Re: The report of my death was an exaggeration

[Maturin]

Except that it is very hard to find out whether any random computer in a shop is made of hardware that is supported, especially for laptops. Tech specs are a joke these days.

I couldn't agree more, especially for the latter part. You want to know a certain hardware specification, and all you get are pictures, videos and "...only the highest quality is handselected for our finest swill..." blablabla. I hate that! It really pi*cough*
I just wanna put some positive perspective into the room. Since because of the laptops I'm wrong, but on self assembled desktop machines you have choices, so I'm 40..60% wrong/right?

If we only purchase hardware that is already supports FreeBSD, we soon will find ourselves limited to only old, outdated hardware.

I see your point. But don't oversee many drivers we have here originally came from Linux.

We users need to pressure hardware companies to develop native drivers for FreeBSD.

Fooh. Tough call. I think you're a bit too idealistic, optimistic with that attidute. It took almost thirty years e.g. NVidia started to provide drivers for Linux on their Webpage, and since shortly for FreBSD, too.
I may be not on the current situation, but AMD still don't (at least not for their GPUs)
How many percentage does Linux have on the market? How many FreeBSD? (which brings us back to topic)
But as I said in my first post: I am optimistic. I think/believe/feel more people join FreeBSD.

At the same time, the "complaints" need to be properly directed - towards the manufacturers of the hardware.

Good one. No, seriously, sorry. In theory you are right, but again you oversee two things:
1. Nobody cares about complaints. Only numbers sold count. If they see their numbers drop, and the number of a competitor rise, and figure out it's because he not only offers good hardware but also provides drivers for FreeBSD/Linux, then they will provide drivers of their own within few weeks.
2. Ever tried to complain about something at a company?
"Have you read our FAQs?"
"Yes, even I knew it before it was nothing but marketing BS questions I cannot believe any of those was ever actually asked by a real customer I read them all anyway, yes, that's why I'm searching for almost half an our on your stupid webpage for something like to contact someone. Aha! Link found to contact formular... at least better than nothing."
"FAQs"
several minutes later of and, after "ask a friend" or "why not ask the forum?":
"welcome to our chatbot! How can I help?"
Gimme a nailgun!!
I had (too) many of those.
From Apple's support I still have a bogroll of E-Mails containing:
...
"[lots of formal garbage] Hello, my name is [anybody #34] How may I help you? [more formal garbage.]"
"Well, I already told your collegue, and it's all in the mails below I already wrote. If you just read it."
"[lots of formal garbage] Please tell me again! [more formal garbage.]"
"...explain, eloborate, blablabla...."
"[lots of formal garbage] Okay. I will give my life to give you the fastest and best solution you have ever experienced [more formal garbage.]"
After several days:"[lots of formal garbage] My name is [anybody #35] We are happy we could help you. Can we close this ticket now? [more formal garbage.]"
"No."
"[lots of formal garbage] Hello, my name is [anybody #36] Why not? [more formal garbage.]"
"Because I did not receive any answer yet."
"[lots of formal garbage] Hello, my name is [anybody #37] How may I help you? [more formal garbage.]"
...
If you find yourself fantazising of driving a truck full of ammonium nitrate and nitromethane into a parking lot in Cupertino, California you know you need to stop and face reality: it's useless.
A couple of days later you think of bombs again when you receive the usual
"Please give us five thumbs up for our support"-mail

Short:
Stop buying certain products, or at certain companys is the only real power consumers have.
And don't you underestimate that!

 

[Jose]

Yes, sometimes happens, but analize that cases, one user came with actitude (and words) like "how to do this"
"dont, I dont wanna do it" ,the 90% of the users that are "rejected" cames in with a bad actitude

We also get a number of Linux zealots bent on showing us the error of our ways. They usually post questions that aren't really questions, but rather statements of how Linux does X so much better, and how can you live without X?

 

[Jose]

Claiming that this dominance is caused by a brand name effect, and not by technical considerations, is claiming that people who spend somewhere between tens of millions and tens of billions on computers make decisions based on emotions and familiarity with brand names. That seems far fetched, and is quite insulting to decision makers. If someone told the folks who just bought El Capitan, or who run tens of millions of FAANG servers on Linux (and few or none on FreeBSD) that they ignored technical merit, and spent $600M or $1B/year based on brand name familiarity only, they would get very mad, justifiably so.

