It's "consumer-grade UNIX" at best. The userland is around 12 years out of date. Not sure why you got downvoted, that's trivial to verify. Just look at the dates on each tool. It varies, but most of it has remained untouched, and compared with a current FreeBSD or Linux system, it's missing all of the bugfixes, new command-line options and features developed since then. Its compatibility and interoperability with other UNIX systems has been compromised as a result. It's all too common to have to special-case MacOS because it's missing trivial options. Like "readlink -f".

It's a shame because you're absolutely correct that in the early days of MacOS X, its UNIX base was a big selling point. But that clearly hasn't been a priority for quite a number of years.

https://en.wikipedia.org/wiki/...

Get people hooked, and then extract money with microtransactions.

What's old is new again. They are calling it "DLC" nowadays.

Computers are our slaves. They are told to perform tasks, and they obey their instructions to the letter, uncomplainingly and without recompense. They are also not human or even alive, and the terminology used here has no bearing upon real-life slavery. They are dumb machines which do the jobs they are told do.

For IDE you might have a point (I'm not familiar enough to comment). For other buses, e.g. I2C or SPI there is a very clear difference between the master, which issues commands, and the slave which acts upon the command and provides a response. We could get all worked up about the "bad" terminology in use here, but I really struggle to see it as a problem of any particular significance. There are far more important (and genuine) problems in the world more worthy of our attention. These are non-problems created by overprivileged people with far too much time on their hands, very little brain, and no real purpose in life other than to cause unnecessary trouble for others. Will the world be a better place if we waste tens of thousands of man-hours to erase a bit of engineering history? Not one little bit. Will any real-world slavery be stopped, or any residual trauma from people with slavery in their families past be eased? Not at all. The only people benefitting here are the SJWs who revel in wielding their capricious and arbitrary power over others in the name of virtue. They are bullies, and not nice people at all.

This is an incorrect understanding of what "master" means. A "master copy" does not mean immutable. It means it's the primary and authoritative source of record, from which all other copies and versions are derived. It does not, and never has, meant an immutable and final form. It might be used to mean that in the audio world, but that's not what the rest of the world uses the term to mean. The way it is used in git is perfectly standard, and has been used in this context in business terminology the world over for centuries.

Err, yes it does. "Blacklist" came from banking where customers who were debtors who defaulted on their loans were recorded on a list with a black border. A literal "black list". It has nothing to with skin colour. It's still used in the banking world today, albeit no longer in physical form. It is a means for accepting or rejecting future business with a particular person, and so it is quite natural to adopt the same naming conventions for dealing with network packets. After all, the "blacklist" terminology long escaped the banking world into the common vernacular and is use for all sorts of diverse purposes today.

It's stuff like this that makes it impossible to take virtue-signalling SJWs seriously. They disrupt our lives for things which make zero improvement to the world we inhabit. If "blacklists" were historical tools for racial oppression of people with black skin, they might have a point. But there is no basis in historical fact for making these changes. They achieve nothing and are utterly pointless.

Not just in the US, I see it in UK acadaemia as well. There are millions of pounds for funding shiny new equipment. But funding the ongoing running costs, consumables, maintenance, technical support staff and the like? Not so much. So you end up with huge capital expenditures going to waste or being underutilised, not because there is any fault with the kit, but because there's no dedicated staff to maintain and support it, train people to use it, and pay for the ongoing running costs. I've worked in some places that do a very good job, but in other places there's prestige to be gained by getting the funding to buy the biggest and shiniest toy with all the extra knobs on. But the people who get the funding and do the purchasing are rarely the people who actually have to use the kit, the actual requirements often come second. In many cases, making do with something a quarter of the price would do the job just as well, and be much more affordable when bits need replacing. Hell, buy two up front and still have half the cash left over for consumables and staffing. Unfortunately, funding and budgeting is insane and makes no sense, and many of the people involved aren't even sufficiently competent to run their own household budgets.

Arecibo is a great shame, but I think it's fair to say it's had a good long run, and would have had to be decommissioned at some point. After the damage that's been inflicted and the generally poor state it's in now, it may well be much cheaper to start over than repair it. Assuming there's still a demand for it, hopefully there will be the possibility for a new one to be built again in the future. But you are likely to be completely correct that had it been funded properly, routine maintenance should have prevented the problems from happening in the first place.

