This item is an advertisement. I hope slashdot got paid good for it... 8-P

It's the City People. They live so close together that any who can detect odor normally, stop making babies. So that gene pattern dies out... ;-P

A lot of people are still believing what "everyone knows". Like, it is too expensive to fix any bugs because the software will be obsolete in a year.

But, that stuff is from the 1960's when the "new computer" system was, it's self, thrown out in a year. Except maybe for mainframes that were bigger than a car.

It is now the 21st century, and software is often expected to last 30 years, like an aircraft carrier is. And, if your software fails, everyone in the world will know it was you.

The cost of fixing bugs goes up by about a factor of 10 as each stage of development is completed. So how is it cheaper to leave the bugs for the end???
The next time someone says it is cheaper to not handle bugs, laugh in their face! 8-}

Actually, most of the time it's due to calculating the cost incorrectly. ;-}

(Or just going by what "everyone knows".)

When your calculations of a solution to a set of equations gives "imaginary" numbers, as in "the square root of a negative number", this means that your equations are missing something.

Usually it is missing a "dimension" or a "degree of freedom". Probably something in the specifications of the units, as used in the equations.

For instance, using calculations meant for DC electricity on an AC circuit can give imaginary expressions. This is because the full equations need to have a time (or "phase") expression, which is missing in DC equations. But it is simpler to use, on regular stuff, and ok as long as it is reliable (with a fudge factor).

Not then. You tested it before sending it to the publisher.
But the publisher had no one that knew code.
Then you were supposed to get a copy of the end result, which of course was too late.

Regular books had a proof-reading step, but magazines often didn't.

A lot of stuff on the telephone BBS systems was about these typos.

Being wrong is not the same as lying. And at the start of the pandemic, everyone was wrong. Firing everyone would not be smart. 8-}

Does this mean that those sites that were effected still have the old invalid URLs imbedded after four years?

I wonder what other invalid URLs they have and how much those bad fetches slow down the user experiance?

One problem is that the people trying to talk to dolphins have always refused to consult with science fiction writers... 8-}

And for the ones who don't know:

    The space travellers all carry around a book titled "How to serve man". How could they -not- be friendly?

8-}

No, the file is not bit-perfect.
The chance of a fault is not zero. Even if the chances of a fault are low, it could still happen the very first time you try it (that's how chance works). And the larger the file gets the worse your chances are.

Also, The calculations are not the same. Last I checked, the LRC in an HDLC packet (on which Tcp/IP is based) was only 16bits. Even if the new versions use 32 bit LRC it is still not much.

People don't talk about this because they don't know what to do about it. But that doesn't mean it's not a risk.

Often they -don't- find IT people. That's why the service is so bad.

It's not always a good thing, to have the first of a new technology. 8-}

Maybe, but Open Source does have it's own problems...

Also, Commercial Software does not always mean totally closed source. It may be quite possible to get at least part of the source, if you sign an agreement not to sell copies of it to anyone else. In some development systems it is even part of the install.