I largely agree with you, but have found that our industry is often driven by groupthink. I'm sure you remember the days of "no one ever got fired for buying IBM (and later, Microsoft.)"

I ironically experienced this early in my career when I could not convince management to use Linux. It couldn't be any good if it was free. It drove my out of the sysadmin career track and into software development because I just couldn't stand trying to support Windows servers. Sun machines were so expensive that I was frankly terrified whenever I had to work on one.

And the tendency to substitute the designation of a thing by a specific brand name is mostly a US thing. Even if the argument was true (which I don't think it is) then it wouldn't explain what is happening in the rest of the world.

This is also common in my native Uruguay, and I suspect Argentina as well. A ballpoint pen is a "birome", and a gas stove is a "primus".

 

[ralphbsz]

I largely agree with you, but have found that our industry is often driven by groupthink. I'm sure you remember the days of "no one ever got fired for buying IBM (and later, Microsoft.)"

I mostly disagree. The effect you describe may exist when non-computer-savvy people buy computers. And given their lack of knowledge, going with the largest and most reputable supplier even makes perfect sense.

I spent nearly 20 years working at IBM (not in sales nor in a customer-facing role), but I do understand why people buy IBM, and don't get fired for it: Because for the most part (much more than any other vendor), IBM stood behind its contracts. If you negotiate with IBM, enter a contract, and the resulting system doesn't work the way it is described in the contract, then IBM will usually move heaven and earth to make it work, or hold you harmless. Because it has a reputation to defend: People who buy IBM should not get fired, because they after all made a safe choice. IBM knows that its reputation (a.k.a. brand name) stands for "trustworthy". Now, this doesn't always work. For example that philosophy stopped working where IBM sold consumer gear (that is, the PC), because then there is no negotiation, no contract, and IBM didn't know what the customer really wants.

Today, for large installations (which are really all that matters from a total number of installed systems viewpoint), the choice of platform is done by very computer- and business-savvy people. If for example Livermore and Amazon use Linux, I fully believe it is because they have evaluated it to be better for their needs and wants.

I ironically experienced this early in my career when I could not convince management to use Linux. It couldn't be any good if it was free.

And to be honest, early on it wasn't any good. I was using HP-UX, SunOS/Solaris and AIX in the early 90s. The Linux of the mid- to late 90s was a hobbyist's toy. It worked like a toy. But that changed relatively quickly, and by about 2010, Linux had successfully eaten all other Unix's lunch.

This is also common in my native Uruguay,

Sadly, I do not have an Uruguayan passport. If I had applied for one early enough, I could have, since my mother did have one. But the time for that was about 50 years ago, so today I "only" have a Brasilian passport. I always joke that I am indeed an American ... a South American. And my oldest passport says on the front cover (translated): "United States" ... of Brasil. So I am from the US, even from the US of A or US of B, just not from Trumpolandia.

 

[USerID]

Let's talk about death, since we're on the subject.
It doesn't exist. There's only a state.
Here's a great example that FreeBSD is alive and doing very well in its niche. Everything is written out and laid out in the report.
https://freebsdfoundation.org/wp-content/uploads/2024/05/FreeBSD-2024_surv_web-_Final.pdf
No need to invent anything.
Someone has already written here that this OS is niche. This is a state: good or bad, progressive or not, but this is the current situation.
The only thing that surprises me in this official report is page 15.
As far as I've noticed, people often can't overcome (change, alter) their habits formed over the years. Holding Хorg.
Considering that our average age is far from great, we can be indulgent and loyal to this.

Another thing I've noticed while reading Russian-language forums is that we are often called grandfathers stuck in the past. Those who write this are typical Ubuntu users, Mint users, on a laptop, average age 28-35. Grandfathers there are 47+. This is a correlational judgment, but it is present.
I forgot to add about corporations: they have a lobby in the right circles and a very powerful legal mechanism.

I will give an example that happened recently (2 years ago). In my country, a medical reform was launched. In medical institutions (of all types of ownership), they began to powerfully introduce the so-called "Medical Information Systems". So, a large number of players entered the market for these systems. I will not burden you with nonsense and details, but one company won this "competitive and fair" fight, which was lobbied from the capital by major government officials. Since my country is the most corrupt in the world today (there are official statistics), everything was simply bought: the capital passed down instructions, and everything went along the corruption chain.
This system is https://helsi.me/ .
Big money and connections are tied to it. The other players ultimately sucked.

