I continue to have doubts about whether to blog. As I explained in February, my main doubt is whether this will accumulate: A newspaper or magazine article will be read by thousands, while one is lucky if ten people read an academic paper. So if you want influence, write for the popular media, right? Well, those thousands of newspaper readers will soon forget everything you said, but a few academic readers may well write papers influenced by your paper. People almost never look up ten year old newspaper columns, but they do often read ten year old academic papers. So an academic paper may still have a better chance at long term influence than a newspaper column.
Blogging can be a part of the "process of accumulation" in several ways:
1. Bloggers document history in making which can be invaluable for researchers who later try to make sense of an happening or a period of time.
2. Insights from blogosphere, at least in some areas, coalesce into an accepted body of wisdom fairly quickly. For example, the debate about desirability of DRM. Such debates could influence outcomes.
3. Debates about emotional topics, like existence of God or abortion rights, would likely make people veer to rational viewpoints, as progressively younger population see viewpoints otherwise disapproved by their religion or family.
4. More nuanced understanding of phenomena like terrorism, that affect large populations whose opinions drive the way their governments or other institutions act.
Together with easily accessible and constantly evolving wiki content or social booking marking, it could well be the most powerful process of accumulation ever seen in the history of mankind.
As a reader, you must realize that I am biased towards hoping you continue to blog. That said, here is my response.
1. Do you want credit for your ideas? Academic papers will assure that your influence will be traced back to you. A blog may inspire many people (Some of whom may even write academic papers), and your insights may spread without giving you credit (I think this occured with respect to you and prediction markets already). If you want certain credit from future scholars in your field then you should concentrate more on academic papers. If you want to maximize your long term influence then a blog, though higher risk, is probably the better way to go about things. I say it is higher risk because there is a large track record of academic papers influencing others. Blogs haven't been around, so they don't have the track record. Therefore, even if many ideas have been generated by blogs, there haven't been many major changes which are traced back to an original blog post.
2. As other people said, it shouldn't be "or". You should look into even more ways to influence the future. Books are also read long after they are published. In terms of spreading long term ideas, there are people who read books who don't read blogs or academic papers. If you get the ideas spread widely enough then you are more likely to cause a long term change. You can start writing your blog in a way such that it will be easy for you to incorporate ideas from blog posts into a future book. The blog could also help create enough of a stir around the book such that it would be more likely to be considered important 10 years from now. That said, if you stopped blogging, you have enough friends who do blog that the advertising point isn't too relevant.
As a regular reader, I have found your posts help me to clarify my thinking habits and encourage modularity-- benefits that have immense real-life utility apart from the academic. I would request you to continue sharing your insights with readers.
"So the fact that each blogger and reader today feels like he is slowly gaining insight does not mean we are part of a process by which humanity accumulates insight. We could just each be relearning and re-expressing what many of our ancestors knew."
The question isn't are we advancing humanity by constantly learning new insights as individuals. The question is what happens if we stop. It's like bailing out a leaking boat, no matter how much water you bail it's going to keep on leaking, but if you ever stop you sink.
At the end of the day it's more important to teach than to discover. Humanity could probably survive and maybe even prosper if academic progress suddenly ceased. However the consequences of us failing to communicate what we already know are simply catastrophic. We need people to dispel knowledge just as much as we need people to discover knowledge.
The question is which group you want to be a part of, while the boat metaphor does point out the futility in teaching I would also point out that for every person who helps bail the boat does float higher.
Academic papers largely influence ... other academics. In a cycle that often seems set up to do little else than impress one another within a insular niche. Ok, I'm exaggerating the skeptic's side of the coin, but it does seem rare that an academic paper has a truly large impact on the wider world (though I'm sure inevitably infuence is a power law). If you want to influence the world, talk to the world. Blogging is one route. Also, I agree with the comment by Eliezer Yudkowsky that blog posts are not like newspaper articles. The best posts will be continually rediscovered through links and search engines. You've created one of the most vibrant and participatory blogs I've seen covering a serious subject: many kudos. One example of your breadth of reach: I noticed that tech celeb Marc Andreessen has you listed among his "interesting writers": see http://blog.pmarca.com/ .
