Is Open Source Really So Much More Secure By Nature
-
Okay, it looks like I'm wrong about the laptops. Linux is still significant, but not leading to the degree I thought.
However, the numbers being shown don't seen to add up.
-
@stacksofplates said in Is Open Source Really So Much More Secure By Nature:
@scottalanmiller said in Is Open Source Really So Much More Secure By Nature:
@siringo said in Is Open Source Really So Much More Secure By Nature:
@coliver said in Is Open Source Really So Much More Secure By Nature:
@Obsolesce said in Is Open Source Really So Much More Secure By Nature:
@coliver said in Is Open Source Really So Much More Secure By Nature:
@Obsolesce said in Is Open Source Really So Much More Secure By Nature:
@scottalanmiller said in Is Open Source Really So Much More Secure By Nature:
Linux is number one in laptops
Where are you getting your statistics from?
That doesn't include servers and completely ignores Android. You're proving @scottalanmiller's point here. You have to specify a specific market that Windows is a leader in, specifically PC/Laptop. On the whole of the industry Android/Linux is the leader.
If you read what I quoted, i responded to a very specific piece of what he wrote...
Then I linked some stats to show that specific text of his I quoted was wrong, and then asked where he got his info, because I can't find anything to show otherwise of that specific thing I quoted.
Pay attention. Context matters. The quotes help with that.
Yep, you're right missed the quote.
This is great. I see too often people not acknowledging a mistake they make & the discussion ends up a turd fight.
Well done Mr @coliver.But he wasn't wrong. Even if he missed the quote, because @Obsolesce didn't pay attention the context and did exactly the thing that he was accusing @coliver of having done.
The assumption here is that I said laptops, but didn't mean it. But that @Obsolesce said desktops and did mean it. That his context matters, and mine doesn't.
But he didn't just quote me, he said to check the quote. If you check the quote, it doesn't match what he was responding with.
So let's dig into this. Total "PC" shipments in 2019 were about 261m, right? And ChromeOS is roughly 30m. And Raspberry Pi is a single computer that was doing 600,000 per month in a slow month. So that's about 7-10m of that one PC alone. That's 37m out of 261m in just those isolated cases. That would be 14.2% right there.
So before we count a single PC built for Linux from parts, or any other SBC maker, or any Windows machine purchased and converted to Linux... all the cases, that form the majority of the use cases that people talk about or that people we all know use here and there.... we already have a bigger market share than supposedly exists for Linux.
So one way or another, these numbers don't jive. Even if we simply work from "how many runs Linux but not Windows or MacOS" machines out of the pool of "total machines sold" and don't count what we assume is the main body of them, even before we talk about the ability to measure... something is wrong. The metrics cannot be accurate.
Then the question also is... how can anyone gather these metrics? For example, how are my three Linux desktops, and Linux laptop, getting counted, they aren't reported to anyone. But at the time that they were purchased, did someone decide to count them as Windows?
-
It's also had to follow market share because there are two ways to measure market... one is what is sitting out there, the other is what is being sold.
For example, Mac has a small percentage of the Japanese market "in use", like say 10%. But in Q4 they sold so many machines that they raised that to something insane like 40%. That 30% increase in one quarter means that what is being sold was approaching 90%+.
So do we look at them as having 40% or 90% of market share?
If you are a software maker, you primarily care about what is sitting out there. If you are an OS maker or a hardware maker, you mostly care about what is being sold.
-
@scottalanmiller said in Is Open Source Really So Much More Secure By Nature:
@Obsolesce said in Is Open Source Really So Much More Secure By Nature:
It said laptops in the image I included.
I keep looking, but I see PC and desktop, but no laptop.
I did underline it in red in the image.
-
@scottalanmiller said in Is Open Source Really So Much More Secure By Nature:
Then the question also is... how can anyone gather these metrics?
Then how could you make the statement you did? You've provided no evidence other than speculation.
-
@scottalanmiller said in Is Open Source Really So Much More Secure By Nature:
And ChromeOS is roughly 30m. And Raspberry Pi is a single computer that was doing 600,000 per month in a slow month. So that's about 7-10m of that one PC alone. That's 37m out of 261m in just those isolated cases. That would be 14.2% right there.
So before we count a single PC built for Linux from partsYou said laptop. ChromeOS isn't only on laptops, raspberry pis aren't laptops. I don't personally know anyone who has built a laptop from parts. You're not even describing the thing you claim to be.
-
I'm also wondering why Scott is including R Pi's as PCs? They aren't. Scott's been telling us the PC spec is XYZ for a decade, clearly the R Pi doesn't qualify for that spec. Nor does the M1 based stuff.
So those 600K units/month are meaningless to this discussion.
I really only see Chromebooks and OEM units sold with a Linux OS pre-installed in these numbers, assuming they are pulling industry numbers.
If they are pulling numbers from some website that sees billions if not trillions of hits a month to use as a basis of %, then we would see the In Use numbers.
-
I don't know that you shouldn't include SBC's in those numbers because of how they exclusive operate a full desktop experience and pretty much everything a user could need on a rather extreme budget.
But it is not a traditional desktop by any stretch with a motherboard, cpu, ram and generally speaking expansion modules - PCI, ePCI etc.
