I cleaned 6 women's bathrooms in a college dorm, 5 days a week for a couple of years in college.
I found their toilet habits to be pretty good. The sinks and the shower were hairy, and the sinks often had a lot of makeup residue.
They also had a habit of puking in the sinks instead of the toilet, which was much harder to clean because you had to scrape the puke out since the sinks had a fairly fine steel mesh over the drains.
This was in the 1980s at the peak of the Bartles & James wine cooler phenomenon, so their trash on Mondays also smelled like sickly sweet vinegar.
The one weird day was trying to finish a bathroom when a girl walks in drunk (at about 3 PM) and says she needs to shower. I'm just finishing the last sink and only need to mop the floor, I ask if she can wait about 5 minutes or go downstairs (which was like a total distance of less than 40 feet). Since she's drunk, she's chatty and also doesn't care her robe isn't staying closed. I'm trying like hell to get out of there before someone sees what's happening because I worry the story will be "pervert janitor".
I'm basically mopping my way out and she says she's gonna shower now, so off comes the robe. She reaches in and starts the shower, turns and says bye, and then gets into the shower.
I ran into her a week later and she laughed about the shower, but she also turned out to be kind of a tragic figure who unloaded some pretty ugly sex assault stories on me.
FWIW, I wish RFC1918 had included a couple of weird and unappealing "isolated"/24s which would have gotten less use than 192.168.0.0/16 and 10.0.0.0/8 or even 172.16.0.0 (which seems to be the least used in my experience).
These lone/24s would be have been ideal to break up for interior interfaces or for use on isolated management networks that can't overlap with other interfaces.
But hasn't it kind of worked that way? YouTube is filled with people I've never heard of who get hundreds of thousands if not millions of views, so the idea of putting together a YouTube channel and making it big doesn't seem totally unrealistic because a fair number of people have actually done it.
I also don't really follow the how you get paid part, but I'm pretty sure that YouTube did kind of arbitrarily change monetization rules kind of recently which cut the income of a fair number of people with large-ish views.
I can't tell if this was YT's plan all along -- let anything but porn on, make monetization generous, and once "being a YouTube personality" hit some critical mass, start keeping the money for small-time personalities now that the concept of self-promoted YT channels has its own momentum.
It should have been an obvious existential risk for anyone using YT as a platform for income because obviously YT can change the rules and claim your earnings anytime.
I feel like we do a marginal job on enforcing project scope. The actual engineers hate to tell clients when something is legitimately out of scope because we bear the brunt for bad project planning (suddenly we're tainted by it). PMs often can't communicate the technical reasons why X is out of scope, and the sales people just want to give it away to keep the customer happy.
But I'm not sure this is a PM problem than a general capitalism problem of contract scope and enforcement in a dynamic environment.
I think generally speaking marketing tends to decrease transparency and increase obfuscation as new product performance shows fewer gains.
If new products had better general performance than old products, product naming would tend to track the performance improvements as was seen in the Mhz era.
But as producers run out of simple performance improvements, they have to start trying to create a new demand profile, often one that's kind of made up and not representative of actual performance.
IMHO, most capitalists prefer lack of product transparency because it usually forces buyers to buy more than they wanted to avoid the problem of not buying enough.
I have worked with PMI certified managers and its been a mixed bag. The best one was probably not going to "get very far' because he wasn't "management material" but he kept projects going and dealt with stall issues better than the other two. The ideal project manager from a project perspective.
The worst one was just a resume padder and a schmoozer who didn't track projects well and often added out of scope work because it helped her schmoozing.
The current one is pretty good at project management, but is often too invested in embedding herself in organizational decsionmaking which causes her to detatch from project details. This is acceptable on smaller projects, but on larger projects it makes her seem uninformed.
If I was just screening Americans for a position today, I'd have to worry about a bunch of vendor certifications and whether they were useful or just empty treadmilling, past experience, and possibly educational background (which grows less relevant with experience). That's a whole minefield of bullshit, and various levels of lying about skills and experience, some of which is standard puffery, some of which is deliberate dishonesty.
With foreign workers? I have no good idea on how to vet them. I feel like everyone involved in foreign employment has an incentive to lie and very little way to have their statements verified. Plus employers are usually willing to believe what they're told because these employees are cheaper and I'm sure they been told of their compliance and work ethic.
American companies have already zeroed out their cost for training for the most part anyway. They demand employees with years of experience and routinely dump anyone who doesn't have whatever skill they look for this week.
You laugh, but people will probably figure out how to do "voice ai squatting" on Amazon, selling bogus "combination" products that match this (ie, some product that is called "milk eggs cheese" but isn't any of them).
