MIT grads struggle with lighting a light bulb

Thread Starter

WBahn

Joined Mar 31, 2012
29,976
A number of years ago, I was given a link to a website that was quite enlightening (no pun intended). I will have to look for that site, but the teaser video is right here:

http://www.youtube.com/watch?v=8ve23i5K334

Now, a little background is in order. Apparently the year before this video was made, some Harvard researchers trying to look into what is being taught versus what is being learned asked a bunch of Harvard economics and related grads at commencement some really, really basic questions that had a tiny, inconsequential twist to them. What they found was that a large fraction of people couldn't deal with the twist. So the next year they devised the experiment in the above video and went to an MIT commencement.

They have gone one to do other experiments involving kids of various ages and reveal some extremely unexpected, yet very common, ways in which what we teach is completely misunderstood by so many.

Their point, of course, is that we are teaching things wrong. While I agree, they seemed to be stuck on pedogocial issues that ignored what I think would solve a lot of these things -- people should get an engineering degree unless they have spent some serious time designing and building and testing and demonstrating lots of different electrical projects across a broad spectrum of the EE world. The projects don't have to big or elaborate, they just have to be real.

When I first saw the video, my first thought was, "Okay, so here is an engineering grad (and most were EE grads, but not all -- which isn't mentioned in this video snippet) that has never got their nose out of a book long enough to open a cheap, old-fashioned flashlight to see how it works. They would have seen right before them that the bulb is almost always touching the positive terminal of the battery directly and that a wire connects the side of the bulb to the negative terminal. That's the kind of thing that most people that WANT to become EE's discovered sometime around 5th grade. I don't recall exactly how old I was, but I distinctly remember taking the batteries and bulb/reflector assembly out of a flashlight case and making it light by using a single wire to complete the connection.

Makes you wonder how many of those MIT grads (and replace the school name with any school name you want) went into EE because they had a passion for electrical engineering, and how many of them went into EE because someone convinced them that companies throw lots of money at anyone with an EE degree, particularly from someplace like MIT.
 

#12

Joined Nov 30, 2010
18,224
That makes me feel better about the fact that I put my nephew through a 2 year tech school and he didn't believe you can oil an electric motor. At least I didn't pay MIT prices!
 

ErnieM

Joined Apr 24, 2011
8,377
I'm not very suprised. While BSEE degree is dusty but still in force and I can barely remember any practical hands on labs. I think we had maybe 2 lab courses 2 or 3 credits a piece.

We were taught a ton of math and physics. We never took even a flashlight apart to see how stuff works.

I do remember one long night on Brooklyn where myself and my partner learned an op amp circuit's gain will improve greatly once the power is actually applied, but the real lesson learned is a circuit may still have an output when it has no power.
 

Thread Starter

WBahn

Joined Mar 31, 2012
29,976
Anyone think MIT, or any other school, would even be embarrassed by this?
Yes, I think they would. In fact, I am very good friends with an MIT alum and he said they were pretty embarrased. They (engineering schools in general) really are working with the intent of producing well-educated engineers and they believe that their place is to instill knowledge and that skills will more-or-less naturally flurish as a consequence. There is an attitude that "we are educating engineers, not training technicians." It's a misplaced, myopic viewpoint that, I believe, stems largely from the prevelance of so many educators having so little real-world engineering experience, let alone real-world hands-on experience with circuits.

As was noted by someone else, the opportunity to get hands-on education in most engineering programs is very limited, and growing more so. This is for a variety of reasons. The cost of running a lab course versus the cost of installing simulator software on the school's network pushes heavily in favor of using simulation instead of hands-on labs. The finite time that you have for courses, coupled with the engineer vs technician mindset favors minimizing lab courses in favor of additional lecture courses. I was fortunate in that the place I got my undergrad had strong traditions in practical engineering and many of the professors in many of the departments came from industry, so we got a pretty good offering of both required and elective lab courses. But even there it is changing along the national trend (it was already going down that road before I got there and has picked up the pace since).
 
Last edited:
This can't be the first time the faculty and administration have seen such results. Maybe the first time on youtube. You can not be around these students that long without someone noticing that there is a certain lack of understanding. Maybe they think it is a small enough percentage...
 

Thread Starter

WBahn

Joined Mar 31, 2012
29,976
Oh, they see it from time to time. But you usually see it a student here and a student there and the strong temptation is to write it off as a handful of really poor students. The 'good' students are good enough at learning how to parrot the techniques presented in class enough to do well on the exam problems that, in almost all cases, are matched to the techniques taught in class. Instructors all too easily get caught up in focusing on the mechanics of the topics they are supposed to be teaching to even try to also explore bigger questions of whether the students can actually think. But even when you do, it primarily serves to dishearten you because you quickly realize that the system is too mired in its own ideology and pedagogy to respond in a meaningful way. So you try to do what you can in the courses that you teach, but you run the real risk of getting poor teaching evaluations from the students for asking them questions about things not covered in class. Those teaching evals have a very real effect on your performance review and whether you will keep your job. So you have a decision to make.


