|
Chalks posted:Surely using any physical attribute to estimate based on statistical averages has the same issue? I'm a little surprised that there's no distinction between "sex" and "gender" in a medical scenario, but I guess recognition of that sort of thing is more recent than many ancient systems big organisations like that would be working on. you misunderstand. this is my entire point. sex/gender is not a physical attribute. neither has any real bearing. sex has correlations with size, anatomy, hormones, etc, but it is not a guaranteed predictor. you're performing medicine-by-proxy when using sex or gender for these calculations. a big part of this push is moving people toward using the actual data they want (size, serum hormone levels, anatomy, blood pressure, etc). there's only a handful of scenarios where the sex is clinically relevant: *you are doing something to the actual genitals *you are prescribing a medication that interacts differently with XX people vs XY people *you specialize in care for intersex people *you specialize in care for transgender people in most other circumstances, there's better ways to get the data you want. for example, traditionally they have male and female ranges for kidney function tests, but while these are statistically correlated, what they've found is that the correlation was down to birth sex being a proxy for weight and so the tests were inaccurate for larger women and smaller men. outside of the specific ways that certain drugs interact with estrogen vs testosterone, and the fact that uterus-havers get pregnant and have periods, the differences between the sexes are mostly cosmetic (yes, i'm classing things like shape and size of bones as cosmetic here). consequently, because programmers generally don't know poo poo about this kind of thing your average emr (us included up until now) basically codified a lot of bad habits of medicine by using the patient's sex for all sorts of clinical decisions. Worse still, we didn't have any standardized way to store the gender identity, or to hide a patient's legal name if they had a preferred name, which both meant lots of embarrassing honest mistakes where patients were addressed wrong, and also giving information to malicious people who then used it wrong (like in the article I linked above). basically, if you're developing software for a particular field you drat well better understand the subject matter in depth or you're going to get something wrong and her people. we had a workflow where you couldn't admit someone into the delivery ward if they were male, which is patently absurd because transmen can get pregnant even on T if they still have their uteruses. why would the software ever need to check someone's sex on admission for delivery? surely if someone is having a baby, this will be obvious to the doctors and you don't need an advisory from clippy going "hey, did you check whether this babyhaver is a lady?". sorry if this is a lot of moderately off-topic information. i have a lot of pet-peeves in this area because as an out transwoman everyone and their loving brother just treats me like im supposed to have the answers to all their questions when they need to do development related to sex. just because im trans doesnt mean im obligated to be ur educator ppl 
|
#
?
Nov 30, 2016 21:04
|
|
|
|
| # ? May 23, 2024 11:10 |
|
|
it's cool seeing people talk about using fpgas for poo poo because i work at a place that makes them
|
|
#
?
Nov 30, 2016 21:12
|
|
|
HappyHippo posted:it's cool seeing people talk about using fpgas for poo poo because i work at a place that makes them altera/intel, xilinx, or one of the randos also why are your tools so bad
|
|
#
?
Nov 30, 2016 21:13
|
|
|
Shaggar posted:that's the one thing that kind of sucks about dapper is if you want to customize this poo poo you have to build custom mapper implementations (idr the name of the interface) and you have to do it basically per class. there are no attributes to assist the mapper. Yeah in an ideal universe I'd change the name of the column in the database but that would require going back to approx. 1997 to not break anything Thanks for the links on how to do custom mappers, if that thing looks too much like a lot of loving faff I'll just change the name of the entity class to... i dunno, MnemonicE or something 
|
|
#
?
Nov 30, 2016 21:14
|
|
|
c tp s: accidentally looping from 0x10000 to 0x11111 when i meant to loop from 0x10 to 0x1f
|
|
#
?
Nov 30, 2016 21:19
|
|
|
LeftistMuslimObama posted:the two genders are: maybe
|
|
#
?
Nov 30, 2016 21:20
|
|
|
LeftistMuslimObama posted:the genders are:
|
|
#
?
Nov 30, 2016 21:21
|
|
|
just thought about the ~2300 line shell script that gets invoked by jcl that i'm partially responsible for maintaining and lol'd a little how goes programming the computers?
|
|
#
?
