Maybe, just maybe, so much power over life and death shouldn't be given to for-profit organisations?
Because then you end up with crap like this.
Continuing that line of thought, you can't pay or promote individual docs for their work, or they could falsify their results. The only option available is a union payscale. Similar to military doctors whom get paid pretty much on salary.
Sadly enough, I've had a simple drive replacement tied up in meetings and other office politics for months. Write up a proposal for change, sit in meetings where various department heads without a clue discuss the potential hazards, write up the rollback process [...]
The saddest part is that a drive replacement is not a change. It's operational maintenance, or, depending on company policy, emergency response.
At one place I worked it was a capital purchase, with all that implies. After all it is a depreciating asset, and they had capital budget to burn, and after all, the drive was part of a capital purchase at one time, wasn't it? Took months. Luckily the error spewing drive didn't fail until I got the replacement, nothing as dramatic as JWSmythe's story.
I've known a lot of datacenters, where their "tech" was the security guard. Sure, I trust him to follow simple instructions (Find the machine labeled X. Reboot it.) Sometimes even those simple instructions are screwed up. There's nothing like getting a database server rebooted in the middle of the day, when you asked for some other server to be rebooted.
Boss: Hey vlm how come "eject" is marked as a mission critical / disaster recovery software component on all our linux boxes. For that matter, why do all our 1U servers have the cheapest cdrom dell ever made as an option? We could save like FIVE WHOLE DOLLARS by getting rid of those cdroms in our standard build.
vlm: Well, see, when you tell a drone which reset button to hit, you run "dish eject" which causes all still living boxes to spit out their drive tray, so you tell the guard to hit the power on the "one that isn't working". (If you don't know what dish is, it runs the same command on multiple machines).
SOME rack mount servers have a handy dandy controllable LED light on them. Personally I wish a mandatory standard part of the ATX power supply standard was a multicolor LED inside the power switch itself "OK all the boxes that are working have changed color to red, hit the only switch that is not red"
DAN appears to be a poor reinvention of flight control software for aerospace from the 70s/80s. Those whom don't know their history are doomed to poorly repeating their past.
I don't know that history, and would (sincerely) love to have you elaborate.
Well, any description as short as your DAN description would be pretty much identical, so I won't bother. But here's an example: Take your five space shuttle main computers from the late 70s, built by different people and at least two completely separate programming teams. Feed in the same positioning sensors and gyros and all that. A circuit watches the elevon (elevator/aileron) output signals. A deviation of more than X percent from the average output means power gets cut to that computer.
Works great if the failure rate is low enough that there's only one problem to avoid. Fails pretty bad if there's multiple problems on multiple processors.
There are also deadlock like problems, look at what happens if all the processors output completely random output because of a weird input situation, you'll average at 50% output (maybe) and that is meaningless for an "eject the external tank" signal.
Another fun way to combine multiple outputs, which is really old school in aerospace but would probably be new for IT, is combining multiple outputs. For example, you don't test a new autopilot design by plugging it in and praying, you hook up two old tested working pilots each with 1/3 control authority and one new mystery box also with 1/3 control authority and then run them. If the mystery box blows up, the two other boxes have 2/3 of the control authority so the plane doesn't crash. You chart recorder the ratio of the three autopilots and hopefully each always contributes about the same unless there's a really good reason not to (maybe your new one does log instead of linear interpolation). This is more fun for engine computers... You run cylinder 1, and cylinder 1 only, on the new ignition system and send the plane up.
This is all REALLY old stuff for the CS and EE-control theory guys. This is all REALLY new stuff for the IT guys.
they do not honestly consider it to be in their enlightened self-interest to contribute to the poor, caring not for societial mores or others suffering. They are strange people, but they exist by their own right.
No, they exist because someone created an unfree market by getting rid of the guillotine. In the old days, they had fun for awhile until it was off with their heads. They didn't breed too successfully. The libertarian solution would be to allow the guillotine thus periodically cleansing the population of sociopaths.
Also I don't think it terribly enlightened to position yourself as the wealthy well fed dude with lots of food, surrounded by starving poor. That strikes me as fairly stupid. For example, I do not vacation in Haiti or Somalia. A stable socialized support system isn't there to benefit the poor, its the most effective way ever invented to prevent the rich (the guys in charge) from getting the guillotine.
