> "It turns out that expert systems are just another way to program"
Yes I agree with this, if you are providing the expert system with premises, although not specific to a particular problem, then you are effectively programming it.
I wonder if we could discover and obtain premises through some automated system, then the deductive process could still apply, leaving us with an understanding of how the solution to a goal was reached via the deductive argument. Do you know if this has been attempted?
I've had very good experiences with React over the last year but was curious to find out what the community thought of the criticism here and if anyone has experience with the alternative promoted https://github.com/Matt-Esch/virtual-dom
I think one of the difficulties is that in a short space of time it's hard to get everyone up to speed on a complex body of code. So I can understand why training courses are somewhat forced to revert to smaller examples that everyone can get in their heads quickly - otherwise you would spend a huge chunk of time understanding the code base.
For me it wasn't really about the code itself, it was more about learning the process of breaking problems down and then applying different refactoring techniques to increase the abstraction.
One of the best parts of the class was the discussions/code reviews and Sandi really does knows her stuff. My classmates and I had many deep discussions with her on a wide range of topics and she was always able to explain things in a clear and concise way.
Perhaps if you are a senior ruby developer of many years experience this course wouldn't be the best investment but I would definitely recommend it for junior/mid range developers.
Was curious to look this up - doesn't appear to have been much movement (see below from 2011)
"It is estimated that there are approximately 200 billion lines of COBOL code in production, with 5 billion lines of new code added to these highly valued core production systems each year. As much as 70% of business data worldwide is stored on mainframes and approximately 80% of daily business transactions are processed in COBOL. There are about 1.5 - 2 million developers globally developing and maintaining COBOL code today.
Yes I can't see how that would be sustainable in the current economic state. Perhaps the retirement age would have to increase in line with the years added to the average lifespan.
Obviously the economic/social impact would be enormous and would require many difficult decisions but difficult != impossible. These are challenges that would surely be worth overcoming?
It seems that as the amount of data being produced continues to expand at an unprecedented rate, it will become essential to master Hadoop which seems to be the gold standard for managing big data - would anyone disagree with this?
You are right. A 1000sqft place in London will cost you at least $3500 a month ($42k a year). In certain locations outside of SV programmers have lower salaries and higher cost of living.
I guess its a supply/demand thing. SV has high demand for engineers and not enough supply of employee talent, therefore salaries increase to compensate. (Although judging from this thread they haven't increased enough..)
Europe does not match the talent demands of SV (maybe because fewer big internet companies based there), therefore salaries don't need to increase as much to attract talent. The demand is a lower proportion of the supply...
Link for anyone not familiar with pyenv/virtual env usage: https://www.jackhoy.com/web-applications/2017/02/12/setting-...