This only would work if you check every line of source code, even the dependencies and build chain, and then build it yourself. See xz utils backdoor or heartbleed, etc.
If we are talking about bigger projects with hundreds of thousands or millions of downloads, than this may be true. But smal scale projects have so few people actively looking through them that even to automatic scan done by the playstore has a higher chance of catching malware. It doesn’t even have to be bad intent, two years ago there was a virus propagating trough the Java class files in minecraft mods which reached the PCs of quite a few devs before it was caught.
I don’t dislike FOSS, a lot of the apps I use come straight from github, but all this talk about them beeing constantly monitored by third parties is just wishful thinking.
My whole point is that you can not point to a 3rd party checking for you and claim that it secure because someone else already checked. And I brought two examples which contradict this claim.
I’m not sure you’re understanding the argument: you cannot monitor closed source, therefore, you have at least as many eyes looking at my random crap on github as you do on the random crap some companies are doing.
And you didn’t understand what I said. While you can not monitor closed source at the code level, you definitely can monitor the apps behaviour. Even the automatic threat protection from the playstore protect function is worth more than the measly amount of people looking through smaller projects codebases.
I hate Google with a passion, but with all their control over android devices, they are more than capable of scanning apps for malicious behaviour and automatically removing them. These few apps in the article are the 0.01% of malicious apps that their algorithm didn’t detect.
There is no guarantee that the released app is exactly the same as the source code when getting it on Google Play. You’d have to decompile or compile from source and try to compare.
The thing is we only know about these vulnerabilities in such great detail because the projects are open source. God knows what kund of vulnerabilities are hidden in closed source software.
A very classic lie, disinformation, used to spread anti-libre software. Anti-libre software bans us, not only me but everyone else, from removing malicious source code.
Very disingenuous of you to fight a strawman and proclaim victory by claiming that I said things which I never did. But if that’s what floats your boat. But for everyone else, try to find any mention of anti-libre software in the original claim.
What are you talking about? You are digging yourself in a trench against me for some reason and you dig deeper every time. I have no idea what your agenda is, but I am stopping participation in it.
I don’t know about you but I have always been a free software advocate, see
Exactly. Neckbeards love to pretend open source magically has no security vulnerabilities, and that the ability to inspect the source means you’ll never install anything nefarious.
I expect all of them to have read the source for every single package they’ve ever installed. Oh and the Linux source too, of course
Yes, opensource doesn’t magically fix all vulnerabilities. But it is for sure way better then closed source, where you don’t have a way of auditing the code
Another classic lie. ‘Open source’ misses the point of libre software. Anti-libre software [malware] bans us [everyone else] from removing malicious source code.
Yes, of course. However, when it’s open source, at least somebody is capable of checking those things, even if it is not you. Somebody in the community is capable of doing so.
Yes, that is true, but let’s not pretend that just because some one is theoretically able to, that all source code is constantly monitored by 3rd parties.
Oh, absolutely, that’s true. Definitely smaller projects have less audited code, and even bigger projects can have bugs. Heart bleed ring a bell, LOL. However, when open source software has a bug and it is discovered, it is fixed by somebody in record time, whereas in closed source software, you don’t know that there is a bug that can be exploited and it definitely won’t be fixed until it’s reverse engineered or something or exploited.
If you download apps from fdroid, at the very least you can be sure that the binary is 100% generated from the provided source code, the devs can’t pull a switcheroo like submitting an altered version of app (e.g. inserting malware) that doesn’t match the published source code.
You’re right, I should clarify better. When I say open source, what I mean is totally open and totally free to contribute to, like the MIT or patchy licenses. Source viewable is a whole different can of worms and not what I mean, so I should be more specific in future.
And this right here is why you use open source apps.
This only would work if you check every line of source code, even the dependencies and build chain, and then build it yourself. See xz utils backdoor or heartbleed, etc.
The whole point is that at some point somebody can check, and you can have a higher level of trust in that than proprietary software.
And if someone does something like this then it has to be disguised as an innocuous bug, like heartbleed, they can’t just install full on malware.
It’s a different beast entirely.
If we are talking about bigger projects with hundreds of thousands or millions of downloads, than this may be true. But smal scale projects have so few people actively looking through them that even to automatic scan done by the playstore has a higher chance of catching malware. It doesn’t even have to be bad intent, two years ago there was a virus propagating trough the Java class files in minecraft mods which reached the PCs of quite a few devs before it was caught.
I don’t dislike FOSS, a lot of the apps I use come straight from github, but all this talk about them beeing constantly monitored by third parties is just wishful thinking.
