Quoting Baltimoron: Osmium, you're not seriously arguing that Behe's oeuvre is just on the level of one of these bad papers that muck up the total volume of scientific papers, are you? I think even the most earnest student's effort would exceed the value of Behe's hackery by orders of magnitude.

They're not student papers, but regular papers. Often paper-writing duties are shared between a student-advisor pair, so you could consider most papers as student papers on some level. (In the journal reading club, they were *selected* by students.) They don't muck up the literature, because they are the foundation of it. Thousands of people working, out of which good ideas will nucleate. But you should view every effort with scepticism. Behe is BS (and doesn't have much of an oeuvre), but the mechanism that weeds through the "sloppy-bad" will work just as well on the "bad-bad." Having your meme die out, the death of your idea, happens when you get ignored.

Quoting Baltimoron: And, was Bob Wright any less slippery during his reading, or did the audience challenge him like John Horgan?

Bob was great, and Bruce Feiler and he sparred on a few things, like the overall importance of Philo in

The "View Diavlog" link is broken. Click here to watch. Note to admins: It appears as though every Apollo diavlog has this problem -- the apollo. part of the URL associated with the "View Diavlog" link is always missing; e.g., here, it points to http://bloggingheads.tv/diavlogs/23185 when it should point to http://apollo.bloggingheads.tv/diavlogs/23185 A bug in the CMS or some other automated procedure that creates the thread page for diavlogs, perhaps?

A few notes: -First of all, this was very good. I like the 'sci sat' take on Apollo Project. Worthy of the front page... -Which makes me think: by being so good these two further make me realize just how arbitrary the distinction between "real" 'heads and "lowly apollo project" heads is. These guys were just as as good as any number of "real participants": they're articulate, and fluid. And from what I gather, their credentials are no less worthy than any number of people who have appeared on the front page. So the distinction between "real" and "apollo" heads has never appeared more arbitrary, elitist, and stupid. -Likewise, the cloaking is now completely pointless. Either do enough masking so that the people are essentially unrecognizable, or be rid of it altogether. These two look almost normal . . . well, in terms of the lack of masking -If apollo project is for commenters, why have two of the six participants not been commenters at all?

And how is it possible that these diavlogs aren't longer? May I ask? This was great. Josh and Ian did a great job. I wanted to know more about their work. Blood batteries? Talking about vampires...! In terms of the scientific literature, journal clubs are pretty common during training. I think that the critical question "what's wrong with this paper?" must precede the final one "what did we learn from it?", but both are necessary. Unfortunately, once you are out of academia, it becomes much more difficult, if not impossible, to keep up with that exercise. I agree that the boundaries between the some of the Apollo and some of the regular diavlogs aren't clear in terms of quality of the presentations or the diavloggers. Please come back with a follow up on this one.

Wow. This is one of the best diavlogs I've watched ever. Many thanks to both Josh and Ian. A couple of quick points. I disagree completely with the Osmium's dismissal of bad papers. All papers are bad, in the sense that they represent a particular stage of knowledge, each stage being necessarily imperfect. I take his general point that some papers push the envelope while others simply rehash stuff or have no simple clear point. My major problem with his general 'hard-ass' approach, is that research and research writing is an uneven exercise at best. Perhaps I'm getting the argument wrong, but why should failures be less illuminating than successes? Absolutely outstanding. Thanks and kudos to both. I'd love to see Osmium host a regular series on this sort of topic.

Quoting kidneystones:
[...]

Perhaps I'm getting the argument wrong, but why should failures be less illuminating than successes?

[...]

From what Ive read, journals having a bias against publishing negative findings is a major problem (Though I wonder if the online depositories have helped alleviate this problem). Imagine how much time and money is wasted replicating experiments needlessly that have already been done several times already to show there is no correlation between X and Y, because people have no idea said experiment has already been carried out. I think when they mean bad papers, they are not talking about negative findings, so much as papers with flawed methodologies or papers that are opaque.

What kind of Characteristics would these batteries have?

Quoting Starwatcher162536: From what Ive read, journals having a bias against publishing negative findings is a major problem

I don't know how this would work. Would a paper describe all/of some the experiments that the authors tried in attempting to get a specific result? Good papers (good = a clear and complete description of an experiment/set of experiments, and a [reasonably] complete analysis of the results) present the context of the work/method/experimental set up, specifically why a method etc. was used. With a working knowledge of a field you should be able to glean what would/did not work. Writing about/analyzing what didn't work seems, at minimum, very difficult. At least I think this is true for the work that I ve done-- maybe this would suit some fields much better than others, but, to me, a thorough analysis of 'what doesn't work' seems daunting.

Small, very small-- just two little fibers coated with a film containing a few enzymes. It's possible to generate enough power from the oxidation of glucose to transmit a signal (~0.6V or there about), so in principle you could remotely, say, monitor your glucose level, or changes in blood flow-- maybe anticipate an embolism.

Quoting irw: Small, very small-- just two little fibers coated with a film containing a few enzymes. It's possible to generate enough power from the oxidation of glucose to transmit a signal (~0.6V or there about), so in principle you could remotely, say, monitor your glucose level, or changes in blood flow-- maybe anticipate an embolism.

Wow! That sounds so interesting! What stage of development is this technology at?

Quoting irw: Small, very small-- just two little fibers coated with a film containing a few enzymes. It's possible to generate enough power from the oxidation of glucose to transmit a signal (~0.6V or there about), so in principle you could remotely, say, monitor your glucose level, or changes in blood flow-- maybe anticipate an embolism.

Wow, cool. When might we see an artificial pancreas for diabetics? (Kidding... mostly.)

-First of all, this was very good. I like the 'sci sat' take on Apollo Project. Worthy of the front page...

-Which makes me think: by being so good these two further make me realize just how arbitrary the distinction between "real" 'heads and "lowly apollo project" heads is. These guys were just as as good as any number of "real participants": they're articulate, and fluid. And from what I gather, their credentials are no less worthy than any number of people who have appeared on the front page. So the distinction between "real" and "apollo" heads has never appeared more arbitrary, elitist, and stupid.

-Likewise, the cloaking is now completely pointless. Either do enough masking so that the people are essentially unrecognizable, or be rid of it altogether. These two look almost normal . . . well, in terms of the lack of masking .

Good points, all! Who will be the first "'head" to take on the real challenge of the Apollo project? Mickey? Bob?

And how is it possible that these diavlogs aren't longer? May I ask?

Arguably, the 20 minute limit frustrates organization, because it just doesn't seem long enough to discuss multiple topics in that period of time. But then, two able interlocutors could turn the Apollo format into the novella equivalent of the elite diavlogs' novel format.

Quoting Baltimoron: ... the novella equivalent of the elite diavlogs' novel format...

Very creative way of looking at it, but I see it more like the headlines format, especially when it's a topic I'm interested in.

irw writes...[...] Thanks for the reply. First, I support the exercise in analysis in a group setting. My question is whether we learn only from having our expectations met. I do almost all my own reading in the social sciences, but I do some history of science. Isn't it true that scientists often/sometimes get results that don't support a hypothesis? Or that suggest/confirm a working hypothesis has fundamental flaws? Darwin spent an immense amount of time testing the solubility of seeds. The result? He discovered that the dispersal of seeds in sea-currents could not explain the spread of plants. His theory of pangenisis turned out to be wrong. I'm sure you could cite hundreds of similar examples. My own work now calls into question some canonical assumptions. There must necessarily be flaws in this exercise. Indeed, if we're trying to push the envelope, we're perhaps going to be more wrong than we are right. It seems likely (to me) that you and Osmium are principally trying to establish which papers are clear, simple and have a single solid point. That's a worthy exercise. However, I'm arguing that there's often as much to be learned from poorly-constructed papers

Quoting mvantony: Nice job, Josh and Ian. Aside from the length of the diavlog and the (to me, silly) masking, it was indistinguishable from a "real" diavlog.

There should be a poll asking whether people think the length should be allowed to be longer if desired. I think most people would opt for the option of longer discussions if there was enough to talk about, both on the talkers side and the viewers side.

I like the short talks, even for the "professionals". I agree with Mvanthony that the masking is a mistake. I didn't like it when it was discussed abstractly, and I like it even less now that I've seen the results: annoying and distracting.

Quoting kidneystones: irw writes...[...]
Isn't it true that scientists often/sometimes get results that don't support a hypothesis?

