updated - Physics education and real world perspective

I was absolutely horrified to read about this story, about how more than 200 girls were kidnapped in Nigeria by a radical Islamic group because they had the temerity to show up for a physics test (emblematic of getting a Western education).  I fervently hope that there is enough public outrage about this to get some positive action there, though it's hard to be optimistic.

Stories like this should remind many of us how petty our concerns (departmental rankings; referee comments; grant reviews; tenure decisions; grad school admissions decisions) really are, even if they seem stressful and important.  It is terrible that in the 21st century there are still large segments of the world where modern learning is expressly forbidden, on pain of death.

Update:  Uggh.  It would appear that the abducted girls have been forcibly married off.  There are so many things wrong here that it's hard to know where to begin.

Informal survey: How important are departmental rankings?

Coincident with the annual US graduate school admission season, I've had a few conversations in recent days where the topic of departmental rankings has come up.  I've written about this general topic before (here and here, for example).  I want to perform a non-serious (in the sense that it's a self-selected survey population that may well be atypical) survey here of those applying to grad school, currently in grad school, or recently (say within the last 4 years) completing grad school:  How important were official rankings (e.g., US News; NRC) of graduate programs in your grad school application process (where you decided to apply) and in your eventual decision?  

Recurring themes in (condensed matter/nano) physics: spatial periodicity

A defining characteristic of crystalline solids is that their constituent atoms are arranged in a spatially periodic way.  In fancy lingo, the atomic configuration breaks continuous translational and rotational invariance (that is, it picks out certain positions and orientations in space from an infinite variety of possible choices), but preserves discrete translational invariance (and other possible symmetries).  

The introduction of a characteristic spatial length scale, or equivalently a spatial frequency, is a big deal, because when other spatial length scales in the physical system coincide with that one, there can be big consequences.  For example, when the wavelength of x-rays or electrons or neutrons is some integer harmonic of the (projected) lattice spacing, then waves scattered from subsequent (or every second or every third, etc.) plane of atoms will interfere constructively - this is called the Bragg condition, is what gives diffraction patterns that have proven so useful in characterizing material structures.  Another way to think about this:  The spatial periodicity of the lattice is what forces the momentum of scattered x-rays (or electrons or neutrons) to change only by specified amounts.

It gets better.  When the wavelength of electrons bound in a crystalline solid corresponds to some integer multiple of the lattice spacing, this implies that the electrons strongly "feel" any interaction with the lattice atoms - in the nearly-free-electron picture, this matching of spatial frequencies is what opens up band gaps at particular wavevectors (and hence energies).  Similar physics happens with lattice vibrations.  Similar physics happens when we consider electromagnetic waves in spatially periodic dielectrics.  Similar physics happens when looking at electrons in a "superlattice" made by layering different semiconductors or a periodic modulation of surface relief.

One other important point.  The idea of a true spatial periodicity really only applies to infinitely large periodic systems.  If discrete translational invariance is broken (by a defect, or an interface), then some of the rules "enforced" by the periodicity can be evaded.  For example, momentum changes forbidden for elastic scattering in a perfect infinite crystal can take place at some rate at interfaces or in defective crystals.  Similarly, the optical selection rules that must be rigidly applied in perfect crystals can be bent a bit in nanocrystals, where lattice periodicity is not infinite.

Commensurate spatial periodicities between wave-like entities and lattices are responsible for electronic and optical bandgaps, phonon dispersion relations, x-ray/electron/neutron crystallography, (crystal) momentum conservation and its violation in defective and nanoscale structures, and optical selection rules and their violations in crystalline solids.  Rather far reaching consequences!

End of an era.

As long as we're talking about the (alleged) end of science, look at this picture (courtesy of Don Monroe).  This is demolition work being done in Murray Hill, NJ, as Alcatel-Lucent takes down a big hunk of Building 1 of Bell Labs. 

This building and others at the site were the setting for some of the most important industrial research of the 20th century.  (Before people ask, the particular lab where the transistor was first made is not being torn down here.)  I've written before about the near-demise of long-term basic research in the industrial setting in the US.  While Bell Labs still exists, this, like the demise of the Holmdel site, are painful marks of the end of an era.

John Horgan: Same old, same old.