Absolutely. Linux is slowly but surely losing the main selling points which led to its initial adoption and success. Being that it was an open Unix-like system which was adaptable to meet any purpose. A key part of that was being simple and understandable by anyone who cared to look at it. With that gone, it's just as opaque and unpredictable as Windows, but with the added disadvantage of having poorer hardware support. It's now on a par with Windows in terms of reliability, so I got to the point where I switched back to Windows after 20 years and run Linux in a VM as needed.

The main problem here is that when Linux was first developed it simply had to copy pre-existing designs. Be that POSIX, BSD, Solaris, X11, even Windows. The design work was basically done, and the primary task was reimplementation. Implementing stuff is fun and absolutely of merit, but it's a lot simpler than design work. The new stuff, most of freedesktop's work, lacks the good design of the components it replaces, and in many cases doesn't take into account the state of the art and ends up being regressive. D-BUS, for example. Intended for desktop use, now crept into systemd at the heart of the system. Not only is it not specified very well, there were and are alternative IPC libraries and tools which are hands down better, but weren't considered. The quality of the design and implementation is what's letting us all down here. And due to the political battles which have been fought and won, we basically have to take it or leave it. I chose to leave, and now use FreeBSD in preference to Linux. Not perfect, and certainly not totally immune to all the brain damage, but still better than what Linux has chosen to become for the time being.

I found another annoyance this week. If I use remote desktop to connect to a Win10 system with a fullHD display, and I'm using a nice 4K display, it scales everything up, and it looks really nice. But if I disconnect and then go to the other system to use it directly, all the applications retain the scaling, and everything is massive with no way to revert it back to the original scaling for the fullHD display. Not the end of the world, but it's just another weird scaling defect.

No... you haven't actually read the question. The GPL is a *distribution licence* and the point you linked to is about *distributing* a combined work of GPL and non-GPL code. It is not an end use licence and does not cover what you can do on your own system. Stop the ranting for a moment, and think about it...

I see. So rather than having any desire to actually understand the problem domain and the tradeoffs different systems make, you make this uninformed response... The specific terminology the different systems use is frankly immaterial. However, the data integrity guarantees and performance implications are not. If you actually cared to objectively assess how snapshots work on different systems, you would bear that in mind.

Please, go back and read what I said without making assumptions. I'm the author of a number of GPL projects, and a long-standing contributor to many others. I'm not advocating for violating copyrights or the GPL.

What I am asking of you is that you read the GPL licence text and tell me where the licence permits software authors to place usage restrictions upon their code above and beyond what is written in the GPL. Because, as I said, it's a distribution licence only, and has no remit to place restrictions upon end use of the code. Assuming that the GPL has more powers than it really does is just as bad as assuming it has less.

So, can you actually address the question that was asked, please. Because I'm still interested in an actual answer if you want to have an adult conversation about it. Does the kernel GPL-only symbol restriction have a legitimate legal basis, or is it just hot air?

You've got the context wrong. There's no suggestion here of relicensing CDDL sources under the GPL. The CDDL licence is file-based, meaning it doesn't cover the project as a whole, only the specific files under the CDDL. Have you read the CDDL? It's all spelled out pretty clearly.

You didn't address the actual question about "GPL-only symbols". All the stuff about "GPL enforcement" is irrelevant; the court cases all involved the distribution of modified GPL sources from linux to busybox, but that's not what's being discussed here.

The GPL is not an end-user licence, it's a distribution licence. It can't constrain the activities of third-parties distributing and using other projects' software. And it can't constrain what happens on an installed system. It is limited solely to the right to distribute the sources of the GPL project under copyright. You can't add extra clauses to the licence. And, to the best of my knowledge, that isn't the case here. So, where exactly does the "GPL-only" symbol stuff have actual legal weight? Where and how is that actually enforceable?

Can you point me to the exact part of the GPL2 licence which permits different licences to have different rules upon linking?

I'm strongly suspicious that it's all hot air, and that it's completely unenforceable.

In ZFS terminology, that's not a "snapshot", it's a "clone". You create a writable clone from a an immutable snapshot. Snapshots, by their very name, are supposed to be a snapshot of the filesystem state at a certain point in time. They are not intended to be modifiable.

While the Btrfs workflow might seem superficially more convenient, being a one-step process rather than a two-step process, it has two major drawbacks. Firstly, it conflates two separate actions, and doesn't provide the same immutability guarantees. Secondly, the Btrfs design has some very unfortunate performance consequences. ZFS performance is not affected by the presence of snapshots. Or clones. This cannot be said of Btrfs.