Why does the radiology equiment need to be given access to the internet anyway such that it would ever get infected by conficker?
So they can update the bugs in the software, as required by the FDA. Even if there are no known bugs, what if they later discovered the billionth patient would get irradiated to a crisp, they need to prove to the FDA they could theoretically deploy a patch.
Also some very expensive embedded hardware (not exclusively medical) phones diagnostic data home for troubleshooting. You're not going to print out a one million line trace file, are you?
That would cause the hospital to loose the medicare/caid funding and consequently have to turn away or eject patients that would be a huge cost to them that would otherwise get treatment.
This assumes that medicare/medicaid patients are, and always will be, your most profitable. If your assumption ever turns out to be wrong...
What's your rate of infections by viruses? If you're running a MSWind network, that might be a fair test. If it's zero, you may be doing pretty well.
This wouldn't work on a Linux or Unix network, as there basically aren't any viruses to probe the network, but on MSWind they might do a fair job of testing you.
Its just an anecdote, but a casual acquaintance I met at HOPE 2006 in NYC or something, worked at a hospital and their solution to preventing LAN/network based infections was to create the semi-mythical one PC vlan.
So the linux side spoke dot1q and had a zillion interfaces and spoke smb via samba or whatever the heck the biomed device used. There was also some confusing discussion of mac address filtering, he was guite proud that any ole sysadmin could do iptables at OSI 3 and up, but he was doing all manner of layer 2 MAC filtering. For example, all the "whatchamacallits" had the same first three bytes of the MAC address, the OUI, and they were only allowed traffic from THE mac address of some remote monitor thingy.
He had a lot of semi-automated script and claimed if anyone swapped the server LAN card his switches would all autoupdate or something based on a vast pile of perl.
Some of the most sensitive "LAN" devices used a "LAN" of a crossover cable instead of running thru wifi and etherswitches. Obviously a little hard to monitor. So he used a hub (not a switch) and some dual ported RMON probe thingy to watch the sensitive "net".
He also did this mac address lockout thing such that only certain macs work in certain ports...
The point of this rambling recollection is you put a regular Fortune 500 IT dude in a hospital IT dept and you get security holes galore, a hospital IT guy merely needs some new outlook, some new ways of thinking.
I sure hope they're talking about a random soldier rank pv2 E1 whom happens to be named "Obama" not the much more famous guy at the tippy top peak of the military chain of command of the worlds largest most powerful military. You'd think the C-in-C might have enough power to take care of this without begging other countries for help?
I'm not trying to debate right or wrong here, trying to focus on an utter lack of efficiency and competence. We're not talking about rewriting the worlds biggest most corrupt banking system or the worlds biggest most corrupt medical-industrial complex. Just a tiny freaking prison in the middle of nowhere.
An article by the NY Times analyzed cables released which indicated the US is having difficulties in fulfilling Obama's promise to close the Guantánamo Bay detention camp and is now considering incentives in return for other countries accepting detainees, including a one-on-one meeting with Obama or assistance with the IMF. '
and how many people are always changing minor things without change control because they feel this is their baby and they can do anything?
That's because the development server IS the production server, for whatever reasons. Its not a maintenance procedure problem, but a design problem way upstream of scheduled maintenance.
The other scenario is when you're breaking individual (or world wide) new ground. It works when a huge team can spend months debating the route and design for some new railroad tracks, however an operating engineer needs full and instant discretion of how and when to work the throttle and brake levers.
There's some things that aren't well defined. You grow your OSPF areas until they blow up, then you make multiple areas. How do you know when they'll blow up? Well, pretty much you don't, unless you estimate so low that you go bankrupt on hardware costs...
Works, although excruciatingly slowly for planned work. The collision of excruciatingly slow proactive planned work, and reactive trouble tickets, always is a source of utter hilarity. Usually the end result is you only do planned proactive paper shuffling for meaningless stuff "lets change the background color to be 0.001% darker" and ram thru development as part of a trouble ticket with no oversight at all (well, to make our big customer happy, we've decided to completely redo our database schema and stored procedures this afternoon as part of the ticket).