Selfishly, I want u to keep up the blogging, because its so much fun for me.
If u r interested in helping humanity, blogging is a new medium of publishing, and may turn out to accumulate, or u could make it so that it would accumulate, on which I have some suggestions for how to change blogs so they will be more likely to accumulate, which I will tell u if u r interested.
But safety is in the old method of publishing, which we know has led to accumulations.
Ur post only mentions one personal aspect of u, which is that blogging is fun for u, and it is of great importance to do what u enjoy as ur work if u possibly can, as life is short, and is getting shorter all the time for each of us, and u only get 1.
If u want to stay an associate professor, tenure is no longer enough, because of tenure review policies that can take tenure away, so u need to find out from ur colleagues what u need to do to keep tenure, if that is a goal of urs.
U seem to have come up with one really big idea, prediction markets, which in a world where there r few big ideas is a big advantage to u.
If u decide to publish, getting collaborators is very helpful, I have found, as it helps in so many ways.
I'd advise u to ask ur colleagues what u should do.
I'd like to see more of the thoughts of interesting thinkers captured and made searchable and available on the internet. Perhaps by making conferences, seminars, and classes universally recorded, tanscribed, and put onto the internet. Also some sort of process where individual thinkers could record their daily life conversations, have them efficiently transcribed, screened for personal information, and then put on the internet.
Not to criticise (you already know I am totally in favour of you continuing to blog), but what are you trying to acheive in this post? Surely only those that are regular readers of this blog, and thus admire your work, will respond?Some will furnish you with extra arguments, to be sure, but that falls prey to selection bias. I really want you to continue blogging, but if you want to answer your question honestly, then we're not the people to ask.
I'm currently writing an economics textbook. I have found blog posts written by professors to be one of my best sources of material. Because such posts are often designed to be read by non-academics they are more understandable and interesting than academic articles. I strongly suspect that other textbook authors have also realized the benefits of mining professor blogs. And through textbooks, many professor blog posts will join the process of knowledge accumulation.
I think the archival methods for webbed content are already good enough that knowledge strongly tends to accumulate, not get lost. Most of it in effect gets ignored, and currently the mostly-ignored material isn't archived terribly aggressively. But the fraction of things that people actually care about seems to be archived (and repeated and re-summarized and so forth) reasonably well. The indexing and citation technology isn't great, but it's serviceable already, and I'd expect natural engineering improvements will tend to make it better.
I have a stronger impression about this in software-related fields than in economics-related fields. Partly it's just because I know more about software, but partly it's just because there seems to be more history to judge from. The Internet has been an important channel for communicating information related to software for a long time. Net distribution of some things like commercial documentation have only taken off more recently, but software academic preprints and materials related to free software projects like Linux have been distributed on the net since at least the early 1990s. It doesn't seem to me that information gets lost except to the extent that people feel that it's irrelevant. I go to the the UT Dallas library often (most recently about two hours ago...) and my unscientific impression the computer books dated 1995 are at least as diminished as net-distributed information of the same age.
One caveat is that small pieces of information don't necessarily get formal credit even when they made a difference. Even when we know how to do it, friction may keep it from being done, or from being done quite right. I once spent many hours rearranging the acknowledgements for other people's work in a sizable free software system into a single summary file, and near the head of the file is my apology for undoubtedly making oversights and errors. And I once pointed out an error in a book's proof of a complexity bound, and when the author later published a paper containing a correct proof, there was no hint that someone else had pointed out the error in the book. That wasn't because we haven't figured out how to acknowledge things like that, it just didn't get done.
On the plus side, small pieces of information also take less time to create. I would guess that you can fire off several dozen reasonably thoughtful web posts or substantive mailing list messages in the time it takes to prepare a paper. Maybe this is a psychological minus: if you spend four dozen hours writing two dozen thoughtful web posts, and twenty two of them fizzle, it might seem less influential than spending four dozen hours on a politely received conference paper. But I'm not sure it is.
