System and numerical libraries and compilers are of course written specifically for the machine. But user-level apps (and a lot of scientific computing uses finished apps) are ported across multiple systems.

Portability is not as big an issue as it was a generation ago, as most supercomputers basically are Linux machines today, and made to more or less look like a typical Linux installation from a user-application level, with a POSIX API; pthreads, OpenMP and OpenMPI; a standard set of numerical libraries; and often even gcc-compatibility in order to minimize the effort of porting. A notable exception is GPU-based machines (that are in the minority today, despite the OP assertion); they don't have a common API to write for, so using them is substantially harder at a user-level.

And at a user level (but unike system libs) porting or coding time very much matters. Let's say your project is going to need a month of wall-clock computing time during the course of a year or two. If switching to a GPU-based system would shrink that by 50% - two weeks - then the effort to move your model code, app, and libraries had better take less than two weeks of work or you're going to waste project time, not save it.

Don't piss off the rich guy.

This isn't just misleading advertising. Many journals put the dates right on the paper in the publication. I've got one on my desk right now that says "Recieved: 29 November 2006 / Revised: 2 April 2007 / Accepted: 12 April 2007 / Published [...]".

Note that the second gap is much shorter, because if the initial submission was good, all that is needed is enough time to verify that the revision addresses all the issues raised by the reviewers during the first gap. If that's what the "4 days" refers to, it's a non-issue.

OTOH, if the first gap was only 4 days, there is cause for alarm. The publisher has to do a little preprocessing to figure out who would be appropriate reviewers, write them, wait for them to respond, try again if too many say no (or don't bother responding at all), then get the paper to the reviewers and allow them enough time to review it (often a month or more).

You vastly overstate the actual situation.

Reviewers are not expected to reproduce the author's work, but they are expected to check whether the authors did their homework, conformed to well known facts, provided sources for other claims, identified and justified assumptions, followed good experimental procedure, drew logical conclusions from their observations, and stated enough information in the paper so that readers can determine that all that stuff was actually done.

Unfortunately, reviewing is almost always voluntary work (something academic scientists are expected to do as part of their job, but far from top priority), so there is a tendency to get lazy and do a not-so-careful review, or delegate it to a graduate student, with a result that things that should have been caught sometimes aren't.

Working against that is the fact that the publisher usually gets 3-5 reviewers, so some catch what others miss.

You are correct, however, that going through this formal phase of peer review doesn't make a paper's claims true, and the post-publication peer review never really ends for a good paper. For example, people are still chewing on stuff Einstein published 100 years ago.

Don't know about Cell, but lots of journals have the authors submit the paper (including revisions) with all the images collected at the back, presumably a holdover from pre-electronic typesetting/layout techniques. So the case of one image appearing with two different captions, it's at least possible that the error was made by the publisher rather than the authors.

Four days isn't long enough to hear back from reviewers whether they're willing to do it. Something is *seriously* wrong with this picture.

(No pun intended.)

Actually, a lot of it was because earlier in that decade our legislators foolishly repealed some laws that had been established during the Great Depression to prevent exactly this kind of thing.

It never worked that way. Private enterprise has to charge you enough for their products and services to make a profit after all their inefficiencies and screwups.

LoL.

But I'm sure there's some kind of fakery involved.

...posting to Slashdot.

I think I'll go investigate this "outside" that I keep hearing about.

They should have kept their mouths shut and sold the finished product to North Korea.

SOP, whether working for a government or another company: bid low, and count on problems and changing requirements to make it profitable before it's done.

That's only a quarter of a million dollars per inch... I'm sure lots of billionaires would find that an irresistible deal!

Yeah, because private enterprise never screws up.

When you establish a single court for an issue, they roam further than when you have a whole system of courts on it. Individual opinions become greatly magnified in there effect, as you essentially give up peer review.