John Horgan writes about science for National Geographic.  You may remember him from his book, The End of Science.   His thesis, 17 years ago, was that science is basically done - there just aren't going to be too many more profound discoveries, particularly in physics, because we've figured it all out and the rest is just details.  Well, I'll give him this for consistency:  He's still flogging this dead horse 17 years later, as seen in his recent column.  I disagree with his point of view.  Even if you limit yourself to physics, there are plenty of discoveries left to be made for a long time to come - things only look bleak if (a) you're only a reductionist; and (b) you limit your interest in physics to a narrow range of topics.  In other words, possibly looking for supersymmetric partners at the LHC might not be a great bet, but that doesn't mean that all of science is over.

A video interview for an online nano course

Two of my colleagues (Dan Mittleman and Vicki Colvin) put together a Coursera class this past year, "Nanotechnology:  The Basics", and as part of that they interviewed several Rice faculty about different bits and pieces.   They spoke to me about nanoelectronics, but the conversation ended up ranging into a discussion of hype in science and the importance of communicating to a general audience.  The video is now up online here, and the hype/science presentation discussion starts at around 18:38. 

Recurring themes in (condensed matter/nano) physics: hybridization

Suppose I have two identical systems, such as two copies of a mass attached to a spring (anchored to an immovable wall).  Each system by itself has some characteristic response, like a frequency of motion, and those responses are identical because the independent systems are identical.  Now consider coupling the two systems together, such as linking the two masses by another (weak) spring, and ask what the total coupled system response looks like.  With classical oscillators like our example, we would say that we find the "new normal modes" of the coupled system - instead of writing separate equations to describe Newton's laws for each mass separately, we can do some kind of change of variables and consider new coordinates that combine the motions of the two masses.  When we do this, we end up again with two characteristic frequencies (basically two effectively independent oscillators), but now the frequencies differ a bit, one being higher and one being lower than the original independent oscillator frequency.  You can generalize this to \(N\) oscillators and find \(N\) new normal modes with a band of frequencies, with the bandwidth determined by the strength of the couplings.

This coupling+splitting the modes crops up again and again.  You can use this to make an electronic band pass filter by capacitively coupling together a bunch of \(LC\) circuits.  You can make a mechanical band pass filter by elastically coupling together a bunch of mechanical oscillators.  You can make a miniband of electronic states by coupling together a bunch of quantum wells that each individually would have identical bound states.  You can make a band of electronic states by coupling together the single-particle atomic states of a periodic array of atoms.  The stronger the coupling, the wider the resulting bandwidth.  This basic idea (taking noninteracting systems, coupling them, and re-solving for new effectively noninteracting modes) is so powerful that when it doesn't work sometimes, it can be jarring.  

Any advice: LaTeX and makeindex

Readers - For a long time now I have been working on a very large LaTeX document (actually built out of a number of sub-documents) that I will discuss further in later posts.  I greatly desire to create an index for this document, and I know about the LaTeX package \makeindex.   The question is, does anyone know of a good frontend application that can make the creation of the index less tedious?  The brute force approach would require me to go through the document(s) by hand and insert a \makeindex tag every time a term that I wish to index appears.  For an index containing a couple of hundred entries, this looks excruciating.  What I would love is an application where I identify the terms for which I want index entries, and it then automatically inserts the appropriate tags (in a smart way, not putting tags inside LaTeX equations, for example).  While this would be imperfect, it would be easier to start from an over-complete index and pare down or modify than to start from scratch.  Yes, I am sure I could use perl or another scripting language to make something, but I'd rather not reinvent the wheel if someone has already solved this problem.  Thanks for any suggestions.

Recurring themes in (condensed matter/nano) physics: boundary conditions

This is the first in a series of posts about tropes that recur in (condensed matter/nano) physics.  I put that qualifier in parentheses because these topics obviously come up in many other places as well, but I run across them from my perspective.

Very often in physics we are effectively solving boundary value problems.  That is, we have some physical system that obeys some kind of differential equations describing the spatial dependence of some variable of interest.  This could be the electron wavefunction \( \psi(\mathbf{r})\), which has to obey the Schroedinger equation in some region of space with a potential energy \( V(\mathbf{r})\).  This could be the electric field \( \mathbf{E}(\mathbf{r})\), which has to satisfy Maxwell's equations in some region of space that has a dielectric function \( \epsilon(\mathbf{r})\).  This could be the deflection of a drumhead \( u(x,y) \), where the drumhead itself must follow the rules of continuum elasticity.  This could be the pressure field \( p(z) \) of the air in a pipe that's part of a pipe organ. 