There's so many ways storage and backups and could be leveraged to mitigate this, although I would imagine for most sites with Windows it would revolve around existing storage platform snapshots or VM backups to offline storage or in isolated security realms, not "zfs" by itself.
I have clients that run hourly incrementals of all VMs and their backup runs in an isolated security realm (firewalled, some physical network isolation and no shared or trust-related Windows security relationship).
I really wish we could stop with foods being either "good" or "bad" for you. My guess is even if you actually get the science to say if something is good or bad, the chances are that it's really only very marginally good or bad for you at reasonable/non-OCD intake levels, not so good or bad that it will swing the health of a normal person.
Even foods/beverages that are demonstrably good or bad for you aren't either in very small amounts. Sugar isn't good for you, but if I ate a glazed donut once a year? It's not going to change anything.
I'm sure there's some marginal value in looking at high-volume consumption foods like coffee, but at this point people have been drinking it for a couple of centuries and tons of it over the last century and we don't have a plague of people dying from coffee poisoning.
Other than the obvious lack of utility for "good' and "bad" labels, all it does is encourage people to over-consume "good" foods, needlessly avoid "bad" foods, all magnified by a marketing tsunami of food companies touting their products as beneficial.
I would bet in Atlanta the IT department is a bunch of people hired for their race or connections. I've seen this in several government IT offices in places way better managed than Atlanta.
The relatively high wages of IT and the "good career prospects" make it a tempting spot to place associates of politically influential leaders. The backwater nature of most small government unit IT contributes, too, as higher ups tend to see them as safe jobs to give away because they're generally not unionized and the (putative) required skill set makes justifying arbitrary hires pretty easy. The lower wages by corporate pay scale standards also means there's less outside competition if they're actually posting the jobs.
Usually there's a handful of legitimate hires who manage to keep the lights on, but they're often lifers with few other options, too, or there's a major political split along race or some other dangerous line.
I did a project once and of the 6 "admins", 3 were Asian and 1 spoke English so poorly his coworker literally translated from English for him. The other 3 were Caucasians who fought with the Asians at every turn and I think actively tried to subvert them with intentional technical screwups.
I think this is right. The worse the loan terms are, the more desperate the borrower will be to eliminate the debt.
But to be honest, I bet there's a non-zero percentage of people looking at this as a great source of speculation capital on something that looks no-lose.
The whole student loan situation is a total mess. The universities push up tuition and expand their bureaucratic fiefdoms, dorms become luxury condos, and idiot students self-indulge in luxury living.
Meanwhile, all this debt accrues and its insulation from ordinary bankruptcy probably has it over rated as a security, meaning its used as leverage for much more and riskier debt, just like mortgages. And we all knew how that went.
Plus, student loans are just a giant drag on the economy. The young people starting to earn adult wages are crippled by debt and spend much less on economic growth activities. Houses don't get remodeled, old cars get driven, fewer trips, less total consumer spending -- all those activities pump up the economy and have a multiplier effect. Student debt service seems like a tax.
I keep thinking a student debt crisis will happen, but it never does.
I feel like Reddit gets a worse rap than it deserves. The top/big/default subs are often filled with shitposts, but this is a tiny fraction of the total subs, most of which I fin to be generally spam/troll free and with reasonable dialogue. And there are an almost limitless number of subs (or that's how it seems).
In most way it reminds me a very decent re-imagining of USENET as a website.
My criticisms of it are more that you can't mark subs or threads read and then go back and see only new posts or comments, which makes following large posts and their replies troublesome.
Facebook has been able to spend its way out of some competing social media trends (ie, buying Instagram) and somehow buck others they couldn't (Snapchat), but mostly they were negotiating from a position of strength due to their network effect.
Now that their actual business model is exposed -- "You tell me, I sell you" -- and they're facing real risks of large-scale disaffection or defection to other platforms, of course they're fans of regulation. Broad social media regulations to keep their existing competition in check and to keep out the next big thing which might turn them into MySpace.
I haven't been contacted by any of my financial institutions about their adoption of new blockchain based technologies to manage any of my accounts or transactions despite all its inherent advantages as a ledger alone -- I'm not even talking about the currency component.
It's a lot of hype and not a lot of anything else yet. It's just a way to hype stocks or products, not anything providing real utility yet.
Tyler Cowen, an economist, wrote an interesting paper about how we've hit the great stagnation. All the low-hanging fruit of economic innovation has been realized, at least in the US. Women participate in the workforce, we've mostly maximized the productivity enhancements of technology (assembly lines, basic robotics, office automation), we've tapped into (and in some cases, nearly tapped out) cheap and simple energy sources, and so on.