Nov 30, 2016 21:25
|
|
|
LeftistMuslimObama posted:you misunderstand. this is my entire point. sex/gender is not a physical attribute. neither has any real bearing. sex has correlations with size, anatomy, hormones, etc, but it is not a guaranteed predictor. you're performing medicine-by-proxy when using sex or gender for these calculations. a big part of this push is moving people toward using the actual data they want (size, serum hormone levels, anatomy, blood pressure, etc). Sorry, I mean I assume there's a medical term for the generic attributes of I referred to as "sex" (such that would be useful in diagnosis) and I just figured that medical systems would use more medicalised terminology. And sorry if you thought I was interrogating you, I was just expressing surprise that a medical system wasn't using medical terminology from day one. I've always imagined that such systems would use all sorts of specific terminology to avoid ambiguity. I obviously understand that statistical information obtained from basic attributes like age/race/sex aren't guaranteed predictors, but often more general information will be used prior to doing the physical testing to establish the real data, as you say.
|
|
#
?
Nov 30, 2016 21:28
|
|
|
Bloody posted:altera/intel, xilinx, or one of the randos altera is now intel my friend hard to say. my guess would be that traditionally fpgas competed against asics, and the usability of the tooling doesn't really factor much into the math when deciding between those options, so fpga vendors never really prioritized it. but maybe there's some other reason.
|
|
#
?
Nov 30, 2016 21:29
|
|
|
carry on then posted:just thought about the ~2300 line shell script that gets invoked by jcl that i'm partially responsible for maintaining and lol'd a little bad. i have two theoretically identical hardware modules that are mostly behaving identically except for in a few very reproducible cases
|
|
#
?
Nov 30, 2016 21:30
|
|
|
HappyHippo posted:altera is now intel my friend my pet theory is that hardware engineers dont know how to make software so the associated software is bad. it infects board layout and schematic capture heavily too
|
|
#
?
Nov 30, 2016 21:31
|
|
|
Bloody posted:my pet theory is that hardware engineers dont know how to make software so the associated software is bad. it infects board layout and schematic capture heavily too well the people working on the software are software engineers. i think they're more concerned with the guts though. stuff like placement and routing are very much nontrivial, and they affect the resulting metrics that do factor into the decisions to pick fgpas over the alternatives.
|
|
#
?
Nov 30, 2016 21:39
|
|
|
Bloody posted:my pet theory is that hardware engineers dont know how to make software so the associated software is bad. it infects board layout and schematic capture heavily too
|
|
#
?
Nov 30, 2016 21:42
|
|
|
Ciaphas posted:Yeah in an ideal universe I'd change the name of the column in the database but that would require going back to approx. 1997 to not break anything you could also use a view or a proc and alias the column name. I use procs so that's how i'd probably handle an old thing I couldn't change. theres some helpful info here: http://stackoverflow.com/questions/8902674/manually-map-column-names-with-class-properties on how to use the customtypemapper or how to write your own.
|
|
#
?
Nov 30, 2016 21:44
|
|
|
Chalks posted:Sorry, I mean I assume there's a medical term for the generic attributes of I referred to as "sex" (such that would be useful in diagnosis) and I just figured that medical systems would use more medicalised terminology. wasn't yelling at you in particular. more my idiot coworkers. but basically the whole problem is that the generic attributes you refer to as "sex" are actually not as cleanly divided as you would think (intersex people to start with, but there's a million other things that make any kind of strict definition of "birth sex" dumb). everyone assumed for a long time that there was a good definition of sex, and a whole lot of medicine was built around it. the healthcare it world is drastically behind the times on fixing this (in fact the elevator story for the companywide project to fix this stuff says that this positions us as the market leader in sex and gender related healthcare). in fact, the sex that goes on your birth certificate (what republicans would like to call your "genetic sex" or "real sex" or whatever) is just decided by the doctor looking at the freshly-born baby and going "it has a dick, that's a boy" or "it has a vagina, that's a girl". if the genitalia are ambiguous, they generally pressure the parents into operations to make them look like one sex or the other, then they pick that one. they don't do genetic testing or anything. there are plenty of XX people out there who had "male" put on their birth certificate and vice-versa. and for intersex people they just slap on whichever is closest. essentially, the "medicine" related to sex is bad from the very beginning, so the medical community is, by-and-large, trying to get away from its use for clinical decision making. to get grossly personal, i have a dick and under some definitions that makes me "male", but i also have almost no testosterone in my blood and lots of estrogen, which under other definitions makes me "female". which one is accurate comes down to how you want to identify me, pretty much, and the same is true about how i was identified at birth. so like i said, if you're programming in something domain-specific, make sure you know your poo poo inside-and-out before you start coding. being good at writing code is like 5% of what is involved in being a good software engineer, tops. if the space shuttle was programmed the way healthcare software was every single loving shuttle would blow up before it got to the launch pad 
|
|
#
?