The thread that unites these diverse systems is that, in the absence of boundaries, these problems allow a continuum of solutions, but the imposition of boundaries drastically limits the solutions to a discrete set.  For example, the pressure in that pipe could (within reasonable limits set by the description of the air as a nice gas) have any spatial periodicity, described by some wavenumber \(k\), and along with that it would have some periodic time dependence with a frequency \(\omega\), so that \( \omega/k = c_{\mathrm{s}}\), where \(c_{\mathrm{s}}\) is the sound speed.  However, once we specify boundary conditions - say one end of the pipe closed, one end open - the rules that have to be satisfied at the boundary force there to be a discrete spectrum of allowed wavelengths, and hence frequencies.   Even trying to have no boundary, by installing periodic boundary conditions, does this. This general property, the emergence of discrete modes from the continuum, is what gives us the spectra of atoms and the sounds of guitars.

How should philanthropists and foundations fund science?

This article from the NY Times discusses the funding of science research by wealthy individuals and, by extension, philanthropic foundations set up by those folks.  It brings up an issue that I phrased in the form of a question as the title of this post.  I'm not going to offer any simple overarching answer, but I do want to make a couple of observations (strictly my opinion, of course):

• Many wealthy people and foundations support medical research.  This makes a lot of sense - generally philanthropists want to Help People, and supporting research that directly affects medical care and quality of life is a completely sensible choice.
• A smaller number of people and foundations support research in the physical sciences and engineering; like those who support medical research, they want to Help People, and they realize that supporting basic research and the education of technically skilled and creative people is a great way to do that.
• Both groups, however, face the challenge that any investment that they make is basically a drop in the bucket compared with what governments can do.  NIH puts tens of billions of dollars a year into medical research.  NSF's annual budget is around $7B.  
• In my experience, the philanthropic supporters of science are well aware of this - if they want to make sure that their money makes a difference, they need to invest in supporting things that are not what government agencies are already doing.  Their challenge, then, is to identify areas (and eventually institutions and people) where their investment will really move the needle.  Of course, they need to be able to tell wheat from chaff.  Peer review is a customary way to do this, and that is often how the government agencies (most of them, anyway) make judgments, but peer review tends toward being risk averse.  An alternative is to have a dedicated science board to do reviews and make decisions.  This, too, is tricky, particularly if the members aren't exact subject matter experts.  How much weight should be placed on prior track record?  Researchers who are senior and already have major awards, etc. can be a lower risk - they have already demonstrated that they can do great work.  On the other hand, if someone is already extremely well supported (as such people often are), how much difference will philanthropic support really make?  It seems like a very tricky decision process, particularly depending on the amounts involved.  
• There is no question that having grants with wide flexibility (e.g., Packard; presumably MacArthur) can be wonderful.  At the single investigator level, there is also no question that there can be real benefits from being able to concentrate on actual science - that's an argument for funding support large enough that it allows investigators to lay off writing other grants to some extent.  (That's one aim of things like the Howard Hughes Investigator program.)  

Feel free to comment below.  Perhaps additional philanthropic support is one way to get more "scientific mavericks" (as long as that doesn't mean funding of unfounded crazy stuff).  

Taking a few days, + a philanthropy suggest for Google or Intel

Last post for a few days.  I want to make a suggestion, though.  Hey large tech companies, like Intel or Google, or for that matter Sematech or the SRC, or foundations like Keck, Moore, Packard, MacArthur:  You may have heard that NSF seems to have put shared physical sciences research infrastructure on the back burner.   I firmly believe that this is a bad decision that will have lasting negative consequences for many people.  I've written before about how much impact on science and engineering research and education there would be if companies (or individuals, I suppose, if they were sufficiently wealthy) would step forward and endow shared research equipment and staffing at universities.   Now is the time, when there is likely to be a real federal gap here.  I'm serious, and I'd be happy to talk with any interested parties about how this could be done - just email me. 
Update:  this is highly relevant.

Coolest paper of 2014 so far, by a wide margin.

Sorry for the brief post, but I could not pass this up.

Check this out:  http://arxiv.org/abs/1403.1211

I bow down before the awesomeness of an origami-based microscope. 

March Meeting wrap-up

I've been slow about writing a day 3/4/wrapup of the APS meeting because of general busy-ness.  I saw fewer general interest talks over that last day and a half in part because my own group's talks were clustered in that timeframe.  Still, I did see a couple of interesting bits.

