FreeBSD Contributor Mocks Gloomy Predictions for the Open Source Movement (acm.org) 94
In Communications of the ACM, long-time FreeBSD contributor Poul-Henning Kamp mocks the idea that the free and open-source software movement has "come apart" and "will end in tears and regret."
Economists and others focused on money — like my bank — have had a lot of trouble figuring out the free and open source software (FOSS) phenomenon, and eventually they seem to have reached the conclusion that it just makes no sense. So, they go with the flow. Recently, very serious people in the FOSS movement have started to write long and thoughtful opinion pieces about how it has all come apart and will end in tears and regret. Allow me to disagree...
What follows is a humorous history of how the Open Source movement bested a series of ill-conceived marketing failures starting after the "utterly bad" 1980s when IBM had an "unimaginably huge monopoly" — and an era of vendor lock-in from companies trying to be the next IBM: Out of that utter market failure came Minix, (Net/Free/Open)BSD, and Linux, at a median year of approximately 1991. I can absolutely guarantee that if we had been able to buy a reasonably priced and solid Unix for our 32-bit PCs — no strings attached — nobody would be running FreeBSD or Linux today, except possibly as an obscure hobby. Bill Gates would also have had a lot less of our money...
The essay moves on to when "that dot-com thing happened, fueled by the availability of FOSS operating systems, which did a much better job than any operating system you could buy — not just for the price, but in absolute terms of performance on any given piece of hardware. Thus, out of utter market failure, the FOSS movement was born."
And ultimately, the essay ends with our present day, and the phenomenon of companies that "make a business out of FOSS or derivatives thereof..." The "F" in FOSS was never silent. In retrospect, it seems clear that open source was not so much the goal itself as a means to an end, which is freedom: freedom to fix broken things, freedom from people who thought they could clutch the source code tightly and wield our ignorance of it as a weapon to force us all to pay for and run Windows Vista. But the FOSS movement has won what it wanted, and no matter how much oldsters dream about their glorious days as young revolutionaries, it is not coming back; the frustrations and anger of IT in 2024 are entirely different from those of 1991.
One very big difference is that more people have realized that source code is a liability rather than an asset. For some, that realization came creeping along the path from young teenage FOSS activists in the late 1990s to CIOs of BigCorp today. For most of us, I expect, it was the increasingly crushing workload of maintaining legacy code bases...
What follows is a humorous history of how the Open Source movement bested a series of ill-conceived marketing failures starting after the "utterly bad" 1980s when IBM had an "unimaginably huge monopoly" — and an era of vendor lock-in from companies trying to be the next IBM: Out of that utter market failure came Minix, (Net/Free/Open)BSD, and Linux, at a median year of approximately 1991. I can absolutely guarantee that if we had been able to buy a reasonably priced and solid Unix for our 32-bit PCs — no strings attached — nobody would be running FreeBSD or Linux today, except possibly as an obscure hobby. Bill Gates would also have had a lot less of our money...
The essay moves on to when "that dot-com thing happened, fueled by the availability of FOSS operating systems, which did a much better job than any operating system you could buy — not just for the price, but in absolute terms of performance on any given piece of hardware. Thus, out of utter market failure, the FOSS movement was born."
And ultimately, the essay ends with our present day, and the phenomenon of companies that "make a business out of FOSS or derivatives thereof..." The "F" in FOSS was never silent. In retrospect, it seems clear that open source was not so much the goal itself as a means to an end, which is freedom: freedom to fix broken things, freedom from people who thought they could clutch the source code tightly and wield our ignorance of it as a weapon to force us all to pay for and run Windows Vista. But the FOSS movement has won what it wanted, and no matter how much oldsters dream about their glorious days as young revolutionaries, it is not coming back; the frustrations and anger of IT in 2024 are entirely different from those of 1991.
One very big difference is that more people have realized that source code is a liability rather than an asset. For some, that realization came creeping along the path from young teenage FOSS activists in the late 1990s to CIOs of BigCorp today. For most of us, I expect, it was the increasingly crushing workload of maintaining legacy code bases...
I mean without open source (Score:5, Insightful)
How would fledgling little companies like Apple possibly compete against the Microsoft domination that was windows vista?
Re: (Score:2)
Re: (Score:2)
By owning their own hardware/software stack. Else they would be reduced to Microsoft's delivery people.
Re: I mean without open source (Score:3)
At the time, much of that stack came from FreeBSD and OpenBSD.