Nov 30, 2016 21:46
|
|
|
LeftistMuslimObama posted:doctors hate every emr because they resent actually being held accountable for doing the right thing and documenting proof that they did the right thing. doctors wish they could operate more like the nypd. anyone who is forced to use a piece of software to do their job is going to hate that software regardless of its quality or merit. you can't take it personally.
|
|
#
?
Nov 30, 2016 21:46
|
|
|
Dr Monkeysee posted:anyone who is forced to use a piece of software to do their job is going to hate that software regardless of its quality or merit. you can't take it personally. basically. imagine how eschaton or any of the vs/.net devs on this forum feel on a day-to-day basis the way people bitch endlessly about their products. i could probably write a utility to categorize your porn hoard and someone would find some way to resent the ability to tag their movies as both "bbw" and "ass2mouth".
|
|
#
?
Nov 30, 2016 21:49
|
|
|
actually i really like visual studio
|
|
#
?
Nov 30, 2016 21:48
|
|
|
Shaggar posted:you could also use a view or a proc and alias the column name. I use procs so that's how i'd probably handle an old thing I couldn't change. theres some helpful info here: http://stackoverflow.com/questions/8902674/manually-map-column-names-with-class-properties on how to use the customtypemapper or how to write your own. My SQLfu (well really database-fu) is weak so I didn't even think of stored procs. Part of why I'm loving around with dapper, get a little closer and more experienced with DB stuff than EF lets me get Thanks (ed) Oh yeah SQL lets you rename results if you're not using * doesn't it. poo poo. that would have been the easiest answer of all, just rename the MNEMONIC column to NAME in the query i'm so terrible!!!!!! Ciaphas fucked around with this message at 21:51 on Nov 30, 2016 |
|
#
?
Nov 30, 2016 21:49
|
|
|
Bloody posted:actually i really like visual studio
|
|
#
?
Nov 30, 2016 21:49
|
|
|
LeftistMuslimObama posted:lol. i was at a go-live once and had an er doctor say straight to my face "i shouldnt have to explain to anyone why i do what i do, and ill be damned if i waste time checking boxes in some machine". then he got fired a month later because it turns out he had a way higher complication rate than any of the other doctors. also it's been my experience that every time someone loudly proclaims they don't need any oversight it's because they're actually terrified of oversight cuz they secretly know they're bad at their job.
|
|
#
?
Nov 30, 2016 21:49
|
|
|
LeftistMuslimObama posted:you misunderstand. this is my entire point. sex/gender is not a physical attribute. neither has any real bearing. sex has correlations with size, anatomy, hormones, etc, but it is not a guaranteed predictor. you're performing medicine-by-proxy when using sex or gender for these calculations. a big part of this push is moving people toward using the actual data they want (size, serum hormone levels, anatomy, blood pressure, etc). I would think a lot of these new correlations were discovered from EMR usage and not previously well understood prior to heavy EMR adoption. Without the huge amounts of clinical data generated by EMRs it would be a lot of guess work trying to figure out what the real measures involved in an outcome are. tbh that's the thing I like the most about EMRs. every provider-patient interaction is a clinical trial.
|
|
#
?
Nov 30, 2016 21:57
|
|
|
Ciaphas posted:My SQLfu (well really database-fu) is weak so I didn't even think of stored procs. Part of why I'm loving around with dapper, get a little closer and more experienced with DB stuff than EF lets me get oh jesus. don't ever use select * in your code. also don't ever use entityframework cause its trash
|
|
#
?
Nov 30, 2016 21:59
|
|
|
select * is for when youre screwing around in a db you're not familiar with, never use it for anything else
|
|
#
?
Nov 30, 2016 22:01
|
|
|
Dr Monkeysee posted:anyone who is forced to use a piece of software to do their job is going to hate that software regardless of its quality or merit. you can't take it personally. This is an anonimized piece of feedback my team has received a couple of years ago after delivering some major feature: valued customer, 2014 posted:Com'on guys! Do you really have a routing team!? WTF are you doing!?!? Your number 1 problem is this random routing... It's been a shining star in the work we've done. I still don't know who that guy is or who he worked for as it was sent fairly anonymously to my team's public feedback email address, but I hold it dear.
|
|
#
?
