Microsoft co-founder Bill Gates and Linus Torvalds, the creator of the Linux kernel, have surprisingly never met before. That all changed at a recent dinner hosted by Sysinternals creator Mark Russinovich.
The worlds of Linux and Windows finally came together in real life, and Dave Cutler, Microsoft technical fellow and Windows NT lead developer, was also there to witness the moment and meet Torvalds for the first time. “No major kernel decisions were made,” jokes Russinovich in a post on LinkedIn.
[Image: Bill Gates and Linus Torvalds meet for the first time. https://platform.theverge.com/wp-content/uploads/sites/2/2025/06/1750435121315.jpg?quality=90&strip=all]
Bill Gates and Linus Torvalds have apparently never met in person before, despite their pseudo-rivalry.
Both Torvalds and Gates are nerds… Gates decided to monetize it and Torvalds decided to give it away.
But without Microsoft’s “PC on every desktop” vision for the '90s, we may not have seen such an increased demand for server infrastructure which is all running the Linux kernel now.
Arguably Torvalds’ strategy had a greater impact than Gates because now many of us carry his kernel in our pocket. But I think both needed each other to get where we are today.
If it wasn’t them, it would have been other people.
Computer science doesn’t rest on shoulder of a “Great Man”
What Torvalds did was inspire a like-minded community to come together and work toward a collective good.
On a shoe-string budget they constantly threaten Gates’ empire.
Gates on the other hand chose to enclose the intellectual commons of computer science and sell them at a profit.
He extracted a heavy toll on all sectors of human activity. And what did this heavy burden buy us ?
Really NOT MUCH ! It squelched out collaboration and turned programming greedy, it delivered poor bloated software that barely worked and then stagnated for 20 years. It created a farm stall for us to live in, their innovation today is only explained as a series of indignities we will have to live with, because of platform dynamics we really, literally cannot escape the black hole that is windows for they have captured the commons and have made themselves unavoidable, like the Troll asking his toll.
Frankly I have to mention one thing - while BG was in MS, the Windows world was kinda fine. He left before even Windows 7. He left after Vista, and Vista wasn’t very good, but what’s important - MS didn’t only do evil.
I mean, yeah, not “fine” fine, but when you are saying “and then stagnated for 20 years”, Bill wasn’t in MS for most of those 20 years.
I agree that platform dynamics suck, but I also very well remember from my childhood that I wanted platforms. Everyone wanted platforms. Everyone wanted platforms like ICQ, not too opinionated and de-facto interoperable, or like Geocities, but people wanted platforms.
It was just plainly unavoidable. Everyone wanted webpages to be dynamic applications and everyone wanted platforms.
Yes, both are traps of evolution.
Say, dynamic pages I wanted would be more like embedded content in its own square, as it was with Flash. Just instead of Netscape plugin API and one proprietary environment it could involve a virtual machine for running cross-platform bytecode, or even just PostScript. Java applets were that idea, sort of (no sandboxing), as always Sun solved the hard problem perfectly, but forgot to invent a way for adoption. Maybe it could be allowed access to cut buffers and even the rest of the page. But that would be requested. This would prevent the web turning into something only Chrome can support.
Say, platforms I wanted would be more like standardized unified resources pooled. Storage resources and computing resources and notification servers and indexation servers for search, possibly partitioned to accommodate the sheer amount of data. Maybe similar to Usenet and NOSTR. With user application being the endpoint to mix those into a “social network” or some other platform. Universal application-agnostic servers, specific user applications.
I’ve said this before here, but techy people vastly overestimate both the ability and the patience of the typical user, and it’s the reason so few people use FOSS products.
Products from big tech aimed at private individuals are designed to be as simple to use as possible, which is why they’re so popular.
Nah, I have worked in IT education and in helpdesk. Average user doesn’t have a better time getting into Microsoft products, it’s not easier for them than FOSS. The reason for Windows domination is Microsoft spending money and lobbying power to put it in front of every user.
When I actually started doing hobby projects, I remembered that feeling with Windows 9x where you learn to avoid “wrong” actions which have a potential of hanging your PC. You don’t even think about it. Just get used that you don’t move the cursor after clicking there, you don’t click here again after a first double click, and other such.
