 |
|
| FriendFeed is the easiest way to share online. Learn more » |
Daniel Lemire › Comments
Modern Biology = Cargo-Cult Science (continued) - http://www.blog.sethroberts.net/2009...
his lab has all the trappings of modern science. But the planes don’t land.
- Daniel Lemire
MathJax: Open Source Ajax-based math display into your HTML pages - http://www.mathjax.org/
Rajarshi Guha and Pierre Lindenbaum liked this
Math display for all browsers MathJax is an open source, Ajax-based math display solution designed with a goal of consolidating advances in many web technologies in a single definitive math-on-the-web platform supporting all major browsers.
- Daniel Lemire
The Endeavour - http://www.johndcook.com/blog...
"Amazing."
- Daniel Lemire
Marginal Revolution: Online Education and the Market for Superstar Teachers - http://www.marginalrevolution.com/margina...
Peter Miller, Dimitrios Diamantaras, Dan Morrill AKA Techwag and 3 other people liked this
Yes: "I have argued that universities will move to a superstar market for teachers in which the very best teachers use on-line instruction and TAs to teach thousands of students at many different universities. "
- Michael Nielsen
As a fairly high-touch teacher, I can't say that this prospect appeals.
- D0r0th34
I'm certainly not making a judgement about what appeals - I greatly prefer a small-group, high-interactivity approach, all other things being equal. But most big Universities have deliberately moved toward a model where lectures are delivered in a low-interaction way to a large audience. That's an approach where it makes a great deal of economic sense to try to scale to ever-larger audiences, which seems likely to result in a winner-takes-all kind of market.
- Michael Nielsen
I'm not sure it was entirely deliberate -- "unmindful" is the word I'd choose. But yes. I just wonder if there's going to be a student/parent backlash at some juncture. Are they really getting what they're supposedly paying for?
- D0r0th34
Having talked to a lot of administrators in Australia about this, yes, I think it was deliberate, at least there. A huge chunk of funding comes simply from student-hours taught, and so they try to ramp up numbers as much as possible.
- Michael Nielsen
I don't see much evidence of this in the UK but then few UK universities have been effective at putting high quality course materials online or teaching into a wider market
- Cameron Neylon
the US certainly has a lot of large-lecture intro courses at big unis, but we have some countervailing pressures to (perhaps?) keep us a little more honest: notably, SLACs, small state schools, and other smaller, high-touch schools.
- D0r0th34
Note that some of this is, shall we say, old news. When I attended UC Berkeley (1962-68), undergrad classes were generally either Very Large (200+, frequently 500+), taught by superstar teachers (including most of the Nobel laureates), or Very Small (<40, frequently 20-30), with lots of interaction, taught by combinations of junior faculty and grad TAs. In my fading memory, it worked great...within limits.
- Walt Crawford
Caveat: A few too many Very Large classes were taught by faculty who were only superstars in their own minds. But back then, we had Fybate Notes, so 90% of students in the dud courses just read the lectures rather than seeing them "live" (if reading from your own years-old lectures can be considered live).
- Walt Crawford
Cameron - How does funding in the UK work? Do Universities get paid per student? If so, I'd be very surprised if there wasn't upward pressure on class sizes, leading to massive low-touch classes where there's not a whole lot of difference between sitting in a lecture hall and watching on a screen.
- Michael Nielsen
Dorothea - something interesting about the leading SLACs is their (typically) enormous fees: essentially, you pay for what you get, a nice individualized, personal learning experience. It's a completely different economic model than the massive Universities cramming 500 or 1000 students into a hall, and probably one that's a lot more immune to the kind of thing described in the original post.
- Michael Nielsen
absolutely. the small state schools are the compromise option: more individualized than the big schools, less $$$, less breadth of subject matter, arguably less prestige, sometimes less quality (though for the most part I don't agree with that; there's no more deadwood at a small state school than at Big Research U).
- D0r0th34
it may be worth remarking that from where I'm sitting, our small state schools are kicking butt and taking names in undergraduate research compared to our two Big Research Us. Coincidence? I think not.
- D0r0th34
The model reminds a lot of the MAGIC group: http://maths.dept.shef.ac.uk/magic... "The MAGIC group runs a wide range of postgraduate-level lecture courses in mathematics, using Access Grid videoconferencing technology."
- Dan Hagon
I have been arguing for people to stop lecturing altogether: http://www.daniel-lemire.com/blog... Lectures in 2009 make no sense. They are a relic of the past. But as lectures disappear, I don't think the role of the "teacher" will also disappear, it will just transform. I hope we will go back to a form of apprenticeship.
- Daniel Lemire
Daniel, Amen!