• There was a great talk by Zhenchao Dong about this paper, where they are able to use the plasmonic properties of a scanning tunneling microscope tip to perform surface-enhanced Raman spectroscopy on single molecules (in ultrahigh vacuum and cryogenic conditions) with sub-nm lateral resolution.  The data are gorgeous, though how the lateral resolution can possibly be that good is very mysterious.  Usually the lateral extent of the enhanced optical fields is something like the geometric mean of the tip radius of curvature and the tip-sample distance.  It's very hard to see how that ever gets to the sub-nm level, so something funky must be going on.
• I saw a talk by Yoshihiro Iwasa all about MoS₂, including work on optics and ionic liquid gating.  
• I went to a session on the presentation of physics to the public.  The talks that I managed to see were quite good, and Dennis Overbye's insights into the NY Times' science reporting were particularly interesting.  He pointed out that it's a very challenging marketplace when so much good (or at least interesting) science writing is given away for free (as in here or here or here).  He did give a shout-out to Peter Woit, particularly mentioning how good Peter's sources are.

I probably should wade into focus topic organizing again;  it seemed like this year there were more issues with parallel sessions about similar topics and some lack of cohesiveness within some of the contributed sessions.  (This isn't meant as a criticism of those who invested their time to do this, which is absolutely appreciated!  I am well aware how hard it is, particularly as the meeting keeps growing.)

The end of the National Nano Infrastructure Network? Federal support for shared facilities.

The National Nanotechnology Infrastructure Network is, as their page says, "an integrated networked partnership of user facilities, supported by the National Science Foundation, serving the needs of nanoscale science, engineering and technology".  Basically, the NNIN has been a mechanism for establishing nodes of excellence at sites around the US, where people could travel to use equipment and capabilities (high resolution transmission electron microscopy; sophisticated wafer-scale electron beam lithography; deep etching) that they lack at their home institutions.  Crucially, these shared facilities are supported by skilled technical staff that can train users, work with users to develop processes, perform fee-for-service work on occasion, etc.  The most famous sites are the Stanford Nanofab Facility and the Cornell Nanofab.  Over the years, the NNIN has been instrumental in an enormous amount of research progress.  Note that this effort is distinct from Major User Facilities (such as synchrotrons, neutron sources, etc).

This year, there was a competition for a Next Generation NNIN - the call is here.  The idea was very much to broaden the network into characterization as well as fabrication, and to reach new, growing communities of users in areas like bio, the environment, earth sciences/geo.  After a proposal process that boiled down to two teams (one with 18 universities; one with 20), very extensive full proposals, reverse site visits, written responses to reverse site visits and reviews, etc., the NSF decided not to make an award.  It would appear that there will be another call of some kind issued in fall, 2014.  For now, what this means is that the NNIN is ending.  Cornell, Stanford, and the other sites face major cuts in funding for staff and support for external users.  (Full disclosure:  I was the Rice rep on one of the teams.)

This whole issue is very complex, but it raises a number of questions that would benefit from a discussion in the community.  What should be the pathway to federal support for shared facilities and staffing, particularly tools and techniques that would be prohibitively expensive for individual universities to support via internal funds?  Should there be federal support for this?  Should it come from NSF?  How can we have a stable, sustained level of research infrastructure, including staffing, that serves the broad scientific community, in an era when funding is squeezed ever more tightly?  If the burden is shifting more toward individual universities having to support shared infrastructure basically with internal funding and user fees, what impact will that have?  Comment is invited.

UPDATE:  Here is a story that Science is running regarding the decision, or lack thereof. 

March Meeting, Day 2

This is a meta-post - I'm writing it while sitting in the back of a session on presenting science to the public.  A brief list of some of the neat things I heard yesterday:

• I saw a very nice talk by Jelena Vuckovic about doing nonlinear and cavity optics, with (self-assembled InAs) quantum dots as the emitters, and the cavity being formed in 2d photonic band gap systems.  The latest work looks at nonlinear effects like photon blockade, and makes contact to some work involving "circuit" quantum electrodynamics (see here).
• I went to a talk by Ken Golden, who taught me sophomore differential equations, and he gave a fascinating presentation about applying rigorous math (percolation theory, treating microstructured composites like effective media) to the challenging problem of understanding melting polar sea ice.  As a side note, he showed a great picture that is an example of the "quasistatic limit" - long wavelength surface ocean waves don't "see" individual ice floes, but instead propagate in an effective medium.
• There was a great invited session about oxide heterostructures.  Mobilities are improving (under the right conditions) to the point where some ways of learning about the band structure through electronic transport are now becoming possible.  Very impressive was a talk by Shahal Ilani, where he presented a very compelling view of the importance of structural domains ("ferroelasticity") in the underlying strontium titanate - when those domains are under control, transport becomes much more clean, revealing the apparent existence of a magnetically interesting phase at high carrier density and high in-plane magnetic field.