My tentative impression is that to the extent that getting your ideas out there is your goal, it's actually worth spending time on both the short stuff and the long stuff. I get that impression especially from technical discussions in software (where the short stuff is typically on a mailing list, not a weblog): someone who only writes publication-quality preprints seems to be less inflential than someone who only writes short stuff, but there are various important things which can only be expressed in longer and more careful publications, so to be really influential you should probably write both.
To the extent that academic credit and reputation are the goal, I'm not sure in general, but I would guess that at least when an idea happens to become really successful, people will often take the trouble to chase back to the origins even if they're humble un-peer-reviewed short writings. People show some interest in finding the origins of even things which predate the Internet, and timestamped digital text makes things a lot easier to hunt and check.
(And, regardless of whether my guesses on communication are sound, I hope you find *some* satisfactory way of communicating your ideas, ideally not just your big ideas but also things like the "I cover counterarguments that most teachers don't cover..." text from your tenure statement.)
"People almost never look up ten year old newspaper columns, but they do often read ten year old academic papers. So an academic paper may still have a better chance at long term influence than a newspaper column."
But people never look up ten year old lunchtime conversations either. And yet I am pretty sure that you are professionally more productive because you work at a place where you can have informal conversations - conversations which are never recorded in any way that can be looked up ten years later. I doubt you would be as productive if you lived alone on a mountaintop with no contact with anyone else other than by reading and writing journal articles. And in turn, I doubt you would advance human knowledge as rapidly.
So, that something will not be looked at ten years later, does not mean that it does not contribute to the advance of human knowledge.
"If I'm mainly the equivalent of a newspaper columnist, rather than a part of a community of modular thinkers, this is to me a waste."
I would say that the primary culprit in the failure of many newspaper columnists to contribute to the advance of human knowledge is not the format of the newspaper column, but the content that the columnist supplies. To put it bluntly, many newspaper columns are insignificant. They are insignificant as columns and they would remain insignificant in any other format.
"Yes, blog posts can influence people, but the question is whether that influence feeds back much into more related intellectual insights that accumulate over the long run."
Robin,
I guess much depends here on whether you're only interested in academic, intellectual influence, or in broader cultural influence. For accumulating hard knowledge and theory, blogging might be a waste of time, because the 5 people in the world who might be interested in adding to your work probably don't read your blog. They read papers. But, if your blog readership is sufficiently intelligent, your blog is a great idea-generating and testing place.
For influencing cultural change, blogs are comparable to newspaper columns -- outside the academic world, people really do form their opinions partly based on the points they hear expressed in the media. Many bloggers are more interested in this function, the influencing of public opinion, than in adding to the collective knowledge base of humanity. This is not such a bad goal -- social change happens partly because of academic insights into human behavior, and partly because of public sea changes -- it seems to take both.
A key difference is that academics organize into a network of specialists with social norms requiring them to cite related previous work, and to situate their publications so that they can be found by others working on similar topics.
The link is the killer app of the web, including blogs. Those links are exactly what allow people using search engines to find your work, no matter what it's about. If the key power of academic works vs popular media rested in the citation, then some of that has already bled over onto the web, blogs included.
So by virtue of being on the web, where you can link and be linked to, you already have a distinctly important advantage over newsprint - one that just happens to resemble the same advantage academic papers have over newspaper columns.
After all, you're not trying to weigh a decision of whether or not you should abandon your newspaper column, are you? How much do you think the relative costs of writing newspaper columns vs blogging has to do with that decision, in contrast to the usefulness of citation in the form of linking?
If the influence of academic papers vis-avis hypermedia is a long term process, then isn't it a little early to make a definitive judgment on the usefulness of blogging?
In the meantime, work on improving the usefulness of links continues (semantic web, tagging architectures, http://microformats.org/ microformats]], social networking, etc.)
Unless blogging is useful to you in vetting topics that you convert into academic papers, you should stop; the blog is not otherwise advancing the field's collective push. Tyler Cowen, Bryan Caplan, et al will presumably read your academic papers. But if you are personally unsanitary or unsavory and cannot knock around ideas with your colleagues at Cafe Strada, perhaps you should keep blogging to achieve the aforementioned instrumental benefits.