- Melanie Reed
Michael, funding is per student up to set maximums, beyond that no more money and there are hard limits annually on how many places are available to bid for. So there is pressure to be efficient and maximise numbers but it isn't open ended. Also very few general courses in UK, most are subject and stream specific so most have relatively small numbers. Maximum I ever taught was 120,...
more...
- Cameron Neylon
Cameron - thanks for that! It's very interesting, and suprises me. On your last point, the original post was talking about "superstar teachers", and it's clear from context that he meant teachers who are extremely good as teachers, not researchers. (His argument is a standard one in economics about winner-take-all markets like music, sport, etc: superstars with even a slight edge in the...
more...
- Michael Nielsen
Yes, but you need to create an impression that such people exist and deserve to be promoted first. Which means they need profile, for teaching, outside their own institution. I would guess this is most effectively kicked off by a few "famous" researchers doing some hard work on teaching and then that provides a known niche in which others can also excel. I couldn't name a single person...
more...
- Cameron Neylon
In the US, Richard Muller has become well known for his course "Physics for Future Presidents", largely off the back of scaling technologies (iTunes etc). When you have the means to scale lectures, it creates a winner-take-all situation, and you expect a market for superstars to emerge. E.g., the record player / gramaphone really helped create the current winner-take-all situation in music, turning it into a far more star-driven market.
- Michael Nielsen
My sense of what is happening at UF is rather like Cameron's last comment -- the so-called "superstar" teachers come from the ranks of researchers, not those who solely teach (per their contract -- the "lecturer" position). In fact, greater kudos are granted to researchers who can also teach than are given for teachers who also do research.
- Mickey Schafer
Michael, I rather like that notion of scalability...it's interesting and a different way for a teacher to consider "students" -- more in the sense of audience. In my recent conversations with non-academics, it seems that their use of the web is not this far-flung ambient surfing that many who hang out here are accustomed to. Instead, they have "go-to" places; and would prefer that major...
more...
- Mickey Schafer
The impression of a winner-takes-all situation may in fact be illusive. Clearly there will be popular expositors and the internet gives them a platform to reach a much wider audience than a single lecture theater. However the internet also allows individual learners to ability to consume material very specific to their personal learning interests - possibly far beyond a traditional...
more...
- Dan Hagon
In a blog post a few weeks back, I arrived at a similar "winner-takes-all" conclusion from a different direction: http://openresearch.sebpaquet.net/2009...
- Seb Paquet
@dupuisj My challenge is serious. In 10 or 20 years, we won't need buildings to house books. No library. Who is going to need librarians?
Steve Koch liked this
if you accept the premise, it depends whether librarians add value to containers (physical books) or to discovery, use and preservation of content
- Richard Akerman
these are some of my thoughts from 2008 on the strains on the profession and approaches it can take to adapt http://scilib.typepad.com/science...
- Richard Akerman
Interesting post. Thanks. I liked it.
- Daniel Lemire
libraries - particularly those on college campuses - have a lot more roles than just as warehouses of books. If you accept the statement that we won't need print materials in 10-20 years (which I don't), you still need library-as-place for study areas, for meeting rooms, as a place to house the people who acquire, organize, and provide access to the non-print materials.
- Christina Pikas
if books are online instead of in "book warehouses," the odds are they will be difficult to use (think about all the different ebook platforms that exist now, compared to ONE print-book platform), then my guess is SOMEONEs will need to know how best to use these newfangled book gadgets and train those who don't know. Just as we now value people who show researchers how to find stuff in...
more...
- Stephanie_Thankful
Stephanie: Oh! My! So... librarians offer tools that are hard to use (Scopus, ScienceDirect...) thus insuring they are needed. Not a great plan. That has been my beef for many years with librarians: why stick with such poor tools? When Google Scholar came about, I was freed from them and I have not looked back. Google Scholar (and similar tools) improve every month... So, relying on these hard-to-use tools to justify jobs is a losing proposition on the long run.
- Daniel Lemire
I think there is a challenge in perceptions of library as place between students and faculty. This is a big issue for my library, as the NRC campus more or less consists entirely of "faculty" and (I believe) we need to completely reinvigorate the idea of library as place for them (I haven't been at the library long enough to know whether it was ever a gathering place, but it's definitely empty now).
- Richard Akerman
I don't know the details, but engineering library at U. New Mexico is clearing out the stacks and has a grant (NSF?) for a building a data visualization center (huge flat panel screens, computers for image processing, etc.) I don't think they'll have virtual reality cave, but that would be a big draw for me. Combined with helping with digital archiving and other repository services, I'm...
more...
- Steve Koch
There is a good point that in some sense libraries are about pooling resources to get a better shared resource than any individual could get. This was books and articles. In some public libraries it's now e.g. electric metres. There's no reason that shared service and expertise can't be giant screens, or compute resources, or reputation management.
