maradydd: (Default)
First, the OMGWTFBBQ. Via Pharyngula, the Hastings Center Bioethics Forum, and TIME Magazine: pediatric endocrinologists Maria New and Saroj Nimkarn are advocating prenatal treatment with the glucocorticoid dexamethasone to "reduce behavioral masculinization" of female children.

Yes, you read that right: they want to expose pregnant mothers to one of the most potent, adverse-effect-prone steroids out there in the hopes of molding unborn girls into models of femininity.

I'll give you a few minutes to find where your lower jaw rolled off to and get a glass of water -- throwing up in your mouth a little is bad for your teeth. When you get back, I'll expand a little on the current standards of practice, and then we're going to go over some organic chemistry.

Back now? Great. First, PZ got one important fact wrong: the American Academy of Pediatrics and other noteworthy medical organizations have absolutely not condoned or endorsed this practice. The "consensus" to which PZ refers is an agreement that the study of dexamethasone as a preventative for congenital adrenal hyperplasia due to 21-hydroxylase deficiency should be conducted "via IRB-approved clinical trials through research centers large enough to obtain meaningful data" and with follow-up studies. This is, in my opinion, a reasonable position. CAH gets press because one of its effects can be ambiguous genitalia, sometimes aka "intersex", but its effects on aldosterone (one of the steroids your body produces) can lead to dehydration, hyponatremia, hyperkalemia, metabolic acidosis and death, in infancy. "Ambiguous" is also used, erm, ambiguously; it doesn't only mean "large clitoris", it also includes things like the urethra and vagina opening into a common cavity and causing severe urinary tract problems.

If there is sufficient reason to believe that prenatal dexamethasone can keep children whose genes prevent them from producing 21-hydroxylase alive, or make it possible for them to avoid difficult, expensive and painful surgery to restore urinary function, that is a valid avenue for research conducted under the auspices of an institutional review board. Attempting to tweak girls' personalities to make them more girly is way, way out of bounds, and New and Nimkarn should be censured for even suggesting the idea.

But what I really want to talk about is steroids, and what you, dear reader, do and don't already know about them.

"Steroid" is a really, really broad term. It's as broad as "sugar" or "alcohol". (The categories also overlap, which can be confusing; there are sugar alcohols and steroid alcohols.) When you think of "sugar" you probably think of that grainy white stuff you put in your coffee, and when you think of alcohol you probably think of booze -- but the picture is actually much bigger. All monosaccharides and disaccharides are sugars, including the ribose and deoxyribose that form the backbone of your RNA and DNA. Ethanol is the alcohol we drink, but it's just one of the aliphatic alcohols, which also include isopropanol (rubbing alcohol), methanol (can blind or kill you if you drink it!), xylitol (used to sweeten chewing gum), mannitol (baby laxative), ethylene glycol (antifreeze!), and glycerol (aka glycerin). I won't bore you with all the various non-aliphatic alcohol families, but there are a lot of them. So, also, with steroids.

Steroids are emphatically not just what dumb jocks inject to get really ripped really fast. (Those are certain anabolic steroids.) Just as "alcohol" refers to organic molecules with an -OH bound to a carbon atom and "sugar" refers to a particular type of carbohydrate building block, "steroid" specifically means "molecule with three six-carbon rings and one five-carbon ring in a particular arrangement". (That four-ring core is called a sterane, if you were curious.) And, wow, are there ever a lot of them. Cholesterol is a steroid. So are androgens (including testosterone), estrogens (there's more than one), and progestagens (humans only have the one, progesterone). But unless you're on hormonal birth control, taking estrogen or testosterone replacements, taking progesterone as part of fertility treatment, or otherwise tweaking your own sex hormones, if your doctor prescribes you a "steroid" it is almost certainly going to be one of the corticosteroids.