Nov 30, 2016 22:04
|
|
|
MononcQc posted:This is an anonimized piece of feedback my team has received a couple of years ago after delivering some major feature: it's from rapgenius
|
|
#
?
Nov 30, 2016 22:07
|
|
|
like almost certainly http://genius.com/James-somers-herokus-ugly-secret-annotated or at least some genius fanboy
|
|
#
?
Nov 30, 2016 22:08
|
|
|
Bloody posted:c tp s: accidentally looping from 0x10000 to 0x11111 when i meant to loop from 0x10 to 0x1f same but writing 0x00007200 instead of 0x7200 into a memory-mapped graphics register, setting the next register over instead
|
|
#
?
Nov 30, 2016 22:09
|
|
|
Bloody posted:it's from rapgenius it isn't. Rap Genius have dedicated support people to talk to them and they won't send lovely feedback through some email addresses to specific teams.
|
|
#
?
Nov 30, 2016 22:17
|
|
|
LeftistMuslimObama posted:wasn't yelling at you in particular. more my idiot coworkers. It's cool, I really appreciate you taking the time to respond to my dumb posts I feel like I have more of an understanding of the issues now so maybe if I end up doing any work in that sort of field I'll be able to avoid being terrible! Bloody posted:actually i really like visual studio I absolutely adore visual studio (although performance and stability wise it's on a bit of a downward slope in recent versions) We use azure and visual studio team services at work and the integration is cool as hell. A really great IDE all over.
|
|
#
?
Nov 30, 2016 22:20
|
|
|
LeftistMuslimObama posted:so like i said, if you're programming in something domain-specific, make sure you know your poo poo inside-and-out before you start coding. being good at writing code is like 5% of what is involved in being a good software engineer, tops. do you have any interest in writing a proper article on the topic of correct database normalisation of this sort of information? I mean you basically already have done this, minus some capitalisation I find this to be a fairly important point because as software developers we basically control the interface between normal people and everything important that they ever want to do in life, and simple little assumptions and simplifications can make people's real lives unimaginably difficult
|
|
#
?
Nov 30, 2016 22:33
|
|
|
qntm posted:do you have any interest in writing a proper article on the topic of correct database normalisation of this sort of information? I mean you basically already have done this, minus some capitalisation sadly, i think doing that would violate my nda. 
|
|
#
?
Nov 30, 2016 22:36
|
|
|
well it wouldn't need to mention specifics or anything, just generic advice for correctly handling gender/sex/etc. in a database context
|
|
#
?
Nov 30, 2016 22:39
|
|
|
i use eclipse, because my company created eclipse and is so married to it it's kind of amusing i also don't like it by and large
|
|
#
?
Nov 30, 2016 22:39
|
|
|
the cloud keeps getting more ridiculous
|
|
#
?
Nov 30, 2016 22:45
|
|
|
why on earth is postgres taking 39 seconds to return all 30k rows in this table that seems mindbogglingly slow
|
|
#
?
Nov 30, 2016 22:45
|
|
|
redleader posted:my butt keeps getting more ridiculous a truck full of drives speeding down the highway has good bandwidth
|
|
#
?
Nov 30, 2016 22:48
|
|
|
Bloody posted:why on earth is postgres taking 39 seconds to return all 30k rows in this table that seems mindbogglingly slow ah. right. mcafee is doing something on my computer so everything is melting down and nothing will work until mcafee is sated
|
|
#
?
Nov 30, 2016 22:52
|
|
|
|
| # ? May 23, 2024 11:10 |
|
|
qntm posted:well it wouldn't need to mention specifics or anything, just generic advice for correctly handling gender/sex/etc. in a database context is it talking about software in a medical context? me nda covers it. i signed that poo poo back when i was signing up for a tech comm job that doubled my poverty cj income. im not super interested in testing it now though. let me talk to my boss about it tho. do you have like a blog or something you wanted to post it to? i think the fear is that our competitors might read it and benefit from knowledge we developed, silly as that sounds. our management are a paranoid bunch.
|
|
#
?
Nov 30, 2016 22:52
|
|