March Meeting, Day 1

Observations from the first day of the APS March Meeting:

• There has been a lot of progress and excitement in looking at layered materials "beyond graphene".  It's interesting to see a resurgence of interest in transition metal (Ti, but more frequently W and Mo) dichalcogenides (S, Se, Te), a topic of great activity in bulk materials growth in the 1970s and early 80s.  There are clearly a lot of bright people working on ways to grow these materials layer-by-layer, with the long-term idea of making structures somewhat like semiconductor heterostructures (e.g., GaAs/AlGaAs), but with the richer palette provided by these materials (exhibiting charge density waves, strong spin-orbit effects, complex band structure, etc.).  Molecular beam epitaxy of these materials with high quality is generally very hard.  For example, Mo and W are extremely refractory, requiring electron beam evaporation at temperatures exceeding 2500 C, and sticking at the sample surface without much diffusion.  Whoever really gets layer-by-layer, large-area growth working with diverse materials is going to make a big impact.
• I saw Heinrich Jaeger give a great talk about granular materials by design.  These are entirely classical systems, but they are extremely challenging.  If you think about it, they are not crystalline (no long-range symmetries to exploit in modeling), they are non-ergodic (the constituent grains are kinetically limited, and can't explore all possible configurations), and nonlinear (the interactions between particles are short-ranged and very strong).  Very interesting.
• I caught two talks in the session looking at silicon-based quantum information processing.  It's possible to create and manipulate dangling bonds on the Si surface (localized states that can trap electrons) and look at how those bonds interact with each other.  Very neat.  Looking at particular individual impurities, with the right system (erbium in Si), you can couple a single impurity to a single-electron transistor charge sensor.  Then, you can manipulate that impurity with optical techniques and use the charge detection to determine its state.  Very impressive.
• The session on secrecy in science was very good.  The ability to manufacture viruses by design is genuinely frightening (though it loses some menace when the words "Pandemic - millions of deaths?" are projected in Comic Sans).  The discussion of intellectual property was great and the role of universities merits its own blog post.  Lastly, I was unaware of the WATCHMAN project, which is a very interesting neutrino physics experiment that as an added bonus should allow the international community to detect rogue nuclear reactors meant for weapons development.

Upcoming blogging: APS March Meeting + recurring physics themes

This coming week is the APS March Meeting in Denver.  I'll probably blog about some of the talks, but I may not give detailed recaps as I've done in some past years - it's hard to get a complete picture of a meeting that's grown so vast (and I have a ton of work I somehow need to get done over that week).  Topics that are clearly hot, based on the meeting program:  topological everything (insulators, superconductors, Majorana structures, etc.); layered everything (graphene, transition metal dichalcogenides, including optical properties); oxide heterostructure materials; quantum information; cold atoms to examine particular condensed matter problems incl systems out of equilibrium; unconventional superconductivity (incl quantum criticality, pnictides, cuprates, etc.); plasmonics.   I'll admit that I'm curious about the future of the New Media in the communication of science to the public. 

I've also been thinking about doing a series of posts really aimed at the public about recurring themes that crop up in physics.  This will require a bit of thought to make the writing really accessible to a general audience, but it could be fun.  This story by Adam Frank was very well done and inspirational in terms of what such a series could be.

Ballistic electrons in graphene nanoribbons at room T: whoa!

The de Heer group at Georgia Tech has a paper in this week's Nature where they present some results on graphene nanoribbons that are quite unexpected and exciting.  Rather than exfoliate graphene from graphite, or grow it via chemical vapor deposition, the Georgia Tech group creates graphene via the controlled transformation of silicon carbide.  In this latest work, they used a vicinal substrate (meaning that it is cut slightly off-axis from a high symmetry direction, so that the surface has regularly spaced atomic terraces).  When this substrate is annealed in a particular way, graphene forms across the surface.  Interestingly, on the plateaus, the resulting material appears to be semiconducting (based on tunneling measurements made by scanning tunneling microscopy (STM)), while the steps reconstruct and form sloping sidewalls that have 40 nm wide ribbons that are metallic graphene (as seen through tunneling and photoemission).