Dexamethasone is, as I said above, a glucocorticoid -- a member of the family of corticosteroids that can affect immune function. (In the interest of space, I'm going to skip the other family, the mineralocorticoids.) It is, not to put too fine a point on it, the nuclear option of corticosteroids. Long-term use -- which, for glucocorticoids, means more than a week -- causes the adrenal glands to start shutting down; stopping glucocorticoids abruptly after this has happened can cause an Addisonian crisis, which can be fatal. Even long-term use as directed frequently causes Cushing's syndrome, which has a whole raft of nasty symptoms including rapid weight gain, high blood pressure, insulin resistance, severe anxiety, and psychosis. As if that weren't enough already, long-term use also causes osteopenia, a lowering of bone density that is the precursor to osteoporosis.

Given the degree of side effects involved with long-term dexamethasone usage -- and the several weeks of treatment involved in the New and Nimkarn study constitutes "long-term" -- the "behavioral masculinization" paper rolls over from "horrible" to "sheer, unrestrained evil". They are literally advocating putting pregnant women through multiple weeks of chemical torture -- not to save lives, but in pursuit of a behavioral "ideal".

If you think this is anything even remotely resembling right, I invite you to spend a month on dexamethasone -- without medication to mitigate side effects, remember we can't give benzos to pregnant mothers because they might adversely affect the fetus! -- and find out what it does to you. The stretch marks alone -- which look more like "I lost a fight with a cage full of tigers" than "boo, cellulite" -- will last a lifetime; the psychological damage from finding out just how deep your capacity for violence and self-hatred can run may fade, eventually.

All that said, there is one extremely valid prenatal use for dexamethasone. If you're about to give birth to a premature baby younger than 34 weeks, one injection of dexamethasone 24-48 hours prior to birth will help the baby's lungs produce the surfactant which it needs to be able to breathe. (Multiple doses used to be the standard, but -- big surprise -- it turns out that the beneficial effects of multiple doses are no higher, in any statistically significant sense, than of a single dose, and the adverse effects on both mother and fetus with multiple doses are worse.) Consider the difference, though: one injection versus several weeks of dosing, sharp increase in likelihood of survival versus reinforcing social norms. It's like day and night.

What it all comes down to, in the end, is this: be an informed patient. Ask questions. When you're prescribed a medication, the minimum you need to know is:
  1. What exact medication is this? Don't accept a category as an answer. You wouldn't hire a contractor who told you she was going to build your cabinets out of "wood"; you wouldn't hire a florist who told you he would make your anniversary bouquet out of "plants".
  2. How long will you be on it?
  3. What is the intended benefit of taking this medication?
  4. What are the potential or likely adverse effects for the timeframe in which you'll be on it?
  5. (if applicable) What are the potential interactions with any other prescription medications, over-the-counter medications, supplements, herbs, &c you take?

Doctors have a lot of training, and they do learn how to perform risk analysis, but at the end of the day, you are the one who gets to decide whether the potential benefits of any medication are worth the risks involved. You can't know the benefits or the risks unless you know exactly what you're putting in your body. Ask, and don't put up with bullshit non-answers.
maradydd: (Default)
Suppose you're designing a protocol, and you're deliberating over whether to use XML, YAML, JSON, s-expressions (!) or some other data representation format for it.

The question you need to ask yourself is, "have I written an EBNF definition for my protocol yet?"

If the answer is "no," drop everything you are doing. Yes, everything. Step away from the keyboard. Get a pencil and paper, or go to the whiteboard, and work out your EBNF first.

Then, and only then, you may decide what to use as a data representation. Deciding what data format to use before you have determined the grammar of your protocol is like deciding what language to use before you have figured out what algorithms you're using.

Protocol structure is a design decision. Data representation format is an implementation decision. Do not intermingle the two; it will only end in tears, yours or someone else's. Probably yours.
maradydd: (Default)
If you live in Europe, happy Labour Day!

Also: not swine flu. I note with pleasure that the overwhelming majority of respondents to my poll sensibly chose "No" (though "tickybox" has quite the crowd of supporters!), and with even more pleasure that a few of you were game-theoretically clever enough to vote "Yes" -- since, if it had been a real betting pool with real money involved, y'all would have made out like bandits if I had been so unlucky. :P Yay not Patient Zero.