In fact, my biggest gripe with the more recent versions of macOS is that they have not updates those elements, made mac-specific implementations without documenting them, or replaced them wholesale.
Macs were quite popular among the tech-savvy 10-20 years ago. Today? Not so much.
Re: I mean without open source (Score:2)
Re: (Score:2)
Re: I mean without open source (Score:2)
Re: (Score:2)
Lack of elegance (Score:2)
For most of us, I expect, it was the increasingly crushing workload of maintaining legacy code bases...
The lack of elegance means that a lot of churn increases the crushing workload of maintaining legacy code.
windows (Score:2)
Re: (Score:3)
Not surprising. Notepad and Command Prompt in Windows 11 are awful, plus Command Prompt has nasty screen corruption bugs. It's 2024, and these two apps worked perfectly in Windows NT 3.51 . The sad truth is that for Microsoft, lately the quality of their Windows product is of much lower priority than the umpteenth pointless reshuffling of the UI, or the user-hostile "innovations" like telemetry, ads in start menu, settings etc. or the solutions in search of problems like the Copilot.
Anyway, this has hardly
Re: windows (Score:2)
command prompt was always a terrible app that did the bare minimum to support whatever Microsoft needed. it didn't play fair with anything else. in fact iirc it doesn't even support ANSI colors.
Microsoft has been working for a while in the new Terminal app which is much better, and it's integrated to windows components and external stuff. in the same app, in different tabs, you can have a cmd window, a ssh session, and a WSL2 command line. and it finally supports full Unicode.
if anyone thinks FOSS isn't imp
Re: windows (Score:3)
They got people who have to fight in internal politics to get something in and nobody with clear vision steering the ship. They need to fight for that to not get their funding cut so they're highly motivated and would need a wrangler, a captain.
They can't agree how the start would open up from the side so you need to patch 11 to get a movable taskbar etc, the default start is seen(fought for) just as an advertisement space for user attention.
Underneath things have gotten better all along mostly, but the ui
Re: (Score:2)
Another good example of this is copilot. They had cortana already, but for some reason need this new branding and the reason is just internal funding funnels.
Being game-related, and having a sexy avatar, meant Cortana was bad branding. They definitely needed new branding because people were mocking them for their old branding. Wintendo, indeed.
Osx isn't much better, but a bit better in this sense - have you lately tried to find screensavers that still actually work for osx? Good luck with that.
That's become a problem with Linux too, if you run Wayland anyway. Fifteen years later it still doesn't work worth a shit. X is too hard to maintain, now gaze in awe at our new system which is too hard to maintain!
Re: (Score:2)
Fifteen years later it still doesn't work worth a shit. X is too hard to maintain
I think a good part of the problem is that the people who have been relentlessly complaining about how hard X is to maintain and so they needed to make Wayland is better are the ones who made that mess out of it in the first place.
X is a protocol. Sure a bit janky in places and some little used corners but a protocol nonetheless. The server is the code they were hacking on.
Re: (Score:2)
I think a good part of the problem is that the people who have been relentlessly complaining about how hard X is to maintain and so they needed to make Wayland is better are the ones who made that mess out of it in the first place.
I don't think that's entirely accurate. Their complaint was that X was a mess when they got there, and I don't have any problem believing that.
On the other hand, they worked on it for many years and didn't make it less of a mess, so they definitely get to share the blame.
Re: (Score:2)
I'm not inclined to take them at their word entirely.
All software is a mess. The reaction of any programmer on opening an unfamiliar codebase is "ugh" and it only goes downhill from there. How many times have all of us thought "I should just rip this up and start again".
The problem is of course that the world is messy, and code cannot be cleaner than the domain it models. You can make a from scratch clean version by ignoring the complexity of the world and doesn't actually manage to fit a good number of edg
Re: (Score:2)
I'm inclined to trust them about X being a mess because nobody else is stepping up to become the new them, and because so much software is a mess.
I'm not inclined to trust them about their level of competence because... *gestures at Wayland*
I will keep using X so long as it remains viable. Whether this means someone picks it up again or Wayland eventually becomes usable, I can't predict. But one of those things is going to have to happen because there's no third option apparent, and Linux is only growing in
Re: (Score:2)
I'm not disputing it's a mess, but my point is think their assessment of what constitutes a critical Mrs is flawed.
