Flutterby™! : Higher Education... 2006-04-03 06:36:06.481722+00

Higher Education...

2006-04-03 06:36:06.481722+00 by ebradway 17 comments

Recently, I've held my tongue on Flutterby in discussions about the value of education. As I have advanced in my education, I've begun to get a better handle on "what it's all about..." See comments for my essay...

I used to marvel over programmers with BS and MS degrees who couldn't code their way out of a paper bag. The general consensus around Flutterby was that higher education was a process of extorting funds from people and granting worthless credentials. While that may still be the case to a certain degree, I've come to understand the purpose behind higher education (at least in the sciences)...

An Associates Degree is a simply a nice name for technical training - usually sold by an institution that packages the training in the wrapper of a "college". Not much more to add there...

A Bachelor of Science Degree (mine is in Applied Mathematics) does not really imply any particular ability. The goal of the usually four year program is to provide the graduate with the vocabulary to intelligently communicate in their field of study. There is alot of emphasis, especially at "liberal arts" schools in providing the graduate with the vocabulary and writing skills to intelligently communicate in their field of study and a reasonable breadth of other fields. Note that the BS holder is NOT significantly more capable in the field. This is why Pascal and Java are used to teach Comp Sci students about programming concepts when they aren't necessarily even used in the industry. The same goes for fields like mathematics. I learned about alot of math that I'll never use and the math that is being used, I'll still have to learn. However, I can now pick up just about any math textbook or journal and glean understanding from it. All the BS implies is that the holder can glean some understanding from literature published in the field.

The Master of Science Degree (mine will be specifically in Geographic Information Science) still does not imply any particular ability in the field. Although, by this point, the student will have had about four years' (two upperclass/undergrad and two MS) worth of courses in which specific abilities are used for course work. Most MS courses follow a similar format to the undergrad course - trudging through a syllabus designed around a particular text - with the addition of some fairly significant project or research work. These projects do a good job of forcing the student to attain some level of competency in the field. However, the real goal of the MS is the completion of the thesis. The thesis involves focusing on a particular hypothesis to be evaluated. To do this, the student needs to be able to read the literature in the field to make sure that the hypothesis hasn't been tested. The two most important chapters of the MS thesis are the literature review and the methodology and analysis. The literature review demonstrates that the student can check the existing literature for similar work (vocabulary and reading skills are a given). The methodology and analysis demonstrate that the student can take information gleaned from the literature review and apply it to the hypothesis in a method that is considered sound. Note: The MS implies that the holder can thoroughly review the literature and apply knowledge gleaned from it.

At the Doctor of Philosophy level (PhD), the student works closely with an advisor, much like an apprenticeship, to learn how to add to the literature and how to convey knowledge of the field to others. This is why a PhD is required to teach at the University level. The PhD culminates in a dissertation - a document that provides a unique contribution ot the field. Any coursework a PhD student takes at this point is focused on that specific contribution. Over several years, the student applies themselves 100% towards that contribution (i.e., crunch time). Frequently, the dissertation is also broken into several articles to be published in journals - mostly because the dissertation itself is never published and is difficult to search.

The focus of this entire process is the body of knowledge represented by the literature. This is akin to the web. You can go to Google and search of all kinds of stuff. You can do the same in any field at a good research library. However, each article is also bolstered by "hyperlinks" to other publications. At the MS level, the student has demonstrated ability to apply information taken from this body of knowledge. A the PhD level, the student has demonstrated the ability to add to this body of knowledge.

I several of Isaac Asimov's books, the human race split into separate sub-species: spacers and earthers (I may have the names wrong). The spacers tended to live several thousand years whereas earthers lived the norm of about 72 years. The spacers tended to be very arrogant and never understood why earthers could advance their knowledge at all if they only lived to be 72 - and all of it died with them. Spacers felt the need to prove everything for themselves and would spend the first 1000 years, or so, of their lives proving the fundamentals of science.

The same thing can be seen in software development. Linux has advanced quickly because the programmers are free (and encouraged) to share their knowledge. At the very least, the source code is available to glean information. Closed-door shops, like Microsoft, seem to spend decades reinventing technologies and fighting millaise that the collaborative community quickly bypasses.

It is through this collaboration of information embodied by the literature that science advances. Higher education simply trains the student how to use the literature and contribute to it.