That's why there is "Game Theory", which is mathematically calculated "for themselves". And if the corruption machine (lobby in government agencies) is also working, then winning this government tender is almost impossible.
They monopolized the market for medical information systems. Moreover, they captured all areas: from clinics to family doctors.
They grabbed 80%+ of the market. Moreover, according to the history of their activities, they did not work their asses off: their product at that time was at the level of a shitty beta version.
But the product was launched in production!!! We tested it to the health of people. We collected cash from subscription fees and built an empire. Although there were already long-tested and stable systems. I myself have had my fill of this shit. The registry - hung, therapists - hung, laboratories - hung. More precisely, their databases in cloud data centers hung. It was terrible. My work was reduced to regularly clearing browser caches. And if the gateway automatically switched to a backup provider (via opensense, failover), then their system incorrectly identified the account and all the work of doctors fell into the abyss!

This business most likely belonged to someone from the near-political elite. Then this system was bought by "Kyivstar" (the largest mobile operator) along with the patient database. Although this is a gross violation of the confidentiality of user data (cards, diagnoses, statistics of medical procedures, pharmaceuticals, etc.).
That is, patient data was transferred to private hands. Moreover, it was leaked to the data centers of merchants. There were scandals and articles on the Internet, but the crowd obeyed, no one twitched.
And the program for medical systems began as a "state" one. I gave an example of state lobbying (pushing) its private business projects through state mechanisms.
There is always a beneficiary in such procedures! Business and politics are very closely intertwined.
In this sense, we are still on the sidelines. But this is our sidelines, our niche. The least cluttered and polluted.

That is why I often write as an anarcho-BSD member.

 

[wolffnx]

Let's talk about death, since we're on the subject.
It doesn't exist. There's only a state.
Here's a great example that FreeBSD is alive and doing very well in its niche. Everything is written out and laid out in the report.
https://freebsdfoundation.org/wp-content/uploads/2024/05/FreeBSD-2024_surv_web-_Final.pdf
No need to invent anything.
Someone has already written here that this OS is niche. This is a state: good or bad, progressive or not, but this is the current situation.
The only thing that surprises me in this official report is page 15.
As far as I've noticed, people often can't overcome (change, alter) their habits formed over the years. Holding Хorg.
Considering that our average age is far from great, we can be indulgent and loyal to this.

Another thing I've noticed while reading Russian-language forums is that we are often called grandfathers stuck in the past. Those who write this are typical Ubuntu users, Mint users, on a laptop, average age 28-35. Grandfathers there are 47+. This is a correlational judgment, but it is present.
I forgot to add about corporations: they have a lobby in the right circles and a very powerful legal mechanism.

I will give an example that happened recently (2 years ago). In my country, a medical reform was launched. In medical institutions (of all types of ownership), they began to powerfully introduce the so-called "Medical Information Systems". So, a large number of players entered the market for these systems. I will not burden you with nonsense and details, but one company won this "competitive and fair" fight, which was lobbied from the capital by major government officials. Since my country is the most corrupt in the world today (there are official statistics), everything was simply bought: the capital passed down instructions, and everything went along the corruption chain.
This system is https://helsi.me/ .
Big money and connections are tied to it. The other players ultimately sucked.

That's why there is "Game Theory", which is mathematically calculated "for themselves". And if the corruption machine (lobby in government agencies) is also working, then winning this government tender is almost impossible.
They monopolized the market for medical information systems. Moreover, they captured all areas: from clinics to family doctors.
They grabbed 80%+ of the market. Moreover, according to the history of their activities, they did not work their asses off: their product at that time was at the level of a shitty beta version.
But the product was launched in production!!! We tested it to the health of people. We collected cash from subscription fees and built an empire. Although there were already long-tested and stable systems. I myself have had my fill of this shit. The registry - hung, therapists - hung, laboratories - hung. More precisely, their databases in cloud data centers hung. It was terrible. My work was reduced to regularly clearing browser caches. And if the gateway automatically switched to a backup provider (via opensense, failover), then their system incorrectly identified the account and all the work of doctors fell into the abyss!