Anyhow I'd love to, but ain't no one paying me to do that and I've got shit that needs doing in my spare time (involving among other things aa relative with Alzheimer's). Plus there seem to be aa lot of paid devs in key places who want X to fail now anyway.
I'll take the same path as you. While X is viable, I'll continue to contribute to various utilities.
Re: (Score:3)
I chuckle every time I see this sentiment. Users with little to no experience developing for X question whether or not X needed to be replaced. The fact that few, if any, of the X.org devs have expressed opinions that Wayland is unnecessary (as drinkypoo has stated) really says a lot. And those same X users expect X.org devs to have to maintain the multiple layers of jank necessary to keep
Re: (Score:2)
The fact that few, if any, of the X.org devs have expressed opinions that Wayland is unnecessary (as drinkypoo has stated) really says a lot.
What? I didn't state that Wayland was unnecessary. You need to work on your reading comprehension skills, and the fact you thought that and then went on to say it brings your entire chain of logic into question because what else did you fail to understand on your way to this point?
What I said is that Wayland as a project was started because the people who were maintaining X said it was necessary, and fifteen years later their solution still doesn't work very well, despite the fact that it doesn't even try t
Re: (Score:2)
The fact that few, if any, of the X.org devs have expressed opinions that Wayland is unnecessary (as drinkypoo has stated) really says a lot. And those same X users expect X.org devs to have to maintain the multiple layers of jank necessary to keep X propped up on modern desktops.
They're mostly the same people, and they are not neutral observers in this. A career in software tells me that all devs almost always think the code is a mess and want nothing better than to rip it apart and replace it.
The other th
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Screensavers are a legacy feature for monitors that don't support power saving modes. I don't really understand the love for keeping the screen on.
Re: (Score:2)
They're also a legacy feature for CRTs. When your monitor is not susceptible to phosphor burn-in, you don't need a screensaver.
Re: (Score:3)
Most later CRT monitors had sleep functions but even newer OLED screens can burn in
Re: (Score:2)
Screensavers are a legacy feature for monitors that don't support power saving modes. I don't really understand the love for keeping the screen on.
On Unix, the screensaver software is commonly the component responsible for turning off the display power when idle. It's also responsible for locking the console when walking away from it. I use xscreensaver specifically, so this is actually [reasonably] secure. (Most of the alternatives for X fail at security.) I find it amusing to have animations behind my login window when I come back to the system, but I could of course use the same software configured to display a black screen or a still image instead
Re: windows (Score:2)
Every screen technology can experience burn in.
Re: (Score:2)
Re: (Score:2)
yea somewhere in my garage is a little 10 inch screen with ADT PRO burned into it from running it so much on my apple //c over the years
Re: (Score:3)
In my workflow, I am constantly dragging windows to screen edges or corners to place them. I imagine that Microsoft has decided that since they gave you these features they no longer need to place applications correctly, but I will also tell you that I have similar problems on Linux. Firefox windows in particular are misplaced on launch, and reliably so, and what's more they are placed low just like you're complaining about Windows doing. Everything else works OK, so FF must be the culprit in this case...
Re: windows (Score:2)
well, I run a dual monitor setup but I like my top monitor to be very secondary. GTK apps are like "wow! so much room for activities!" and expand vertically into the second monitor. granted, they are 720p monitors, but the windows aren't "dense" in the first place. there is a lot of wasted blank space that could be compressed to accommodate the windows.
Re: (Score:2)
I run dual or triple monitors with Windows 10 (soon to be "upgraded" to 11) for work, where one display is the laptop screen. In the office I use dual 20" 1080p monitors (plus the laptop, which I think is also 1080p but I don't actually know, which strikes me as funny now I think of it - display scaling makes it irrelevant in my use case) and at home I use a 40" 4k TV, which has the same real estate as four 20" 1080p monitors (both in size and resolution). What different windows are configured to do often p
Re: (Score:2)
KDE was doing that. My second monitor is the tv in the living room for watching movies. KDE would always place new windows on that screen even when the tv was turned off! Google turned up a tweak for the config file to fix that.
Re: (Score:2)
Some TVs act like they're on (in terms of the interface, that is) when they're off. Some don't. I just switched from using a 40" Sony 1080P 3DTV to a 42.5" LG 4KTV. The Sony was in the first camp, the LG is in the second. Systems would see the Sony when it was off, but only see the LG when it's on. I think it even doesn't communicate with the PC when the input is switched, but don't feel like checking now.