Oh, and this video is something like ten to fifteen years old.
 

strantor

Joined Oct 3, 2010
6,782
This video and following discussion makes me depressed. Just started college and now I feel like the next 4 years are literally going to be devoted only to obtaining a piece of paper.
 

Thread Starter

WBahn

Joined Mar 31, 2012
29,976
I would be depressed in general but not in particular. If YOU want to get a good education, then I am still quite convinced that a very good education can be had at nearly any college. You will have to take on more of the responsibility at some than others and you will have to take on more of the responsibility today than in the past. But if you go there to learn and not just check off a bunch of boxes, then you will do fine.

The upside is that you will be able to look at a large fraction of your classmates and be very comforted by the fact that they represent the bulk of your competition.
 
Last edited:

MrChips

Joined Oct 2, 2009
30,702
Also you should be in a better position to engage with the knowledgeable technical staff and be able to tackle more challenging projects, all to your benefit.
 
Last edited:

Veracohr

Joined Jan 3, 2011
772
This video and following discussion makes me depressed. Just started college and now I feel like the next 4 years are literally going to be devoted only to obtaining a piece of paper.
What you get out of your education is up to you. If your school focuses on theory and doesn't give you much practical, hands-on learning, do it yourself.

I'm pretty sure my school isn't going to teach me vaccum tube technology, but if I want to try making a tube guitar amp one day, that part of it is up to me to research.

Of course, you could always switch to a school with a more hands-on technical program.
 
I would certainly not let this video get you down. As others have said, you get what you put into it. It does prove that even if you are "good" enough to get into a really good school like MIT, you can still get a really mediocre/poor education if you so choose.

Many of us went to college that stressed a lot of theory over "hands on" education. There are a lot of opportunities to make up for the missing components. And, unless some specific research opportunity lands you a real job afterwards, you will have a lot of learning to do once you get out into the the workforce.

Learn how to approach complex problems...
 

BillB3857

Joined Feb 28, 2009
2,570
I was moving up the corporate ladder of a large aircraft manufacturer, even though I didn't have a degree in anything. As a 2nd level supervisor, I was given the task of interviewing potential maintenance electricians. My boss had been after me for years to get my tail back in school and get a BS in anything. Shortly after assigning the interviewing task, he got on my case again about getting back to school. I told him that he had almost convinced me earlier and I had even contacted some local schools about schedules, etc., but he had assigned this new task and it had changed my mind. He couldn't understand how that added duty could possibly impact me going back to school, since it would be weekend college. I simply told him that he had me interviewing college graduates and if that is what college did to you, I didn't want any part of it. After a few choice words, he told me that he fully understood what I meant, but if you refuse to relinquish common sense during the "educational process", advancement would be much easier. He was right! After enrollment at a local university, I was able to advance three more rungs up the ladder, which would not have happened had I not been working toward my degree. Ironically, the degree is not in the electrical/electronic field. I double majored in Management and Psychology (They are one in the same, aren't they?) AND, i graduated a year after I retired!
 

#12

Joined Nov 30, 2010
18,224
You already have the practical experience, which puts you head and shoulders above your classmates. When the chemistry teacher says something stupid like, "Electrons don't have any mass." you will be as smart as I was and know not to correct the sadistic bastard so he can try to humiliate you in front of the whole class.

When you walk into your first math class and find out the person next to you is graduating at the end of that class, you can secretly smile abut how feeble your competition will be.

Above all, you get as much out of it as you put into it.
My calculus teacher never found out I was writing and running transformer winding programs on the mainframe while the other kids were partying.
 
Last edited:

strantor

Joined Oct 3, 2010
6,782
I have no doubt that I will get more out of it than the majority of my classmates, but I hope it shows in some area that employers look for. I am in it for personal fulfillment, but I'm in it for the money as well. I'm not going to MIT or anything - I'm probably going to be at the back of the line for "who has the most impressive diploma" but hopefully the years of technician experience will put me out ahead of the jobseeking hordes. wishful thinking? I wish not.
 

Thread Starter

WBahn

Joined Mar 31, 2012
29,976
I simply told him that he had me interviewing college graduates and if that is what college did to you, I didn't want any part of it.
I can very easily understand how you might come to that conclusion. I've interviewed a number of people over the years and if my total impression of the value of college came from that group as a whole, I would have a pretty jaded impression as well.

Now, to be fair, we were bottom feeders in the grand scheme of hiring. We couldn't afford to identify the top talent and make them offers in the fall for a position starting in the spring. We didn't hire someone until our business was such that we had high confidence that we could support another engineer for at least a few years, which meant that by the time we reached that point we needed someone that could start immediately, which means people that did not make the cut to already get offers from the larger recruiters.