Okay, but that’s a different claim than that you have to personally vet and compile every single thing you use, which is what I was responding to.
Open source isn’t perfect, but it is objectively and obviously better than closed.
My whole point is that you can not point to a 3rd party checking for you and claim that it secure because someone else already checked. And I brought two examples which contradict this claim.
I’m not sure you’re understanding the argument: you cannot monitor closed source, therefore, you have at least as many eyes looking at my random crap on github as you do on the random crap some companies are doing.
And you didn’t understand what I said. While you can not monitor closed source at the code level, you definitely can monitor the apps behaviour. Even the automatic threat protection from the playstore protect function is worth more than the measly amount of people looking through smaller projects codebases.
I hate Google with a passion, but with all their control over android devices, they are more than capable of scanning apps for malicious behaviour and automatically removing them. These few apps in the article are the 0.01% of malicious apps that their algorithm didn’t detect.
There is no guarantee that the released app is exactly the same as the source code when getting it on Google Play. You’d have to decompile or compile from source and try to compare.
Using F-Droid is good alternative.
The thing is we only know about these vulnerabilities in such great detail because the projects are open source. God knows what kund of vulnerabilities are hidden in closed source software.
Yes, but we don’t know what we don’t know. There are many problems like that in open source too, and even if we can look nobody does.
Therefore I find it problematic to say that just because you use open source programs you’re safe like the parent tried to.
Yes, important to keep in mind that software being open source doesn’t automagically make it secure™.
Still, I think it’s important to stress that the benefits of open source outweigh the risks when it comes to security (imho).
I agree with that.
I don’t agree with how it has been presented by the grandparent here as if open source somehow automatically makes it secure.
🚩🚩🚩
A very classic lie, disinformation, used to spread anti-libre software. Anti-libre software bans us, not only me but everyone else, from removing malicious source code.
Very disingenuous of you to fight a strawman and proclaim victory by claiming that I said things which I never did. But if that’s what floats your boat. But for everyone else, try to find any mention of anti-libre software in the original claim.
‘Open source’ is created to subvert libre software.
What are you talking about? You are digging yourself in a trench against me for some reason and you dig deeper every time. I have no idea what your agenda is, but I am stopping participation in it.
I don’t know about you but I have always been a free software advocate, see
But anyway, I’m not interested in whatever this discussion derailed into.
Then, you would know anti-libre software bans, not you, us separately and together from proving its claims and removing malicious source code.
Exactly. Neckbeards love to pretend open source magically has no security vulnerabilities, and that the ability to inspect the source means you’ll never install anything nefarious.
I expect all of them to have read the source for every single package they’ve ever installed. Oh and the Linux source too, of course
Yes, opensource doesn’t magically fix all vulnerabilities. But it is for sure way better then closed source, where you don’t have a way of auditing the code
I have never seen anyone make that claim.
Lots of arguments saying it’s an improvement, but never that it magically fixes everything.
Who does? Feels like you’re just talking about inexperienced “btw i use arch” kinda skiddies
Another classic lie. ‘Open source’ misses the point of libre software. Anti-libre software [malware] bans us [everyone else] from removing malicious source code.
Yes, of course. However, when it’s open source, at least somebody is capable of checking those things, even if it is not you. Somebody in the community is capable of doing so.
Yes, that is true, but let’s not pretend that just because some one is theoretically able to, that all source code is constantly monitored by 3rd parties.
Oh, absolutely, that’s true. Definitely smaller projects have less audited code, and even bigger projects can have bugs. Heart bleed ring a bell, LOL. However, when open source software has a bug and it is discovered, it is fixed by somebody in record time, whereas in closed source software, you don’t know that there is a bug that can be exploited and it definitely won’t be fixed until it’s reverse engineered or something or exploited.
Being open-source is not sufficient, but necessary.
If you download apps from fdroid, at the very least you can be sure that the binary is 100% generated from the provided source code, the devs can’t pull a switcheroo like submitting an altered version of app (e.g. inserting malware) that doesn’t match the published source code.
With the new changes to the repo management, that’s not going to remain true for much longer.
‘Open source’ misses the point of libre software.
They used to be synonymous. Of course now we have conbtributor licenses
It was invented to derail libre software.
deleted by creator
You’re right, I should clarify better. When I say open source, what I mean is totally open and totally free to contribute to, like the MIT or patchy licenses. Source viewable is a whole different can of worms and not what I mean, so I should be more specific in future.
Contributing isn’t the point. AGPL helps us keep control of our own computing.