While things like editing config files were … more normal for the average person even, you’d have a paper manual generally. For everything, kitchen appliances and anything technical you could buy too. You wouldn’t expect everything to just work without reading it. Freezes and crashes were worse.
Windows won because most people didn’t know of anything else.
And it is still true today. Windows has the lion share of the market because we were raised with Windows and the vast majority of people don’t want to learn a new OS.
Luckily they learned from it and redesigned the kernel from scratch – hold on, my producer’s telling me that no, it’s still the NT kernel under there. Outstanding.
Barf. Or maybe, just maybe, we have other shit to do rather than spend hours trying to figure out how to do one thing in Gimp. It’s great that YOU’RE passionate about tech. Some of us have other hobbies. Imagine that holy shit
Buddy, if I open Photoshop it’s gonna take me hours to learn how to do one thing too, what a horrible example lmao. There’s like so many easy slam dunks you could’ve said too.
Also, I never mentioned Photoshop. Open any standard drawing app that was developed recently: Procreate, Infinite Paint, Krita, Fresco. Look how straightforward it is to start working. Look at the Ui. It doesn’t get in the way.
Edit: oh no the FOSS evangelists are not feeling it. I get it. I use a lot of FOSS apps for work. That doesn’t mean we have to be evangelical in our defense of FOSS. Recognize there are issues and we can work to fix them. Don’t get so defensive, Lemmy. My god.
You should not expect to use a tool (edit: competently) without spending time learning how to use it. Photoshop has a learning curve too, even if it’s an easier one.
As someone who’d learned Photoshop and, eventually, learned GIMP (just because it was easier to run after eventually switching to Linux), trying to argue that Photoshop has an industry stranglehold because it – apparently – is just so much more intuitive than GIMP is absolutely wild. No one I knew learning Photoshop was finding that the UI or layout just magically clicked (or even swiftly got less impenetrable, as time went on).
Yeah, it’s very obvious that some of the people responding here don’t interact much with non-tech people, and they have DEFINITELY never worked IT.
Most people aren’t interested in learning the more intricate things. And if you try to force them, they’re not going to get more interested as they learn, because they literally are not interested in tech. They want to accomplish a task, if that takes a bunch of learning just for one thing, they’ll go a different route, or pay someone else to do it for them.
Keep in mind this status quo is already the result of decades of oversimplification. I am not saying everyone needs to compile the Linux kernel in order to have a computer. I’m saying you should have a basic level of familiarity with the computer you’re using, same as any other tool.
You should know how to check and top up your engine oil, change a tire in an emergency, etc, if you’re going to own a car.
You should know how to safely handle, operate, store, transport, and clean your firearm if you’re going to own a gun.
You should know how to empty the chamber or bag, clean the filters correctly, what not to suck up and how to troubleshoot if you do, if you’re going to own a vacuum.
You should know how to operate it, when and how it should be cleaned, and what not to do while it’s running, if you’re going to own an electric range.
You should know the difference between a web browser and your computer’s filesystem, the difference between RAM and storage, and that you can Internet search most errors to judge whether you’re comfortable trying to fix them yourself or not, if you’re going to own a computer.
There will ALWAYS be a point where it’s more worth paying someone else instead of learning something yourself. But it’s about the cost-benefit analysis, and the threshold for what’s considered “intricate” is a depressingly low bar where computers are concerned. As I’m sure you are well aware.
That has to be one of the most out of touch takes I’ve seen in a while. You’re basically saying that things should be intentionally more complicated, and you expect the result to be people just power through and getting used to things being that way, instead of just stopping.
To add to subignition’s point, there is a value in learning useful software. More complicated software means that there is a learning curve - so while you are less productive while learning how to use it, once you gain more experience, you ultimately become more productive. On the other hand, if you want the software to be useful to everyone regardless of his level of experience, you ultimately have to eliminate more complex functionality that makes the software more useful.
Software is increasingly being distilled down to more and more basic elements, and ultimately, I think that means that people are able to get less done with them these days. This is just my opinion, but in general I have seen computer literacy dropping and people’s productivity likewise decreasing, at least from what I’ve observed from the 1990s up until today. Especially at work, the Linux users that I see are much more knowledgeable and productive than Apple users.
…No. I am saying that too much abstraction of how something actually works is detrimental to the end user. It’s not about making things intentionally more complicated, it’s about removing the need to understand key components of something ultimately causing more harm than good.