It's not to say there aren't marginal improvements, but they come with such large marginal costs that it's debatable whether they produce real productivity improvements and certainly not the productivity improvements their previous generations did when they were adopted.
Office computers are a great example, when I started working in a large office there was no email or calendaring and there was a ration of about 4-5 workers for every 1 support worker (principal job to deal with the administrivia of the worker -- scheduling, typing, copying, etc). That ratio went to something like 20:1 or smaller 5 years after email and I'd say a good chunk of what was left was semi-symbolic for senior executives or more like a personal assistant for a specific group capable of contributing in some small way to the work product itself, not just the coordination.
But since then (last 20 years, give or take) there haven't been any real increases in productivity other than small marginal increases in capability -- bigger documents, more storage, etc, but nobody is really more productive and the support costs of the technology have increased with the dependence on it.
In terms of computing technology, I think it really has stagnated in the last 5 years. Virtualization and SSDs seem to be the last big breakthrough technologies that actually improved productivity and those appear in many ways to have hit productivity improvement ceilings. Virtualization seems to be getting more complex mostly on a feature-addition basis and I've talked to customers who have an entire cluster deployed for management tools for their...cluster.
I think you can really see it in the prices of 10+ gb ethernet and SSDs. SSDs have gotten a lot cheaper but I'm still seeing prices in the thousands for enterprise SSDs. It took less than 5 years for ethernet to go from 10/100 to ubiquitous 1 gb. It's taking way longer for 10+ gb to get adopted because the vendors have held the line on pricing for that and enterprise SSD because they have nothing else to offer. It's the end of the line.
Sure, there are ways to max them out further, but now you're back to huge marginal implementation costs -- new cable plants for 10 gb, massive bus reorganization for improved flash performance, all with productivity enhancements that may not even cover their marginal costs.
Software and operating systems seem stuck in stylistic and rent-seeking iterations without even claimed productivity enhancements. It's all driven by vendor profit enhancements, tracking, advertising, not productivity utility. A brand new PC/OS with a flash disk doesn't deliver anymore average productivity than a 5 year old model and the actual worker productivity itself probably hasn't improved in 10-20 years.
We're on our way to making the area outside a gated community look just like it does in India, so I assume this will make the people here love that gated living just as much.
I can just imagine a laundry list of reasons, topped with the catch-all of "poor management".
You probably need a scoring system to derive a cash-strap score. Add 1 for being in a large urban area with increased operating costs for labor and supplies. Add 1 for leadership dominated by academics from non-management fields. Add 2 for ill-advised expansion plans (new buildings/facilities, etc) predicated on rising tuition *and* enrollment. Add 2 for financial planning which does not properly account for tuition discounting designed to boost enrollment. Add 2 for private colleges specializing in niche or narrow areas with limited economic potential (a music college would seem to be one of them).
Around here anyway, most of the small liberal arts colleges seem to be doing well even with sky-high tuition. Most people don't pay the full freight but general management seems good and many are in semi-rural locations where costs are lower. They also have something of an exclusivity component, only offer undergrad degrees but serving as well-known feeders to elite graduate school programs. Pretty much everyone I know who went to one also holds a grad school degree. And super loyal alumni who have helped build up giant endowments which smooth income and allow for cheap capital expansion funding (combined with highly sensible capital expansion).
My guess is that the Chinese won't fare well here. They could be duped by general trends of increasing tuition in the US and college attendance numbers into thinking it's a no-lose proposition. It may be a trend based solely on duping other Chinese in China -- I'd wager there's some status attached to US degree, so if you can become a semi-legitimate diploma mill (ie, your students actually come to the US and the school holds accreditation) you may end up with a limitless supply of Chinese hoping for a cut-rate advantage back home where among provincials a re-worked Harvard logo that says "Miss Smith's Secretarial School" carries the same weight as the real thing.
I'll also add -- how will self-driving cars change any of this equation?
It's not going to happen overnight -- hell, Uber just killed a pedestrian! -- but it does seem like we're on the cusp of a self-driving cars. It seems like that would radically change the equation, making it really practical to own at least fewer cars if not for a lot of people to not own one at all.
My gripe with this always is that the changes to make driving extremely unbearable happen quickly -- in a matter of less than a decade, I've seen the carrying capacity of a dozen arterial local streets reduced by half in favor of traffic calming and bike lines.
Yet the so-called improvements in public transport happen over MANY decades, and only in the smallest of increments. IMHO there are obvious ways to make bus lines faster (semi-express routes, where the bus stops every 5 blocks, not every block) but then there's all the litany of complaints about old people, handicapped, etc, who would be inconvenienced so these never happen. And it's not like the housing pattern changes are going to happen in my lifetime. I'm not ever going to move to a train accessible area and buses will never improve in my close-in area, and the market doesn't care, my in-city house has doubled in value in 20 years.