While disintermediation has its benefits, I gather that those benefits do not synch with your goals.
Blogging can be a part of the "process of accumulation" in several ways:
1. Bloggers document history in making which can be invaluable for researchers who later try to make sense of an happening or a period of time.
2. Insights from blogosphere, at least in some areas, coalesce into an accepted body of wisdom fairly quickly. For example, the debate about desirability of DRM. Such debates could influence outcomes.
3. Debates about emotional topics, like existence of God or abortion rights, would likely make people veer to rational viewpoints, as progressively younger population see viewpoints otherwise disapproved by their religion or family.
4. More nuanced understanding of phenomena like terrorism, that affect large populations whose opinions drive the way their governments or other institutions act.
Together with easily accessible and constantly evolving wiki content or social booking marking, it could well be the most powerful process of accumulation ever seen in the history of mankind.
As a reader, you must realize that I am biased towards hoping you continue to blog. That said, here is my response.
1. Do you want credit for your ideas? Academic papers will assure that your influence will be traced back to you. A blog may inspire many people (Some of whom may even write academic papers), and your insights may spread without giving you credit (I think this occured with respect to you and prediction markets already). If you want certain credit from future scholars in your field then you should concentrate more on academic papers. If you want to maximize your long term influence then a blog, though higher risk, is probably the better way to go about things. I say it is higher risk because there is a large track record of academic papers influencing others. Blogs haven't been around, so they don't have the track record. Therefore, even if many ideas have been generated by blogs, there haven't been many major changes which are traced back to an original blog post.
2. As other people said, it shouldn't be "or". You should look into even more ways to influence the future. Books are also read long after they are published. In terms of spreading long term ideas, there are people who read books who don't read blogs or academic papers. If you get the ideas spread widely enough then you are more likely to cause a long term change. You can start writing your blog in a way such that it will be easy for you to incorporate ideas from blog posts into a future book. The blog could also help create enough of a stir around the book such that it would be more likely to be considered important 10 years from now. That said, if you stopped blogging, you have enough friends who do blog that the advertising point isn't too relevant.
As a regular reader, I have found your posts help me to clarify my thinking habits and encourage modularity-- benefits that have immense real-life utility apart from the academic. I would request you to continue sharing your insights with readers.
"So the fact that each blogger and reader today feels like he is slowly gaining insight does not mean we are part of a process by which humanity accumulates insight. We could just each be relearning and re-expressing what many of our ancestors knew."
The question isn't are we advancing humanity by constantly learning new insights as individuals. The question is what happens if we stop. It's like bailing out a leaking boat, no matter how much water you bail it's going to keep on leaking, but if you ever stop you sink.
At the end of the day it's more important to teach than to discover. Humanity could probably survive and maybe even prosper if academic progress suddenly ceased. However the consequences of us failing to communicate what we already know are simply catastrophic. We need people to dispel knowledge just as much as we need people to discover knowledge.
The question is which group you want to be a part of, while the boat metaphor does point out the futility in teaching I would also point out that for every person who helps bail the boat does float higher.
Academic papers largely influence ... other academics. In a cycle that often seems set up to do little else than impress one another within a insular niche. Ok, I'm exaggerating the skeptic's side of the coin, but it does seem rare that an academic paper has a truly large impact on the wider world (though I'm sure inevitably infuence is a power law). If you want to influence the world, talk to the world. Blogging is one route. Also, I agree with the comment by Eliezer Yudkowsky that blog posts are not like newspaper articles. The best posts will be continually rediscovered through links and search engines. You've created one of the most vibrant and participatory blogs I've seen covering a serious subject: many kudos. One example of your breadth of reach: I noticed that tech celeb Marc Andreessen has you listed among his "interesting writers": see http://blog.pmarca.com/ .
Definitely one day (in the near future) blog will become a serious source for the academic.
Selfishly, I want u to keep up the blogging, because its so much fun for me.