Re: (Score:2, Insightful)
Coders where better 30 years ago.
Re: (Score:1)
Bettor spellors two.
Re: windows (Score:2)
They had a lot less to deal with
Re: (Score:1)
Granted, Eleven is terrible. But it's terrible compared to XP SP2 and later, Vista, and Seven, because those were all built on a mature
Trojan Horses (Score:5, Interesting)
Some of OSS is being destroyed from the inside , eg the Microsoft employed (along with others) Poettering and his abysmal, bug ridden and security risk systemd replacement for init on linux. Don't tell me that the MS culture is not having some influence on the design of this abortion now even if it didn't originally. The unix philosphy for system components (do one thing well) has been comprehensively binned in the case of systemd for an octopus of a system doing things pid 1 has no business doing.
Re: (Score:3, Insightful)
Re: (Score:2, Informative)
Void Linux was one of the first distributions to adopt systemd, then they dropped it. So it is possible to get rid of systemd.
Re: (Score:2, Informative)
You must have heard of it by now, it's debian without systemd, choose your init system upon install. There's already several forks of Devuan. Its definitely gaining ground. You can also "sidegrade" Debian over to Devuan, which rips out systemd, but there's not much point now with Devuan being available.
Re: (Score:2)
While you can run Debian without systemd, it seems to be getting harder to do so. I am now doing new installations as Devuan and the next dist-updates for my remaining Debian systems will also go to Devuan.
When the Debian technical committee let itself be corrupted into approving systemd, and what followed, that was a real low-point for FOSS. Fortunately, enough people were not stupid enough to go along with it.
Re: (Score:2)
Re: Trojan Horses (Score:2)
So it is possible to get rid of systemd.
But not all the shims needed to correct systemd dependencies.
Re: (Score:2)
That you for demonstrating you don't have a clue beyond how to make a fool of yourself.
Re: (Score:3, Insightful)
SystemD(amage) is about Embrace, Extend, Extinguish. What do I win?
When you simply assert a conclusion rather than making an argument, people can easily come along and make you look like an idiot -- if you didn't do that all by yourself.
Re:Trojan Horses (Score:4, Interesting)
SystemD(amage) is about Embrace, Extend, Extinguish.
I don't often reply to ACs, but I'll make an exception in this case. You're spot on correct. That's what SystemD is. It checks all the boxes for EEE, straight from the Microsoft spellbook. If it looks like a duck, swims like a duck, and quacks like a duck, then it's a duck. Had I mod points, I would upvote.
Re: (Score:2)
You are just one of the propaganda-liars. In your case probably an "useful idiot", i.e. not even getting paid for spreading crap.
Re:Trojan Horses (Score:5, Informative)
Don't tell me that the MS culture is not having some influence on the design of this abortion now even if it didn't originally.
It always did, even if Lennart didn't have any input from Microsoft in its early days, which I do not take for granted given Microsoft's history.
systemd was always at best an attempt to turn Linux into Windows.
The absurd irony of this is that Windows has multiple ways to do everything because of cruft upon cruft, so it wasn't even a faithful attempt.
The good news is that it is still totally possible to have a Linux system without systemd. Devuan handles the problem areas that you run into if you run Debian and install sysvinit, for example. It requires dozens of little tweaks throughout the system to undo the enshittification, and they handle these for you. There are other distributions as well, of course, but some of them are also deeply wacky. Debian is notable for giving you minimally-mangled software.
Re: (Score:2)
Some of OSS is being destroyed from the inside , eg the Microsoft employed (along with others) Poettering and his abysmal, bug ridden and security risk systemd replacement for init on linux. Don't tell me that the MS culture is not having some influence on the design of this abortion now even if it didn't originally.
LOL. Man the Poettering haters are off the deep end with conspiracies. He's been hated on for decades and his contributions labelled a disaster, but now suddenly it's Microsoft's fault and "destroying OSS from the inside".
The unix philosphy for system components (do one thing well) has been comprehensively binned in the case of systemd for an octopus of a system doing things pid 1 has no business doing.
It's just as well there's no OSS software that doesn't run Systemd. None. Not a single one. Every OSS project is forced to use Systemd. Yessiiirrreeee.
