Doing any sort of large-scale computational fluid dynamics or finite element simulations may require a great many cores. For example, you might want to conduct a very detailed simulation of the air flow around a vehicle, airplane, structure, etc. to have a basic understanding of its aerodynamics before spending time and money testing an actual prototype in a wind tunnel. You might also want to look at how very complicated, soft-body structures deform due to a variety of external stimuli. Such information would be crucial for certain materials science applications. Chemical reaction and acoustic simulations may also require a great deal of computing power, especially if you want to have a high spatio-temporal resolution.

Essentially, there are plenty of physical and theoretical science applications that can benefit from massive processing capabilities. There is a lot of fundamental science that is also performed in simulation before any actual tests occur.

Your generalization would be true if I was just one of a handful of students who worked up to general engineering principles from rudimentary physics knowledge. However, I can point to hundreds of my peers at MIT who did the same thing, many of whom likely have a much deeper understanding than I do. Given my exposure to the curriculum at CalTech and Stanford, I feel rather confident in stating that engineering students at those schools weren't just given equations and told to memorize them. Instead, they slogged through a series of derivations of those principles and had to build up their own understanding of the meaning behind those derivations. I'm sure that others can chime in about their experiences at other top-tier institutions, such as Berkeley, CMU, and the Ivies.

As an aside, undergraduate research assistantships are becoming more commonplace at some institutions. I agree that most undergraduates will probably not come out publishing papers in prestigious journals or conferences. However, that does not mean that they don't enhance their knowledge and understanding of various concepts.

In short, there are thousands, if not tens of thousands, of engineers out there with educational experiences that either partly or fully mirror my own. Consequently, you really need to be cautious when you make sweeping generalizations like engineers only spend their time memorizing formulas without reflecting on how those formulas came to be.

There are plenty of incorrect assertions and generalizations made in this post. It honestly reads like a dogmatic diatribe.

As EE/CS undergraduate students, my classmates and I learned the fundamental physics behind various phenomena, not just the high-level equations. That is, we learned why, for example, transistors function they way that they do and why we can rely on simplified equations to characterize their behavior. Most of what we were taught is still covered in the MIT undergraduate curriculum (see courses 6.002, 6.012, 8.012, 8.04, and 8.044).

As EE/CS graduate students, my lab partners and I were responsible for furthering the state of the art. During these years, we had to understand why, for example, our experimental results diverged from our model predictions and how to revise those models accordingly. In some cases, we invalided long-standing, widely taught models and proposed new ones. If we didn't understand the fundamental physics behind these models, we wouldn't have made the contributions that we did. We also wouldn't have had our work published in Science, Nature, and PNAS.

I don't even need to lengthily address your comment that engineers aren't smart. There are plenty of people on Slashdot that can thoroughly invalidate that claim.

The child poster who first replied to your comment was not me, the parent poster.

To expand upon my original comment: I am not interested in paying for and using software that is tied to platforms and services that I do not want or need. In my opinion, Origin is a good example of such a platform for two reasons. The first is that it is a glorified game-launcher application. If I've purchased a physical copy of the game, I should not need to install and use Origin to simply run the game, especially if its integration with Origin is minimal. Secondly, Origin is a digital storefront dedicated solely to a small catalog of EA products.

In contrast, I find platforms like Steam to be useful. For the most part, I'm able to launch and play games outside of the Steam service. As well, Steam offers a broad selection of products from a number of publishers. The fact that they offer massive sales throughout the year is also appealing, though tangential to the discussion.

Since EA has refused to release any of its newer games on Steam or other distribution platforms, there four options: (i) don't play the game, (ii) pirate the game and use a crack to get around the Origin requirement, (iii) pay for the game and crack it to get around the Origin requirement, or (iv) pay for the game and install/use Origin. Option (iv) is unappealing, as I do not wish to use Origin. Option (iii) is the one with the best intent; however, it is an unlawful choice due to circumventing the application protections. Moreover, in giving money to EA, I am reinforcing their use of Origin. Option (ii) is also unlawful. In this case, there are three possible side effects: (i) EA starts more tightly integrating their games with Origin, making cracking much more difficult or impossible, (ii) EA stops targeting computer gamers, or (iii) EA opens up their catalog to compensate for lost sales. This last side effect, while appealing, is unlikely.

Since EA started bundling their games with Origin, I have consistently chosen option (i) and will continue to do so in the future. If I had an overwhelming desire to play the game, which is not likely to happen, I would either go with option (ii) or an altered version of option (iii). I don't disagree with your assertion that this is an entitled viewpoint. However, it is not one on which I have acted.

