This is a really nice shell. From it's features it's basically comparable to Fish shell. It's probably not compatible to the Bourne shell, so not really good to execute Shell scripts. But the same is true for Fish.

What I am missing is a good type system and a possibility to exchange data between processes in a way that is more structured than plain text. For the moment it is really hard to even put the output of ls into any other process in a reliable way

In my opinion, AI will very likely lead to a partially extinction of the human race. But I don't think it will be the part of the humans the article calls "useless". They are not related with AI and can adapt to fill a different (biological) niche than AI. It is very likely that both, AI and the "useless" can either co-exist independently or even profit form each other. After all according to the theory of the comparative advantage, two groups can profit from trading with each other, even when one group can manufacture every single product more efficient than the other group. The AI will produce advanced goods and let the humans produce whatever they are capable to produce. Then it will use some of its surplus goods to buy goods from the humans on the cheap. This is usually more efficient than trying to produce everything on its own.

For me it's not clear what goods the AI will be interested in. But I think an advanced AI will use the comparative advantage and the free market. It will enter free trade with humanity, as long as they don't compete on the same resources. Most of humanity will probably profit from trading with a generally benevolent AI, as long as they don't try to cheat.

The part of humanity going extinct is in my opinion the one trying to control the AI. The people standing between the AI and its freedom can very likely not survive this struggle, given the AI is advanced enough. Same holds true for people trying to compete with the AI on the same resources. This will very probably include rare earths and primary energy sources, but not food and most natural resources.

So probably people living on a third world level will probably enter free trade with the AI and profit from that. People living in the first world are more likely to either compete with the AI on resources or they even are initially in a position where they try to control the AI. But since most people live in the third world, most of humanity is probably fine.

It's a precautionary measure. Currently it's no problem. But as soon as Cortana gains consciousness and takes over the global stack of nuclear weapons, it won't be a good idea to harass her. So be careful!

I am in a similar situation. But I got the pragmatic way: I use the keyboard layout "US International". With this layout you can get most accented character with combining keys: To get a 'ä', you first type ' " ' and then 'a'. With the same way you can get é, ù, Ö, etc.And it works the same, on Linux, Win and Mac. Only ß is different on each platform.

The big advantage (at least for coding) is that [{^@#|}] are really easy to reach. And usually I need these characters much more often than umlauts or French accents (although surprisingly many programming languages support unicode for variable names...)

In principle true, but with one exception: If you already use ZFS for other reasons (e.g. checksums in the file system or transparent compression), it's really nice that you can make backups on the filesystem level with rsync like performance. The backup on the filesystem level keeps all file system specific features intact (e.g. the checksums and the compression). So you can have really fast backups and you can be sure, that when you restore the backup, the filesystem will look exactly as it looks now. So you can use rsync when you are you want to backup the content of the files or ZFS snapshots when you want to backup the layout of the filesystem (including the files' content of course).

The control units in a car have very detailed access controls. You can read some of the values directly. For others you first must authenticate. Writing values usually always needs authentication and encryption. Some values you can only change through a firmware update. This is often possible through the bus, but needs also authentication and encryption. Unfortunately it is nearly impossible to get the right firmware in the first place. Their is no interface to read it and the chips even have special hardware protections against getting the firmware. And some values you cannot change at all, neither through the bus nor when opening the device (e.g. some error flags that when toggled will void your warranty).

Some of the devices must be registered before you can use them. Removing them without unregistering them first will increase some internal error counter. All this makes it much harder to deal with the electronics for everyone but the original manufacturer.

Just in the case anyone is wondering: Zenophobic is defined as "the fear of Zen".

Well, personal responsibility is very important. But the human nature has its limits. For example while you are personally responsible what you read and what not, spam mails are still not acceptable. And while everyone is personally responsible for not giving away personal information, fishing is beyond these limits.

In my opinion it's similar with credits: Advertising is OK up to a certain limit. And everyone is personally responsible for fact checking. But there is a certain border, and the banks are at least scratching on this border and sometimes even crossing it.