Okay sarcasm aside I want to take a time out and call you stupid for equating OSS with necessitating Unix philosophy. Two different concept
Re: (Score:3)
I am personally happy with systemd. When Debian switched to it (from SysV init), my system booted faster because of parallelism, I could reliably start and stop services, and creating a new one didn't involve writing a shell script with hundreds of lines of boilerplate. And speaking about Poettering, I have no problem with Pulseaudio either. My audio setup is somewhat complex, and it does exactly what I want it to do.
Sure enough, these tend to stray away from the "UNIX philosophy", but so what? And systemd
Re: (Score:1)
I am personally happy with systemd. When Debian switched to it (from SysV init), my system booted faster because of parallelism
I run servers. They measure their uptime in years, not hours.
I don't need parallelized boot because I only do it once every few years. Relevant XKCD [xkcd.com] for why System===D was a complete waste of time.
As for my workstation which also only reboots every few months, it takes about 90 seconds to get to the login screen with SystemD and before I was forced to switch by Debian it took about 2 minutes. Big fucking deal. I walk into my office, reboot my computer, make a cup of coffee, walk back to my desk and it's up.
My laptop is running FreeBSD--no System===D there. It takes about 2 minutes to boot. That's why I don't walk into meetings 3 minutes late and bitch about my laptop taking "forever" to boot.
I could reliably start and stop services
Huh--I do that using rc in FreeBSD just fine.
and creating a new one didn't involve writing a shell script with hundreds of lines of boilerplate
Cool story. Creating new services doesn't require me to write a new shitty init system.
My audio setup is somewhat complex, and it does exactly what I want it to do.
Difficult to compare such vagaries, but I have a wired headset, bluetooth headset, pixelbuds, 6 HDMI monitors with speakers, a webcam for video conferencing, a webcam focused on a project board so I can show connections being made on a breadboard, and I have a bluetooth surround sound system for music. It does exactly what I want it to do when I want it to do it. I have no idea what my audio library is, but a dpkg -l | grep -i pulse shows it's not pulse audio.
Sure enough, these tend to stray away from the "UNIX philosophy", but so what? And systemd is actually made of small modules that work together
The Unix Philosophy is to "do one thing and do it well". SystemD is "consolidate everything in to a monolithic project, spew out shit code, and when something fucks up hose the entire system". Unix and Linux have worked well for decades without SystemD.
Linux is a monolithic kernel, it didn't stop it from being a huge success.
modprobe, lsmod, and rmmod disagree with you.
SD more closely adheres to this philosophy, and it is fine too, having an active community and all that. Feel free to use BSD if it aligns more with your values and needs.
There is it. Just like in politics. "I really don't like how you're fucking up XYZ."
"MOVE TO AFGHANISTAN IF YOU DON"T FUCKING LIKE WHAT I'M DOING! THEY DON'T HAVE SYSTEMD EITHER!!!!@##1223one."
Could systemd be better, probably, but it makes my computers run well enough.
Spoken like you've never tried init. I've never had init break on me. I've been using Linux since ~1998.
System===D* has failed me so many times since it was introduced I ended up switching our production equipment to FreeBSD to avoid it.
I have some issues with Linux, but the init system is low on the list.
...and yet you're defending SystemD as if it's super important you have it for various nebulous reasons.
Re: (Score:2)
> modprobe, lsmod, and rmmod disagree with you.
Linux is still a monolith even though it supports dynamic linking.
The reason is that it doesn't have a stable API and even less of a stable ABI. To write a module, you have to compile it with the current kernel header files, even a minor change in the kernel will break your modules, that's why DKMS is a thing. And even with that, the kernel devs may at some point decide to break the API you are relying on, without warning. The way you are supposed to have li
today's foss world (Score:4, Insightful)
Re:today's foss world (Score:4, Informative)
"My laptop spies on me,"
Depends what OS you use. Windows definately, MacOS probably for some things, LInux - depends on the distro and what window manager (sorry, not allowed to call them that now, I mean Desktop) you run on it. I run Slackware so no phoning home on mine and yes I have checked using a packet sniffer.
Re: (Score:2)
what window manager (sorry, not allowed to call them that now, I mean Desktop) you run
The two aren't the same thing. A window manager is all you need, but a desktop has a session manager and a file manager which implements a desktop and some other stuff. I run a desktop (XFCE4) but I have swapped out a bunch of pieces of it and decided exactly which bits I want running at all. Hooray for choice.
Back when I had a limited system with little RAM I was all about fvwm1, I had come from systems with Motif and it best offered the same functionality (down to look and feel) for minimal resources. Now
Re: (Score:2)
Depends on what applications you use. Whether we like to admit it or not, the OS doesn't do much for security except cover its own butt. Almost all security is done at the application level.
