It seems like the music licensing companies in many countries are equally evil. This latest move by the Belgians is just business as usual, disgusting as it may be.

But, why do they then choose acronyms that are so easy to make fun of?

ASCAP: Ass-Cap (put a cap in yo ass)
SABAM: Sa-*BAM* (like punching someone in a Batman comic book)

I'm sure there are fun mis-pronunciations for the equivalent associations in other countries as well. Anyone from other countries want to contribute more?

I can somewhat answer the inverse of this question, though: "How many drugs do we reject in clinical trials because some researcher used the wrong animal model to test?"

My memory is a little fuzzy on the exact number from when I worked in the industry, but something like 70% of all drugs that pass Phase 1 trials fail in Phase 2 trials. Phase 1 trials are small and test for safety problems, and Phase 2 trials expand to a larger cohort to test for efficacy -- does the drug work. The pharmaceutical industry loses around 70% of all its drug candidates due to them plain not working. Often times, this is due to it working in mice/rats, but not working in humans. This isn't entirely surprising, since rodents are a good bit different than humans. Also, many animal models simply mimic human disease, rather than actually being related to how the human disease works. For example, many animal models are done over a short period of time, say 30 days or less between the initial insult (do something nasty to the rodent to induce disease-like symptoms). For inflammatory diseases, real human disease may take years from the initial problem (whatever it may be) to develop into full blown disease. When you're comparing an animal model on the time scale of a few weeks to human disease over several years, and the insult is something very very different from what it could possibly ever be in a human, it should come as no surprise that the model is often biologically very different from what is going on the human. This isn't necessarily due to rodents being too different from humans, but could easily be due to the model simply being the wrong model to mimic human disease.

Long story short, many animal models just don't do a good job at representing human disease. This is not news to anyone who has worked with them or been in the pharmaceutical industry. However, not everything is gloom and doom here. There are, actually, many animal models that *DO* do a good job of modeling human disease. The trick is to know which ones are good and which ones aren't for various diseases, drugs, pathways, etc. before you start spending the big money on clinical trials....

Back in the early to mid 90's, when I was in undergrad and using several different unix platforms (AIX, HP-UX, SunOS, Linux, DEC-OS, dumb X-terminals, etc.), different programs on different platforms treated backspace as different things. The talk/ntalk/ytalk command on HP-UX was especially annoying. It would interpret the backspace key as a ^C and kill your talk connection to your buddy across the country using his unix account to chat with you. Imagine typing away, then hitting the backspace key to fix a typo, only to have your connection killed. Sometimes shift-backspace worked (HP-UX, ftp clients, various login prompts), sometimes DEL (usually easy to fix by setting the terminal variable and/or tset), but ^H almost always worked like it should. I believe on a dumb X-terminal on my desk that I had to use to connect to a Sun server during early graduate school, ^H was the only way I could figure out how to actually issue a backspace. So, in agreement with what other posters have said, ^H isn't really important anymore. Not like it was in days of yore, when there were MANY different ways to issue a backspace, you could never be sure which would work, and the backspace key could sometimes do unexpected things. These days, backspace and delete generally do what you'd expect them to do.

To this day, I *still* automatically use shift-backspace to fix typos during login prompts, as well as ^D to forward delete instead of the delete key :) Just don't hold down ^D at an empty unix terminal prompt, or it will auto-exit the connection. This can be especially annoying in an environment when you have multiple console windows open and you hold down ^D on a blank prompt. The first window closes, focus switches to the next window, which in turn closes due to ^D on a blank prompt, etc.. In short order, all your windows are gone, all because you started deleting at the start of the command line and it ran out of stuff to delete. I never understood why this was a "feature". I have, however, adapted and will often use ^D on a blank prompt intentionally to exit a session (it's so much faster than typing "exit") ;P

A bit further off-topic, but there was one program on one system that I used one time where neither the ENTER nor the RETURN key functioned as expected. I was forced to resort to ^M to issue newlines! Ah, the days needing to know all of the alternate ^key commands :)

This makes me think of the South Park rerun that was on last night or maybe the night before -- the one where the kids had to vote for the Giant Douche or the Turd Sandwich as their school mascot. When your choice is between a giant douche and a turd sandwich, it doesn't much matter which congress-critter you vote in -- neither choice will result in the outcome you desire above....

I have used 3D molecular modeling tools for years. You can get by just fine with a 3-button mouse. 3 single buttons, 3 2-button combinations = all the axes you need to translate/rotate/zoom, all continuous. You don't need a mouse wheel at all, just button combinations and intelligent movement of the mouse while the combinations are held. It's really quite intuitive and easy to use.

However, you are correct in that I was never able to zoom, translate, and rotate all in a single gesture. However, 3 separate smooth gestures is no problem at all, and (to me at least) preferable. For fine positioning, it is much better to limit movement to only one of these axes at a time, so that you do not accidentally mess up one of your axis positions while you are adjusting another to get it just right.

Wouldn't this be a sea of cautious *disrespect*?

