Because there are also OCP network equipments, like a switch design from Facebook that lets you do software defined networks easily.

As someone who does VoIP for a living: Software VoIP clients work as well as the underlying network allows -- same as with hardware phones like the Cisco Callmanager. Normally the hardware phones are in a special VLAN, often using layer-2-priorizing (class 3 and class 5) to the next switch, and they sending their packes with DiffServ information (e.g. the signalling with AF31, and the payload packets with EF), thus they get threated in priorized queues through the network. Desktop computers on the other hand are mostly in VLANs that either ignore the DiffServ information, or even actively strip them off, and often the software VoIP client isn't installed in a way to even add DiffServ information.

You can totally fuck off the VoIP phones by misconfiguring the switches and routers in your network, no problem, and then their voices sounds as shitty as the software client. And you can install the software client correctly, and threat the VoIP packets accordingly also in the Desktop LAN, and suddenly the voice quality will be as good as with the hardware phones.

The Germans call them MdB (Mitglied des Bundestages).

Look at astrophysics again. And really look at it.

While we can't observe a new Big Bang (the last one happened 13.7 billion years ago), we can observe the results from the Big Bang. We can for instance observe the redshift of far away stars and galaxies. We can also observe the cosmic background radiation. The Big Bang gives a very consistent description of what we can tell about the cosmos around us, and no alternative explanation comes even close.

But you would probably also claim that the explosion of the Krakatoa volcano in 1883 is "just untestable speculation passed of as fact", because no one survived who saw the actual magma coming out of the mountain. And the reports of a big smoke cloud, tsunamis and earth quakes could have been caused by something else. And we don't even know if they really happened, because all we have are just written reports, like the Bible, right? And going to Krakatoa and finding large layers of volcanic ash which buried the remainings of people and houses and animals and which look as if they were around 130 years old are so very indirect that the actual volcanic eruption still can be called "speculation" in your world, right?

The same goes for Climate science. We have the daily weather report, but the theories that allow us to predict the weather (yes, the single event, and not just the long term average), are "closer to religion than to science"? We have complete daily weather data (yes, the actual measurements done by real humans with real instruments) for some regions of the world starting in the early 18th century, and for most of the world starting in mid-19th century, and drawing them in a diagram and describing the resulting long term average as going upward is "closer to religion than to science"? Please elaborate!

Maybe what you picture for yourself as science has some serious flaws, but that's the problem with the picture of science you have. Not a problem of the sciences.

I wonder where the flaws of Evolution are.

And please, just because someone used the theory of Evolution to explain something, and then the explanation proved to be wrong doesn't mean that there was a flaw in the theory of Evolution. It's also not a flaw in the theory of Whole Numbers if you find a calculation error in your balance sheet. Yes, biologists are people, and they make mistakes, and sometimes someone comes up with some "because Evolution!"-argument, which later proves to be inconsistent with newer findings.

So again: Which are the flaws in the theory of Evolution, that are inherently a part of the theory itself?

... which normally results from the context, because the mass is singular, while the measurements are plural.

I didn't know that resisting arrest carries the death penalty without trial. But surely there is a law about this.

It's not only the IBM mainframe world. Those are the official replacements in Germany. IBM just used it, because it was already standardized.

Actually, only the military does, and only on the teleprinter. The official transcription is still 'ss'.

I don't like Hefeweizen at all. Never have, never will.

Exaggerating and calling people names make sure that your arguments appear well-balanced and thoughtful, and you in no way will be confused with a zealous nutter, right?

But it is quite complicated to kill someone by encryption. Maybe reading him 2048-bit-RSA-keys will cause him to die of boredom?

That was the original question: Which jobs will be replaced by robots (e.g. not given to you)? The whole point of the article is what career to chose if you don't want to be replaced by robots. And the grand parent offered some ideas, which I doubted. While those jobs may be not directly replaced by robots, we just need less and less of them. For your career, it is mainly irrelevant if you get replaced by a robot, or if your job just becomes obsolete. You will get fired.

I have several issues with your analysis.

