Worst Case Scenario: a bit of Fun with Computing

[studiot]

This was inspired by a recent incident in thread posting and takes the form of a competition to think of the worst case scenario.

The scene a computer controlled situation where a user is offered an input box to input instruction, but the computer is programmed to substitute the requirements of the previous user.

 

To start of here is a scenario.

 

A guy goes into a robot surgery clinic for an ingrowing toenail.

She wakes up to discover the previous user had a sex change.

 

[Sensei]

Quote

Worst Case Scenario: a bit of Fun with Computing

In different countries coma and dot can delimiter thousands and fraction part.

It depends on operating system language and country settings and custom application preferences (if it allows override global settings).

Machine can be shared by couple persons from different regions of the world.

f.e. 1,000.0 means 1 thousand for one person in one country, but for somebody else 1,000 means 1.

Imagine situation, stock market investor, or stock broker (freshly hired and not used to different delimiters), will use "1,0 mln" he will try to sell 1 millions of shares in one country, but in 2nd country it will mean 1000 mln = 1 bln of shares. Instant bankruptcy and/or collapse of worldwide stock markets (at least few minutes collapse, like in Flash Crash accident)

 

Some programmers hardcode delimiters they are used to, and some others don't even know that different regions of the world use different delimiters.

 

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

(in the middle of Wikipedia article there is world map by delimiters)

Edited November 22, 2017 by Sensei

[studiot]

2 minutes ago, Sensei said:

Some programmers hardcode delimiters they are used to, and some others don't even know that different regions of the world use different delimiters.

A salutory lesson indeed. +1

[Sensei]

There is another thing that is country dependent.

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

"Another word for one thousand million is milliard, but this is used much less often in English than billion. Some European languages such as Finnish, Swedish, Danish, Portuguese, Spanish, Italian, Dutch, Hungarian, Norwegian, Polish, Russian, French, and German, use milliard (or a related word) for the short scale billion, and billion (or a related word) is used for the long scale billion. Thus for these languages billion is thousand times larger than the modern English billion. However, in Russian, milliard (миллиард) is used for the short scale billion, and trillion (триллион) is used for the long scale billion."

 

Long and short scales

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

 

[DrP]

16 minutes ago, Sensei said:

Some programmers hardcode delimiters they are used to, and some others don't even know that different regions of the world use different delimiters.

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

 

 

I thought the standard decimal mark was the full stop / period. Looks like it doesn't have an international standard (dot, line, comma or comma on line).  Clearly it needs to be standardised. I would argue for the dot as a decimal and a comma for the thousand break, but I guess I am biased because that is what I have always used. I think it the most common, so we should adopt it. The US will just do whatever it wants anyway regardless of the rest of the world having a meeting and deciding on one standard way. 

We employed a French lad once for a year or so and it confused me once when he wrote 1000 as 1.000 and 5.783 as 5,783....  only once though. I was adamant that he used the correct notation...  he argued that it was correct in France...  Fine, but we are in the UK so use our notation or **** off back to France.

 

Edited November 22, 2017 by DrP
sp.

[Strange]

There is a similar problem with dates. Is 10/11/12 the 10th November 2012 (UK) or the 11th October 2012 (USA) or 12th November, 2010 (Japan)

Actually, I don't think people would normally use two digit years in Japan - apart from anything else, the current year is both 2017 and Heisei 29.

[Sensei]

Decimal mark is serious problem while automatically gathering data from websites from various regions of the world.

 

[DrP]

Just now, Strange said:

There is a similar problem with dates. Is 10/11/12 the 10th November 2012 (UK) or the 11th October 2012 (USA) or 12th November, 2010 (Japan)

Actually, I don't think people would normally use two digit years in Japan - apart from anything else, the current year is both 2017 and Heisei 29.

Different cultures, different practice...  although it makes more logical sense to me to have Day/Month/Year. Seems odd to put month before the day then jump to the year...  it seems all mixed up...  I think you done it deliberately to annoy us after achieving independence. lol.  

 

 

[Sensei]

16 minutes ago, Strange said:

There is a similar problem with dates. Is 10/11/12 the 10th November 2012 (UK) or the 11th October 2012 (USA) or 12th November, 2010 (Japan)

Actually, I don't think people would normally use two digit years in Japan - apart from anything else, the current year is both 2017 and Heisei 29.

+1 brilliant example.

In my personal projects I am using reverse format f.e. 2017-11-22 (because when they're in the list in separate rows, columns look nice and clean).

Such format can be also sorted by ordinary Excel/OpenOffice SpreadSheet, using regular string comparison for ASCII, without having to parse it.

Edited November 22, 2017 by Sensei

[Strange]

6 minutes ago, DrP said:

although it makes more logical sense to me to have Day/Month/Year.

Obviously not! Year-month-day is by far the most logical (which is why it is part of an ISO standard) as it allows sorting by date very simply. I always name files in this format, for example.

But you are right that the US system is utterly bonkers.

Edited November 22, 2017 by Strange

[DrP]

1 minute ago, Strange said:

Obviously not! Year-month-day is by far the most logical (which is why it is part of an ISO standard) as it allows sorting by date very simply. I always name files in this format, for example.

For filing maybe I'll concede that...  but when talking about the date today I still hold that it is more logical to state the day first..  Today is Tuesday the 22nd of November -  I have no need to add the year for that.   If some one askes 'what day is it' I say ' it's the 22nd of November' not '2017, November, the 22nd day'. 

[studiot]

When I first worked in Arab countries I had a shock on prices in shop windows.

There is no symbol for zero in Arabic.

So they use the decimal point.

So 50 is written 5.

I thought I was getting a bargain when I tried to buy something.

[Strange]

13 minutes ago, studiot said:

There is no symbol for zero in Arabic.

There is. It's a dot.  

They use a comma for the decimal mark, I think.

[studiot]

Unfortunately I can't show exactly how they represented it in 1976.

My last post was the nearest I can get.

[DrKrettin]

36 minutes ago, Strange said:

There is. It's a dot.  

They use a comma for the decimal mark, I think.

I have a photo of a manuscript from the 2nd or 3rd century which is the earliest known example of a zero, being a dot. On the bottom line near the righthand corner.

[Strange]

56 minutes ago, studiot said:

There is no symbol for zero in Arabic.

So they use the decimal point.

Not quite as potentially confusing as the Babylonians who had no symbol for zero and so left a blank space. 

[studiot]

Yes I seem to remember they put it in the middle of the character height.

The confusing thing is they also have a bent loop for another number (khamsa or 5)

 

http://www.softschools.com/languages/arabic/numbers_1_10_in_arabic/