An important yet neglected aspect of surge pricing model is that it fails to correctly account for decrease in demand.
As such, one can probably argue that fixed prices capture the real market better in the expectation sense.
Have you programmed in Smalltalk? Classes tend to do a lot more (well everything that can be done) compared to C++. Smalltalk has a clean design for the core parts so these hang together pretty well. Nonetheless, inheritance hierarchies are deep and there are many fat classes. So small size in itself is usually not a virtue, clarity in role is on the other hand.
sorry to say this, but this is exactly why MG is dangerous for the society. Because people start perpetuating myths and fictions passed to them in entertaining tone as facts. They start narrating them to kids and gradually turn them into common wisdom. To whoever reads this shit as entertainment. Please don't. Go watch a movie, listen to standup, watch porn, whatever else makes you feel less board. Pop science is perhaps even more important than scientific journals and shouldn't be crapped on the way MG does. I just can't fathom why people are such entertainment junkies anyways.
you and others are ignoring the fact that the guy has been rejected more than a dozen times at technical interviews. He has demonstrated an inability to pass what's the bare minimum: basic programming, algorithms and data structure. If someone with a PhD cannot re-master this material (which he/she has supposedly learned in undergrad) within few weeks of preparation, no matter how fancy the PhD research is, then that in itself is telling about his/her commitment, drive or competence. These claims sound like saying oh I am a best seller novelist yet having forgotten (and unwilling or unable to revisit) the alphabet. At the end of the day, there is no evidence that PhDs have a hard time getting interviews (the author got 20 or so), and if they really ace the interviews, heck even do reasonably well, they will have no problems getting hired. A PhD degree by itself will not get them hired, but it will not jeopardize their chance either, provided that they do well in the interview.
The bar for passing technical interviews is pretty low. All you need to do is to get reasonably competent in one or two mainstream programming languages (I emphasize reasonably, you definitely don't need expert knowledge), say Java or C++. This shouldn't take more than a couple of weeks of focused effort if you already have some degree in CS and know programming. At the same time you have to refresh your memory about data structures and algorithms. Dust off your favorite algorithm book, and go through sorting, searching, binary trees, hashing, elementary graph algorithms and dynamic programming. If you have taken a relevant course before and paid attention to it (you were told this was going to be your most important CS course ever, no?), it shouldn't take more than a day or two to remember this stuff (again assuming that your undergrad CS degree was not a waste of time). Then you need to solve a few exercises to practice, mostly to become confident in yourself. Topcoder, dozens of online sample questions would do. You won't need full coverage of every possibility, but to assure yourself that your brain still ticks and backed by your knowledge you can approach the problem presented to you at the interview. In total you need a week and a half or two weeks to get to top shape to pass interview at the likes of facebook, google, amazon et al.
Another Relevant anecdote: I personally prototyped a complex data transformation in Clojure in about a week. I then had to re-implement it in C++ which took about three more weeks, but was actually able to perform near to the speed of the super fast raid HDD (~1gBps). The clojure code (even after a lot of type hints etc) never got past 60MB/sec throughput, which would be a showstopper for a tool that has to process a lot of data online.
because you probably don't know what to look at after all. Look at the predicated type declarations for example. It's a language with concepts built in.
Facebook works for me very well because I don't quite use the news feed as a social networking feed. I have changed its push model to a pull model. That means, I have pretty much hidden all of my friends and use the news feed to only read updates from magazines and pages I am interested in. I however visit people's profile that I am reminded of every now and then to see whats going on with their lives. The closer I am to someone the more frequent her/his profile gets a visit. To make this tractable, I every few months go through a round of deleting people I don't care about (like someone I've met for a few minutes and I am sure I will not see again, or if I do I will mostly likely pretend I don't know the person).
knowing these are not going to guarantee good performance. But not knowing them is a clear indication of incompetence. A programmer who doesn't know basic algorithms and data structure is a bad programmer.
" you're describing girls who want to get into a field full of cool people."
No I am not. And you are grossly oversimplifying the process typically involved in vocational decision making. Seldom people go to a field based on pure passion, when most other things about it is at odds with their interest.
"I agree that female teenage culture has a lot to answer for in dissuading girls from studying technical subjects"
such as what? sounds like blaming the victim. Female teenage culture may happen to be just fine, and the techie culture is what alienates women from participation.
well, maybe the big part of what kills women's interest in programming/CS is the unnecessary nerdiness that comes attached with it. As if watching starwars or Dr. who or reading marvel comics is essentially related to competency in CS. Note that this is largely a US phenomena. I am not making a judgement about the value of said material, just that they need not be bundled and presented as part and parcel of computing.
"Someone's physicality (Let's move on from just thinking about sports only) can/does have relevance to someone's aptitude for CS, or, their ability to do well."
it doesn't matter what you saw there. the point of TFA is there are many equally (if not more) hardworking candidates that don't find their ways to these schools because of existing prejudices and non-transparent admission policies.
no you shouldnt. look at any medical science, exercise physiology, psychology etc. papers. All you see is data gathering and basic statistic analysis (anova, t-test etc.). Basic stats is what they teach at the undergraduate level to scientists and engineers.
He is totally right and I have seen many many examples of it in grad school. The fact is, US's quantitative GRE is a joke even for mid-high school students in other countries. I take it that he says to become a good scientist in many fields, you don't need Tensors, stochastic PDEs, calculus of variations, group and field theory, functional analysis, path integrals etc. You can get by only mastering single variable calculus, statistics and some basic linear algebra.
full of crap. It sounds like when an office clerk in a publisher house demands abandoning latin words, puns, ironies figures of speech etc. because he is too dumb to use them properly. There are poets and there are mba-style bullshit writers. Great C++ code can read as beautifully as poems in style and elegance (c.f. Elements of Programming by Stepanov).
it's a shortcut to learn basic network programming with sockets, without spending too much time on it. Anyone more than casually serious about it should consider looking at Richard Stevens books, Unix network programming and TCP/IP illustrated (esp. volume 2).
what makes this stuff seem "high-performance" TCP to you? This is standard socket programming. High performance TCP server is when you actually roll out your own TCP stack.
Felt like expanding a bit more:
Note that this is a result of evolution in the language: What's referred to as modern style of C++ that has emerged from the works of people involved in the STL/boost tradition and for the large part informed the new standard. It's a noticeably different style from what you would see in a codebase written in mid-90s. It's arguably less object-oriented than most Java systems, and adopted certain concepts from functional programming, while exploring several innovations hitherto unseen in other languages.
What also enables this is advances in modern architecture. 64bit processors allow for large address spaces and large stack frame sizes, which obviate the need to allocate every single bit on the heap. You can often see sizable C++ programs measuring around 100KLOC or larger which use new/delete in a handful of places. Automatic storage is, well, automatically managed. For when free store semantics is needed, smart-pointers abstract the complexity of several common cases where automatic-management is not adequate.
Using value semantics, streams, lambdas, regular expressions, and a good library for command line flags processing (e.g., boost) C++ can even be a perfectly apt choice of language for scripting. The development effort need not be much longer than python but with no optimization the result can be easily 10 times as fast.
none of the posts I read here mentioned the true advantage of C++ over other languages: It's the only language that offers enough flexibility to write 1) generic, 2) correct and 3) efficient code. With C you get only 3. With Haskell you get 1 and 2. With ML you get some of 1 (yes I contend that C++, in the hands of a competent programmer, can be as good as ML or even better for writing generic code) some of 2 and some of 3. C++ allows you to have cheap abstractions (using them does not require paying significant penalty, unlike Scheme or Smalltalk). Java and similar managed languages excel in neither of these criteria in particular.