Summary
In 2014, I wrote a piece enumerating some pros and cons of publishing altruistically relevant content in formal journals. The present piece presents a more personal spin on the topic, explaining why I myself tend not to submit papers for formal publication. Some of my reasons are good, and some are more reflective of my emotional whims.
Contents
Research paper vs. research notebook
As of the time when I'm first writing this piece (late 2016), I'm a few months away from 30 years old. I've been interested in altruism-related issues since age 13. Suppose I have until age 70 to keep working on these topics. That means I'm roughly (30-13)/(70-13) = 30% of the way through my altruistic lifetime. But the gaps in my knowledge about how to best reduce suffering remain enormous.
My writing serves several purposes, but often the main purpose is to make as much progress as I can on hard questions, some of which have basically never been studied before—such as how to reduce the suffering of wild invertebrates due to natural causes. Often, I'm mostly trying to figure things out for myself and leave behind a paper trail of my findings in the form of a page on my website. My goal is usually not to write a cohesive, cogent article that advances a particular argument in which I have a great deal of confidence. 
My goal is usually, instead, to compile a bunch of relevant information that I found or ideas that I had in an organized way for my own future reference and for others to benefit from. I note my confusions and uncertainties, mark points that need further research, and suggest speculative possibilities that I haven't yet investigated thoroughly. Thus, my articles are sort of a public research notebook.
Most of what I write is not of publishable quality. It's often not sufficiently original or researched in sufficient depth. But shallow investigation often makes sense for my purpose, which is to cover a lot of ground and have as many insights as I can before I die, while maintaining some degree of high quality. I'm not optimizing for tenure or grant money.
Helpful practices disallowed in formal publications
One practice that I like that's unacceptable in published writing is to extensively quote my sources, even when mere factual material is being explained. I think this practice is both faster and less error-prone than paraphrasing. It's what you would do if you were trying to pass along notes to a friend about some topic. I'm uncertain why it's not allowed in published writings—perhaps because it's "cheating" by appropriating the words of others without contributing anything original. (But why is there originality value in merely rewriting someone else's words in a new way?)
In addition to quoting words from source papers, I like to include screenshots of tables and figures from source papers. These allow readers to quickly see numbers for themselves without having to pull up the original articles. And since a picture is (sometimes) worth a thousand words, including these figures is better than merely trying to paraphrase the results using text. While formal publications can reproduce figures in limited circumstances, they can't do this nearly as much as I feel is optimal for the convenience of readers. (Textbooks do a better job at reproducing figures and tables from other sources than novel research papers do.)
I also prefer to add page numbers whenever I cite information from a specific part of a paginated document. While some journals allow this, it's often more standard to omit page numbers in the sciences, and some journals will remove them. Rekdal (2014), p. 572:
A philosophically minded colleague of mine whose sources are often thick theoretical books had an article accepted in one of the most prestigious journals in nursing science. One of the reviewers recommended that he remove most of the page numbers from his references, which he subsequently did. [...]
The poor advice my colleague received on this occasion is not an isolated case. A few years ago, I received the final proofs for a book chapter I had written, only to discover that the proofreader had removed all the page numbers in my references, except one accompanying a direct quotation.
Citing non-scholarly sources
Those writing for academic journals are often discouraged from including many citations to non-scholarly sources. Maybe the purpose of this rule is to prevent inaccurate material from entering scholarly journals? But I think that non-published material is often useful, especially in domains where few or no scholars have written about a topic. For example, suppose I want to mention the business operations of different companies in an industry. Maybe there's scholarly literature about this, but I'm more likely to find relevant information from news articles and company websites. Or suppose I want to learn about the practices of small dairy farmers. Maybe there are some scholarly surveys or books about this subject, but it's also likely that one can find a lot of useful information from self-published blogs by the farmers themselves, from informal videos on YouTube, etc. Indeed, what farmers write about their own techniques may be more accurate than what a journal article reports about them.
In some disciplines, like moral philosophy, there's in my opinion not really such a thing as expertise. I think the informal opinions of people writing on forums, in comment sections, in YouTube videos, etc. are often as valid as the opinions of prestigious ethicists, since they're all just expressions of moral intuitions in various guises. Why should citations of academic philosophy articles be allowed but citations of forum discussions not be?
I prefer to cite a wide range of sources of varying "quality" and let readers decide how much weight they want to give the different statements. The scholarly practice of deliberately ignoring most of the written information in the world seems divorced from reality. The next time you find a stock trader or management consultant who only uses scholarly literature to draw conclusions, you should let me know.a That said, there may be some value in having certain segments of information in society (academic publishing, as well as high-quality journalism and such) that have higher quality-control standards than other segments, to reduce the tendency for myths to propagate.
Editing articles
Another benefit of owning my own articles rather than publishing them is that I can update them constantly. As I learn new information, I often find tidbits that are relevant to some past article I wrote. I can quickly edit the old article to add in the new info, all on my own within a few minutes. With a published article, I probably can't update it at all except in cases of outright errors, and contacting the necessary authorities to do so is effortful.
de Lacey et al. (1985), p. 885: "A correction tucked away in some corner of a subsequent issue is unlikely to be read." Of course, the situation is somewhat improved if electronic journals link to the correction from the original article (or, better yet, make the correction directly to original article).
Static publishing is, in my opinion, a relic of the pre-Internet era and is one of many ways in which academic practices are antiquated. If you need to refer to a static version of a web page, you can use WebCite. So it seems to me like Internet publishing of HTML pages is strictly superior to static publishing of articles. If academic articles could be edited in real time, we would have fewer persistent errors and less need for up-front peer review. Publishing online is my form of peer review, since I can fix errors when readers point them out.