The mobile world kinda is "anti-FOSS" (Score:2)
Yes, mobile platforms today are mostly published with source code, but you are not supposed to be able to compile and run it yourself. Instead companies like Google and Apple see you more as a product than a customer.
The time would be ripe for a new ecosystem... however there is one major difference.
Back in the 1990s PCs were well specified. If you wrote an operating system for one computer, put it on a bootable disk, you were very likely to boot it of any other PC. There was no need to port Linux from Comp
Re:The mobile world kinda is "anti-FOSS" (Score:4, Interesting)
Similarly, (again from recent test and experiences) Whatsapp, for instance, also flags usage on phones without Google Mobile Services and locks you out. Hey, the program installed fine with a message (you won't have access to some functions). Ha ha... that's not really true, what they mean is THEY won't have access to certain functions.
I think the conclusion is that the big players are closing ranks to allow information interoperability with each other and no private services from outside that ecosystem. Very dystopian.
Re: (Score:2)
I think the conclusion is that the big players are closing ranks to allow information interoperability with each other and no private services from outside that ecosystem. Very dystopian.
Agreed. They definitely do not want you to have any control over your systems.
Re: The mobile world kinda is "anti-FOSS" (Score:2)
then try to login to Microsoft Hotmail/Outlook/website it's flagged as "suspicious activity"
Of course. The goal is to wed you to one unique hardware "token" that identifies you, allows your access to resources and cannot be easily duplicated. You should have just taken the Gates Covid vaccine with embedded tracking chips when it was available.
How many people use platforms which support true multi-user accounts (without some deep config. hacking or side loading an environment)? You are your phone and your phone is you.
I must be driving my bank mad by running secured X from my overseas desktop to
Re: (Score:2)
So... I'm not against using a hardware token. True authentication of my incoming login request is something I want, even with the big bad giant companies. I want my account to only be accessed by me. So in the case of token to identify my hardware, I don't object. I still object to the reality that MS and all the others are reaching into my pockets to take whatever I have there if it'
Re: (Score:2)
True authentication of my incoming login request is something I want,
Even when you are trying to go on line anonymously? They have your hardware fingerprint. If you try to put up "ad blockers", you must be an evil person. We need to know who you are.
They're not "ad blockers". They are tracking blockers. Because they can still serve up all the ads they want. They just can't link me to my bank balance, credit report, or zip code. And then redirect me to the DeLux Model web pages with all the high prices. Because people that live down the road from Bill Gates just can't be all
Re: (Score:2)
For instance I have various web based cloud control panels, another for voip configuration.. it's extremely important I and only I have access to those, and for both the cloud and voip accounts, I can switch Ip addresses, I use multiple vpns and IP address
"Device" (Score:2)
You need "device" to have "app" so "payment method" can be attached to your "subscription", "human". Add your phone number for your security. SMS (texting) fees may apply. Check with your "carrier". Enter your "provider" credentials for credit to view this "content." "Like" and "subscribe". Are you not not a robot?
: Cloudflare spins spinning
Please wait while we loop again
still not verify'ng.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Overall, it feels like there's a lot of stuff available for free, but [...] if I try to step outside the guardrails, somehow that's not supported and crickets.
It's always been if you don't like it, submit a patch. But what I find especially BS about your comment is that it's otherwise a description of how the closed source stuff works, not the open; some of it does some of that stuff, but you can turn all of it off in exactly the way you can't with closed software. My computer doesn't do any of that stuff you complain about. I turned off search suggestions, for example. My laptop doesn't spy on me, I run Devuan Linux, not Windows 7 or later. My FOSS applications
Re: (Score:2)
It sounds a lot like you don't know how to use your software and hardware and are just making sure we all know it.
There is a lot of that going around. Remember that most computers these days are used by non-experts and many do not know they are not experts. Dunning-Kruger Effect at work.
Re: (Score:1)
Re: (Score:2)
Some of them, yes. Others are the minority that gets it in a sea of bullshit, hype and fanboyism.
The evernet will kill open source (Score:4, Insightful)
You don't have to give anyone the code if they always just connect to your server. They only have terminals and they like it that way.