It's my own personal belief that we are on the verge of a major expansion of science. The World Wide Web encapsulates the basic goals of the scientific literature. Electronic publication allows for greater numbers of journals and faster time to publication. Hyperlinks allow citations to come alive - not just references to the bibliography at the end - but actual links to the source documents.

The literature review that used to take months of scanning cryptic indeces and digging through dusty stacks in the reserach library can be accomplished quickly online. The indeces have been electronic (and now with Google Scholar are becoming readily available) and full-text of the publication can be had instantly. This means more thorough literature reviews are possible in much less time. This means research will be better focused and publication will come faster...

#Comment Re: made: 2006-04-03 08:28:21.480535+00 by: meuon [edit history]

I'm currently working for a Ph.D... comments withheld to protect a job, that I'm learning to like. - But you could have said "Wet Paper Bag".

Quote away. What I said wasn't really earth-shattering - but it's not really well understood even by people marketting higher education.

Surprisingly, I agree with some of what you say: "At the PhD level, the student has demonstrated the ability to add to this body of knowledge." True. College works as a (lossy) high pass filter. But I don't agree that "Higher education simply trains the student how to use the literature and contribute to it", because "the literature", especially in software, has become disjoint from the products of the universities.

Instead I'd argue that the education system all the way through is, like most religions, repeating rituals that the participants no longer understand the sources of, believing that the end goals are still served by these rituals. In many cases they are, but the inefficiencies are starting to show. I think that you need only look as far as your notion that "Closed-door shops, like Microsoft, seem to spend decades reinventing technologies and fighting malaise" to see the problems with a system that takes eight or ten years of indoctrination to allow people to "contribute to" "the literature", rather than the Linux development model, where contributions are evualuated by an open community, and if they're of sufficient value even when they're rejected, the project forks.

Both systems have value, and I continue to work primarily on closed source software, but I think the analogy shows quite a bit more than you originally intended.

And I've read a number of PhD theses, and quite a few papers in the published literature, and I think that much of the culture of the university results in papers in which obfuscation and difficulty in understanding are prized ("whoah, I can barely understand this guy, he must be smart!"), so simple one or two page ideas that may warrant a half a page of end-notes end up as twenty or longer page papers that spend pages citing everything that the members of the reviewing bodies may have had a hand in to sway them on political grounds if (or when) they don't understand the technical merits.

Pursuing the PhD sounds like the perfect path for you, and I believe that it'll give you an environment to look to things that excite and interest you. But keep your eyes open for those places where the rituals and obeisances that you're asked to perform aren't serving the ends that they're claimed to, and when you make those compromises make sure that you're doing it consciously.

I realized that perhaps there's another thing at play here: I'm on a few mailing lists about digital identity systems. On these mailing lists there are single digits of people who are actually doing something, and tens of people who are constantly pointing out the reasons something is a bad idea, usually because it doesn't accomplish something that the designer didn't think was important.

I have no patience for this, in general most of the solutions do indeed have problems, but I'll happily hitch my horse to the first reasonable solution which looks like it'll have a few implementations, and I'll happily be one of those few.

It occurs to me that the processes of academia are probably a lot about mitigating that sort of bullshit, and about building concensus when everyone, no matter what their final contribution to the system, feels a need to get heard.

This isn't a well thought out theory, and there's a lot of ways I want to explore it because it's one of my big failings in trying to deal with the larger world, but maybe that'll trigger some ideas.

Can you give me an example of such a ritual? I wrote a paper back in 1990, when I took a class on Psychology of Religion in which I drew parallels between religion and academia. However, other than the "pomp and circumstance", I think most people in academia do not put alot of effort into rituals. Of course, you can attack the basic premise of lecture-based education but I haven't come across anything that works as well en mass.

Maybe in a non-peer-reviewed journal - but in a peer-reviewed journal, the reviewers are other experts in your field. And what you are refering to is the normal way of documenting the literature review. Prior to the internet, this was how ideas managed trackbacks - links to source ideas. The purpose of the review is to show how the idea developed out of existing, published ideas, so you don't have to redo the work in the cited papers and others trying to grasp your ideas can go back and pick the elements you are building upon.

This precisely is what the MS degree teaches you to do. In addition to developing more depth in experience and knowledge in your field, you learn how to acquire knowledge from the literature and apply it to your own problem. Learning to properly search the literature and cite it is the focus.