Finally, if any of you happen to have photos of me in the act of Science!, and would like to see them alongside an interview with me in h+ Magazine, would you be so kind as to send them my way? The deadline is Wednesday. Since I now have my Arduinos again, I will be endeavouring to spend some time this weekend on some prototyping for the Open Thermocycler Project, but it would be nice to have something more recognisably biological.
maradydd: (Default)
Len's Principles of Programming #15: "Do not engage in edge play with the stack. The stack has no safeword."
maradydd: (Default)
This one's a first for me: an eBay phishing email in Spanish, purporting to be from, with accompanying fake Spanish website. (For the love of God, if you click on that link, do not attempt to sign in with actual eBay credentials. You probably shouldn't click on it anyway, though it doesn't appear to attempt to do anything evil other than phishing.)

COSIC has quite a few Spaniards; if I run into any of them before I head back to the States next week, I'll run the email by them and see if there are any amusing grammatical mistakes. (I will laugh my ass off if it turns out they're using, say, Mexican Spanish as opposed to Castilian Spanish.)
maradydd: (Default)
Advice to people at concerts who think it's a good idea to chatter away over the music: when the people behind you counsel you to kindly shut up or continue your conversation elsewhere, consider the fact that since they are behind you, they have a perfect shot at the tendon in the back of your knee, and if you have any interest in walking back to your car after the show, you should probably follow their suggestion and shut your fucking cakehole.

I'm just sayin', is all.

For the curious: no, I didn't actually cripple anyone tonight, much as I wanted to. I did, however, wait for the song to end, then lean over the jackass's shoulder and cheer at the top of my lungs directly into his ear. That got the point across.
maradydd: (Default)
Dear pretty much everyone,

If you try to contact me on (insert IM service of your choice here) and do not receive an immediate response, DO NOT FUCKING KEEP PINGING ME.

My status indicator does not read my mind and I typically do not bother to set it. The fact that I am set "Available" should therefore not be taken as gospel truth, either of my presence or of my availability to talk. I might have left the house. I might be in the shower. I might be fixing a meal, or eating one. I might be having hot monkey sex on the bathroom counter. I might have fallen asleep and left the laptop on.

I might be in the middle of a code binge, and if so, that has doubly unpleasant meaning for you: if I take the time to reply, it knocks me out of my groove, and if you keep pestering me, it knocks me out of my groove, so either way I want nothing more than to blow your head off with my 12-gauge. Your petulant whines of "but you weren't saying anything!" don't matter. If I do not answer after one contact attempt, assume I am not there. Maybe I'm simply not there. Maybe I'm there, but I just don't feel like talking. Or, horror of horrors, maybe I'm there but I just don't feel like talking to you. Moods change; I am generally pretty antisocial, but as long as you don't continually give me reason to be antisocial in your direction, I'll feel chatty at some point.

Again: if I don't answer after one try, assume I'm not there. For all intents and purposes, it will be true.

A proto-FAQ, for questions I expect may come up:

But what if there's something I really, really need to tell you ASAP?
Email me. The address connected to this livejournal is checked frequently -- more often than I check my cellphone's voicemail -- and you can flag it "return receipt requested" so that my ISP will let you know I've read it. If "tell you" is "get your opinion on" or "have you solve for me", well, that might take longer.

I am also not particularly bothered by one-line pieces of information (useful information -- articles you think I'd find interesting fall into this category) which do not demand an immediate reply; see But, Meredith, why do you stay connected while you're working if you hate being bothered so much? for more.

But I'm lonely!
Then go out and do something with your friends. Find a new hobby, or new people who enjoy a hobby you already have. Paint your walls. Paint somebody else's walls. Volunteer for a homeless shelter, a nursing home, a hospital, or your local parks department. If all else fails, go to the animal shelter and adopt a dog.