You're correct: I'm currently not a salaried employee. I also hope to never be salaried again, let alone work for a company that bars me from overtime simply because I'm considered a "computer professional" in the eyes of the government.

As an aside, I can definitely empathize with those who are salaried employees. I had to deal with being labeled a salaried employee all throughout graduate school, despite my contract saying otherwise, and basically miss out on $100k to $125k/year (USD) in overtime; pretty much everyone else in the EECS department was in a similar situation. Suffice to say, when I had the chance to join a start-up company as a fairly compensated employee, I jumped at the opportunity.

When I was a graduate student, a 50- to 60-hour work week was basically a vacation, given that I routinely put in 70 to 85 hours per week. Moreover, it wasn't unheard of for students to basically not leave the lab for an entire week, let alone only sleep 5 hours per day on a couch during that time, while some important experiment was being conducted.

Nowadays, a 60-hour work week is the norm for me, and I've come to enjoy it. I have around three "productive" days where I work a total of 39 hours, two "semi-productive" days where I work a total of 18 hours, and an additional 3 hours that I spread out over the week for administrative tasks and meetings. While it would be nice to cut back to just 40 hours per week, I nearly double my salary by working those additional 20 hours.

The base Model S is $69,990 (USD) according to Tesla's website and Wikipedia, not $35,000 to $40,000 (USD). With federal tax credits, the base price comes down to $63,570 (USD). With state incentives, it becomes a bit more difficult to qualify the final price: essentially, you can either get a $2,500 (USD) rebate (California), a $6,000 (USD) income tax credit (Colorado), a $5,000 (USD) income tax credit (Georgia), or a $4,000 (USD) rebate for the car and a $3,000 (USD) rebate to offset the cost of electric vehicle charging stations (Illinois).

Considering what I paid for the Model S P85+, I do wish that the base price had been as low as what you originally claimed.

As an academic, part of the problem with starting wonderful open journals and conferences is the fact that there are very few incentives for us to spend our time to build up the reputation of the publication. Although being editor-in-chief or associate editor of a journal is nice to have for a tenure review, some universities weight it less than the number of publications produced, the prestige of the publication venue, how many students you have advised, how much grant money has been brought to the university, and how much publicity your work has received. Since so many of my colleagues are focused on maximizing these metrics, they have very little time for much else when starting their careers. Moreover, even when they have tenure, they still have to chase grant money to sponsor all of the students in their labs; when I was in graduate school, my adviser seemed to be flying around every two or three weeks to meet with program managers to get even more money.

Another item of note is that it is much easier to get support to start a conference if you align yourself with one of the major academic publishers, e.g., IEEE or Springer. Provided you can meet your attendance quota, these publishers provide much of the infrastructure and initial funding to host such events.

If you're using Word or OpenOffice, that might be a problem. If you're using LaTeX, it's not, provided that you're a reasonably quick typist and have memorized the standard mathematical commands. I ended up typing all of my lecture notes for my statistics Ph.D. classes without much of a hassle. In fact, most of the students in my classes came to me for portions of my lecture notes, as I was able to capture all of the important comments that the professors would make in haste while continuing on with a derivation or proof.

As for a comment on the article, since very little information was given about their testing protocols there may be some inherent bias in their findings. Specifically, their testing methodology seems to hinge on showing that short-term conceptual recall rates decrease when using laptops. That is, the authors don't bother addressing long-term retention and generalization.

As an actual researcher, let me state that your post has little to no bearing on reality. That is, open-access journals do not prevent an individual or group of individuals from artificially inflating various publication metrics. Moreover, agencies look at much more than those metrics, e.g., research output, research impact, past publication venues, and the number of students who are supported and are expected to graduate under a grant, when deciding how to dole out funding.

All that's missing is some mention of hosts files.

I just wanted to notify you, informally, that you've infringed upon my patent that details a process for complaining about patents. I'll make sure that my lawyers send you the appropriate notice paperwork by the end of next week.

Come tomorrow, I guess I should stop by the Department Chair's office and let him know that he should revoke my endowed scholar position, let alone the positions of my colleagues, as we're all apparently fools.

I chuckled heartily over this, especially considering that Autodesk hasn't released a native 3DS Max binary for OS X.

Private entities rarely, if at all, focus a majority of their efforts into pure research, unlike the national labs. Funding pure research, which is one of the few actions that the US Government at least does halfway correctly, is ultimately essential if we are to progress the state of the art and thus create new fields and products that are ripe for commercialization.