This business most likely belonged to someone from the near-political elite. Then this system was bought by "Kyivstar" (the largest mobile operator) along with the patient database. Although this is a gross violation of the confidentiality of user data (cards, diagnoses, statistics of medical procedures, pharmaceuticals, etc.).
That is, patient data was transferred to private hands. Moreover, it was leaked to the data centers of merchants. There were scandals and articles on the Internet, but the crowd obeyed, no one twitched.
And the program for medical systems began as a "state" one. I gave an example of state lobbying (pushing) its private business projects through state mechanisms.
There is always a beneficiary in such procedures! Business and politics are very closely intertwined.
In this sense, we are still on the sidelines. But this is our sidelines, our niche. The least cluttered and polluted.

That is why I often write as an anarcho-BSD member.

for that reason I gone to change my course in bussiness, I gonna make me a farmer..make a bakery shop
just the only to survive , I'am sick
even the linux jobs are fake here,all is a bussiness of a recruiters and agencies

 

[bgavin]

Except that it is very hard to find out whether any random computer in a shop is made of hardware that is supported, especially for laptops. Tech specs are a joke these days.

There are many top grade freebies you can add to the Hirens Boot CD that will tell you every bit of hardware in the booted machine.
If they won't let you check out the machine, take your business elsewhere.

For my support business, I audit customer machines for the Device IDs of installed hardware and drivers.
I have a vast array of drivers collected over many years.
I wrote a utility to recursively search all the .INF files (windows drivers) to find near or perfect matches for the Device IDs.

I'm here on FBSD because I admire greatly the operating system.
Otherwise, FBSD doesn't make me a dime, whereas VMware and Windows do.

 

[jbo@]

[...], but rather statements of how Linux does X so much better, and how can you live without X?

Sorry did you mean wayland? It is clearly superior in all possible ways.

SCNR

 

[Jose]

And to be honest, early on it wasn't any good. I was using HP-UX, SunOS/Solaris and AIX in the early 90s. The Linux of the mid- to late 90s was a hobbyist's toy. It worked like a toy. But that changed relatively quickly, and by about 2010, Linux had successfully eaten all other Unix's lunch.

I 100% agree. Linux was totally not ready for prime time in the late '90s and probably into the early aughts. So why did it get so much momentum? I can only offer two anecdotes from early in my career. First one is from a large and well-known software firm that's still very large and in charge nowadays. I heard rumors of a Freebsd port of our flagship product that was never made available externally. The reason why was our infamous CEO had decided that Linux was "the future" and had issued an edict from on high that it would be supported in a year. Also at this company there was a strict prohibition on sharing source code with any third parties. Thus the programmers tasked with the port could not share code with the Linux project which lacked several features without which our product just would not work. I heard stories of pointed hints dropped into public mailing list messages.

Some years later I was working at a medium-size software company that is now gone thanks to several acquisitions. My bailiwick was to fix bugs in the current product (as opposed to working on new features.) This meant I got to see a lot of the older branches of the code. To my surprise, I found several Freebsd #ifdefs. I did some software archaeology and was able to get the Freebsd port to compile. I was unable to share it with anybody, of course, on pain of getting fired. Besides, it was a dead end because all the code was gone from the latest versions. I also found out that port had been a love project from mostly programmers who were long gone. So why was it never "productized"? Because by that point "everyone's using Linux" and no one's ever heard of this Freebsd thing.

These anecdotes reinforced the feeling that our industry is completely incapably of resisting fads. There's a mad scramble whenever something new and shiny appears. Everybody rushes onto the bandwagon regardless of the technical merits of whatever the fad du jour is. I'm hoping people move on from AI soon so I can look at it in peace.

Sadly, I do not have an Uruguayan passport. If I had applied for one early enough, I could have, since my mother did have one. But the time for that was about 50 years ago, so today I "only" have a Brasilian passport. I always joke that I am indeed an American ... a South American. And my oldest passport says on the front cover (translated): "United States" ... of Brasil. So I am from the US, even from the US of A or US of B, just not from Trumpolandia.

That's a bummer. You could've been an "Oriental" too. That's what we're sometimes called because the full name of our little country is "The Oriental Republic of Uruguay."

 

[Espionage724]

I first heard of Linux when a PC Gamer magazine advertised Ubuntu (at or around 6.06). Some time later I had Compiz cube and wobbly windows on Intel 950GMA showing off to friends, and dozens of free Ubuntu/K/X and I think even Edubuntu CDs Prior I used Windows exclusively since 98.