- Richard Akerman
@mrgunn I'd put him pretty squarely in the comp sci camp.
Computer scientists are the fifth column in libraries...
- Richard Akerman
librarians are the fifth column in computer science
- D0r0th34
In my experience, CS people have been both among the most open and welcoming on the one hand and on the other hand, the most oddly aggressively anti-library. Fortunately, there aren't really any of the later at mpow. At least not to my face ;-)
- John Dupuis
one suspects they've had to deal with a few (or more than a few) anti-CS librarians
- D0r0th34
I'm not anti-librarian, and I don't know about anti-CS librarians... but I have spent part of my undergraduate years struggling to find what I was l was looking for. It looked to me like the librarians made sure it was difficult by sticking with archaic software. As recently as 2004, CISTI had user interfaces from hell. Then came tools like Google Scholar and I was "liberated". I don't...
more...
- Daniel Lemire
He's got a good point about archaic interfaces. Seems to me like some of the tortuous processes one is required to go through is partly due to the byzantine licensing agreements that are more or less forced on libraries. Open Access FTW!
- Mr. Gunn
I'm not sure Google Scholar got special licensing agreements. What they did however is build their own tools and crawlers.
- Daniel Lemire
Google Scholar absolutely does arrange for license to crawl scholarly publishers, Daniel. Mr. Gunn, yes, grotesque agreements are part of the problem.
- D0r0th34
@Daniel What do you think about CISTI Discover http://discover-decouvrir.cist...
- Richard Akerman
The (long tail) End of the Book - http://synthese.wordpress.com/2009...
"Paper books have benefits over e-books. So does a typewriter or a plume over a word processor. A sailboat is often better than a motorized boat. And so on. And yes, of course, ebooks might be the nail in the coffin for librarians. At the very least, they will need to seriously reinvent themselves, if that's possible.But the future is here, and it includes ebooks. Whether it includes librarians, that is much less certain."
- Daniel Lemire
The Endeavour - http://www.johndcook.com/blog...
"Forget surveys. Think corporate data. Silly things like income can be defined in twelve different ways inside the same corporation."
- Daniel Lemire
It doesn’t pay to be the computer guy - http://www.johndcook.com/blog...
"In some surveys, you see that older people (40 something) are leaving IT en masse. We also have fewer students in our IT programs. Finally, it has gotten somewhat harder, in my experience to recruit experienced IT people. The net lesson, here, is that IT as an industry is just not all that important. Having an IT engineer twice as good as your competitor will not help your bottom line, much. Yet, I don’t entirely buy the “rapid change devalues skills”. I haven’t used Windows in about a decade, but I still help family members with their Windows problems. I understand the foundation. It does not matter which Operating System you use, my knowledge is transferable. Obviously, there might be, one day, a paradigm change in how Operating Systems are designed. But still, a lot of my knowledge will carry over, I’m sure. The fact that things change, a little bit each year, maybe a bit more than other industries, is not necessarily a negative. It means that there are new things to learn all the..."
- Daniel Lemire
Software Archeology - http://www.johndcook.com/blog...
"@White 10 What do you 20 goto 50 30 bad? 40 goto 10 50 mean 60 gotos are 70 goto 30"
- Daniel Lemire
The Endeavour - http://www.johndcook.com/blog...
"Cool. I’ll know where to look if I need such inequalities! (Alas, in my research, I always end up with unknown or custom distributions.)"
- Daniel Lemire
Office 2007 documents are zipped XML - http://www.johndcook.com/blog...
"I believe this storage format was initiated by OpenOffice (actually Sun Microsystems)."
- Daniel Lemire
PhD Proposal, Pass! - http://beshiros.wordpress.com/2009...
"Congratulations. I hope you will tell us more about your research in the future."
- Daniel Lemire
The Endeavour - http://www.johndcook.com/blog...
"Gotos, C++, Fortran, really small fonts… you are going to give me nightmares."
- Daniel Lemire
Since the primary effect of software is to facilitate bureaucratic complexity, software projects must fail (funny) http://www.berglas.org/Article...
Greg Guitarbuster liked this
I think that argument is deeper than it lets on.
- Geoff Wozniak
I agree.
- Daniel Lemire
"We now see why it is so critical to society that software projects fail. The boundless creativity of politicians and bureaucrats to develop new and more complex regulation is bounded only by the bureaucracy's inability to implement them. The absolute size of the bureaucracy is constrained by external factors, so the only effect of automation can be to increase bureaucratic complexity. "
- Todd Hoff
The Noisy Channel - http://thenoisychannel.com/2009...
"Good. Now whenever I’m annoyed at Google, I’ll know where to voice my complaints! ))))"
- Daniel Lemire
A question on MDX performance - http://www.dbms2.com/2009...