It's just one fork away. (Score:3)
User Power (Score:2)
Re:User Power (Score:4, Interesting)
UNIX was reasonably priced, in that FreeBSD lite existed. But the documentation sucked and the community was unfriendly about it, so Linux was able to grow up and take over free Unix-land. I had already done multiple SunOS installs (including on a 3/260, where you had to relink libc just to get DNS resolution) when I first tried FreeBSD, and found the install documentation to be incomplete and misleading. When I tried to get help from FreeBSD users they acted like I was an idiot for asking simple questions about things like which partition letters to use for what, and I was coming from a BSD-based system where the conventions were similar but slightly different and the documentation did not offer any explanation whatsoever. One small stack of Slackware floppies and I was up and running and never had a reason to look back.
I dabbled with OpenBSD, which had functional documentation and also an actual reason to exist (its development ethos) but by that time the world had moved on to Linux. Virtually all Unix software was being developed there in whole or in part (especially the free stuff, but also a lot of the closed commercial software had Linux support by then) and making it work on a BSD had become difficult in many cases. It didn't help that all commercial Unixes were SysV by then, either. Running a *BSD had become the plucky hobbyist's choice, and Linux was mainstream Unix. Today you can choose to buy systems from IBM with it running on the metal, when once they used BSD for their experimental Unix forays (e.g. for ROMP.) BSD may not be [officially] dead but it's definitely on critical life support, and I specifically blame the community. I've written dozens of howtos for Linux which helped snowball it along in whatever tiny part I can be said to be responsible for, and I did it because others did the same.
Anyway the point of this probably excessively long post is that there are lots of different things that drive OSS success. In my case I was familiar with the superiority of the design of Unix over that of Windows, so I was driven to run the superior system on my equipment. It was not about zero cost because it was already easy to get software without paying for it over the internet by that time, in fact the people facilitating that were more competent and somehow apparently more driven than the ones getting paid to distribute it legally because it was in many ways more convenient for the end user. It was about quality and choice. And it was not about experimentation, either; It was strictly about quality. I had plenty of experience with Windows exploding without apparent reason, and conversely, with Unix chugging along without problems for literally years on end, since we had so many systems not connected to the internet yet and it was also at the time possible to put a system on the internet and not have it owned in seconds unless it was running IRIX. ;)
Correction (Score:2)
BSD lite existed. It begat 386BSD, which begat FreeBSD, which existed by the time I tried to run a BSD on a PC.
Money (Score:2)
Yes when focussed on money everybody has a lot of trouble figuring out the free and open source software. That's because they myopically interpret free as gratis, having a zero price, alas that is not the connotation of free here.
Mind you nobody - not even banks - has any trouble with OSS when it comes to making use of it and laughing all the way to the bank because of the cost savings...
classism is the underlying problem (Score:3)
Classism leads to corruption which leads to incompetence. We see this time and time again. People just aren't ethical enough to keep their hands out of the cookie jar they've been 'entrusted' with.
Our greed is killing us.
Home use and dotcom (Score:4, Interesting)
I was able to use Unix in college & wanted to have it at home instead of DOS. I used some programs people ported to DOS from Unix. People wanted the tools locally so they didn't have to be on the servers. I was helped quite a bit by editing & running LaTeX on my DOS PC instead of the college VMS system. I only needed to do the final printing on VMS. After college, I continued with DOS and the Unix ports. I bought Minix for my 286 because there really were not any Unixen for it.
I was eventually able to buy a 486 but Unix were still > $1000. Luckily, 386BSD was being published in Dr. Dobbs. Unfortunately, it would not run out of the box on my system. But Linux did. I could finally use the full GNU tool set on my system. All that work I did at home with Unix tools lead to a sysadmin job with Unix.
I compiled lots of free Unix programs for the systems we had: SunOS, Irix, HP-UX, Solaris, AIX and put it in NFS /usr/local. Why did we do that? The tools provided with the Unixen didn't work as well or were not available. Our users programmed in Fortran and C. Documentation was in LaTeX and edited with emacs. It would have been great to have an easy to use spreadsheet or word processor at a reasonable price on Unix. Instead, we bought Macintoshes & MS Office for those that needed it. Mosaic came out & it ran on Unix and Macintosh. DOS and Windows didn't have a standardized TCP/IP stack so Mosaic wasn't there w/o 3rd party software.