But Meredith, why do you stay connected while you're working if you hate being bothered so much?
Because I am a deeply antisocial person with a family and a social life. IM is one of the fastest ways for my sister to get in touch with me if one of my parents is in the hospital, or for a friend to say "hey, a bunch of us are going to the movies later, give me a call before 7 if you want to meet up".

IM is also a great source of work assistance for me. I'm deeply grateful that I can reach out to [ profile] allonymist, [ profile] yoctohedron, [ profile] cipherpunk, [ profile] ti94 et al when I need someone else's perspective on a code problem, and I happily return the favour when it's within my means to do so -- and by the same token, I leave them the hell alone when they say "can't think about that right now, too busy".

That, and sometimes I just want the person-who-works-at-home version of a water-cooler break. But that happens on my schedule, not yours.

How long should I wait to ping you again, if I've tried once and you haven't answered?
Wait about an hour and try again. If I don't answer, all the above continue to apply. Note that passive-aggressive remarks, of the "why arent u there?" or "*sigh*" variety, will decrease my likelihood of responding.

But you never want to talk to me.
Then you've probably annoyed me with your constant pestering and should leave me alone for at least a couple of days. Note that if you can see me online at all, it means I haven't blocked you and am still up for talking with you at some point, but on my schedule, not yours. See But I'm lonely!, and get a dog.
maradydd: (Default)
[ profile] cristalia has a call to arms regarding the incident at this year's Hugos wherein Harlan Ellison groped Connie Willis's breast onstage.

I'm not especially involved in the SF community anymore, mainly due to lack of time and effort on my part, but I've also heard reports from [ profile] thewronghands and other tech women about inappropriate groping/wrist-grabbing/whatever from people in the tech field. Now, I have a standard response to things like this: if someone gropes me, I punch the offending party in the face. I don't care if it's a friend, an acquaintance, or a total stranger. A friend might get the courtesy of me saying "Hey, cut that shit out," but if they keep it up, they're going to get punched. Someone at #s decided it would be funny to tickle me, and ended up getting clocked in the throat. Someone at [ profile] elegantelbow's New Year's Eve party decided it was okay to fondle the scarification piece on my right shoulder, and got an elbow in the ribs for his trouble. This kind of thing doesn't happen to me often -- perhaps I'm just not the most gropeable person in the world -- but the response by now is just instinct.

However, I know there are a lot of people out there who, for some reason or another, don't feel comfortable punching an assaulter in the face. Maybe you've never hit anybody before and think it wouldn't work. Maybe you're worried about getting hit back. Maybe you were brought up to believe that Nice Girls don't do that sort of thing and haven't trained yourself out of it yet.

Well, I am not a Nice Girl, so here is my offer: if I am in your immediate vicinity and someone gropes you, I will punch them in the face for you. All you have to do is alert me -- quickly -- to the problem and the offending party. A nice loud "$name, get your hands off my $bodypart!" should suffice. This serves two purposes: one, it lets me know who to let have it, and two, it draws attention to the asshat in question and directs the condemnation of the rest of the room straight to the offending party. Shame -- particularly shame in the heat of the moment -- is a powerful disincentive toward sexually offensive behaviour. People grope other people because they think can get away with it. Okay, yes, Harlan has drawn the ire of a large part of the SF community after the fact, but unless some form of lasting censure arises from this groundswell, he has gotten away with it. Had someone raised a ruckus at the time, he would have had to deal with a rather more acute form of embarrassment than what he's being subjected to now; the condemnation of peers who are in front of you is a lot more cutting than the condemnation of peers who are far away.

So, SF folks, do what you can after the fact; I applaud that.

But next time, someone punch the asshat in the face, okay?
maradydd: (Default)
Tonight's steaming double shot of hatred gets flung into the twin gaping maws of the American National Standards Institute and the International Standards Organization. Pull up a chair, all of you -- I may be about to rant about a programming language, but it'll make sense even to the non-geeks in the house, I promise.
So, you were saying, Ms. Furious... )


maradydd: (Default)

September 2010

12131415 161718
26 27282930  


RSS Atom

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags