TU SIGUIENTE PASO

Ignite Velocity Amsterdam 2015

O'Reilly · 9 lecciones · Gratis en YouTube

LECCIÓN 3 / 9

John Willis - Ignite Velocity Amsterdam 2015

←Siguiente

Sobre este curso

we are recovering from a network connectivity issue this morning hopefully regular flight operations will be resolved shortly four hundred

thousand passengers stranded when all flights went down due to a software bug in our hyper-modern in our hyper-connected modern world i dare to ask the question which will kill you

sooner cyber terrorism of software bugs let's take a little look into this and kind of see if we can answer that question 1980 no rad faulty circuit problem not captured by the software

system checks enabled a false report to state that there was missiles coming in 0000 was apparently replaced with to a phone call to the White House at 225 am 1983 Soviet nuclear early warning system

malfunctioned and reported that we that the Soviets were under attack from America thank heavens the software engineer at the time used his gut feeling and reported it as a false alarm

1985 theory act 25 radiation therapy machine delivered a hundred times in the intended dose of radiation due to a miscalculation in software nobody died on that one but there was not a very

good outcome 1994 in Scotland chinook helicopter crashed killing all 29 passengers on board initially the pilot was blamed for the error however in further investigations they found that

it was a faulty software bug that actually caused the downing 1998 NASA's 655 million dollar Mars climate orbiter disintegrates as its trajectory as his trajectory went off site apparently

somebody didn't read their contract and the calculation was done in pound seconds instead of Newton's there is nobody in outer space to push a reset button nASA has recently said but they

are looking for any volunteers if anybody would like to take up that role in 2000 radiation therapy planning software deliver different notions which resulted in massive overdoses there was

five people reported dead however off record they expect that there was approximately 21 people that were actually killed 2003 a race condition bug causes blackout across eight states

in the US and Canada 50 million people were affected at 256 power plants went offline or due to a software bug 2003 computer software blunder at st. Mary's mercy medical center in Michigan cost

the lives of 8,500 patients these patients families were informed and they were told that they were no longer alive they actually were alive it was just that there was a software bug that

reported all these people dead and as I said informed Social Security and also their families software bugs what can i say 2005 Michigan Department of Corrections experienced the computer

programming glitch which led to the early release of 23 prisoners apparently they weren't too concerned because these at least weren't murderers as they stated in the news report so start at

the moment we've looked at real deaths by software bugs 78 Falstaff's 8500 none of these so far has been caused by cyber terrorism so in our hyper modern connecting world as we all know software

will eat the world mez marc andreessen said in the wall street journal as software becomes more and more important and runs all our modern day systems I beg to ask the question should we be

concerned about cyber terrorism or should we be concerned about the quality of our code because in a hyper motor connected world with large distributed systems the quality of code becomes not

just important but it actually becomes a necessity to us and we cannot be left in these scenarios because there are human lives at stake here and the more actively become the more human lives are

at stake and the more abstract did we become in our code and our education of code the more we are going to be putting ourselves in these positions software bug should not take down a system we

currently are looking at our infrastructure and we're building our infrastructure to handle faults to handle fault tolerance to handle failures if a container goes offline or

a node goes offline it shouldn't really affect anything so I beg to say that we need to architect our software for failure we need to accommodate for failures we also need to teach fault

tolerant code design patterns we cannot be expecting things to not to always run and how do we do this well education is the key sometimes we have to admit that we don't know everything we have to ask

for help we have to learn from each other and we have to share our information and share our bugs and share our problems thank you

SIEMPRE HAY ALGO MÁS

Sigue alimentando tu curiosidad.

Explorar más cursos
De la comunidad, para la comunidad

¿Encontraste un gran curso?

Compártelo. Nuestro equipo lo revisará antes de añadirlo al catálogo.