"Mondrian from Pentaho will “translate” your MDX queries to SQL. However, I am an academic, so I do not know whether you can actually use Pentaho Mondrian for production. (It is fine for my students though.)"
- Daniel Lemire
A question on MDX performance - http://www.dbms2.com/2009...
"I realize that the question is about benchmarking solutions supporting MDX. Nevertheless, there is a possible confusion. MDX is a language. If you have performance or scalability problems, they are probably not due to MDX. MDX has some limitations, and I suppose it is possible to hang yourself with it… but if you use it properly for the type of problems it was designed for, then MDX, in itself, has nothing to do with the performance of your OLAP system. All in all, I think it is an odd question."
- Daniel Lemire
Simple backup software - http://www.johndcook.com/blog...
"What is wrong with rsync?"
- Daniel Lemire
The Endeavour - http://www.johndcook.com/blog...
"What a great subject!!! (And closely related to a current interest of mine… random hashing…) Lucky you to contribute to such a great book. I envy you."
- Daniel Lemire
The Endeavour - http://www.johndcook.com/blog...
"Even non-technical books lose value quite fast."
- Daniel Lemire
Nothing is “Miscellaneous” - http://synthese.wordpress.com/2009...
"I think there is a difference between classifying books on shelves, and retrieving digital documents on a laptop.The Dewey system is a workaround for our inability to index physical objects in rich ways. I can't very well do full text search on physical books, can I?In the digital age, we no longer need Dewey, except maybe at the public library... until we move to e-books... which should happen with a decade or so."
- Daniel Lemire
Free ebook: Data-Intensive Scientific Discovery (RT @JohnDCook) - http://research.microsoft.com/en-us...
Rafael Sidi, Mike Chelen and Christina Pikas liked this
Presenting the first broad look at the rapidly emerging field of data-intensive science Increasingly, scientific breakthroughs will be powered by advanced computing capabilities that help researchers manipulate and explore massive datasets. The speed at which any given scientific discipline advances will depend on how well its researchers collaborate with one another, and with technologists, in areas of eScience such as databases, workflow management, visualization, and cloud computing technologies. In The Fourth Paradigm: Data-Intensive Scientific Discovery, the collection of essays expands on the vision of pioneering computer scientist Jim Gray for a new, fourth paradigm of discovery based on data-intensive science and offers insights into how it can be fully realized.
- Daniel Lemire
My Biased Coin - http://mybiasedcoin.blogspot.com/2009...
"Welcome to our little private club. Think about how their standards must be low: I was there before you! Kidding aside, I just need to find a way to *stop* receiving the paper version of CACM. Frankly, in 2009, I'm at a point where I prefer to read PDF or HTML articles."
- Daniel Lemire
Free PDF Download of book on Statistical Learning (2008) RT @JohnDCook - http://www-stat.stanford.edu/~tibs...
Geoff Wozniak liked this
The Elements of Statistical Learning (2nd edition) by Hastie, Tibshirani and Friedman (2008). Springer-Verlag. 763 pages.
- Daniel Lemire
Make a decision - http://sethgodin.typepad.com/seths_b...
Not deciding is usually the wrong decision.
- Daniel Lemire
Scientific data sharing - http://www.dbms2.com/2009...
"I must point out that your discussion is American-centric whereas it shouldn’t. If the UK (say) funding agencies require data sharing (and they do), this changes the game for everyone, including the Americans. I don’t see people sharing the data just “nationally”. As an aside, I recently wrote a book chapter which has some relevance here: On the Challenges of Collaborative Data Processing http://arxiv.org/abs/0906.0910 In this chapter, we ask “collaborative text editing lead to Wikipedia, where can collaborative data processing lead?” At the very least, I feel that we are asking the right question. (As to answering it, it gets tougher.)"
- Daniel Lemire
The Noisy Channel - http://thenoisychannel.com/2009...
"I’m really happy (for you and in general), even though I prefer text to television."
- Daniel Lemire
Finally updated the academic site - http://beshiros.wordpress.com/2009...
"Good. But what is the deal with the big truck?"
- Daniel Lemire
Free PDF Download of the Book A=B by Petkovsek, Wilf and Zeilberger (1996) - http://www.math.upenn.edu/~wilf...
"A=B" is about identities in general, and hypergeometric identities in particular, with emphasis on computer methods of discovery and proof. The book describes a number of algorithms for doing these tasks, and we intend to maintain the latest versions of the programs that carry out these algorithms on this page. So be sure to consult this page from time to time, and help yourself to the latest versions of the programs.
- Daniel Lemire
This book is awesome. Used the methods to sum sum several nasty series during my Ph.D.
- Matt Leifer
Other ways to read this feed:
