The male/female dynamic to me appears to very closely mirror the adult/child dynamic and I'm not sure why more people don't frame it this way.
Such a comparison is "saying the quiet part out loud", and so it's only said explicitly by /pol/acks who don't care about optics and only intend to maximally offend. It's rhetorical suicide in the same way that saying "we want women to have the right to kill babies" would be.
I can't express these thoughts in a more coherent manner right now, so here's an array of tenuously connected musings vaguely related to the subject of women's role in society. If this sucks, let me know.
- On the whole, the concrete utility that women provide to men is sex and reproduction, their biological prerogatives. This affords them a certain intrinsic privilege, but it must be disheartening to know that much or possibly all of your value to society is fundamentally animal in nature, divorced from your sapience. The golden goose is well cared for, but it is still caged.
- The most egalitarian societies have been those where women are economically productive. If every woman refused to go to work tomorrow, what would actually happen?
- Fertility rates seem to be bimodally distributed between <1.8 TFR liberalized societies and >3.0 TFR patriarchal societies. Much concern is had over sub-replacement fertility in the West, but is is possible to increase fertility up to "just" replacement levels?
Preaching to the choir. This isn't a discussion about how evil the Dems are, and while I don't necessarily disagree with you it's irritating to see the applause lights come on with nothing to prompt them.
How do you use voting systems on social media/fora? (likes, upvotes/downvotes, etc.)
Personally, I almost only upvote/like posts when I'm sorting by newest - as that's when my upvote has the most marginal value to promote a post - and I virtually never downvote/dislike. Past the point where my individual vote has substantial value to the algorithm, I simply don't care to leave my petty display of opinion. For those of you who do use your votes, here are some questions to reflect upon:
- Are you selective with your votes or do you vote on most/all posts you see?
- Do you find yourself upvoting people you disagree with due to the quality of their argument, or vice versa?
- Do you downvote people you're arguing with or do you leave judgement entirely to the masses?
- Do you remove the auto self-upvote on your posts/comments?
(This is a repost of my comment from last weeks thread, hopefully it gets seen this time around)
So, I'm at college now. After dozens of minutes spent researching schools, many sleepless nights spent putting off homework, and endless effort spent on not giving a shit, I'm here, and I'm...not disappointed in myself, exactly, but I absolutely could have done better.
As I alluded to, I didn't really care all that much about being attractive to colleges in high school, but I now regret it, at least a bit. I had a top percentile SAT and plenty of AP and dual enrollment credits, but my lack of extracurriculars and thoroughly mediocre GPA sunk my application to the point where I only truly got into one of the five schools I bothered to apply to. While I didn't really have a "dream school", my current university is on its face significantly lower-ranked than my top pick; by median SAT scores, my preferred school is about 200 points higher than my current school with a particularly prestigious CS program (my major) to boot.
Regarding my current situation, I ask a couple questions of the Mottizen public:
1.] How important is your alma mater for job opportunities with a CS degree? While I want to transfer to my preferred school for a variety of pragmatic and personal reasons, I do have a not-insignificant scholarship at my current institution. It wouldn't ruin me financially to forgo it as my college fund should cover the brunt of it, but it is a counterbalancing factor. (N.B: I have now seen this thread partially answering this question. I'd still like to know how much it matters for CS specifically.)
2.] What are job prospects for a CS degree looking like in the next 5-10 years? I've heard that the CS bubble has popped and I'm fine with not having a junior position handed to me on a silver platter, but I wouldn't mind switching tracks to a different engineering degree (and consequently removing much of my incentive to transfer) if CS is headed for the shitcan thanks to AI/oversaturation/whatever.
So, I'm at college now. After dozens of minutes spent researching schools, many sleepless nights spent putting off homework, and endless effort spent on not giving a shit, I'm here, and I'm...not disappointed in myself, exactly, but I absolutely could have done better.
As I alluded to, I didn't really care all that much about being attractive to colleges in high school, but I now regret it, at least a bit. I had a 99th percentile SAT and plenty of AP and dual enrollment credits, but my lack of extracurriculars and thoroughly mediocre GPA sunk my application to the point where I only truly got into one of the five schools I bothered to apply to. While I didn't really have a "dream school", my current university is on its face significantly lower-ranked than my top pick; by median SAT scores, my preferred school is about 15 percentile points higher than my current school with a particularly prestigious CS program (my major) to boot.
Regarding my current situation, I ask a couple questions of the Mottizen public:
1.] How important is your alma mater for job opportunities with a CS degree? While I want to transfer to my preferred school for a variety of pragmatic and personal reasons, I do have a not-insignificant scholarship at my current institution. It wouldn't ruin me financially to forgo it as my college fund should cover the brunt of it, but it is a counterbalancing factor. (N.B: I have now seen this thread partially answering this question. I'd still like to know how much it matters for CS specifically.)
2.] What are job prospects for a CS degree looking like in the next 5-10 years? I've heard that the CS bubble has popped and I'm fine with not having a junior position handed to me on a silver platter, but I wouldn't mind switching tracks to a different engineering degree (and consequently removing much of my incentive to transfer) if CS is headed for the shitcan thanks to AI/oversaturation/whatever.
Funnily enough, I did just that a couple of months ago. I was trying to find someone's old reddit comment and happened to notice one of their comments on /r/themotte. My curiosity was piqued, and here I am now.
(Bit mundane for a first post, I promise I'm working on a couple more substantial contributions)
- Prev
- Next
My main concern isn't whether a ~2.0 TFR is attainable, it's whether it's sustainable. The reactionary route of mostly/entirely restricting women to the homemaker role seems to result in fertility rates higher than really needed, and all other approaches seem to converge on sub-replacement fertility. For as much as we've avoided Malthusian collapse, the prospect of population growth outpacing productivity is theoretically sound, and I don't want to push our luck much further.
For nearly all of human history, populations were kept stable despite TFRs of 4.5+ by massive infant mortality, appreciable maternal mortality, and more death in general. The social technologies that ensured fertility was kept that high are now mostly unneeded, potentially harmful, and crippled by the Pandora's box of contraception. In their absence, the paradigms that have emerged haven't been any more adaptive, to say the least. I have no idea what the optimal arrangement for fertility in industrial society might be, but I'd bet it won't be as simple as retvurning to tradition.
More options
Context Copy link