Cap'n Refsmmat

Could you give me some examples of links which no longer work? If you hover your mouse over the "2 hours ago" text, you should get the exact date and time. Relative times are for some reason fashionable now. I just found the setting for this and enabled it, for posts made within an hour of each other.

I had a relationship fail because my partner was afraid this would happen, although it didn't. The fear turned out to be worse than the baggage itself. Instead of being afraid you'll never get over your exes, and letting that fear dictate what you do next, move forward and try to make your new relationship the best it can be.

As this claim is speculation, unfounded in science, it belongs in the Speculations section; however, the Speculations rules state: So I will skip a step and close this thread directly. Please do not post unfounded speculation in the science forums, or as answers to scientific discussions.

Yes, this behavior shouldn't have changed much between versions.

I don't expect the spammers are coordinated enough to remember that SFN converts their links to plain text, and avoid us in the future. I'm guessing it's semi-automated and they're just presented lists of sites to register on. The software only knows the last page on SFN you visited, but has no way of knowing if you subsequently left to view another site -- it doesn't get any live information from you while you're looking at the page. So if I close SFN after making this post, I'll be listed as viewing this thread for another ten or twenty minutes until my session times out.

I tried disabling links for new members recently, but apparently that just disables the button in the editor toolbar, and you can still paste in text with links. So that's not working. We have a CAPTCHA (Google's reCAPTCHA 2) to filter out bots, plus a custom question in registration. Every registration is sent to the IPS spam service before it's allowed through. Unfortunately the spam service checks registrations, not posts, so those who make it through are not subsequently checked. There are plugins available to check registrations against a different spam service as well. We used one of those before the upgrade, but it still didn't catch everything. Unfortunately there aren't well-supported plugins to check the text of posts for spam. Anything we write ourselves would have to be maintained through upgrades and software changes for several years, which would be a burden. So we're a bit stuck. If someone could write us a plugin to send 5000 volts to spammers over the Internet, that'd be great.

It doesn't delete it -- it makes the rest of the post bold, as you can see in studiot's example. It assumes the tag ends at the end of the post. I don't recall IPB 3.6 having this behavior, and I'd rather it leave the tag untouched than assume it should apply to the rest of the post.

Wow, seven years since I started this thread and it's still alive. Agreed, this book is excellent, along with its sequel Dark Sun about the hydrogen bomb. Brilliantly written. Admittedly my own reading has started to lean towards history instead of science. Some interesting science books I've read in the past few years: Primates and Philosophers, by Frans de Waal and others, in the form of an extended essay by de Waal and responses to his points. de Waal is a primatologist, and the book is about the origins of moral behavior in humans, with de Waal pointing out basic elements of sympathy and empathy visible in the great apes, and the respondents entering extended argument about to what extent this suggests human morality is entirely evolved instead of invented. The Better Angels of Our Nature, by Steven Pinker. Pinker is probably the best scientist I've read at writing an extended and well-researched argument. Again on the theme of morality, and why violence has declined over the centuries. The SFN crowd may also enjoy these less sciency books: Anti-Intellectualism in American Life, by Richard Hoftsadter. Written in 1964, giving a long history of anti-intellectual ideas (including anti-science ideas) in the United States since its colonization. Broken up into discussions of religion (including more history of American religion than I had ever encountered before), politics, culture, and education, and remarkably even-handed for a book on this subject. (It doesn't consist simply of a bunch of snark about what uncultured morons Americans are.) Double Star, by Robert Heinlein. A fun sci-fi story about an actor hired to double for the solar system's most prominent politician, unexpectedly finding himself in deeper than he expected. Amusing to see the 1950s vision of the future: spacecraft easily fly to Mars, but government records have to be kept on microfilm on the Moon -- no Internet or advanced computers.

Blurgh. Does it maybe think that B in square brackets should mean the bold BBCode, even though BBCode is going away in favor of the allegedly-WYSIWYG editor? A test: this will be bold if BBCode still works I adjusted the super/subscript icons, so hopefully you should be able to access them. edit: yes, it does think that. Why does BBCode still work?! And why did the tag work in your post without the corresponding closing tag?

After a bit of further research, it seems the current session expiration time is 24 minutes: if you haven't done anything in 24 minutes, PHP will expunge your session records. studiot, maybe you could do a quick experiment by logging in, setting a 25-minute timer, and seeing if you're logged out when you return? If so, that's something I can tweak in our configuration. I'm not sure whether it changed since the previous version, but we did change PHP versions, so maybe something is different now.

I just tested logging in without "Keep me logged in" checked, and the cookie was set to expire in November. So there shouldn't be such a rapid timeout. Are you on an ISP which frequently changes your IP address? If your IP changes, you'll be logged out. That was the case with the previous version too, so maybe you got unlucky with an IP change this time. I can disable the IP address check, though I'm not sure of the security implications.