Plus transit officials here are crazy for light rail, which is great if you can build it fast, but it's a few billion dollars per line, the lines take 5 years build once they even start digging, and the most recent line is delayed like 5 years because of pissing and moaning over the right of way (which manages to avoid by 1.5 miles a long-term dense urban housing area).
The net effect is that bus improvements are forgotten about (a slow, tedious route for poor and old people with time on their hands?) because transit officials are salivating over trains and zero progress is made in balancing the reduced ability of cars to move around.
So much of this debate is unrealistic in the US. On one hand, the anti-car faction has a fantasy that every city can be transformed into downtown Copenhagen in 5 years. We'll all live happily in flats and townhouses and take trams and trains everywhere and not own cars. The super pro-car faction wants 10 lane highways everywhere and 1,000 space lots for every new building. Neither side acknowledges the other at all. European cities evolved dense over 1,000 years and in the 20th century farmed their poor out to tower blocks in the suburbs. American cities started dense and then walked away from them in the 1940s and 1950s for suburbs, leaving their poor behind in ghettos.
I think a big part of traffic problems is that urban planners have become ideologically opposed to cars and have begun to array urban planning tools against cars to make driving difficult. We get "traffic calming" which translates as lanes removed and parking removed in favor of dedicated bike lanes (it's also snowy and below freezing about 4 months out of the year).
The hope is people will find driving so difficult they give up cars for bikes or transit without considering that both are a poor substitute for cars in many cases -- distance, poor transit systems, weather, need to carry packages, etc.
I'll grant them that suburban car-centric planning is a disaster, but mostly I consider it just pseudo-planning. To this day there are suburban shopping areas where it's like 5 large tenants built buildings and lots and whatever adjoining space was left becomes a "road" which results in absurdities like requiring 4 left turns to get anywhere.
I just figure there has to be a middle way that's not so anti-car it makes things impossible but not so pro car you wind up with a wasteland of roads.
That's all nice reasoning and I'm sure something like it showed up in a presentation at Microsoft when they pitched this.
But the real reason is of course that top executives are demanding improvements in Edge's usage numbers. They've tried PR, they've done the battery life "Pepsi challenge", they've reset preferences during major updates, they have nag screens reminding you that entire villages in India rely on Edge for clean water and medicine, they even made it better at searching for porn.
But still nothing moves the needle quite like algorithmically jamming it down people's throats with no other option. Microsoft badly wants to close the door on Windows, forcing users into a closed system of rented (or ad supported) desktop operating systems, email, cloud services, and so on. They want to be Apple but they consistently play to their own internal needs (backstopping failing internal projects or money hoses like Office) with Windows when they could be enhancing the user experience.
The DOJ should have broken MS up. An operating system company and an applications company. Both would have had much better incentives to improve for the sake of users.
I'll bet the real savings was in shipping, and not just the obvious local delivery (bottler to retailer) aspect. I'll bet that the savings might have been enough to eliminate entire bottling plants that serviced areas once too distance to ship glass bottles to.
I just worked on a project at a power plant. All the plant operation systems are air gapped from the Internet. That being said, the "business network" used all over the plant for non-plant operations is connected to the internet and it's not hard to see how the two networks could be bridged, accidentally or via "nearline" methods with USB sticks, connecting a plant system to the internet temporarily for updates and so on.
I think for the separation to be truly effective you almost need spy agency level protocols and facility design to be part of it. At this plant, there's only one wiring plant and one set of network closets which houses switching for both plant and business networks, so accidental cross-connects are possible. Plus the control room has business network jacks and PCs which makes it possible for thumb drives to easily move between systems. And of course internet-connected wifi is plant-wide.
I think to improve air gapping, there should be totally distinct wiring plants and network closets and any area critical for plant function shouldn't have any internet connection cabling at all and no wifi, either. You can't stop LTE, but they could possibly ban via policy even having cell phones in critical plant areas.
TL;DR -- plants do air gap, but it's kind of sloppy and presents a lot of opportunity and accidental cross-connection hazards.
I'm just pointing out the huge continuum between "first flight off a carrier" and someone flying for 20 years.
They don't let just anyone take off and land $70 million dollar fighter planes from a ship at sea. These guys have years of instruction and hundreds of hours of flight training before they ever set foot on an aircraft carrier.
So even a "green" pilot is neither inexperienced or uneducated, but I highly doubt the military would have allowed a newly minted carrier pilot be involved in intercepting these things. The odds are extremely in favor of these pilots being very experienced Naval aviators.
