I think you are either trolling or you fundamentally don’t understand, what you’re talking about.
Nothing is obfuscated. You can download each and every code file, audit it, and build the binaries from exactly that code. You can even compare the binaries to the ones provided by major distros thanks to reproducible builds.
Just because you don’t understand code, doesn’t mean it’s obfuscated. Following that logic, even a loaf of bread is “obfuscated” because you don’t understand sour dough.
That’s not what I’m saying. Yes, it’s open source and you can build the binaries itself. I’m saying that the process is obfuscated or complicated because instead of text log files, you have to use journtalctl to view them.
Then again, someone said it may be text files with markers so I have to look into that
Are you really sure, you’re using “obfuscation” right? Because that implies that someone intentionally makes something harder to read to hide something. That’s not the case here. Nothing is hidden, it’s all there, the formats are well defined and easy to read.
Yeah, of course, it’s all there in binary. For programs of course that’s not a problem, but for data that you may need to look at any time, it is. It’s harder to interpret both for humans (significantly) and both for any program that want to make use of it (unless they use the specific library that came up with the format, and by that also pulling in all its libs transitively)
Binary data is not much less obfuscated than the system files of windows. It’s all there, you can read it
For programs, that is not a problem.
This is a problem for data.
Why? Because you very rarely need to read the program’s “content”, and when you do, you’ll instead go look at the source code anyways. But for binary data files there is no source code that is the equivalent of the contents in readable form.
If you want to read it as a human in your text editor, good luck with making sense of it. If you want to read it with your program it’ll have to pull in a tree of dependencies out of questionable necessity, and any of that dependencies could have a severe bug or a security vulnerability that affects your program and it’s users. And the only reason you needed to import that lib is to be able to parse this binary format. It’s not even a common one like an archive format, but a totally custom made format of systemd.
And then there’s another problem. You may be able to make sense of the binary data with your bare hands and a text editor, but you better not edit it that way, because you may mess up the delicate offsets, or you may wanted to replace a value (e.g. a string, out some kind of list) with a longer one but you can’t because of the former problem.
Binary is ok for programs, and you know what, it’s also fine for data in transit (network) and of course archives.
But for data, whether it’s a log file or configuration, or some other that would be totally fine in text format, it’s just annoying, limiting, and overcomplicated.
I think you are either trolling or you fundamentally don’t understand, what you’re talking about.
Nothing is obfuscated. You can download each and every code file, audit it, and build the binaries from exactly that code. You can even compare the binaries to the ones provided by major distros thanks to reproducible builds.
Just because you don’t understand code, doesn’t mean it’s obfuscated. Following that logic, even a loaf of bread is “obfuscated” because you don’t understand sour dough.
“…because you don’t understand sourdough”
Made me spit up my coffee.
I’m not sure I totally agree with him, what I’m sure is that I don’t agree with you, because you clearly haven’t read his comment.
That’s not what I’m saying. Yes, it’s open source and you can build the binaries itself. I’m saying that the process is obfuscated or complicated because instead of text log files, you have to use journtalctl to view them.
Then again, someone said it may be text files with markers so I have to look into that
Are you really sure, you’re using “obfuscation” right? Because that implies that someone intentionally makes something harder to read to hide something. That’s not the case here. Nothing is hidden, it’s all there, the formats are well defined and easy to read.
Yeah, of course, it’s all there in binary. For programs of course that’s not a problem, but for data that you may need to look at any time, it is. It’s harder to interpret both for humans (significantly) and both for any program that want to make use of it (unless they use the specific library that came up with the format, and by that also pulling in all its libs transitively)
Binary data is not much less obfuscated than the system files of windows. It’s all there, you can read it
So literally every program on your machine is obfuscated. Linux kernel? Obfuscated. Wayland? Obfuscated. And even VIM: obfuscated.
You’re creating problems where there are none.
Did you read my comment in it’s entirety?
For programs, that is not a problem.
This is a problem for data.
Why? Because you very rarely need to read the program’s “content”, and when you do, you’ll instead go look at the source code anyways. But for binary data files there is no source code that is the equivalent of the contents in readable form.
If you want to read it as a human in your text editor, good luck with making sense of it. If you want to read it with your program it’ll have to pull in a tree of dependencies out of questionable necessity, and any of that dependencies could have a severe bug or a security vulnerability that affects your program and it’s users. And the only reason you needed to import that lib is to be able to parse this binary format. It’s not even a common one like an archive format, but a totally custom made format of systemd.
And then there’s another problem. You may be able to make sense of the binary data with your bare hands and a text editor, but you better not edit it that way, because you may mess up the delicate offsets, or you may wanted to replace a value (e.g. a string, out some kind of list) with a longer one but you can’t because of the former problem.
Binary is ok for programs, and you know what, it’s also fine for data in transit (network) and of course archives.
But for data, whether it’s a log file or configuration, or some other that would be totally fine in text format, it’s just annoying, limiting, and overcomplicated.
Again, that’s not what obfuscation means.
Also, what exactly is the difference between cat and journalctl? You can’t read a text file without a program either.
Of course, raw text files are more common, but what you’re drawing up here is a mixture of old man yells at cloud and tin foil hat territory.