But as I look back on my college education, I can't think of very many courses that I would consider to have been largely a waste of time. I also learned a huge amount of both theoretical and practical stuff. Now, a lot of the learning was from stuff I did on the side, like designing boxes to ring telephones or dim lights for the student theatre group or designing and building the buzzer circuits for the physics bowl competition. But in most of those side projects I was actively applying what I was learning in class and, as I got into more advanced classes, I was always discovering ways that I could have done those things better. I certainly had a wide spectrum of instructors, but only a tiny number (perhaps three) that I can honestly say were all but worthless. On the other end, I had at least a dozen, perhaps even two dozen, that pushed me and challenged me. Almost all of them tolerated my constant questions and took time outside of class to delve into them even when they went well beyond the content of the course they were responsible for teaching.

In fact, I got the impression from most of them that they enjoyed doing our conversations in their office after class, which is precisely how I felt whenever a student would come to me the same way. So my advice from that is don't be afraid to engage your instructors outside of class about anything technical; certainly some won't give you the time of day, but I suspect it won't take long to find a long list of ones that will and whom can pass on a lot of useful knowledge.
 

t_n_k

Joined Mar 6, 2009
5,455
Wbahn's comments highlight the value of quality mentoring. It would hopefully be unsurprising if recent graduates are eased into their initial employment with a close eye on their development and productivity. One wouldn't wisely appoint a recent electronics graduate to head up a design team with critical commercial consequences - unless the skill set and relevant experience mix was exceptional. Again, good employers [like good educators] provide quality mentoring & supervision where independence & maturity are developed progressively. It would be interesting to ask electronics employers what their expectations would be for a new recruit - how soon would they expect to get some return on their investment in that staff member. Three months? A year?

I imagine this would vary from one country to another. Engineering in some countries will not rate as highly as other professions.
 

Thread Starter

WBahn

Joined Mar 31, 2012
29,976
It would be interesting to ask electronics employers what their expectations would be for a new recruit - how soon would they expect to get some return on their investment in that staff member. Three months? A year?

I imagine this would vary from one country to another. Engineering in some countries will not rate as highly as other professions.
It will also vary a lot by employer, both in terms of the type of industry but also in the size of the company.

Companies with developed product lines or large companies tend to take more of at least a quasi-formal training program where you are expected to learn the companies products and/or design flows and you might not be expected to produce anything productive for months are even a couple of years, though that is probably pushing it for all but a tiny handful.

For the company I worked for, we needed productive output almost immediately. But that did not mean that we expected new hires to sit down and be productive based only on what they walked in the door with. The company does full-custom mixed-signal ASIC design. When I was hired, I had zero IC design experience of any kind. I had never laid out a transistor or used any kind of a layout editor. But we had a customer that needed a metal-only mask set of a chip we were designing so that they could get some dummy wafers fabbed and refine their post-processing steps. So I got about a one-hour intro to our layout editor and what we were trying to do and then I was on my own. As I had questions, I walked next door and asked. But by the end of that week I was submitting a layout for fabrication that I had done 100% of the layout work on. My next step was to work on the LVS of a chip that had just been submitted. We were under the gun because we had about a week before we would not be able to pull the submission in case an error was found. In those days, with our tools, the LVS output was quite a puzzle. So I was shown a few of the ways that errors could be tracked down and then left on my own. I asked a lot fewer questions on this because the basic information was pretty straight forward and I just had to wrap my mind around how to organize it. I actually found a couple of errors in time for the database to be updated and resubmitted (by someone else). The real point is that we could not afford to have people spend a bunch of time on make-work for training purposes. They learned by doing real things on real projects from day one.

There's always (at least for what we were doing) a way to carve out a piece of a project and narrow the difficulty and scope so that the person can learn what they need to while still producing a useful result within the timeline of the project. But that does not mean that this approach doesn't have its downsides. Instead of coming up with a more-or-less standard curriculum for new hires to work through, you have to carve tidbits out on a case-by-case basis. Also, as you went, your skillset tends to grow in a very ad hoc and erratic fashion. I worked there for a couple of years before I did my first place and route because it simply hadn't come up on any of the chips I had worked on to that point. So I went to someone that had only been with us for about six months and had them show me how to do place-and-route.

Aside: For those that might be scratching your heads wondering how it might be that someone could work for an ASIC design house and not do place-and-route for years, it is useful to understand that we did really, truly, full-custom designs -- as in lunatic fringe. Because of the often bizarre (and really cool) things our customers were trying to do, we usually had to layout every polygon on the chip (except for the pads, but sometimes even those) by hand. Most of our digital gates where tightly coupled to the pixel array and so had to be laid out by hand for that chip. For the little bit of digital logic in the periphery it was usually easier to lay that out by handm too (but using standard cells that we laid out ourselves to address noise issues).
 
Top