This is also far from my personal experience, you might not even realize what free software you’re depending on?
Your browser is most likely the most complex piece of software you interact with daily and it is most likely FOSS. The Linux kernel is FOSS and is incredibly robust. Most compiler suites, FOSS. Most programming languages, FOSS. These are all incredibly well written and robust tools. AOSP, kinda FOSS, and the forks like Graphene are definitely FOSS. Hell even a lot of macOS programs are actually FOSS. I could go on and on, there is absolutely amazing work being done on FOSS by incredibly talented people.
There is great paid and proprietary software out there, sure, but no it’s not the majority of top quality software in my personal experience and likely a lot of people’s experiences and it is almost guaranteed to rely on a FOSS library somewhere
But without Microsoft’s “PC on every desktop” vision for the '90s, we may not have seen such an increased demand for server infrastructure which is all running the Linux kernel now.
Debatable, in my opinion. There were lots of other companies trying to build personal computers back in those times (IBM being the most prominent). If Microsoft had never existed (or gone about things in a different way), things would have been different, no doubt, but they would still be very important and popular devices. The business-use aspect alone had a great draw and from there, I suspect that adoption at homes, schools, etc. would still follow in a very strong way.
There were plenty of alternative graphic shells for DOS, too.
For me it’s interesting to imagine what if a multi-user memory protected yadda-yadda serious system replaced DOS, but preserved the modularity and interoperability of components, so that people would still use different graphic shells, different memory compressors\swappers and so on, and then the PC world would be much more interesting today.
That’s what, only in the sense of desktop shells, Unix-likes have raising them above Windows, or at least have until X11 dies. I think that XLibre person, despite their mental instability and wish to seek conflicts, was right to fork it and it’s a good call and that XLibre project will live on. Because yes, RedHat had a policy for X11 stagnating and being deprecated, and they imposed it on the Xorg project itself. I think we’ll see that, oh wonder, X11’s modular architecture (in the sense of extensions too) will prove better project-wise than Wayland’s. Even with legacy, technical debt, obsolete paradigm, all those things people like to mention. This happened too late to kill Wayland, but not too late to save X.
Which is BTW why this meeting involving Dave Cutler is cool again. See, NT is in its architecture more modular than Linux.
I doubt they are going to do any project, but in case they are - would be cool if it were a third OS in the VMS and NT row. Supporting Linux ABI and drivers, but maybe even allowing to use Windows NT device drivers. How cool would that be.
OK, that’s what’s called “пикейный жилет” in Russian, utterly useless talk of the kitchen\taxi kind.
I remember that IBM was famously missing the trend in the late 80s/90s and couldn’t understand why regular consumers would ever want to buy a PC. It’s why they gave the PC clone market away, never seriously approached their OS/2 thing, and never really marketed directly to anybody except businesses.
Microsoft really pushed the idea that regular people needed a home PC which laid the foundation for so many people already having the hardware in place to jump on the internet as soon as it became accessible.
For a brief moment it looked like a toss up between Microsoft IIS webservers serving up .asp files (or coldfusion .cf - RIP) vs Apache pushing CGI but in the end the Linux solution was more baked and flexible when it was time to launch and scale an internet startup in that era.
Somebody else would have done what Microsoft did for sure, had they not been there, and I suppose we could be paying AT&T for Unix licenses these days too. But yeah, ultimately both Gates and Torvalds were right in terms of operating systems and well timed.
If Microsoft hadn’t been around Apple would have probably defined the early PC era. The Apple II was released in 1977, 4 years before IBM decided to enter the home market with the PC.
Or Commodore might have been the one to dominate. They sold about 5 million Amigas.
Or it could have been NeXT after Jobs was forced out of Apple and started a new computer business.
The winner turned out to be Microsoft, but desktop computers were well on their way to being a standard thing long before Microsoft / IBM got into the market.
Both Torvalds and Gates are nerds… Gates decided to monetize it and Torvalds decided to give it away.
But without Microsoft’s “PC on every desktop” vision for the '90s, we may not have seen such an increased demand for server infrastructure which is all running the Linux kernel now.
Arguably Torvalds’ strategy had a greater impact than Gates because now many of us carry his kernel in our pocket. But I think both needed each other to get where we are today.