Yes, absolutely. In my experience discussion of this kind of idea(s) occurs at the research group level or at conferences or written in editorial/perspectives. I don't know how this would occur in a paper; it shouldn't. Or at least it should be limited to a sentence or two in the discussion. A paper should make a short little point and provide sufficient evidence/analysis to support that point. A paper with a statement like "I discovered that the dispersal of seeds in sea-currents could not explain the spread of plant life in this location" would get a strong D from Osmium. And, it should; it needs an ending. It needs a point. It would be much easier (if much easier is possible) to add to/generate grand ideas if each paper reported an accurate little point. I think sifting through the chaff makes it harder. Does that make any sense, or am I still missing. Thanks for you comments etc, much appreciated

irw writes....[...] All right. You and osmium are arguing then, if I understand you, for a particular style of paper-writing that makes depends on researchers being thoroughly up-to-date on all published material in their area of expertise in all languages, I presume, before the piece is submitted for publication. I generally agree, btw, with your approach to research and to publishing. The argument can be made that researchers stand a better chance of winning support and funding by writing in the style you advocate. The problem, however, is that awarding a 'D' to any research paper, even with caveats isn't likely to encourage debate or further inquiry. Cheers!

Quoting irw: A paper with a statement like "I discovered that the dispersal of seeds in sea-currents could not explain the spread of plant life in this location" would get a strong D from Osmium. And, it should; it needs an ending. It needs a point.

Hm, just as I was reading that, I thought maybe that sounded like a good topic. Perhaps just because I know nothing about seed dispersal. A negative result might be good, provided it is large and important enough. However, something like "I discovered that compound X cannot be prepared in an aqueous solution," is definitely not acceptable for a paper topic. Rather, that is information you keep in your memory banks until you can write something BIGGER about compound X. Then you report it as a simple statement in the experimental section. I think I collect small negative results and then use them as spice to sprinkle in a paper.

Quoting kidneystones: irw writes....[...]

All right. You and osmium are arguing then, if I understand you, for a particular style of paper-writing that makes depends on researchers being thoroughly up-to-date on all published material in their area of expertise in all languages, I presume, before the piece is submitted for publication.

Kind of, yes. But that's one of the points of peer-review: if there is a major piece of background literature you're missing, a reviewer should tell you you have to include it and consider it. That's one of the checks/balances that you are up-to-date. The language thing can be difficult, but generally only because the journal is completely unavailable to you, like because it's in Japanese, etc. In that case, if it has an important result, it will eventually be covered in a literature review article in English, and you will get its point from there. So that adds a couple years to the meme's time constant for spreading far and wide, but it still will. People will compete for being the first person to find an important, obscure paper. Doubly so if it's in a far-out language. Fun fact: If you can get ahold of the journal that a

Quoting osmium: Hm, just as I was reading that, I thought maybe that sounded like a good topic. Perhaps just because I know nothing about seed dispersal. A negative result might be good, provided it is large and important enough.

As Messrs. Michelson and Morley will happily attest. It's hard to say where to draw the line between "important enough" and not, but I think in the abstract that a careful experiment providing strong evidence that, say, "X won't happen in these circumstances" is of potential use to other workers, if for no other reason than to save them some time. As to whether it counts as of sufficient worth to be included in a given journal should obviously be left up to that journal's editors, but I would like to see a more open attitude, in general, to the publishing of negative results. Seems like it might be especially useful in the pharmaceutical realm, come to think of it.

Quoting bjkeefe: As Messrs. Michelson and Morley will happily attest.

It's hard to say where to draw the line between "important enough" and not, but I think in the abstract that a careful experiment providing strong evidence that, say, "X won't happen in these circumstances" is of potential use to other workers, if for no other reason than to save them some time. As to whether it counts as of sufficient worth to be included in a given journal should obviously be left up to that journal's editors, but I would like to see a more open attitude, in general, to the publishing of negative results.

Seems like it might be especially useful in the pharmaceutical realm, come to think of it.

Yes, Michelson and Morley is an excellent example. A sense of proportion is important, and like you say, there's no rule--you just have to feel it. An editor is one check, but pride is another. You don't want people to start thinking "That guy publishes dumb stuff." So you ride the line, figure out what's appropriate and what's not.

Quoting osmium: Yes, Michelson and Morley is an excellent example.

A sense of proportion is important, and like you say, there's no rule--you just have to feel it.

An editor is one check, but pride is another. You don't want people to start thinking "That guy publishes dumb stuff." So you ride the line, figure out what's appropriate and what's not.

