by
nodamage
6y ago
> If this behavior is normal, then we need to make it not normal, not continue to compensate for their ineptitude. Fix problems, don't avoid them. This sounds great in theory, but two decades of history of malware on Windows have al
396.
▲
by
nodamage
6y ago
That's not exactly a ringing endorsement, is it? Sure, the malware situation on Android is better than Windows. It's still far worse compared to iOS. https://arstechnica.com/information-technology/2020/02
397.
▲
by
nodamage
6y ago
> The anti-trust agreement Microsoft signed is part of the reason for that massive boom in prosperity. This history doesn't sound right to me. Microsoft languished for a decade after that agreement, completely missing out on the mob
398.
▲
by
nodamage
6y ago
I mean, is it really that far out there if you look at the past two decades of history of malware on Windows and particularly Android? Sandboxing has not prevented the proliferation of malware on Android, why would it be any different on iO
399.
▲
by
nodamage
6y ago
Sandboxing didn't prevent fake versions of Fortnite and other serious malware from spreading on Android, nor does it generally prevent information theft and privacy violation through malicious use of legitimate APIs.
400.
▲
by
nodamage
6y ago
The problem is not everyone clicks on the right link. Non-sophisticated users looking for Fortnite don't even know what Epic is or whether or not they're the official place to get it. https://blog.malwarebytes.com/
401.
▲
by
nodamage
6y ago
> This argument doesn't hold water because it applies as much to the choice of phone as the choice of app store. That's silly. The number of people who might buy a specific no-name budget brand of phone with this problem is muc
402.
▲
by
nodamage
6y ago
There is literally malware spreading by masquerading as the Epic Games Store: https://www.zdnet.com/article/new-lokibot-trojan-malware-cam...
403.
▲
by
nodamage
6y ago
Don't the overwhelming majority of people access Facebook via the app these days? So the official source of Facebook for those people is... the App Store or the Google Play Store. > And I'd posit that gaining access to someone&
404.
▲
by
nodamage
6y ago
It doesn't sound like you actually understand the relevant aspects of the case. Psystar made multiple counterclaims, including a claim of tying between "Mac OS" and "Mac OS-capable computers". A tying claim requires
405.
▲
by
nodamage
6y ago
> The problems with letting all apps advertise external payment systems: I don't think this list is very compelling because external payment systems are already permitted for physical goods and it doesn't seem to be a huge prob
406.
▲
by
nodamage
6y ago
You seem to think if Epic fails to prove consumer harm exists, that their case will not survive summary judgement? This is not true. What Epic needs to do to survive summary judgement is to present a trialable claim that injury to competit
407.
▲
by
nodamage
6y ago
> I never had any malware I think you need to be careful about taking your experience as a sophisticated user that understands how to avoid malware and extrapolating that experience to the general population. There's a reason fake F
408.
▲
by
nodamage
6y ago
I think you're understimating the level of risk associated with people attempting to sideload Fortnite. People aren't necessarily intentionally seeking shady versions (but even if they were, I find this kind of victim-blaming coun
409.
▲
by
nodamage
6y ago
I think you massively underestimate the number of people that will click yes to install random packages on their own phones in order get access to free games or porn or whatever. https://research.checkpoint.com/2019/age
410.
▲
by
nodamage
6y ago
I have read the case in its entirety, see the entire section under B. Market definition: Alleged "Mac OS-capable computers" market . (It's true that Psystar's attempt to establish "Mac OS" as a single-brand ma
411.
▲
by
nodamage
6y ago
> The security in iOS is tied to the sandboxing model - not the app store. I disagree. Sandboxing is designed to prevent your phone from being compromised but it doesn't prevent bad actors from using legitimate APIs in malicious w
412.
▲
by
nodamage
6y ago
You are correct, the case was Apple v. Psystar : https://casetext.com/case/apple-2 Of particular note in that case is the court ruled that Apple's alleged "monopoly" over the market of "Mac OS-capa
413.
▲
by
nodamage
6y ago
I don't disagree that it's arbitrary, but the person I replied to implied Apple changed the rules specifically to block Hey's app, which is not correct.
414.
▲
by
nodamage
6y ago
I mean, presumably the people who really cared about side-loading decided to buy alternative phones instead. So, of iPhone buyers, you're left with two remaining groups: 1. People who care a little bit about side-loading but not enou
415.
▲
by
nodamage
6y ago
I mean, both contribute to the security of the ecosystem. There are plenty of APIs available within the sandbox that are meant to be used by developers to use for legitimate purposes. But those same APIs could be abused by malicious actors
416.
▲
by
nodamage
6y ago
The anti-trust claims that succeeded against Microsoft involved Microsoft forcing other companies to ship IE instead of Netscape. The coercion of third-parties is really where the anti-trust violations came from. Microsoft also held 95% o
417.
▲
by
nodamage
6y ago
It's not. People attempt to justify it by claiming "you will make more money if you sell on iOS", but there are plenty of Android-only apps that are doing just fine. At any rate, just because a platform is popular and will ma
418.
▲
by
nodamage
6y ago
Hey is not a reader app, so the rule for reader apps does not apply.
419.
▲
by
nodamage
6y ago
This seems like splitting hairs. At the end of the day, if the consumer buys an app for $10, the developer gets some portion of that, and the retail store gets the other. Whether that happens in one simultaneous transaction or two different
420.
▲
by
nodamage
6y ago
> having to get a EV code signing certificate ($300+/yr) to bypass windows smartscreen warnings, which are enabled by default Windows was a malware disaster before measures like this were added to the OS, so I would argue this (and
More ›