If it wasn’t them, it would have been other people. Computer science doesn’t rest on shoulder of a “Great Man”
What Torvalds did was inspire a like-minded community to come together and work toward a collective good. On a shoe-string budget they constantly threaten Gates’ empire.
Gates on the other hand chose to enclose the intellectual commons of computer science and sell them at a profit. He extracted a heavy toll on all sectors of human activity. And what did this heavy burden buy us ? Really NOT MUCH ! It squelched out collaboration and turned programming greedy, it delivered poor bloated software that barely worked and then stagnated for 20 years. It created a farm stall for us to live in, their innovation today is only explained as a series of indignities we will have to live with, because of platform dynamics we really, literally cannot escape the black hole that is windows for they have captured the commons and have made themselves unavoidable, like the Troll asking his toll.
Frankly I have to mention one thing - while BG was in MS, the Windows world was kinda fine. He left before even Windows 7. He left after Vista, and Vista wasn’t very good, but what’s important - MS didn’t only do evil.
I mean, yeah, not “fine” fine, but when you are saying “and then stagnated for 20 years”, Bill wasn’t in MS for most of those 20 years.
I agree that platform dynamics suck, but I also very well remember from my childhood that I wanted platforms. Everyone wanted platforms. Everyone wanted platforms like ICQ, not too opinionated and de-facto interoperable, or like Geocities, but people wanted platforms.
It was just plainly unavoidable. Everyone wanted webpages to be dynamic applications and everyone wanted platforms.
Yes, both are traps of evolution.
Say, dynamic pages I wanted would be more like embedded content in its own square, as it was with Flash. Just instead of Netscape plugin API and one proprietary environment it could involve a virtual machine for running cross-platform bytecode, or even just PostScript. Java applets were that idea, sort of (no sandboxing), as always Sun solved the hard problem perfectly, but forgot to invent a way for adoption. Maybe it could be allowed access to cut buffers and even the rest of the page. But that would be requested. This would prevent the web turning into something only Chrome can support.
Say, platforms I wanted would be more like standardized unified resources pooled. Storage resources and computing resources and notification servers and indexation servers for search, possibly partitioned to accommodate the sheer amount of data. Maybe similar to Usenet and NOSTR. With user application being the endpoint to mix those into a “social network” or some other platform. Universal application-agnostic servers, specific user applications.
But this is all in hindsight.
Who’s Gate?
The Typo Monster, he comes out at night, mostly
Now you just need to slay the Apostrophe Monster.
I’ve said this before here, but techy people vastly overestimate both the ability and the patience of the typical user, and it’s the reason so few people use FOSS products.
Products from big tech aimed at private individuals are designed to be as simple to use as possible, which is why they’re so popular.
Nah, I have worked in IT education and in helpdesk. Average user doesn’t have a better time getting into Microsoft products, it’s not easier for them than FOSS. The reason for Windows domination is Microsoft spending money and lobbying power to put it in front of every user.
Maybe true today, but less true in earlier times (90s and early 2000s) when Microsoft was really gaining dominance.
I don’t think you remember how insanely terrible Windows was in the 90s.
When I actually started doing hobby projects, I remembered that feeling with Windows 9x where you learn to avoid “wrong” actions which have a potential of hanging your PC. You don’t even think about it. Just get used that you don’t move the cursor after clicking there, you don’t click here again after a first double click, and other such.
While things like editing config files were … more normal for the average person even, you’d have a paper manual generally. For everything, kitchen appliances and anything technical you could buy too. You wouldn’t expect everything to just work without reading it. Freezes and crashes were worse.
Windows won because most people didn’t know of anything else.
And it is still true today. Windows has the lion share of the market because we were raised with Windows and the vast majority of people don’t want to learn a new OS.
Luckily they learned from it and redesigned the kernel from scratch – hold on, my producer’s telling me that no, it’s still the NT kernel under there. Outstanding.
Most users neither know nor care what that is.
good talk
Big tech designing their products to be overly simple is one of the driving forces behind the average user having poor patience and aptitude for tech.
No, it’s not. We have other shit to do and very limited quality time.
Barf. Or maybe, just maybe, we have other shit to do rather than spend hours trying to figure out how to do one thing in Gimp. It’s great that YOU’RE passionate about tech. Some of us have other hobbies. Imagine that holy shit
Buddy, if I open Photoshop it’s gonna take me hours to learn how to do one thing too, what a horrible example lmao. There’s like so many easy slam dunks you could’ve said too.