Slashdot Mirror
I cleaned 6 women's bathrooms in a college dorm, 5 days a week for a couple of years in college.
I found their toilet habits to be pretty good. The sinks and the shower were hairy, and the sinks often had a lot of makeup residue.
They also had a habit of puking in the sinks instead of the toilet, which was much harder to clean because you had to scrape the puke out since the sinks had a fairly fine steel mesh over the drains.
This was in the 1980s at the peak of the Bartles & James wine cooler phenomenon, so their trash on Mondays also smelled like sickly sweet vinegar.
The one weird day was trying to finish a bathroom when a girl walks in drunk (at about 3 PM) and says she needs to shower. I'm just finishing the last sink and only need to mop the floor, I ask if she can wait about 5 minutes or go downstairs (which was like a total distance of less than 40 feet). Since she's drunk, she's chatty and also doesn't care her robe isn't staying closed. I'm trying like hell to get out of there before someone sees what's happening because I worry the story will be "pervert janitor".
I'm basically mopping my way out and she says she's gonna shower now, so off comes the robe. She reaches in and starts the shower, turns and says bye, and then gets into the shower.
I ran into her a week later and she laughed about the shower, but she also turned out to be kind of a tragic figure who unloaded some pretty ugly sex assault stories on me.
FWIW, I wish RFC1918 had included a couple of weird and unappealing "isolated" /24s which would have gotten less use than 192.168.0.0/16 and 10.0.0.0/8 or even 172.16.0.0 (which seems to be the least used in my experience).
These lone /24s would be have been ideal to break up for interior interfaces or for use on isolated management networks that can't overlap with other interfaces.
But hasn't it kind of worked that way? YouTube is filled with people I've never heard of who get hundreds of thousands if not millions of views, so the idea of putting together a YouTube channel and making it big doesn't seem totally unrealistic because a fair number of people have actually done it.
I also don't really follow the how you get paid part, but I'm pretty sure that YouTube did kind of arbitrarily change monetization rules kind of recently which cut the income of a fair number of people with large-ish views.
I can't tell if this was YT's plan all along -- let anything but porn on, make monetization generous, and once "being a YouTube personality" hit some critical mass, start keeping the money for small-time personalities now that the concept of self-promoted YT channels has its own momentum.
It should have been an obvious existential risk for anyone using YT as a platform for income because obviously YT can change the rules and claim your earnings anytime.
I feel like we do a marginal job on enforcing project scope. The actual engineers hate to tell clients when something is legitimately out of scope because we bear the brunt for bad project planning (suddenly we're tainted by it). PMs often can't communicate the technical reasons why X is out of scope, and the sales people just want to give it away to keep the customer happy.
But I'm not sure this is a PM problem than a general capitalism problem of contract scope and enforcement in a dynamic environment.
I think generally speaking marketing tends to decrease transparency and increase obfuscation as new product performance shows fewer gains.
If new products had better general performance than old products, product naming would tend to track the performance improvements as was seen in the Mhz era.
But as producers run out of simple performance improvements, they have to start trying to create a new demand profile, often one that's kind of made up and not representative of actual performance.
IMHO, most capitalists prefer lack of product transparency because it usually forces buyers to buy more than they wanted to avoid the problem of not buying enough.
I have worked with PMI certified managers and its been a mixed bag. The best one was probably not going to "get very far' because he wasn't "management material" but he kept projects going and dealt with stall issues better than the other two. The ideal project manager from a project perspective.
The worst one was just a resume padder and a schmoozer who didn't track projects well and often added out of scope work because it helped her schmoozing.
The current one is pretty good at project management, but is often too invested in embedding herself in organizational decsionmaking which causes her to detatch from project details. This is acceptable on smaller projects, but on larger projects it makes her seem uninformed.
How do we know how great their training is?
If I was just screening Americans for a position today, I'd have to worry about a bunch of vendor certifications and whether they were useful or just empty treadmilling, past experience, and possibly educational background (which grows less relevant with experience). That's a whole minefield of bullshit, and various levels of lying about skills and experience, some of which is standard puffery, some of which is deliberate dishonesty.
With foreign workers? I have no good idea on how to vet them. I feel like everyone involved in foreign employment has an incentive to lie and very little way to have their statements verified. Plus employers are usually willing to believe what they're told because these employees are cheaper and I'm sure they been told of their compliance and work ethic.
American companies have already zeroed out their cost for training for the most part anyway. They demand employees with years of experience and routinely dump anyone who doesn't have whatever skill they look for this week.
You laugh, but people will probably figure out how to do "voice ai squatting" on Amazon, selling bogus "combination" products that match this (ie, some product that is called "milk eggs cheese" but isn't any of them).