That bolded part is part what I had in mind when I wished last post for a more open attitude about publishing negative results. Another thing I should have added is that I think such a change in attitude might improve the quality of research overall -- it seems to me that those just starting out in a given field, especially when they are under publish-or-perish pressure, might feel less inclined to tweak their results towards some desired "positive" end if they had more confidence that a carefully performed experiment and write-up with "negative" results would be more likely to be viewed by the community as worthwhile (publishable). Maybe I'm being a little dreamy-eyed about all this, but it seems to me that much of what comprises scientific progress is the steady accumulation of

Quoting bjkeefe: Not to make your earlier example ("compound X cannot be prepared in an aqueous solution") have to stand for too much, but imagine if instead of having to {(1) search the literature to see if it could be, (2) fail to find a reported positive result, (3) conduct the experiment yourself to find out for sure}, you could have instead just looked it up. Maybe this is too artificial an example, but do you see what I'm getting at? It ought to be easier to build a body of work following Mr. Holmes's dictum, and not have to wait until you had a truly monumental result, for "negative" findings to be considered useful to others.

Reports about making compound X will tell people how it's been done in the past, and the information about how it CAN'T will likely be there if it has been tried, as a sentence rather than a full topic. Anyone making X should have done their homework, otherwise they will waste time. But I could imagine a different system than the current one, where it's more like a wiki than like papers. Now each paper is supposed to be a fully-digestible unit, so they

BTW, hey IRW. I remember we had a question exactly like this making the gel. I just clicked the link for our paper to see if we wrote it in there. Here is the osmium attachment to the protein:

Synthesis of Os-HSH. The osmium complex [Os(bpy)2Cl2]Cl was synthesized from K2OsCl6 and bipyridine (Fisher Chemical) as described in ref. 15. Purified HSH (20 mg·ml−1) in deionized (DI) water was mixed with 10 mg·ml−1 [Os(bpy)2Cl2]Cl (1:20 molar ratio) and the reaction vessel was immersed in heating oil held to 70°C. The reaction was allowed to proceed for 6–18 h under argon atmosphere and with constant stirring. The synthesis product was filtered over a 10-kDa cellulose filter with 500 times the reaction volume (1 liter) each of DI water and PBS (50 mM NaiP, 500 mM NaCl). The product was lyophilized after 3 days of dialysis against DI water.

I recall we also tried this in buffer solution (PBS) instead of deionized water, and it didn't work. Didn't recall if we wrote that into it, and I guess we didn't. Do you feel bad for people that we didn't report that, or do you think they should have to figure it out? I mean, is that important info?

Quoting osmium: Reports about making compound X will tell people how it's been done in the past, and the information about how it CAN'T will likely be there if it has been tried, as a sentence rather than a full topic. Anyone making X should have done their homework, otherwise they will waste time.

I can see that, sure. (I did think that trying to stretch your example too much would break it.)

But I could imagine a different system than the current one, where it's more like a wiki than like papers. [...]

I had that thought in mind, too. I guess I'm old-fashioned enough to want a bit more of a formal review process -- even for what might amount to no more than an added sentence -- than the free-for-all that wikis can permit. On the other hand, I continue to be amazed at how good Wikipedia is, almost all the time, compared to my initial expectations for it, so I could see this approach working for incremental scientific results.

Even then, though, don't doubt for a second that everyone will notice if you spend a couple years trying to make a career on little ditties like that, without slogging out in the

Quoting osmium: Do you feel bad for people that we didn't report that, or do you think they should have to figure it out? I mean, is that important info?

No. It doesn't belong in the paper. At least not without much more explanation of why it didn't work. If there was a Wiki it could go there. My point is that negative result don't match with what a paper should be-- i.e. evidence supporting a hypothesis. If we go about saying things like 'we think mixing X and Y will not make Z', and the go about showing, in detail, why that's true, then I m on board with negative results. If instead we start making statements like 'we think mixing X and Y will make Z' and btw mixing X and A, or B, or C or D will not make Z' and don't go about showing why they don't work then I m not on-board. Stick that in your wiki/conference/research group disscusion etc. Those kinds of statements --the X and A or B or C will not make Z-- might be very useful to someone, and might even eliminate some reproduction of others results, but they don't belong in a paper with sufficient supporting evidence. Or at least that's

osmium: "uh, and, uh the uhhh.." in head: "i can't hear myself, what am i saying, is this what it's like to be deaf, why can't i hear??" moral: leave one ear uncovered

Quoting bjkeefe: The "View Diavlog" link is broken.

Known bug, unfixable at this time. Thanks for noting it.

Quoting Brenda: Known bug, unfixable at this time. Thanks for noting it.

Okay. Thanks for acknowledging the report.