Also, I never mentioned Photoshop. Open any standard drawing app that was developed recently: Procreate, Infinite Paint, Krita, Fresco. Look how straightforward it is to start working. Look at the Ui. It doesn’t get in the way.
Edit: oh no the FOSS evangelists are not feeling it. I get it. I use a lot of FOSS apps for work. That doesn’t mean we have to be evangelical in our defense of FOSS. Recognize there are issues and we can work to fix them. Don’t get so defensive, Lemmy. My god.
You should not expect to use a tool (edit: competently) without spending time learning how to use it. Photoshop has a learning curve too, even if it’s an easier one.
But, also, who thinks Photoshop is easier‽
As someone who’d learned Photoshop and, eventually, learned GIMP (just because it was easier to run after eventually switching to Linux), trying to argue that Photoshop has an industry stranglehold because it – apparently – is just so much more intuitive than GIMP is absolutely wild. No one I knew learning Photoshop was finding that the UI or layout just magically clicked (or even swiftly got less impenetrable, as time went on).
That really nails it, I think. Tech is a hobby for some, a means to an end for the vast majority.
Yeah, it’s very obvious that some of the people responding here don’t interact much with non-tech people, and they have DEFINITELY never worked IT.
Most people aren’t interested in learning the more intricate things. And if you try to force them, they’re not going to get more interested as they learn, because they literally are not interested in tech. They want to accomplish a task, if that takes a bunch of learning just for one thing, they’ll go a different route, or pay someone else to do it for them.
Keep in mind this status quo is already the result of decades of oversimplification. I am not saying everyone needs to compile the Linux kernel in order to have a computer. I’m saying you should have a basic level of familiarity with the computer you’re using, same as any other tool.
You should know how to check and top up your engine oil, change a tire in an emergency, etc, if you’re going to own a car. You should know how to safely handle, operate, store, transport, and clean your firearm if you’re going to own a gun. You should know how to empty the chamber or bag, clean the filters correctly, what not to suck up and how to troubleshoot if you do, if you’re going to own a vacuum. You should know how to operate it, when and how it should be cleaned, and what not to do while it’s running, if you’re going to own an electric range. You should know the difference between a web browser and your computer’s filesystem, the difference between RAM and storage, and that you can Internet search most errors to judge whether you’re comfortable trying to fix them yourself or not, if you’re going to own a computer.
There will ALWAYS be a point where it’s more worth paying someone else instead of learning something yourself. But it’s about the cost-benefit analysis, and the threshold for what’s considered “intricate” is a depressingly low bar where computers are concerned. As I’m sure you are well aware.
Do you hunt for all of your food and cook it from absolute scratch?
I bet you sometimes use a grocery store.
What are you even talking about? You’re trying to make an analogy here but it’s a really poor one.
It’s actually the perfect analogy, you just can’t see it because you’re stuck in the bubble.
Christ this is such a Lemmy take.
The other option is users just not using tech at all.
That has to be one of the most out of touch takes I’ve seen in a while. You’re basically saying that things should be intentionally more complicated, and you expect the result to be people just power through and getting used to things being that way, instead of just stopping.
To add to subignition’s point, there is a value in learning useful software. More complicated software means that there is a learning curve - so while you are less productive while learning how to use it, once you gain more experience, you ultimately become more productive. On the other hand, if you want the software to be useful to everyone regardless of his level of experience, you ultimately have to eliminate more complex functionality that makes the software more useful.
Software is increasingly being distilled down to more and more basic elements, and ultimately, I think that means that people are able to get less done with them these days. This is just my opinion, but in general I have seen computer literacy dropping and people’s productivity likewise decreasing, at least from what I’ve observed from the 1990s up until today. Especially at work, the Linux users that I see are much more knowledgeable and productive than Apple users.
…No. I am saying that too much abstraction of how something actually works is detrimental to the end user. It’s not about making things intentionally more complicated, it’s about removing the need to understand key components of something ultimately causing more harm than good.
What about the boat loads of marketing - ads - aimed at making you believe those proprietary programs are the best? Clearly you fell for it.
There are shit proprietary software and good proprietary software. There are shit FLOSS and good FLOSS
I’ve used my share of free software. Some of it worked well, but it always felt clunky, and just never as straightforward to use as a paid product.