I've just tried disabling the link button for new members (2 posts or fewer). I'm not sure how it'll work, since I don't really understand the WYSIWYG editor, but maybe it'll inconvenience some spammers.

I am, but I'm not a big fan; they can be hard to read when you've got a bunch lined up, whereas you can still interpret a bunch of boxplots together.

I'm a big fan of boxplots for this type of data, although you have small enough samples that you can just plot each point directly. What kind of analysis do you intend to do once you've removed the outliers? Are you trying to test for significant differences, or model the data somehow? I ask because, to me as a statistician, there isn't really such a thing as an "outlier". Unless some data points arise from a mistake in the experiment, they're all real measurements from the true distribution of possible outcomes. Removing some data can be valid for analysis, but it depends on what your goals are. I'd be able to give better advice if I knew what you were trying to achieve.

The editor is supposed to WYSIWYG, so "preview post" would show you the same thing as you were just looking at. Is there a particular case where it doesn't look right and you need a way to check it? The editor automatically inserts a new paragraph whenever you hit Enter, so there's a break between paragraphs. If you don't want a new paragraph, you can hit Shift-Enter to just get a new line.

Why specifically do you want to detect outliers, and what are you going to do with the data once you identify outliers? There's no one hard-and-fast test that identify outliers, because there's no one reason for something to be an outlier. Do you have reason to believe the outlying results are somehow erroneous, or are they as valid as the others?

I strongly recommend the book Dreamland, by Sam Quinones, about the opiate epidemic, if you'd like to read the full story. I think there are a few reasons: The media goes through obsessions. Now that they've heard about fentanyl, it's the only thing they want to talk about. Drugs like OxyContin have been modified to make them harder to abuse. The DEA has been cracking down on pharmacies that sell enormous amounts of hydrocodone and oxycodone, along with "pill mill" doctors who give prescriptions to anyone who has cash. Doctors are more hesitant about prescribing pills, so it's harder to get pills to use or sell. People who were already addicted are finding it harder to get hydrocodone or oxycodone pills, so they have to switch to illegally produced heroin. And some of that heroin contains fentanyl.

You don't seem to be reading anyone's posts in this thread. Why do you think the FDA is pushing fentanyl? It is not. The people overdosing on fentanyl are getting it illegally, not from their doctors. The FDA was not involved. Banning fentanyl would not help, because people who overdose on it get it illegally, mixed in with illegal heroin. People who make heroin are mixing in fentanyl, because it's cheap and strong, and saves them money, not because the FDA told them to.

Are you referring to these boxes on the home page? If you click the "1 reply" link (or the "Saturday at 12:15 PM" link), you'll be taken directly to that status update, not just to their general profile page. Another one of the many hidden links throughout the new software.

Unfortunately, given the size of the changes between 3.6 and 4.0, manually porting security fixes simply wouldn't be practical. So much of the code was replaced outright that we wouldn't be able to take some changes and not others. What in particular is most confusing about the new system, besides the Unread Content page that has changed so much? Maybe there are tweaks we can make to help everyone get around.

That's a good idea. Unfortunately, the Admin CP only seems to give me the option to replace the Unread Content link at the top right with Unread Topics: (..sigh.) If enough people think it's worth changing out Unread Content for Unread Topics, I can do that, and people who want Unread Content can find it under the Activity tab.

You want to click the little dot or star next to the title of each thread, which takes you to the first unread post in that thread. It's not particularly well labeled. Also, under the Activity Tab, My Activity Streams, there should be an Unread Topics stream which is condensed and easier to follow. I wonder if I can adjust the defaults.

I suppose, if you're worried about being dragged into a fight, the backup option is an anonymous comment on a site like PubPeer. Their commenters have forced a lot of retractions by finding image duplication and manipulation in published articles. The system will notify authors of your comment on your behalf. I've previously written to journal editors about a major error in a paper, though not one arising from misconduct. They seemed confused about what to do, and apparently had never published a letter to the editor before. I expect the reaction would have been more confused if I had alleged misconduct.

Ick. That's the kind of situation where they claim they "improperly quoted" and it's fine because they referenced the original paper, despite passing off its words as their own without any indication. Hopefully the editors don't tolerate that kind of plagiarism.

There usually is a copyright issue, yes, plus academic dishonesty. This kind of thing crops up on Retraction Watch fairly often (there's a whole category for plagiarism). I think the usual course of action is to email the editors of the journals with the evidence. The Committee on Publication Ethics even publishes a flowchart on how editors should respond to plagiarism notifications from readers. Maybe some big commercial publishers do, but none of the publishers I've submitted to do, as far as I know.