If u r interested in helping humanity, blogging is a new medium of publishing, and may turn out to accumulate, or u could make it so that it would accumulate, on which I have some suggestions for how to change blogs so they will be more likely to accumulate, which I will tell u if u r interested.
But safety is in the old method of publishing, which we know has led to accumulations.
Ur post only mentions one personal aspect of u, which is that blogging is fun for u, and it is of great importance to do what u enjoy as ur work if u possibly can, as life is short, and is getting shorter all the time for each of us, and u only get 1.
If u want to stay an associate professor, tenure is no longer enough, because of tenure review policies that can take tenure away, so u need to find out from ur colleagues what u need to do to keep tenure, if that is a goal of urs.
U seem to have come up with one really big idea, prediction markets, which in a world where there r few big ideas is a big advantage to u.
If u decide to publish, getting collaborators is very helpful, I have found, as it helps in so many ways.
I'd advise u to ask ur colleagues what u should do.
I'd like to see more of the thoughts of interesting thinkers captured and made searchable and available on the internet. Perhaps by making conferences, seminars, and classes universally recorded, tanscribed, and put onto the internet. Also some sort of process where individual thinkers could record their daily life conversations, have them efficiently transcribed, screened for personal information, and then put on the internet.
Robin,
YES what your write on this blog feeds back into intellectual insights that accumulate over the long run.
Dear Robin,
Not to criticise (you already know I am totally in favour of you continuing to blog), but what are you trying to acheive in this post? Surely only those that are regular readers of this blog, and thus admire your work, will respond?Some will furnish you with extra arguments, to be sure, but that falls prey to selection bias. I really want you to continue blogging, but if you want to answer your question honestly, then we're not the people to ask.
Robin,
I'm currently writing an economics textbook. I have found blog posts written by professors to be one of my best sources of material. Because such posts are often designed to be read by non-academics they are more understandable and interesting than academic articles. I strongly suspect that other textbook authors have also realized the benefits of mining professor blogs. And through textbooks, many professor blog posts will join the process of knowledge accumulation.
I think the archival methods for webbed content are already good enough that knowledge strongly tends to accumulate, not get lost. Most of it in effect gets ignored, and currently the mostly-ignored material isn't archived terribly aggressively. But the fraction of things that people actually care about seems to be archived (and repeated and re-summarized and so forth) reasonably well. The indexing and citation technology isn't great, but it's serviceable already, and I'd expect natural engineering improvements will tend to make it better.
I have a stronger impression about this in software-related fields than in economics-related fields. Partly it's just because I know more about software, but partly it's just because there seems to be more history to judge from. The Internet has been an important channel for communicating information related to software for a long time. Net distribution of some things like commercial documentation have only taken off more recently, but software academic preprints and materials related to free software projects like Linux have been distributed on the net since at least the early 1990s. It doesn't seem to me that information gets lost except to the extent that people feel that it's irrelevant. I go to the the UT Dallas library often (most recently about two hours ago...) and my unscientific impression the computer books dated 1995 are at least as diminished as net-distributed information of the same age.
One caveat is that small pieces of information don't necessarily get formal credit even when they made a difference. Even when we know how to do it, friction may keep it from being done, or from being done quite right. I once spent many hours rearranging the acknowledgements for other people's work in a sizable free software system into a single summary file, and near the head of the file is my apology for undoubtedly making oversights and errors. And I once pointed out an error in a book's proof of a complexity bound, and when the author later published a paper containing a correct proof, there was no hint that someone else had pointed out the error in the book. That wasn't because we haven't figured out how to acknowledge things like that, it just didn't get done.
On the plus side, small pieces of information also take less time to create. I would guess that you can fire off several dozen reasonably thoughtful web posts or substantive mailing list messages in the time it takes to prepare a paper. Maybe this is a psychological minus: if you spend four dozen hours writing two dozen thoughtful web posts, and twenty two of them fizzle, it might seem less influential than spending four dozen hours on a politely received conference paper. But I'm not sure it is.