Another example, if it takes a month and endless meetings to replace a failing drive during scheduled maint, and a half hour to replace a failed drive at any time, this simply eliminates all proactive maintenance. Much easier / cheaper to burn the power supply out, have a nice long outage, and then replace the whole device, than to get permission to blow dust out of the air filter.
The end result is usually much worse than it was at the beginning.
would be enormously cumbersome if they needed a consensus of administrators to execute.
Thats why you leave changes to the 24x7 onsite operations team not one lone admin doin' his thing in the cube. They're the ones monitoring the systems, seems most sensible if they "push the buttons" on the things they watch. Ideally you have one team that does nothing but watch and one team that does nothing but do, and theoretically they cooperate.
And besides, DAN appears still to be vaporware.
DAN appears to be a poor reinvention of flight control software for aerospace from the 70s/80s. Those whom don't know their history are doomed to poorly repeating their past.
Next up, we'll reinvent the concept of the security office from AS/400, or maybe the idea of hard realtime control.
Maybe someone out there could could reinvent the concept of the watchdog timer so the "DAN" cluster doesn't go into deadlock? Naah, we'll let them "discover" it themselves, the hard way.
Spoken like someone who has never had to live on a fixed income, but almost certainly will someday, IF he is lucky.
My grandmother gets a COLA pay raise from SS almost every year. I haven't kept up with inflation since 2001 in other words I've been poorer every year relative to inflation.
Statistically my anecdote is pretty much the normal situation. SS recipients have a fixed income, relative to inflation, and the median income of wage earners relative to inflation has been decreasing since the 70s.
I cam only dream of moving up to a fixed income instead of a declining income.
if SS is anything like the civil war era northern army pension system, this would still be a substantial cost savings over doing it manually,
Yeah, the Social Security Administration records are maintained with quills and scrolls. They need $500 million to upgrade to computers otherwise they will have to go back to using stone tablets.
Folks disabled during the war of northern aggression generally were documented very early with fountain pen on paper (say, heart disease due to typhus or getting shot), but the old age pensioners generally got typewritten notes. I don't believe any of them survived to even the early electromechanical punchcard era.
I was thinking more in terms of workflow, apply, provide proof, approve/disapprove, appeal, pay out till they die at which point they worked remarkably fast to close the account. There can be quite a remarkable stack of paperwork, especially if multiple appeals are involved due to a contested disability, address changes, in and out of nursing home/hospitals, etc.
Can you give an example of a chemical anomaly? Not trolling, just interested.
Basically stuff we can't reliably engineer yet due to materials science mysteries and the failure of mechanical engineers. Look into things like controlled and predictable supersaturation, superheating of liquids above their boiling point, supercooling, etc. Beyond a pretty obvious "its gotta be clean stuff in a clean smooth container" no one has any solid engineering data on exactly how to reliably make a beaker superheat pure distilled H2O at STP to exactly 104.1 C before explosive boiling sets in.
Also some weird stuff happens in flames and generally speaking non-equilibrium reactions. Thats where the fullerenes aka buckballs were found after centuries of burning stuff and saying, eh, its just as bunch of ash.
Basically anywhere working as a chemist would be a huge pain due to uncontrolled / hard to reproduce conditions. Those are the long term answers that no one has solved for decades.
The short term answer is go visit your local PHD candidate or researcher and ask them what they're having trouble figuring out today. Why is the yield on catalyst bed number 7 so low unless its purged with (something) even though technically that should have no effect? Why won't this blasted grignard reagent form when theoretically its supposed to "just work"? This (insert swearwords) organic compound just won't freaking crystallize.
1) Data acquisition (who has incrementally paid into the system) - this is B.S. accounting because it's all going to come from the general tax fund soon anyway, so why the charade?
They gather that data to verify eligibility and calculated benefit payout. Don't you get an annual statement listing what you earned for the past X years and if you were to become disabled you would get Y dollars per month, etc? I just got my annual statement on Friday.
They send this out annually because if the food store that I worked at in 1991 forgot to credit me with that income, its a heck of a lot easier to set the record straight in 1992 rather than waiting for me to retire or become disabled decades later and ooops I haven't paid in enough "fully qualified" years for whatever benefit.