There's so many ways storage and backups and could be leveraged to mitigate this, although I would imagine for most sites with Windows it would revolve around existing storage platform snapshots or VM backups to offline storage or in isolated security realms, not "zfs" by itself.
I have clients that run hourly incrementals of all VMs and their backup runs in an isolated security realm (firewalled, some physical network isolation and no shared or trust-related Windows security relationship).
I really wish we could stop with foods being either "good" or "bad" for you. My guess is even if you actually get the science to say if something is good or bad, the chances are that it's really only very marginally good or bad for you at reasonable/non-OCD intake levels, not so good or bad that it will swing the health of a normal person.
Even foods/beverages that are demonstrably good or bad for you aren't either in very small amounts. Sugar isn't good for you, but if I ate a glazed donut once a year? It's not going to change anything.
I'm sure there's some marginal value in looking at high-volume consumption foods like coffee, but at this point people have been drinking it for a couple of centuries and tons of it over the last century and we don't have a plague of people dying from coffee poisoning.
Other than the obvious lack of utility for "good' and "bad" labels, all it does is encourage people to over-consume "good" foods, needlessly avoid "bad" foods, all magnified by a marketing tsunami of food companies touting their products as beneficial.
I would bet in Atlanta the IT department is a bunch of people hired for their race or connections. I've seen this in several government IT offices in places way better managed than Atlanta.
The relatively high wages of IT and the "good career prospects" make it a tempting spot to place associates of politically influential leaders. The backwater nature of most small government unit IT contributes, too, as higher ups tend to see them as safe jobs to give away because they're generally not unionized and the (putative) required skill set makes justifying arbitrary hires pretty easy. The lower wages by corporate pay scale standards also means there's less outside competition if they're actually posting the jobs.
Usually there's a handful of legitimate hires who manage to keep the lights on, but they're often lifers with few other options, too, or there's a major political split along race or some other dangerous line.
I did a project once and of the 6 "admins", 3 were Asian and 1 spoke English so poorly his coworker literally translated from English for him. The other 3 were Caucasians who fought with the Asians at every turn and I think actively tried to subvert them with intentional technical screwups.
I think this is right. The worse the loan terms are, the more desperate the borrower will be to eliminate the debt.
But to be honest, I bet there's a non-zero percentage of people looking at this as a great source of speculation capital on something that looks no-lose.
The whole student loan situation is a total mess. The universities push up tuition and expand their bureaucratic fiefdoms, dorms become luxury condos, and idiot students self-indulge in luxury living.
Meanwhile, all this debt accrues and its insulation from ordinary bankruptcy probably has it over rated as a security, meaning its used as leverage for much more and riskier debt, just like mortgages. And we all knew how that went.
Plus, student loans are just a giant drag on the economy. The young people starting to earn adult wages are crippled by debt and spend much less on economic growth activities. Houses don't get remodeled, old cars get driven, fewer trips, less total consumer spending -- all those activities pump up the economy and have a multiplier effect. Student debt service seems like a tax.
I keep thinking a student debt crisis will happen, but it never does.
I feel like Reddit gets a worse rap than it deserves. The top/big/default subs are often filled with shitposts, but this is a tiny fraction of the total subs, most of which I fin to be generally spam/troll free and with reasonable dialogue. And there are an almost limitless number of subs (or that's how it seems).
In most way it reminds me a very decent re-imagining of USENET as a website.
My criticisms of it are more that you can't mark subs or threads read and then go back and see only new posts or comments, which makes following large posts and their replies troublesome.
That's what it sounds like to me.
Facebook has been able to spend its way out of some competing social media trends (ie, buying Instagram) and somehow buck others they couldn't (Snapchat), but mostly they were negotiating from a position of strength due to their network effect.
Now that their actual business model is exposed -- "You tell me, I sell you" -- and they're facing real risks of large-scale disaffection or defection to other platforms, of course they're fans of regulation. Broad social media regulations to keep their existing competition in check and to keep out the next big thing which might turn them into MySpace.
That's just putting lipstick on a pig.
I haven't been contacted by any of my financial institutions about their adoption of new blockchain based technologies to manage any of my accounts or transactions despite all its inherent advantages as a ledger alone -- I'm not even talking about the currency component.
It's a lot of hype and not a lot of anything else yet. It's just a way to hype stocks or products, not anything providing real utility yet.
Tyler Cowen, an economist, wrote an interesting paper about how we've hit the great stagnation. All the low-hanging fruit of economic innovation has been realized, at least in the US. Women participate in the workforce, we've mostly maximized the productivity enhancements of technology (assembly lines, basic robotics, office automation), we've tapped into (and in some cases, nearly tapped out) cheap and simple energy sources, and so on.