I haven't seen an OS advertised like that since. I'm not sure how that could work today, but I thought the PC Gamer ad and CDs were cool advertising! The CDs made it easy; they shipped em to my door for free, and all I had to do was boot one to get hooked. Nowadays you have to be encouraged to download an image and flash it to a USB, or other manual process.

 

[ralphbsz]

I 100% agree. Linux was totally not ready for prime time in the late '90s and probably into the early aughts.

I would see that more differentiated. I was doing production work on Linux in 1995 and 1996. At home, I had a two-node setup (one 386-40, one 486-25), connected via "network", initially PLIP (since supported Ethernet cards were hard to find). And when I say "production work", it means serious data analysis using both Fortran and a packaged software (CERN's PAW), reading and processing data from 8mm tapes by the gigabyte, and such. So was it "ready for prime time"? Not in a professionally managed server cluster setting, in particular since system administration on Linux was very undeveloped, and system upgrades unmanaged and painful (I remember using 30 floppies to upgrade the SLS distribution). Yes, in an environment where users are willing to put in many hours per machine to install and administer. So Linux was both not ready for prime time, and yet ready and being used heavily.

I had actually tried to obtain BSD before, but utterly failed. 386BSD (the Jolitz port) might have existed, but few people knew about it being available as open source. There was also a wide-spread impression that 386BSD could not be run, since it was missing the AT&T copyrighted code (the knowledge about Tahoe was not widespread). You could get 386BSD from BSDi, for the small sum of about $1000 (twice the cost of the hardware). It didn't help that USL (a part of AT&T) was suing BSDI, which made the survival of BSDI questionable. On the technical side, BSDI had de-facto no windowing system available, with the X server being a year or two late to be released, and then only supporting an exotic video card (the Tseng ET-5000) that was nearly unobtainable, even in the heart of Silicon Valley. I needed X to be able to do graphical data analysis (as did thousands of other physicists, which is why PAW was ported to Linux so quickly). So if you wanted a home Unix machine with X, it was Linux, or buying a commercial workstation (DEC, HP, Sun were all available at about $10-20K).

Underlying all this is that BSD was already fragmented at that time, Bill and Lynne Jolitz were at war with all the other factions (I met them later, they lived very close to us, and were involved in school politics, they were highly obnoxious people), with the closure of the Berkeley CSRG the code was already fragmenting.

But your observation is also true: The moment Linux became "viable" (capable of compiling and running large programs, with X support), it became hip and a fad, and everyone wanted to be on that bandwagon. And to be honest, for good reasons: Suddenly there was a good OS available that ran on cheap hardware. My 386-40 cost about $500; the 486 I got used for half that.

So why did it get so much momentum? I can only offer two anecdotes from early in my career. First one is from a large and well-known software firm that's still very large and in charge nowadays. I heard rumors of a Freebsd port of our flagship product that was never made available externally. The reason why was our infamous CEO had decided that Linux was "the future" and had issued an edict from on high that it would be supported in a year.

Exactly, that's what many people were doing. Roughly at that time I was working at Hewlett-Packard (in their research lab). On my desk was a very nice HP-UX workstation; my lab at work had dozens of HP-UX servers. My laptop usually ran Windows, because that was the best way to get compatible interfaces. At home I had a docking station with two PCI slots for it, and I dual-booted it into Linux (and connected it to a rack full of disk drives and RAID arrays, for doing development at home). The same lab was porting Linux to Itanium (now known as Intel Itanic); the day I interviewed there to get my job, I ran into Linus in the lobby (I knew him from earlier interactions). Everyone knew that Linux was the future.

But it was not that BSD was unknown at that point (after all, this was the late 90s, when the Internet already existed, and information flowed freely). I knew one of the NetBSD people, who ran a whole ISP using a Pentium on it. My desk neighbor at work was a contributor to OpenBSD. One of our summer interns was a Berkeley grad student who went drinking with all the *BSD people (he had been in Kirk's and Eric's famous hot tub). So we knew about BSD ... but for serious work, Linux was the only OS available on x86 machines; BSD was seen as purely a vehicle for research and hobbyists, with no corporate backing, and therefore no future.

These anecdotes reinforced the feeling that our industry is completely incapably of resisting fads.

It's more than a fad. When companies such as HP and IBM commit to "making Linux great", and put their huge resources behind it, it becomes the future. Note that this does not guarantee a success (remember for example the Apple-IBM joint venture that went nowhere). But BSD had none of that. On the contrary, it had gyrations of buy-outs, Rob Kolstad, merging with Walnut Creek CD (!!!), and feuding factions.