Having just completed my MS thesis, I realized another important aspect. My thesis - i.e., the actual written document - was a work that was developed over several months based on ideas that I had been working with for several years. It was the first time in my life that I worked on a written document in a way that was similar to programming. I would write an initial version, see how it read and pass it on to my advisor. We would both suggest revisions and they would be made. This was repeated about 10 times. Then the thesis was passed on to the rest of my thesis comittee (two other PhDs). They each read through and not only suggested simple grammatical changes but recommended things I should add to better substantiate my ideas. By the end, I was able to jump around inside the document and add sentences and tweak wording just like I was jumping around inside source modules fixing bugs and adding features.

I'm reluctant to name names. Various people who end up doing reviewing for assorted publications have been, still are, and will be in the future part of my social group and I don't want to single them out because I don't believe that what they do is wrong, or at least isn't something that everyone does. Even though it might be embarassing. However, when the deadline for the publication is coming down and you've got fifty papers to go through on your desk and your work and family obligations besides, understanding every nuance of that paper isn't possible, instead the reviewer falls back on secondary characteristics.

And, of course, from Alan Sokol's famed experiment in post-modern literary criticism to some of the fallout and tit-for-tat that occurred from that, I'm also very clear that this isn't just a phenomenon that happens in computer graphics.

As for the rituals, I tink the PhD thesis is one of them: The number of "look at this cool system I built" papers I've read is pretty high, the number of those that had real insight (in my admittedly subjective opinion) rather than being proof that someone could indeed implement a complex system has been pretty low. At this point I automatically chalk up any Media Lab graduate's thesis to the "they give good demo" category, and carry on.

But I think the rituals go deeper than that. The undergraduate experience is only slightly removed from hazing, an "if you've got the stamina to sit through four years of lecture we'll give you your degree" sort of thing, less about teaching than about justifying some of the sillier notions of what constitutes knowledge. The price fixing and collusion to leave students as broke as possible another.

That's not to say that the rituals and secret handshakes don't have their place. Feynmann was quite conscious that what he learned in school wasn't knowledge, it was a shared language for that knowledge, and as long as that's acknowledged I think it's a good thing.

Overall, though, my experience with those who have PhDs and those who've chosen a strongly academic path has been that there's nothing in that process that makes those people more knowledgeable, or better thinkers, it teaches them the rituals of that shared communication, and in the process pretends that that specific form of communication can substitute for thinking. Given that I also believe that thinking occurs primarily through the manipulation of the language structures we build for ourselves, this meshes in fairly tightly with my experience that many of those with doctorates don't manage to rise above the molds into which they've been poured.

So I view your news with something of the same reaction as I'd have if you chose a ministerial path, or the priesthood. I'm glad for you that it fulfills your needs, and I've known some who've taken that path whom I respect, and there are things which society demands can only be accomplished through that path, but it is the latter limitation that I'd rather see broken down than reinforced.

I really think you miss the point. The "cool demo" thesis is more than that. It's a "cool demo" that implements some .005% more than existing systems in the literature AND documents that .005% in a way that is usable by others. Sure, sometimes the level of complexity of the "cool demo" looks a little hokey, but the PhD isn't awarded based on the degree of contribution to the field. It's awarded based on the student's ability to make any contribution to the field that is sufficiently documented. The thesis or dissertation isn't supposed to be proof that someone can implement a complex system.

This is just like mathematics. You can show that something works by example (i.e., slap together a demo or work out a set of numbers that seem to work) or you can PROVE it. In order to prove it, you have to denote what is given (accepted as already proven) and you have to show adequate detail in the language. If you don't prove something in mathematics, then it really can't be used to advance thinking.

The PhD dissertation is showing that you know how to prove something new in your field. It doesn't have to be earth-shatteringly brilliant. It doesn't have to be terribly complex. It's merely the first stepping-stone in a career that, hopefully, will provide time to work on the brilliant and complex.

If an idea in any academic field is not well-documented, then it really can't be used either. As you said, Feynmann viewed what was taught in school as a shared language for knowledge. And that is the same statement I am making.

Does this process make one a better thinker? Nope. Is this the only process that can lead to good ideas? Nope. But this is a widely acknowledged process for sharing ideas.

And Greenspun's argument about tuition is like saying you shouldn't bother to drive a car because a Ferrari Enzo costs $1M - the benefits of driving the car don't justify the costs. My BS and MS together cost maybe $30K in tuition (Tennessee has rather high tuition and I paid out-of-state in Missouri). My PhD will cost $0. And did I learn $30K worth of stuff? I believe I did.