It's not to say there aren't marginal improvements, but they come with such large marginal costs that it's debatable whether they produce real productivity improvements and certainly not the productivity improvements their previous generations did when they were adopted.
Office computers are a great example, when I started working in a large office there was no email or calendaring and there was a ration of about 4-5 workers for every 1 support worker (principal job to deal with the administrivia of the worker -- scheduling, typing, copying, etc). That ratio went to something like 20:1 or smaller 5 years after email and I'd say a good chunk of what was left was semi-symbolic for senior executives or more like a personal assistant for a specific group capable of contributing in some small way to the work product itself, not just the coordination.
But since then (last 20 years, give or take) there haven't been any real increases in productivity other than small marginal increases in capability -- bigger documents, more storage, etc, but nobody is really more productive and the support costs of the technology have increased with the dependence on it.
In terms of computing technology, I think it really has stagnated in the last 5 years. Virtualization and SSDs seem to be the last big breakthrough technologies that actually improved productivity and those appear in many ways to have hit productivity improvement ceilings. Virtualization seems to be getting more complex mostly on a feature-addition basis and I've talked to customers who have an entire cluster deployed for management tools for their...cluster.
I think you can really see it in the prices of 10+ gb ethernet and SSDs. SSDs have gotten a lot cheaper but I'm still seeing prices in the thousands for enterprise SSDs. It took less than 5 years for ethernet to go from 10/100 to ubiquitous 1 gb. It's taking way longer for 10+ gb to get adopted because the vendors have held the line on pricing for that and enterprise SSD because they have nothing else to offer. It's the end of the line.
Sure, there are ways to max them out further, but now you're back to huge marginal implementation costs -- new cable plants for 10 gb, massive bus reorganization for improved flash performance, all with productivity enhancements that may not even cover their marginal costs.
Software and operating systems seem stuck in stylistic and rent-seeking iterations without even claimed productivity enhancements. It's all driven by vendor profit enhancements, tracking, advertising, not productivity utility. A brand new PC/OS with a flash disk doesn't deliver anymore average productivity than a 5 year old model and the actual worker productivity itself probably hasn't improved in 10-20 years.
We're on our way to making the area outside a gated community look just like it does in India, so I assume this will make the people here love that gated living just as much.
I can just imagine a laundry list of reasons, topped with the catch-all of "poor management".
You probably need a scoring system to derive a cash-strap score. Add 1 for being in a large urban area with increased operating costs for labor and supplies. Add 1 for leadership dominated by academics from non-management fields. Add 2 for ill-advised expansion plans (new buildings/facilities, etc) predicated on rising tuition *and* enrollment. Add 2 for financial planning which does not properly account for tuition discounting designed to boost enrollment. Add 2 for private colleges specializing in niche or narrow areas with limited economic potential (a music college would seem to be one of them).
Around here anyway, most of the small liberal arts colleges seem to be doing well even with sky-high tuition. Most people don't pay the full freight but general management seems good and many are in semi-rural locations where costs are lower. They also have something of an exclusivity component, only offer undergrad degrees but serving as well-known feeders to elite graduate school programs. Pretty much everyone I know who went to one also holds a grad school degree. And super loyal alumni who have helped build up giant endowments which smooth income and allow for cheap capital expansion funding (combined with highly sensible capital expansion).
My guess is that the Chinese won't fare well here. They could be duped by general trends of increasing tuition in the US and college attendance numbers into thinking it's a no-lose proposition. It may be a trend based solely on duping other Chinese in China -- I'd wager there's some status attached to US degree, so if you can become a semi-legitimate diploma mill (ie, your students actually come to the US and the school holds accreditation) you may end up with a limitless supply of Chinese hoping for a cut-rate advantage back home where among provincials a re-worked Harvard logo that says "Miss Smith's Secretarial School" carries the same weight as the real thing.
I'll also add -- how will self-driving cars change any of this equation?
It's not going to happen overnight -- hell, Uber just killed a pedestrian! -- but it does seem like we're on the cusp of a self-driving cars. It seems like that would radically change the equation, making it really practical to own at least fewer cars if not for a lot of people to not own one at all.
My gripe with this always is that the changes to make driving extremely unbearable happen quickly -- in a matter of less than a decade, I've seen the carrying capacity of a dozen arterial local streets reduced by half in favor of traffic calming and bike lines.
Yet the so-called improvements in public transport happen over MANY decades, and only in the smallest of increments. IMHO there are obvious ways to make bus lines faster (semi-express routes, where the bus stops every 5 blocks, not every block) but then there's all the litany of complaints about old people, handicapped, etc, who would be inconvenienced so these never happen. And it's not like the housing pattern changes are going to happen in my lifetime. I'm not ever going to move to a train accessible area and buses will never improve in my close-in area, and the market doesn't care, my in-city house has doubled in value in 20 years.