Google Scholar brings up a PDF file I can obtain without a subscription. So, I searched for "flav" to see what's going on here. Table 1 gives stats for "Total flavanols". The only other hits are in the references, where there are 3 references with flavonoid(s) in their title, and 2 more with flavonol(s). Now, I haven't tried to hunt down and read over these 5 cited articles, but given their titles, I think it highly likely that they do indeed cover flavonoids and not flavanoids.

Now, could flavanoid information have been taken from other citations that don't have flavanol in their title? Sure. But, since they cite 5 papers that appear to be solely dedicated to flavonol/flavonoids, and nowhere mention flavonol/flavonoids in the rest of their manuscript, I'm guessing that "flavanol" really is a typo and that they really did measure, report, and meant to talk about flavonols.

So, the criticism does appear to be valid. The only mention of flavanols in the entire manuscript does appear likely, at least to me, to be a typo meant to convey information about flavonols. Now, whether this was an intentional dishonest typo, or an honest accidental typo, I'm not going to speculate. I'm just confirming that the claim of misreporting flavonols as flavanols appears to have merit.

I feel the need to defend PNAS here, since I actually consider it to be one of the better journals -- *IF* the manuscript has been accepted via the non- Academy of Sciences peer review process. PNAS has two (maybe three?) submission tracks: A) You are a member of the Academy of Sciences, and this results in your manuscript getting accepted more easily than usual. B) You get a friend who is an Academy member to submit your paper for you, which also lowers the bar. I'm not sure if this is really part of track A) or not, so it may or may not count as a separate track. Track C), however, is where someone who is NOT associated with the Academy submits a manuscript. Papers that come in from "external" submissions and make it past the Editor, and then all the peer review, are actually generally quite good.

So, PNAS is a mixed bag. Manuscripts from Academy members can be a bit on the sketchier side due to how the submission/review process works for them, but manuscripts from non-Academy members are often, IMHO, a good bit better than average. As with all scientific manuscripts, you must read them over carefully and determine if they have sufficiently proved their point or not.

Unfortunately, you may have a point about this particular article, since it was "Contributed by Bruce D. Hammock", who a quick Google search shows to be an Academy member.... I have not actually read the TFA, though, so I can't comment on whether it is any good or not.

From TFA (and the summary):

How likely is it that a control system vendor would have the usernames and passwords of their client, used in the actual production system? Maybe they actually do, as part of some sort of remote support agreement, but if this is the case, that's already a bad security practice.

It seems more likely to me that the vendor has a list of default usernames and passwords, and THIS is what was obtained. Perhaps what Weiss *really* meant to say would be be something like: "Someone got ahold of the default usernames and passwords that our vendor uses. Since we never changed them from the default values, it's our own damn fault."

After seeing SO many stories like this, it's usually a case of not changing default passwords. Given that Weiss's statement *could* be read as I have read it, this seems the most likely scenario to me. I'm going to write this one up as stupidly bad security policies until I have sufficient evidence contradicting this assumption.

Wouldn't a digital millennium be 1024 years?

No, that would be a mellibinnium.

I looked it up at www.tbauto.com. Said car is a reproduction 1770 Fardier de Cugnot. From the website:

"The original Fardier de Cugnot has been in the collection of the Le Conservatoire de Arts et Metiers, Paris, France since 1801. Currently on display is one reproduction Fardier on loan from the Deutsche Ban Museum in Nuremberg, Germany, as well as a completely functional, faithful reproduction that was created from the ground-up by The Tampa Bay Auto Museum."

So, the oldest car is on display in Paris (I have not investigated whether it works or not), and the one(s) in St. Petersburg are replicas, so nothing has yet to disprove that the car in TFA is the world's oldest working car.

Not notable? I may be confusing itracker with another tracker of that era, but wasn't itracker the one with the built-in Worm mini-game? It even had 2-player support. I have fond memories of playing 2-player Worm with others in my dorm in the mid 90's.

Aside from Worm, it was actually a very nice tracker, and had a high compatibility rate with playback of other MOD formats. I'd certainly put itracker on the list of notable trackers that advanced the then- state-of-the-art. I think I used it a lot for adding BPM events to the beginning of a lot of my old Amiga mods so that they'd play at the correct tempo on a PC, which used a different NTSC/PAL vertical scan rate timer setting or something like that to determine the beats per minute. It was the old NTSC vs. PAL Amiga problem for MOD file tempos. Or maybe it was whichever tracker used .XM as the native format, my memory is a bit hazy on that.

BTW, If you ever want to convert MOD/IT/XM/S3M/etc. to MIDI, check out the -OM output mode of TiMidity++. It does a pretty good job of it (disclaimer: I wrote a lot of the mod->midi code a long time ago).

unless it involves cats and cheezburgers

How about equivarogue?

This is why we are upset at Apple. They should not be writing it down somewhere. While the subpoena itself may not be considered an invasion of privacy, the act of Apple recording your location (which is now being subpoenaed) most certainly is. Is this hard to grasp?