Failure is Not an Option. It’s a Fact @ DevOps Meetup Zurich

the original title of this session was
failure is not an option it’s a fact
and i even tweeted like that when i
received the speaker’s card to promote
the session very fitting
but then they figure it out and fix it
so welcome to failure is not an option
it’s a fact i’m excel reese
in this session i will talk about
failure that is deeply ingrained in my
life
failure tends to cause a deep emotional
response
and that it’s okay
and some deep emotional feelings will
have strong effects
so strong that they affect our physical
body
but in recent years we have shifted our
view and find that failure can be a
cause
of celebration
so how do we face failure when there is
a dramatic impact on our life or how do
we deal with it in our professional
careers
or the benefits of celebrating it
john bowers is crystal clear about the
deathly effects of small errors he
believes that at some point it was too
difficult or too painful so we decided
to dismiss our natural ability to deal
with failure and replace it with a lower
acceptance level
and now
we are forced to sit back and just
accept this new norm
of good enough attitude
and the results that come with it
sometimes with huge costs
elizabeth gilbert failed at getting
published for almost six years so for
almost six years every single day she
had nothing but rejection letters
waiting for her
and it was devastating every single time
and every single time she had to ask
herself if she used
just quit and a sparrow self the pain
she found
that putting your head down and
performing with diligence devotion
respect and reverence whatever the task
is
was the key to success
astra teller while at x formerly called
kukulex the moonshot factory describes
how everyday they will get excited and
cheer
how are we going to kill our project
today because x believe that the only
way to get people to work on big risky
things audacious ideas is to make it
safe to fail
team kill their ideas as soon as there
is evidence to support it because they
are rewarded for it
failure is a little tricky even more in
i.t
at the project level
failing a project can have devastating
consequences for the entire organization
in this particular paper the authors
examine more than 1400 projects
comparing their budgets and estimated
performance benefits with the actual
costs and results
must
incurred in high
expenses failure and i.t projects have
sung call corporations
black zones are very common in i.t
one
of six projects that overrun their
budget went as high as 200 percent
meanwhile the average of
over budget was almost 70
and how
we react to failure when we think we
should have been successful
michael schrag as part of the mit
lincoln laboratory suffered many
setbacks
in a project where success should have
been an easy slam dunk
it deeply impact his whole career
let’s have a closer look to failure in
it projects
we have testing fail fast never trust a
test that you haven’t seen failed have a
healthy number of unit integration
contract consumer provider ui end-to-end
api acceptance and exploratory tests
and even create good reports
security how often do we analyze run
metrics or heuristics to verify the
provenance quality security
maintainability of our dependencies
open or closed source it doesn’t matter
or how do we manage the contract with
the file because an api is a contract
that we defined
from 23 000 web apis that had breaking
changes between versions
almost none use any deprecation
mechanism so
much failure
sometimes
but sometimes is okay i was confused
until i encountered this issue the april
2011 issue
of harvard business review is all about
failure it is more than 10 years old but
it is as relevant today as it was then
not all failures are created equal
mistakes or failures fall into three
broad categories preventable complexity
related intelligent
so preventable failures most failures in
this category can indeed be considered
but they usually involve deviations from
specifications
in the closely defined process of high
volume or routine operations
unavoidable
failures in complex system a large
number of failures are due to the
uncertainty of work a particular
combination of needs people and problems
that may have never occurred before
some failures can be averted by
following best practices
still small process failures are
inevitable
intelligent failures at the frontier
failures in this category can be vitally
considered good because they provide
valuable new knowledge that can help an
organization live ahead of the
competition and ensure its future growth
discovering new drugs creating a radical
new business designing a new innovative
product
and testing customer reactions in brand
new markets are tasks that require
intelligent failures
i hope this session has given you a new
perspective on failure because things
are not always what they seem
thank you very much i’m excited please
talk to me ask me questions thank you
very much