But you are not wrong, Linux was also a fad. Just like the Web was a fad (I was lucky enough to resist that), Java was a fad (I participated heavily in that one), OO programming was a fad (I actually see it as the best way to think about structuring programs), and today AI is a fad.

There's a mad scramble whenever something new and shiny appears. Everybody rushes onto the bandwagon regardless of the technical merits of whatever the fad du jour is. I'm hoping people move on from AI soon so I can look at it in peace.

Exactly my point. But some of those fads end up leaving a big change in the way we do computing.

 

[bakul]

I picked up 386BSD floppies from a friend in 1992 and even speeded up its ip checksum by a factor of 5 in July 1992. I had X windows running by August (so did many others). A lot of 386bsd related discussions were on comp.unix.bsd and you didn't have to be in the "in group" (I don't know or care if there was one!) -- I had actually met Jolitz after some a 386bsd talk in 1992 but don't recall much now. I was already running SUN 3/50 & had done an earlier BSD port in '80s so BSD was a familiar territory. I had looked at Linux 0.11 but the moment I saw it was following system V I lost interest!

The AT&T suit did a lot of damage to the BSD camp and people couldn't tell rumors from reality.

[Edit: what AT&T *should've* done instead is to say Unix is obsolete and should have released Plan9! They had the heft, the star power, and the budget to make it stick. We would all have had a much better architected and far simpler system! The road not taken]

 

[Beastie7]

Once FreeBSD gets performant WiFi and word gets out a lot of this will change IMO.

 

[jbo@]

Once FreeBSD gets performant WiFi and word gets out a lot of this will change IMO.

My laptop is doing >250 Mbps over WiFi no problems.

 

[Jose]

...You could get 386BSD from BSDi, for the small sum of about $1000 (twice the cost of the hardware). It didn't help that USL (a part of AT&T) was suing BSDI, which made the survival of BSDI questionable. On the technical side, BSDI had de-facto no windowing system available, with the X server being a year or two late to be released, and then only supporting an exotic video card (the Tseng ET-5000) that was nearly unobtainable, even in the heart of Silicon Valley..

BSDI is still one of my favourite Unixes. I ran it on a $4k Dell P90 machine that was still less than half what the cheapest Sun box we could buy would've been. I remember going to ask for money to buy the hardware, full of trepidation. My boss at the time said "oh, is that all?" I guess I didn't have a good feeling for what the VAX hardware he was buying at the time cost.

I remember calling BSDI support because I didn't know how to configure IP addresses properly, and not only did I not get the idiot quiz ("is it plugged in? Is it turned on?"), I got a thoroughly knowledgeable person who immediately diagnosed the boneheaded mistake I had made in the network mask.

The machine was to be an HTTP proxy, and needed no GUI. I sneered at such things back then anyway.

...Everyone knew that Linux was the future...It's more than a fad. When companies such as HP and IBM commit to "making Linux great", and put their huge resources behind it, it becomes the future. Note that this does not guarantee a success (remember for example the Apple-IBM joint venture that went nowhere). But BSD had none of that. On the contrary, it had gyrations of buy-outs, Rob Kolstad, merging with Walnut Creek CD (!!!), and feuding factions.

But why? The BSD license made a whole lot more sense for commercial investment, and BSD would have run just as nicely on that low-powered hardware. Looks like getting the X server to work wasn't that big of a lift either.

But you are not wrong, Linux was also a fad. Just like the Web was a fad (I was lucky enough to resist that)...

One of the ironies of the dot com crash is that the hypesters were actually right. The Internet (and the Web) did change everything. Perhaps we read too much into that and that's why idiotic things like "the sharing economy" get legs nowadays. Personally I think it's the greatest advancement since Gutenberg's printing press. You can't expect those to come along every few years. There's good money to be made suckering people into believing, though.

...Java was a fad (I participated heavily in that one), OO programming was a fad (I actually see it as the best way to think about structuring programs), and today AI is a fad.

Yeah, I'm also on-board with those two. I remember excitedly poring over Java source printouts (!) at a fern bar in San Francisco with a good friend who was also my boss at the time.

 

[covacat]

bsdi's bsd/os used to ship with XIG X server which rocked. there was trial version on freebsd 2.x cd (limited to 2 x clients which sucked).