1. Maintenance workers

Yes, they are all humans, but while you don't replace them with robots, you just need less and less of them. In the 1950ies and even in the 1970ies for instance, a computer had to untergo regular maintenance. The tape drives and the programming card feeders had to be cleaned and readjusted, worn out bearings had to be replaced, all the others had to be lubricated, boards with defective elements were pulled, the elements soldered out and new elements built in, the boards were put back etc.pp.

Those maintenance jobs are almost gone. Today, you let your hard drive run until it fails, then you replace it with a new one. The data is on RAID anyway, and the new hard drive will be filled with data automatically. All the compute boards are now a single main board and some bars of memory, and replacing them is easy. And have you ever repaired a network switch? No, you get a new one from the spare parts storage and just replug everything. Thus a single person now can do maintenance for a whole data center during a shift, when in former times, you need dozens - and that was only for that single mainframe running the central database.

And in general, the main time between failures has gone up for almost every computer component. Most of them don't fail anyway until they get replaced because they become obsolete.

2. Design and engineering

Yes, the actual design of a new component is human work, but design as a career has a big disadvantage: design per se is no steady work. Once done, a single design is finished, and now it can be used over and over again. And there is no guarantee, that a new design is necessary after you finalized the last one. Or at least, there is no guarantee that you get paid for a new design because the old one is good enough. And many tasks in a design bureau are now computerized anyway. No technical draftspersons anymore for the finalization, whose task is now done perfectly by AutoCAD and the like. Drawing an RC-circuit is now a single point and click, and not a 10 min task to get everything rectangular and nicely fitted into the page. Need just the electrical installation of a construction plan? I'll send the approbriate layers to the printer instead of calling the assistant draftsperson. And the fan-in and fan-out of a circuit or a sub-component is now calculated on the fly and the right connectors with the right capacity to PWR and GND are automaticly put into my new chip design. My mother worked as a typograph, and I remember, when I was a child, she was sitting at her desk, cutting the galley proofs to length to arrange them on a page and glued them in place, intermixed with the drawings and the marginals and the footnotes and the headlines. Now this whole typesetting process is highly automated, text flows freely around other typographic objects, and we just point and click to change everything from one-column to two-columns.

3. Programming

For programming in general, see 2. Most of the tedious, but steady parts of programming are now done by prefabricated software components, by libraries, by integrated developing environments, by code generators. We have code profilers, we have test case generators, we have automated versioning. A single programmer can maintain larger and larger code bases. We have large databases of code samples, easily browseable. We have online communities for complicated questions.

4. Construction

Actually, construction needs less and less people. Many parts are prefabricated. Others are standardized, and easily mounted on site. You don't see people building window frames on a construction site. Windows are built in highly automated plants and then shipped on site, mounted with construction foam, and then everything is done. We don't mount individual planks, we have large wooden panels. We don't use the hammer to drive in a nail, we have pneumatic nailing machines. We don't do individual cabling anymore, we do structured cabling, where we just run some 100-pair-cables from site to site and then patch everything in place. In the 1970ies, network wires were individually pulled through the building and then soldered on sockets. Today, we just use the crimp tool, and we buy a big selection of 3ft, 6ft and 10ft Cat-6-cables. And we standardize everything on IP.

5. Art

How many paintings do you have on your wall? How many of them did you actually buy and not just inherited them or got them as a present? We can get about every slightly famous painting as high quality print on canvas. We could just upload our own photographs to some website, and two days later, we get them printed with some "real paint" effect, and framed to hang them on the wall -- no artist involved. If all musicians of the world stopped making music right now, we could listen for the rest of our lives to music without any repetion of a single piece just from the freely available music on the web. Most of us could easily do without any living artist. It's just our decision not to do so. And even if computers would be able to churn out real Work of Art (maybe they do already?), many of us will support human artists just for the sake of it... in principle, we do it right now also. But as ever, being an artist will not feed most of them who want to. Only a few of them will be able to make a living from their art. Becoming an artist will thus never be a regular career. But alas, it never was anyway.

Many of the jobs you describe may not be directly replaced by a robot, but they become more and more rationalized, or they just become obsolete.