Later on, when dotcom happened, everyone was running on Sun systems. We bought millions of $$ worth of Sun equipment. When the market fell, we couldn't afford Sun. By this time, Linux could run everything the Unix systems could. Even better, it came with all the free stuff we used to have to compile. Most of that software started to be developed on Linux and ported to Unixen. SunOS was no longer the 1st supported OS in the Makefile; it had become the adapted OS.
The Pentiums were cheaper than Unix hardware. Engineers could get a Pentium easier than a Unix workstation. They could dual boot it from IT's Windows 95/98 into Linux at work. They could buy them at home. College students could also do that instead of logging into the overloaded Unix servers for classwork. They found that Linux's TCP/IP stack ran faster than Solaris 8's. Solaris 8 was still optimized for 10T and 8MB RAM. Solaris 9 was an effort to catch up & optimize for 100T and > 32MB RAM like Linux did. By that point, some had switched to Linux because it proved to be a better performer. The LAMP stack was developed on Linux and the Unix systems got the port/adaptation.
This was the reason Unix eventually lost to Linux. Companies no longer had the buckets of $$ to spend so they were looking to save. Once enough of the important software was primarily developed (& optimized ) on Linux, Unix became a 2nd tier for web operations.
Re: (Score:2)
DOS and Windows didn't have a standardized TCP/IP stack so Mosaic wasn't there w/o 3rd party software.
Microsoft actually did have their own TCP stack for Windows 3.1, but it was late to the party and had a not very good GUI. Most users ran Trumpet Winsock, which was readily available and cheap (and could be easily used without paying.) The very best stack was from TGV ("two guys and a vax", later purchased by Cisco and turned into a Cable Modem dev/QA lab after toying with and then abandoning the idea of making a competing TCP stack for NT) and it was fast enough that you could reasonably use a DOS/Windows
Re: (Score:2)
DOS and Windows didn't have a standardized TCP/IP stack so Mosaic wasn't there w/o 3rd party software.
Microsoft actually did have their own TCP stack for Windows 3.1, but it was late to the party and had a not very good GUI. Most users ran Trumpet Winsock, which was readily available and cheap (and could be easily used without paying.) The very best stack was from TGV ("two guys and a vax", later purchased by Cisco and turned into a Cable Modem dev/QA lab after toying with and then abandoning the idea of making a competing TCP stack for NT) and it was fast enough that you could reasonably use a DOS/Windows 3.1 PC with a few NICs in it as a router for a small network. It also had a really good config GUI. All of them used the same interface (winsock) so you could swap them out interchangeably.
When Mosaic came out, we had 1 system running NT 3.1 and maybe 1 running Windows 3.1. I think Windows for Workgroups 3.11 was the 1st to have TCP/IP included. I think that was a few months (a year?) away. We were not going to get Trumpet Winsock though, we were a Unix shop with Macintoshes running System 7.
The Macs were mostly on Farallon localtalk over phone wire and we had a gateway to ethernet (gatorbox? Fast-mumble?). I had fun getting Columbia Appletalk Package patched (100 patchdiffs!) and running
Re: (Score:2)
netatalk was a godsend all right. Our Linux server in the geekhouse I lived in back then ran netatalk, samba, and nfs and we had windows, macs, a couple of Suns and one vt100 talking to it...
I think this central premise is false (Score:2)
if we had been able to buy a reasonably priced and solid Unix for our 32-bit PCs â" no strings attached â" nobody would be running FreeBSD or Linux today, except possibly as an obscure hobby.
I don't think that this is true at all. One of the hardest parts of changing from any commercial PC desktop OS to any Unix was and remains that software isn't there for it. Only a small set of popular software has ever been directly ported, and only a limited number of versions were ever implemented before the developers gave up on those ports. For example, we have had Wordperfect Office but only a few versions and never Microsoft Office, Adobe Photoshop but only a couple of versions and only for a couple o
What's next? (Score:1)
If the frustrations and anger of 2024 are different, what are the new areas to reboot and reimagine?
(I miss the revolutionary days and can't believe there's nothing out there to overthrow, while everything today feels so ... settled.)
maintaining legacy code bases ... (Score:3)
I have worked, starting from PDP11s, New code was put in the public domain to avoid the need to maintain it. (Google: DECUS).
It was useful code, and others would not just maintain it but grow the functionality, once it was able to run wild and free.
Trivial utilities grew to be great wild beasts.
The secret of success was to keep the PHBs in the dark.
Netcraft: Poul-Henning Kamp Is Dying (Score:2)
wait, did I get that headline quite right? :)