 

[POSIX.1]

My laptop is doing >250 Mbps over WiFi no problems.

This sentence was non meant to be helpful, isn't it?
What about submitting some specs?

 

[jbo@]

This sentence was non meant to be helpful, isn't it?

It was.

What about submitting some specs?

AX201 with stable/14 (currently "equivalent" to 14.3 "pre-release").
Works out of the box.

 

[ralphbsz]

But why? The BSD license made a whole lot more sense for commercial investment, ...

That's actually a complex question, and a two-edged sword.

For actual end users, who have no plans to modify the OS itself, the difference in licenses initially makes little difference: In GPL you might have to open-source changes you made, in BSD you don't have to; but if you don't make any changes to the OS, and only use it, then that difference is irrelevant.

But another argument is psychologically in favor of the GPL: In the late 90s, it was thought that free software (I'm using that word deliberately here!) would be a great solution to lowering the cost of the OS down to zero. After all, for an end user, you don't have to buy an expensive license+support from a vendor, instead you get college students to program it all for you. The GPL was actually embraced by big corporations for just that reason. That sounds like a great argument, until you notice that those college students have no idea about good design, good quality control, or long-term support. And even 20 years later, for example in 2015, I was still hearing senior executives using that argument. For example, I was tangentially involved in a large and very complex software projects (very specialized stuff) that needed about 100 people to maintain and support (a typical support contract ran $30K per year). At some point some pointy-haired boss decided that the solution would be to open-source it, and let users and competitors do development of new features, bug fixing, and support by reading the source code. The same pointy-haired boss also thought that the solution to configuration and management of this thing being very complex (the manual was several volumes thick when printed) was to write a management GUI.

I started using *BSD (first OpenBSD, then FreeBSD) again in about 2005 or 2007, when it became clear that Linux is going down the hill fast, for a personal use server / appliance operating system. But note that what I do for my few machines at home has no predictive power about what the industry as a whole does.

So going back to our earlier discussion: Open source was also a fad. But strangely, in spite of the fact that people didn't understand how it really works, or what its undesirable side effects are, it ended up succeeding; just like the web and the internet and OO programming:

One of the ironies of the dot com crash is that the hypesters were actually right. The Internet (and the Web) did change everything.

 

[Beastie7]

There is no argument for a viral, commie license.

 

[rootbert]

But another argument is psychologically in favor of the GPL: In the late 90s, it was thought that free software (I'm using that word deliberately here!) would be a great solution to lowering the cost of the OS down to zero. After all, for an end user, you don't have to buy an expensive license+support from a vendor, instead you get college students to program it all for you. The GPL was actually embraced by big corporations for just that reason. That sounds like a great argument, until you notice that those college students have no idea about good design, good quality control, or long-term support.

I would argue that big corporations with time pressured code monkeys produce equally bad code, some projects (in sectors you would not believe if I told you) code have brought me to that conclusion. I would even argue that student's code may be even better because they do it for passion and have to revise their ideas and refactor their code. And of course not all commercial code is garbage, and not all GPL code is garbage.

At some point some pointy-haired boss decided that the solution would be to open-source it, and let users and competitors do development of new features, bug fixing, and support by reading the source code. The same pointy-haired boss also thought that the solution to configuration and management of this thing being very complex (the manual was several volumes thick when printed) was to write a management GUI.

which I find a good idea, it could have created a positive impact on the ecosystem of that software - no idea what software exactly you are talking about but still: it could have worked out.

 

[Jose]

After all, for an end user, you don't have to buy an expensive license+support from a vendor, instead you get college students to program it all for you. The GPL was actually embraced by big corporations for just that reason. That sounds like a great argument, until you notice that those college students have no idea about good design, good quality control, or long-term support.

Yeah, I believed it at the time too. Then I heard horror stories about the "research grade" code in BIND 8... Nowadays we call that sort of thing a "Jupyter notebook" and it's all the rage again. The more things change...

 

[POSIX.1]

The same pointy-haired boss also thought that the solution to configuration and management of this thing being very complex (the manual was several volumes thick when printed) was to write a management GUI.

"Management by GUI"

 

[POSIX.1]

The report of my death was an exaggeration

 

[Charlie Brown]

he claims that FreeBSD is doing well (meaning: not dying), because the number of Google searches for it has a slight uptrend

That was me ! ... in 2021, 2022 and in 2023 I switched to chatgpt!