This argument should get more interesting once I get into the PhD program and my research with my advisor. I say this because I am going into an area that is very "hot" in the commercial world (GIS) but based on very, very old science (cartography) that actually falls under the auspices of what is generally recognized as a soft-science or social-science (geography).

The undergraduate experience is only slightly removed from hazing, an "if you've got the stamina to sit through four years of lecture we'll give you your degree" sort of thing, less about teaching than about justifying some of the sillier notions of what constitutes knowledge.

I recommend you look into the model of the new Olin College of Engineering in Needham, MA. I think you might be pleasantly surprised.

And oh yeah, if you're one of the 75 students accepted into each year's class, your tuition is fully paid for four years.

Diane, that looks more like it. I love some of their statements about curriculum. And my comments about tuition are not to denigrate the idea of paying one's tuition, more to point out that for the most part tuition structures at the undergrad level are more about hazing than institutional support.

Do you know of any art colleges like Olin? Not so much for the money aspect (although, hey, that's nice), but for the other ideas? The younger rat boy is starting to look around, and I'd love to point him toward some place where he'll actually feel like he'll be doing meaningful work.

Olin's program looks cool - but would it scale? They take 80 students a year - period. So they are guaranteed naturally gifted self-learners...

Further, there is a difference between engineering education and other fields in academia. Engineering is a discipline that requires a significant amount of education to enter. I'm not saying that you need the education to do engineering tasks, but you do need it to gain acceptance into the realms of the "professional engineer". This kind of education is similar to law school and medical school - where there is a recognized level of very specific education.

This is very different from academia (or maybe these same... just packaged different). When I was shopping for grad schools, I also considered applying to CompSci departments. Afterall, what I am doing is as much a computer science as it is geographical. Further, CompSci PhD's are generally paid better than Geography PhD's. But I began to ask myself "what does a CompSci dissertation look like?" and "can I really abstract away from the details of implementation into the realm of computer as a science?" I found it easier to continue down the path of geography and cartography, leveraging my computer science skills while not having to deconsruct them to the nth degree.

As far as art programs like Olin - good luck! Olin's model will work because of heavy alumni gifting. Olin's alumni will have the kind of income where they will likely give generously to the school, allowing the "tuition-free" model to continue. I doubt the same funding model will work for the arts... Just a hunch...

As I said, it's not the free tuition bits of the Olin model that pique my interest as much as the emphasis on undergraduate collaboration on large projects, the philosophical attitudes behind "competency assessments", that sort of thing. Granted, most of what I have to go on in making this assessment is a minimal amount of their marketing material, but they seem to be emphasizing different things than most college marketing materials, and I thing that those points of emphasis could transfer.

So, Eric, dropping back to your assertion that I miss the point: Your argument makes sense, that the end goal of the path which leads to the PhD is one of communication and not of technical competence. That certainly meshes with my experience. And if that's so, then where we differ is on the relative values of the ability to communicate versus the ability to do, and I'll freely concede that I weight the balance far further towards the latter than I should.

In terms of art colleges with a similar curriculum model to Olin, the first thing that popped into mind was the Cooper Union School of Art. Oddly enough, they are also a tuition-free school (primarily of art, engineering, and architecture) and very small in size, similar to Olin. Their "home test" for admissions is a known back-breaker and weeds out the unlikelies; Olin has a Candidating Weekend where those who've made the big cut are put through design exercises, presentations, interviews, etc. -- the admissions process at both these schools is unique and does ensure that the students they admit are likely to benefit from the curriculum and contribute to the school. I think it's likely that the small size is critical in sustaining these sorts of curricular models.

Communication is taken for granted by "young turks" (as we've had pointed out to us before) as is the last 1% of functionality (as long as what was needed was in the first 99%).

The goal of academia is teach communication so that the ideas explored can be fully utilized and well understood by future generations. i.e., the comments are at least as important as the code...

So that leaves to answer: how can technical competency be facilited? For me, it's all through self-learning. And I think the only reason I have gained greater technical competency than many others is that I have a higher tolerance for frustration than most...

Eric:"For me, it's all through self-learning" - And that is why Eric (and others) are an exception, rather than the rule in formal education. For Eric, and people like him, the education is a foundation, maybe more like a springboard, that enables an already self-enabling person to do more than they could by themselves.