"Irrational" reasons for self-publishing
I love writing, but I dislike writing according to someone else's requirements. Writing in an assigned way is like being an artist who's told what to paint and using what techniques. It stops being art and becomes a chore. Perhaps one could still express creativity within the boundaries of what's allowed, but this is at least more difficult.
Some stylistic expectations on academic writing are useful, like making sure your core thesis is clearly articulated and your evidence is well organized. But some other expectations are annoying, like omitting many of the things that humanize writing: informal language, speculation, anecdotes, and jokes. Of course, some academic journals are more lenient than others.
One of the biggest inhibitors against my submitting writing for publication is the experience of getting a draft back and being told to make changes that I disagree with. I'm happy to make useful revisions, but in response to most criticisms, I feel: "Yeah, well, you know, that's just like your opinion, man." If I were more rational about the situation, I wouldn't be so annoyed and would just take reviewer/editor comments as more information to help me adopt a different approach next time. But my irrational, "I like it better my way" impulse is also the part of my brain that strongly motivates my writing in the first place, and it has to be kept happy.
Finally, another irrational reason that I dislike journal submissions is that it takes forever (i.e., months) to get a response. Once I've finished writing an article, I'm excited about it and want to share it right away. I also want to get feedback right away so that I can make changes while the material is still fresh in my head. In contrast, submitting to an academic journal is like being asked to keep an exciting secret for a year. (This is less true in cases where you can put a preprint online.)
In general, academia has a lot of rules, some sensible and some not, that one has to play within. I don't like following someone else's rules. I want to do things the way I think is best and most fun.
A steelmanned case for formal publishing
The best argument in favor of submitting articles to journals would probably go something like this: "Yes, it's true you want to cover a lot of intellectual ground, because very few people in the world have yet explored systematically what follows from taking antispeciesist negative-utilitarian ethics seriously. But you don't have to cover all that ground yourself. You should instead focus on building a community of people who will further explore the kinds of issues that you write about, such as wild-invertebrate suffering. By creating an academic field around these ideas, you'll allow more people to follow in your footsteps."
This is a decent argument, but I think it applies more strongly to general movement-building around suffering-reduction research, such as what the Foundational Research Institute does, than specifically to academic publishing.
Many of the topics I write about aren't amenable to being published in journals anyway, at least not top journals. This is sometimes because the topics are too speculative and sometimes because they're too weird and interdisciplinary (e.g., how does cattle grazing affect aggregate invertebrate suffering?). Academic philosophy often discourages articles that are too empirical, while ecology journals are unlikely to accept articles with a normative angle (especially an anti-conservation angle!).
In any event, I'm open to the possibility of formal publishing in the future in cases where it makes sense to do so. And for people who are more rational and less emotional than I am about writing, I would encourage exploring formal publishing as an interesting option to try out.
Why I don't plan to publish a book
Since roughly 2013, people have been asking me if I've thought about writing a book on reducing suffering. While things could always change in the future, so far I've been disinclined to publish a book, for several reasons.
A big reason is that I dislike the permanence of a book. When I write online, I try to be fairly careful but not necessarily extremely careful in checking my facts, ideas, grammar, and the like, because if there's a problem, I can just edit the piece to fix it. In contrast, if I publish something more permanent, I feel I should spend more effort getting everything right, but that degree of perfectionism plausibly isn't an optimal use of resources, unless the book will be very widely read or used for a purpose where accuracy is essential, like if it's a book of guidelines for doctors or civil engineers. I see my writings more as first drafts, where there's always more to be improved later. This is similar to the philosophy of "eventualism" on Wikipedia (Meta-Wiki "Eventualism").
Publishing a book once every few years is like shipping a new version of software once every few years, with no bug patches in between. Yes, you can write errata on your website, but how many readers of the book will look those up? I feel like book publishing made sense in the pre-Internet era where you couldn't immediately update writings, and you had to bundle a lot of writing together to achieve economy of scale. I'm less convinced that this approach makes sense today.
Maybe one could argue in favor of an online e-book in which each chapter is its own HTML page. The main difference of that approach from the way I currently write essays is that the e-book would be a step-by-step presentation that gradually introduces concepts in a clear manner. I think having an e-book like this could indeed add some value as a supplement to my current series of more modular essays, but I'm not sure how much value it would add, and I generally feel like I'd rather just improve the original version of an article than rewrite it. Sometimes other people rewrite my ideas to make them more accessible, and I'm grateful for that.
Fundamentally, I feel like most of my thoughts are half-baked, and I don't have a crisp set of answers that I want to convey to readers. I'm more inclined to keep working on unpolished drafts and inspire other people to also explore these ideas, rather than prematurely converging on conclusions that are likely to change within 10 years anyway as the field of research on how best to reduce suffering continues to make progress.
As far as formal book publishing, I'm emotionally averse to the idea for many of the same reasons I'm not very excited about formally publishing academic articles. I prefer to write the way I want to write, rather than being told to write a certain way by an editor; for me, that sucks the fun out of the activity.
While I haven't investigated the topic in depth, I tend to assume that there should roughly be an "efficient market" regarding publishing books versus articles, i.e., the impact per unit of effort is probably fairly similar in the two cases. Suppose that weren't true and that book publishing was actually much more impactful than web-article publishing in expectation ex ante. In that case we'd expect to see more people publish books until book publishing became more saturated. Of course, this efficient-market hypothesis may be wrong in particular cases, especially within niche subfields in which there are relatively few authors.
Footnotes
- Kyle Bogosian: "In academia, you can't draw any conclusions without having extensive research that is peer-reviewed. But in finance and capital markets, which perform functions that are similar to charity evaluation, analysts are a lot more flexible with their estimates, projections, etc." (back)