Using in situ multiple tungsten STM tips, they are able to measure the conductance of such ribbons as a function of length.  Remarkably, they find that the two-terminal conductance is approximately independent of length (!) over a broad range of lengths (from 0.5 \(\mu\)m to about 16 \(\mu\)m) even at room temperature, and it has the very suggestive value of \(e^{2}/h\), which is what you would expect for a single quantum channel, with one species of the electronic spin.  This kind of violation of "Ohm's Law" is expected when the electrons travel essentially without scattering from one end of a device to the other.  Ordinarily we can't see this at room temperature in macroscopic conductors, because there are many ways electrons can scatter, including inelastic processes involving lattice vibrations.  The authors have a number of other measurements that are consistent with the implication that a single channel is somehow able to propagate ballistically over these long distances at room temperature.  Indeed, they can use additional tips as "passive" scattering centers; placing an additional tip on the wire makes the conductance drop, presumably because that tip is able to cause back-scattering. 

These observations are very interesting, since they suggest that there is some kind of "protected" channel that allows conduction by basically making back-scattering (which would usually contribute to resistance) very disfavored.  The apparent spin polarization (inferred from the conductance value, not measured directly) is also intriguing.  I wonder if the "kink" at the edges of the ribbons where the sidewall transitions to the flat plateaus on either side of the ribbon acts as some sort of source of strong spin-orbit interactions (despite the low \(Z\) of carbon) by distorting the graphene lattice.  In any case, it is nice to see a genuinely surprising graphene result.

What's the deal w/ NIF and fusion?

The National Ignition Facility at Lawrence Livermore National Lab just published a couple of papers (PRL and Nature) about their latest results in inertial confinement fusion.   The idea is to hit a deuterium-tritium fuel pellet with 192 converging high power laser beams and dump enough energy into the nuclei (by various means) that they can overcome their Coulomb repulsion and fuse, releasing a helium nucleus, a neutron, and energy.  Their latest results demonstrate net "fuel gain" - they are able to infer via complicated means how much energy actually got coupled to the D/T (about 10 kJ in a shot), and from the neutrons they can determine how much energy came out from the fusion reactions (about 15 kJ in a shot).  This sounds great, and it's an important physics milestone for the researchers.  However, something gets lost in the press releases:  They dump in about 1.8 MJ from the lasers to get 10 kJ into the fuel.  That's an input coupling efficiency of 0.05%.  Also, bear in mind that they have to rebuild the whole sample holder and everything before each shot.

While the latest results are a nice and critical physics step, it is extremely hard for me to believe that the NIF approach will ever lead to anything resembling a power plant.  For a sense of scale, the NIF annual budget is something close to $1B, while the US commitment to ITER is on the order of $200M/yr, the Princeton Plasma Physics Lab annual budget is around $100M, and the fusion program at Sandia is about $5M/yr.  (For reference, the F-35 fighter program costs about $12B/yr, and the NSF annual budget is around $7B/yr).   NIF is a fascinating physics testbed, a way to study certain processes without detonating nuclear weapons, and the warp core of the most recent iteration of the starship Enterprise.  However, press articles implying that this recent result is a breakthrough toward fusion power are misleading.

Are blogs really changing scientific discourse?

Lately there has been a fair bit of talk (here, for example, or here) about whether blogs, particularly those written by scientists, are actually changing the scientific discourse and the way science gets done (particularly in terms of debating controversies or resolving disagreements).   I was recently asked this by a science journalist, too.

My short answer is, "maybe, sometimes, but mostly 'no'."  (Thus, I am roughly consistent with the old adage that article titles posed in the form of a question are almost always answered by "no".)  The main reason that blogging is, in my view, not having some major transformative effect on science is that the vast majority of scientists do not blog, and a slightly smaller (but still vast) majority do not even read blogs let alone comment on them or ponder writing one.  Blogs are still far and away the exception rather than the rule in terms of how scientific discussions take place.

That being said, when the relevant participants participate in blog discussions (or the equivalent, as on mathoverflow), very cool things can take place.  However, I think the most productive version of this happens either when someone really tries to educate an interested audience (my attempted model here most of the time - a sort of science journalism by scientists), or when informed discussion happens between knowledgeable experts (sort of a virtual version of the kinds of conversations that can happen at good conferences).  I do think that unilateral discussions of controversies can serve a useful purpose.  However, one-sided presentations on the internet are not all peaches and cream, as you no doubt know. 

(A mildly amusing note:  My previous post got a big spike in pageviews thanks to Physics Today tweeting the link.  Thanks, PT!  Hopefully some of those people will stick around.  Of course, my most-viewed post of all time, by about a factor of 3, is still my commentary about whiskey stones.  Clearly I should routinely stake out an aggressive position on some physics point connected to good Scotch.)