Plus transit officials here are crazy for light rail, which is great if you can build it fast, but it's a few billion dollars per line, the lines take 5 years build once they even start digging, and the most recent line is delayed like 5 years because of pissing and moaning over the right of way (which manages to avoid by 1.5 miles a long-term dense urban housing area).
The net effect is that bus improvements are forgotten about (a slow, tedious route for poor and old people with time on their hands?) because transit officials are salivating over trains and zero progress is made in balancing the reduced ability of cars to move around.
So much of this debate is unrealistic in the US. On one hand, the anti-car faction has a fantasy that every city can be transformed into downtown Copenhagen in 5 years. We'll all live happily in flats and townhouses and take trams and trains everywhere and not own cars. The super pro-car faction wants 10 lane highways everywhere and 1,000 space lots for every new building. Neither side acknowledges the other at all. European cities evolved dense over 1,000 years and in the 20th century farmed their poor out to tower blocks in the suburbs. American cities started dense and then walked away from them in the 1940s and 1950s for suburbs, leaving their poor behind in ghettos.
I think a big part of traffic problems is that urban planners have become ideologically opposed to cars and have begun to array urban planning tools against cars to make driving difficult. We get "traffic calming" which translates as lanes removed and parking removed in favor of dedicated bike lanes (it's also snowy and below freezing about 4 months out of the year).
The hope is people will find driving so difficult they give up cars for bikes or transit without considering that both are a poor substitute for cars in many cases -- distance, poor transit systems, weather, need to carry packages, etc.
I'll grant them that suburban car-centric planning is a disaster, but mostly I consider it just pseudo-planning. To this day there are suburban shopping areas where it's like 5 large tenants built buildings and lots and whatever adjoining space was left becomes a "road" which results in absurdities like requiring 4 left turns to get anywhere.
I just figure there has to be a middle way that's not so anti-car it makes things impossible but not so pro car you wind up with a wasteland of roads.
That's all nice reasoning and I'm sure something like it showed up in a presentation at Microsoft when they pitched this.
But the real reason is of course that top executives are demanding improvements in Edge's usage numbers. They've tried PR, they've done the battery life "Pepsi challenge", they've reset preferences during major updates, they have nag screens reminding you that entire villages in India rely on Edge for clean water and medicine, they even made it better at searching for porn.
But still nothing moves the needle quite like algorithmically jamming it down people's throats with no other option. Microsoft badly wants to close the door on Windows, forcing users into a closed system of rented (or ad supported) desktop operating systems, email, cloud services, and so on. They want to be Apple but they consistently play to their own internal needs (backstopping failing internal projects or money hoses like Office) with Windows when they could be enhancing the user experience.
The DOJ should have broken MS up. An operating system company and an applications company. Both would have had much better incentives to improve for the sake of users.
I'll bet the real savings was in shipping, and not just the obvious local delivery (bottler to retailer) aspect. I'll bet that the savings might have been enough to eliminate entire bottling plants that serviced areas once too distance to ship glass bottles to.
I just worked on a project at a power plant. All the plant operation systems are air gapped from the Internet. That being said, the "business network" used all over the plant for non-plant operations is connected to the internet and it's not hard to see how the two networks could be bridged, accidentally or via "nearline" methods with USB sticks, connecting a plant system to the internet temporarily for updates and so on.
I think for the separation to be truly effective you almost need spy agency level protocols and facility design to be part of it. At this plant, there's only one wiring plant and one set of network closets which houses switching for both plant and business networks, so accidental cross-connects are possible. Plus the control room has business network jacks and PCs which makes it possible for thumb drives to easily move between systems. And of course internet-connected wifi is plant-wide.
I think to improve air gapping, there should be totally distinct wiring plants and network closets and any area critical for plant function shouldn't have any internet connection cabling at all and no wifi, either. You can't stop LTE, but they could possibly ban via policy even having cell phones in critical plant areas.
TL;DR -- plants do air gap, but it's kind of sloppy and presents a lot of opportunity and accidental cross-connection hazards.
I'm just pointing out the huge continuum between "first flight off a carrier" and someone flying for 20 years.
They don't let just anyone take off and land $70 million dollar fighter planes from a ship at sea. These guys have years of instruction and hundreds of hours of flight training before they ever set foot on an aircraft carrier.
So even a "green" pilot is neither inexperienced or uneducated, but I highly doubt the military would have allowed a newly minted carrier pilot be involved in intercepting these things. The odds are extremely in favor of these pilots being very experienced Naval aviators.