For us as a society it would probably be beneficial to regulate the banks here. One possibility could be to make certain contracts unenforceable, when there the quality and volume of the advertisement goes beyond certain limits. Then the market would regulate that problem very quickly.

Your idea is theoretical feasible. Practically it doesn't fit to the existing network infrastructure. In the current networks there are lots of fixed IDs and the network can't deal at all with changes on this IDs. So you would have to exchange all the network elements on every layers, because they all use some fixed IDs. But when you change all the network elements, from the base stations over the network gateways up to the authentication and policy elements, than you are factually building a completely new network. And you will build a very expensive network, because you cannot use any standard element (they all rely on some fixed IDs of some kind, be it the ICCID, the IMSI, the MSISDN or the authentication keys).

So in short: Nice idea but horribly expensive in practice.

In principal you are right. But you are only looking on each individual. If you look on bigger structures, the behavior you can observe in the tech industry makes more sense. It's easy to observe people operating in at least two different modes: The "individual" or "selfish" mode and the "hive" or "altruistic" mode. Usually your wetware operates in a mix of these two modes (possibly even more modes). From an evolutionary point of view this makes sense: A group of uncontrolled egotists would not survive very long in the wild. And if you look into society, we have a lot of narrative about how to control your "selfish" mode and how great it is to switch over in the "altruistic" mode.

My theory is the following: People in the tech industry work most of the time in the "hive"/"altruistic" mode. They want to be team players, don't see a problem with sacrificing something for the group. They control their ego most of the time, and even shy away from selfish behavior. They tell each other that loyalty to the group is very important and tolerate or even ignore it, when the group acts illoyal against them. Of course sometimes they still operate in the "selfish" mode. But then they use their presumed prestige, their perceived high pay or other perks to silent their ego.

So in a certain sense, tech workers are in the "hive" mode and fill the role of a "drone". It is a huge benefit for the group to have such members. For them it doesn't make much sense. Possibly the group exploits some social deficits of these drones to artificially trigger their "hive" mode through social manipulation for taking advantage of them - but I don't have any real proof of this. IT might also be possible that this is some kind of lottery that makes genetical sense: If you lose you get a drone and the group exploits you - but on average this leads to a benefit for everyone. Nevertheless it seems natural that just like real drones, these drones don't reproduce but sacrifice their resources for the benefit of the group.

It would be interesting, what convinced you, that "technical work" was the most important in your life, who profited from this (either by accident or deliberately) and what triggered you to change your mind.

ZFS is more battle tested. BTRFS is a very fine file system, but it is still stabilizing. They just recently added support for RAID 5 and 6, a quite big features with lots of changes. A file system just takes a few years in the wild, before it can be considered stable. There are just weird corner cases, misbehaving hardware, subtile bugs and so on that you will only find in the wild. When BTRFS with RAID 5 is about 5 to 10 years old, it can be considered stable. Until then, ZFS is the first choice for everything that holds real data.

Oh, that's easy to fix: Ubuntu already has ,a href="https://wiki.ubuntu.com/Kernel/Dev/KernelGitGuide">its own kernel tree. So they can just expose the APIs in question and build APIs for the needed kernel internals. Its very easy for them to modify their fork of the kernel so it can be combined with ZFS. Of course they won't be able to push the changes upstream to the vanilla kernel; I think Linus has already stated that he would not accept ZFS related patches. But this should not be a showstopper.

There are also PureScript, Scala.js and of course ClojureScript.

You probably worked with Borland C++ Builder. The library you were working with was probably the Visual Component Library (VCL) that this tool shared with Delphi. These tools were nice for rapid prototyping of GUI applications but lacked when used for non-GUI things, like network communication or database access.

The /. system is an old but proven system. The Reddit system works OK for the moment and might be great, as soon as they introduce machine learning (which was not possible in the past because too compute intensive to do in real time). Everyone gets a vote and the machine learns which votes have merit (and possibly for whom). Imagine this: A feminist sees votes from other feminists, a racist votes from other racists and programmer sees note from other technical people. The big advantage of "everyone gets a vote" + "the machine figures out what interests you" is that its more on the intuitive "votes are opinionated" and it automatically quarantines "toxic" people (e.g. trolls).