I'd disagree both with you and the comic.

The problem with publish or perish is that it leads to people gaming the system. If pay rises (or employment) solely hinge on the amount of (however graded) publications you have achieved, this leads to people essentually using parts of books they are writing as articles etc. This in turn leads to a doubling of information, inconsistencies etc, alongside the fact that not every topic can be adequately dealt with in a journal article.

The system actively punishes big projects and thoughts in favour of small, more easily digestible (and thus publishable) projects. If a university only looks at the number of publications and not at their nature, they shift from quality to quantity, which in turn is bad, because it fundamentally undermines the purpose of academia.

edit: I just saw _dps also linked to the gamification bit, but the rest holds true nonetheless (and I do agree with him)

I'm not saying that publish or perish is an optimal solution, just that there's nothing special about it. Plenty of people have "file your TPS reports with cover-sheets or perish", but there's not special jargon for it.

Publish or perish is not just about doing your job, rather it limits you in how to do your job, and that limitation runs counter to what your job often requires.

It's not so much "publish" in itself that's the problem[1], it's how publishing is defined and handled, and that a certain form of publishing is seen as right to the detriment of any other form of published material, whereas the preferred method is often not the best method for a given project.

[1]: though it certainly is a big part of the overall problem of uniting teaching, research and publishing

And I think the point that zimbu668 is making, is that such situations are still not unique to academia. Perverse or misaligned incentives are the norm in quite a few industries.

For example, sales guys are compensated based on sticker price, not profit margin, which means that the guy who sells $2MM of work for $1MM (i.e. at a loss) gets more commission than the guy who sells $0.5MM worth of work for $0.75MM. For another example, grading programmers by lines of code... 'nuf said ;). Another totally ubiquitous example: grading office workers by ass-in-seat time rather than value produced. Another also ubiquitous example: Departments in a company encouraged to spend as much money as they can at the end of the quarter, so their next budget doesn't get cut.

I could go on, but I'll stop here. The point is, even if "publish or perish" refers to misaligned incentives, that still means it's basically the same as 90% of other jobs out there.

Alongside the other major problems with publish-or-perish mentioned in other responses, most people's ability to fulfill their job requirements is not decided by a program committee of 30 colleagues from other workplaces looking-over the reviews given by 3-5 anonymous colleagues, but actually written by their grad-students, based on a bare-minimal reading of your actual work.

A software engineer can commit his code when it compiles and/or when the unit-tests pass. An academic might spend months writing and revising a paper, only to be successively told, "Too many examples", "not enough examples", "there's a bug in your proof", "I can't understand this so it must be brilliant", and finally "too long, didn't read". The paper is then not published, and at minimum the academic must revise the paper and resubmit, taking several more months.

After maybe 6 months, the academic might have succeeded in getting one publication out of one project, and he has to be running multiple papers' worth of projects concurrently throughout the whole year to make sure he produces enough papers each year.

The big deal is that journal publications are a by-product of some kinds of research, and not in themselves proof of good work. Unfortunately, it's the dominant mechanistic (I won't use the word "objective") way of evaluating academic job performance. Being mechanistic, it is easily gamed, and the gaming can corrupt the underlying process (e.g. the generation of "publons" http://en.wikipedia.org/wiki/Least_publishable_unit).

Most academics see themselves as being tasked with expanding the frontier of human knowledge, not with documenting it and persuading other people that it's important (which is seen by many as a vulgar popularity and marketing contest, not an intellectual one).

Expanding knowledge without communicating it is useless, and salaried researchers should know that.

But parts of it can be a zero sum game, just like occurs between the marketing departments of Coke and Pepsi.

This is pretty deep in the thread, but I'll bite because I have a few counterintuitive ideas I'd like to share. Apologies for the lengthy prose.

> Expanding knowledge without communicating it is useless

Perhaps controversially, I'm not sure I agree with this, at least not on the timescale of a singe person's career. Recording new knowledge is absolutely essential, and many important ideas in science were discovered or rediscovered by reading old personal notes of great scientists that never made it into a letter or article. Not even the most ardent publish-or-perish critics argue that researchers shouldn't record their ideas and findings somehow, and make those recordings available.

But actually communicating the work in the present, rather than merely making it available in a recorded form, requires a willing recipient and a crafted message. So the focus on publication conflates the creation of good ideas with the ability to persuade others (and the readiness of others to be persuaded).

You might say "Sure, that's also true everywhere else". Having been on both sides of the fence, my experience is that it's actually much less true everywhere else, precisely because research is so future-oriented. In most lines of work you are trying to get results today/tomorrow/this-quarter, and, while you always need to persuade in order to advance, you don't have to persuade about hypotheticals beyond your lifetime like "This theorem could become the basis for an entire field of mathematics someday". See, for example, the originally unpublished works of Karush (now at the heart of optimization theory) or Lyapunov (now the basis of much research on nonlinear systems). Both pursued their ideas for love of the ideas, and both were met with indifferent audiences.

Changing the subject slightly, I submit that in many cases journal publications have evolved to be more advertising and less communication, particularly in the most prestigious journals which, for Moore-defying reasons, impose page limits and will not allow you to attach large images, data files, or programs. While not nearly as bad as patent descriptions, journal publications are almost never as useful in reproducing results as simply having the work artifacts of the research process: procedure pages from someone's lab notebook, program source code, raw data files unaltered by "smoothing for visualization", sensor/instrument configurations, the 20 line informal proof that inspired the 20 page formal proof that gets to a journal, etc. These kinds of artifacts are recorded by any researcher in the workaday process of doing research. Assuming basic archiving ability, there is no risk of these ideas being lost to society even in a journal-free world.

But if you don't believe me that journal articles are not isomorphic to shared knowledge, the proof is in the pudding: a fellow researcher will happily share his journal article (advertisement) with you, but ask for a copy of the lab notebook pages and you may be in for a chilly response.