My tentative impression is that to the extent that getting your ideas out there is your goal, it's actually worth spending time on both the short stuff and the long stuff. I get that impression especially from technical discussions in software (where the short stuff is typically on a mailing list, not a weblog): someone who only writes publication-quality preprints seems to be less inflential than someone who only writes short stuff, but there are various important things which can only be expressed in longer and more careful publications, so to be really influential you should probably write both.
To the extent that academic credit and reputation are the goal, I'm not sure in general, but I would guess that at least when an idea happens to become really successful, people will often take the trouble to chase back to the origins even if they're humble un-peer-reviewed short writings. People show some interest in finding the origins of even things which predate the Internet, and timestamped digital text makes things a lot easier to hunt and check.
(And, regardless of whether my guesses on communication are sound, I hope you find *some* satisfactory way of communicating your ideas, ideally not just your big ideas but also things like the "I cover counterarguments that most teachers don't cover..." text from your tenure statement.)
"People almost never look up ten year old newspaper columns, but they do often read ten year old academic papers. So an academic paper may still have a better chance at long term influence than a newspaper column."
But people never look up ten year old lunchtime conversations either. And yet I am pretty sure that you are professionally more productive because you work at a place where you can have informal conversations - conversations which are never recorded in any way that can be looked up ten years later. I doubt you would be as productive if you lived alone on a mountaintop with no contact with anyone else other than by reading and writing journal articles. And in turn, I doubt you would advance human knowledge as rapidly.
So, that something will not be looked at ten years later, does not mean that it does not contribute to the advance of human knowledge.
"If I'm mainly the equivalent of a newspaper columnist, rather than a part of a community of modular thinkers, this is to me a waste."
I would say that the primary culprit in the failure of many newspaper columnists to contribute to the advance of human knowledge is not the format of the newspaper column, but the content that the columnist supplies. To put it bluntly, many newspaper columns are insignificant. They are insignificant as columns and they would remain insignificant in any other format.
"Yes, blog posts can influence people, but the question is whether that influence feeds back much into more related intellectual insights that accumulate over the long run."
Robin,
I guess much depends here on whether you're only interested in academic, intellectual influence, or in broader cultural influence. For accumulating hard knowledge and theory, blogging might be a waste of time, because the 5 people in the world who might be interested in adding to your work probably don't read your blog. They read papers. But, if your blog readership is sufficiently intelligent, your blog is a great idea-generating and testing place.
For influencing cultural change, blogs are comparable to newspaper columns -- outside the academic world, people really do form their opinions partly based on the points they hear expressed in the media. Many bloggers are more interested in this function, the influencing of public opinion, than in adding to the collective knowledge base of humanity. This is not such a bad goal -- social change happens partly because of academic insights into human behavior, and partly because of public sea changes -- it seems to take both.
A key difference is that academics organize into a network of specialists with social norms requiring them to cite related previous work, and to situate their publications so that they can be found by others working on similar topics.
The link is the killer app of the web, including blogs. Those links are exactly what allow people using search engines to find your work, no matter what it's about. If the key power of academic works vs popular media rested in the citation, then some of that has already bled over onto the web, blogs included.
So by virtue of being on the web, where you can link and be linked to, you already have a distinctly important advantage over newsprint - one that just happens to resemble the same advantage academic papers have over newspaper columns.
After all, you're not trying to weigh a decision of whether or not you should abandon your newspaper column, are you? How much do you think the relative costs of writing newspaper columns vs blogging has to do with that decision, in contrast to the usefulness of citation in the form of linking?
If the influence of academic papers vis-avis hypermedia is a long term process, then isn't it a little early to make a definitive judgment on the usefulness of blogging?
In the meantime, work on improving the usefulness of links continues (semantic web, tagging architectures, http://microformats.org/ microformats]], social networking, etc.)
Unless blogging is useful to you in vetting topics that you convert into academic papers, you should stop; the blog is not otherwise advancing the field's collective push. Tyler Cowen, Bryan Caplan, et al will presumably read your academic papers. But if you are personally unsanitary or unsavory and cannot knock around ideas with your colleagues at Cafe Strada, perhaps you should keep blogging to achieve the aforementioned instrumental benefits.
While disintermediation has its benefits, I gather that those benefits do not synch with your goals.