But sure, I couldn’t possibly have reached that conclusion on my own, it’s obviously the marketing.
Sounds like you’re cherry picking both; I’ve seen plenty of garbage that costs money as well.
Sure, but if you look at the top quality softwares, the majority of them are paid.
Because money is a big encouragement to make them as flawless as possible. Something FOSS just doesn’t have.
This is also far from my personal experience, you might not even realize what free software you’re depending on?
Your browser is most likely the most complex piece of software you interact with daily and it is most likely FOSS. The Linux kernel is FOSS and is incredibly robust. Most compiler suites, FOSS. Most programming languages, FOSS. These are all incredibly well written and robust tools. AOSP, kinda FOSS, and the forks like Graphene are definitely FOSS. Hell even a lot of macOS programs are actually FOSS. I could go on and on, there is absolutely amazing work being done on FOSS by incredibly talented people.
There is great paid and proprietary software out there, sure, but no it’s not the majority of top quality software in my personal experience and likely a lot of people’s experiences and it is almost guaranteed to rely on a FOSS library somewhere
They are used due to support not quality. Companies need to be able to purchase service and support agreements and very often FOSS has none of that.
It’s a reason. Another reason is all the stuff that Microsoft was found guilty of doing during their conviction for abusing their monopoly.
Debatable, in my opinion. There were lots of other companies trying to build personal computers back in those times (IBM being the most prominent). If Microsoft had never existed (or gone about things in a different way), things would have been different, no doubt, but they would still be very important and popular devices. The business-use aspect alone had a great draw and from there, I suspect that adoption at homes, schools, etc. would still follow in a very strong way.
There were plenty of alternative graphic shells for DOS, too.
For me it’s interesting to imagine what if a multi-user memory protected yadda-yadda serious system replaced DOS, but preserved the modularity and interoperability of components, so that people would still use different graphic shells, different memory compressors\swappers and so on, and then the PC world would be much more interesting today.
That’s what, only in the sense of desktop shells, Unix-likes have raising them above Windows, or at least have until X11 dies. I think that XLibre person, despite their mental instability and wish to seek conflicts, was right to fork it and it’s a good call and that XLibre project will live on. Because yes, RedHat had a policy for X11 stagnating and being deprecated, and they imposed it on the Xorg project itself. I think we’ll see that, oh wonder, X11’s modular architecture (in the sense of extensions too) will prove better project-wise than Wayland’s. Even with legacy, technical debt, obsolete paradigm, all those things people like to mention. This happened too late to kill Wayland, but not too late to save X.
Which is BTW why this meeting involving Dave Cutler is cool again. See, NT is in its architecture more modular than Linux.
I doubt they are going to do any project, but in case they are - would be cool if it were a third OS in the VMS and NT row. Supporting Linux ABI and drivers, but maybe even allowing to use Windows NT device drivers. How cool would that be.
OK, that’s what’s called “пикейный жилет” in Russian, utterly useless talk of the kitchen\taxi kind.
I remember that IBM was famously missing the trend in the late 80s/90s and couldn’t understand why regular consumers would ever want to buy a PC. It’s why they gave the PC clone market away, never seriously approached their OS/2 thing, and never really marketed directly to anybody except businesses.
Microsoft really pushed the idea that regular people needed a home PC which laid the foundation for so many people already having the hardware in place to jump on the internet as soon as it became accessible.
For a brief moment it looked like a toss up between Microsoft IIS webservers serving up .asp files (or coldfusion .cf - RIP) vs Apache pushing CGI but in the end the Linux solution was more baked and flexible when it was time to launch and scale an internet startup in that era.
Somebody else would have done what Microsoft did for sure, had they not been there, and I suppose we could be paying AT&T for Unix licenses these days too. But yeah, ultimately both Gates and Torvalds were right in terms of operating systems and well timed.
If Microsoft hadn’t been around Apple would have probably defined the early PC era. The Apple II was released in 1977, 4 years before IBM decided to enter the home market with the PC.
Or Commodore might have been the one to dominate. They sold about 5 million Amigas.
Or it could have been NeXT after Jobs was forced out of Apple and started a new computer business.
The winner turned out to be Microsoft, but desktop computers were well on their way to being a standard thing long before Microsoft / IBM got into the market.