-
@DustinB3403 said in Is Open Source Really So Much More Secure By Nature:
I don't know that you shouldn't include SBC's in those numbers because of how they exclusive operate a full desktop experience and pretty much everything a user could need on a rather extreme budget.
But it is not a traditional desktop by any stretch with a motherboard, cpu, ram and generally speaking expansion modules - PCI, ePCI etc.
He made it abundantly clear he only meant laptops. If it's not a laptop it's not counted. SBC's are not laptops, not all ChromeOS is laptops.
-
@stacksofplates oh I know, I'm not jumping into Scott's corner, I was just responding to Dash's comment on how SBC previously weren't considered "Desktop computers" by Scott, but are now apparently being lumped in.
If we look at the Oxford dictionary for a "desktop" computer (computer unquoted on purpose) it's
desk·top /ˈdeskˌtäp/ Learn to pronounce noun the working surface of a desk. a computer suitable for use at an ordinary desk. noun: desktop computer "a new low-end desktop" Definitions from Oxford LanguagesWhich if we then look at how RPi's Alpines etc all work and operate, they would also now fall into that category and should be included in such tallies.
-
So, in this source, Laptops seem to be lumped in with Desktops.
https://chromeunboxed.com/chromebook-market-share-sales-growth-q3-2020
All this data I found combined, for Linux to be such a large part of usage on laptops as you are thinking, just isn't possible. It would have to mean that not only do laptops make up a much larger portion of the desktop/laptop graph, but SO MUCH more so that Linux would have to be on so many of them that it would have to outweigh the others like OS X. But these graphs are showing that they are Windows... so it's simply not possible.
The data includes both laptops and desktops together, and still, Linux is only in the LOW single digits.
I cannot find a single statistic anywhere on the internet that would suggest Linux usage on Laptop is even close to being significant in comparison to ChromeOS/OS X/Windows. Even if you mentally manipulate the data to kind of like mold it into support of your own bias, it still makes no sense to think Linux on laptops is anything other than nil.
-
@Obsolesce yeah to summarize, "The year of the Linux Desktop" hasn't arrived yet, regardless of what is being deployed to SBC's (which should be included in such tallies IMO).
-
@DustinB3403 said in Is Open Source Really So Much More Secure By Nature:
@stacksofplates oh I know, I'm not jumping into Scott's corner, I was just responding to Dash's comment on how SBC previously weren't considered "Desktop computers" by Scott, but are now apparently being lumped in.
If we look at the Oxford dictionary for a "desktop" computer (computer unquoted on purpose) it's
desk·top
/ˈdeskˌtäp/
Learn to pronounce
noun
the working surface of a desk.
a computer suitable for use at an ordinary desk.
noun: desktop computer
"a new low-end desktop"
Definitions from Oxford LanguagesWhich if we then look at how RPi's Alpines etc all work and operate, they would also now fall into that category and should be included in such tallies.
We were never talking about desktop computers... maybe some post someone made did, but I thought the whole discussion was around PCs, and Scott specifically narrowed it to Laptops.
-
@Dashrender said in Is Open Source Really So Much More Secure By Nature:
We were never talking about desktop computers... maybe some post someone made did, but I thought the whole discussion was around PCs, and Scott specifically narrowed it to Laptops.
So what the fuck is this about?
@Dashrender said in Is Open Source Really So Much More Secure By Nature:
I'm also wondering why Scott is including R Pi's as PCs? They aren't. Scott's been telling us the PC spec is XYZ for a decade, clearly the R Pi doesn't qualify for that spec. Nor does the M1 based stuff.
-
XYZ specs, would indicate that a PC (desktop or laptop) has a specific set of hardware specifications.
-
@DustinB3403 said in Is Open Source Really So Much More Secure By Nature:
So what the fuck is this about?
Scott and NTG uses Linux on all of their desktops and laptops, so he is desperately trying to say the entire market reflects what he does in his world.
-
Additionally, if you look at the Oxford dictionary definition for a "Desktop computer" it clearly cares not about what the components are inside of the system, so long as the intended use is: "a computer suitable for use at an ordinary desk."
Which a RPi very clearly falls into that category since it's not by design a tablet, cellphone etc.
-
@Obsolesce said in Is Open Source Really So Much More Secure By Nature:
@DustinB3403 said in Is Open Source Really So Much More Secure By Nature:
So what the fuck is this about?
Scott and NTG uses Linux on all of their desktops and laptops, so he is desperately trying to say the entire market reflects what he does in his world.
Yeah that's fine, it's his world view. He may have hundreds or thousands of customers who are purchasing Pi's in bulk, that doesn't change the definition of what a "desktop computer" would qualify as, and the components of what's inside the case clearly don't matter either.
-
@DustinB3403 said in Is Open Source Really So Much More Secure By Nature:
Additionally, if you look at the Oxford dictionary definition for a "Desktop computer" it clearly cares not about what the components are inside of the system, so long as the intended use is: "a computer suitable for use at an ordinary desk."
Which a RPi very clearly falls into that category since it's not by design a tablet, cellphone etc.
Right now, I have 3 laptops on my desktop...
Yes, you guessed right, my desk has to be this big:
-
@Obsolesce That's an insane table lol...