Also if someone steals your SS number and works under it, you can trivially figure out how much money they earn, which is interesting.
From experience doing genealogical research, if SS is anything like the civil war era northern army pension system, this would still be a substantial cost savings over doing it manually, crazy as that might sound.
Also from having been involved in major data conversion projects over the decades, a cost of $2 per account converted would have been an incredible daydream. Just having a final step of a semi-intelligent semi-trained human being review the converted account will cost more. The only thing saving the SS is probably huge scale and frankly all the accounts are pretty much the same story other than personally identifiable information.
Finally from having been around for awhile I know that shock stories like this are based on rolling everything they possibly can into that figure, rounding up, passing along to the next guy whom adds some more (maybe even the same stuff) and rounds up again, repeat until a scary enough figure is generated. So this is probably like three annual department budgets plus training budget plus a lifetime supply of backup tapes plus a couple weeks salary for all front end personnel (assuming they're paid during training).
Can you create an Apt repository that exists for pay for software? As in, software the user has to pay for to use?
Easily, yes. As far as I know, no one has ever tried.
apt-get lives on the client side. On the server side its just a plain ole http or ftp server that has a very precise layout of certain generated files. The apache server or proftpd server doesn't really "know" its acting as an apt repository, its just slinging files.
So your question simplifies into "can I set up a FTP or HTTP server that someone has to pay to access?" Several solutions seem to involve embedding an identifier into the URL, although there would be nothing wrong with securing it at a VPN layer, or very simplistically playing games with ip access lists.
the iPhone started out at $600,... eventually they got to $250 or $200.
Where in the world do you live?
Here in the USA, the box plus some accessories and apps will probably set you back a couple hundred bucks, at which point you own a ipod touch. To turn it into an iphone, you need to sign a two year contract at over $100/month, lets round down to $2500.
In the US I bought an ipod touch for a couple hundred, but there is no way you can buy an iphone for much less than $3000.
Has the iphone dropped in price by a factor of 10 since I last researched it?
And then you get into straight out stupid shit like electric carving knives, wtf? A vibrating knife to cut chicken?
Another application error. You cut leather-like ham with that, not "fork tender" barbecue chicken.
You also missed some economic issues... like the kcup thing, at work if your labor is worth a buck a minute, homemade coffee is terribly expensive compared to practically any other non-drive-thru solution, unless you're making coffee as a hobby.
I'm not that savvy on the melting point of various metals, but how would the rails handle the time spent exposed to the sun?
Float them on pools of liquid lead... "strong enough" once it freezes, mushy enough at temperature to prevent kinks. This was discussed in the sci fi story, or online, don't remember. I always thought the story made more sense if you ran the train thru a tunnel with greenhouses on the roof of the tunnel, so you travel along as you eat your food. Cooling would be a problem, but you'd have nearly infinite solar power, so...
This imaginary Mrecury sounds like a great place to live. What happens when, lets say, the rail needs maintenance or is damaged and needs to be repaired? The whole population gets cooked and then new people can move in once the rail is repaired and get free fried chicken? Sounds awesome.
Park in the nearest tunnel. Sucks to spend that much time in the dark instead of perfect twilight, but it beats death. I don't remember if that was actually in the sci fi story, or some online discussion.
The other thing is that Mercury rotates kind of slowly compared to the earth. Anything other than a major viaduct / bridge and a crane is trivially fast enough to haul a train past the damaged track, or just wait until the track is fixed...
I would have to do the math to make sure, but at some latitude a human being could quite easily outwalk the sunrise... Heck on the equator I think a person could outwalk the sunrise, if you assume it rotates a hundred times slower and the circumference is quite smaller. On the earth you need to maintain a couple thousand miles per hour at the equator, but a factor of a hundred slower rotation plus a factor for smaller circumference, its doable. One trivial solution is not one really long train but many that shuffle back and forth and could theoretically be walked past in a disaster.
One proposed method of colonizing Mercury is to build a city on rails that circles the slowly-spinning planet, always staying in the shady site where temperatures are cool enough for human habitats.
With this story, such an idea doesn't seem nearly so far-fetched.
Charles Stross story? I read some sci fi story along those lines in the 80s.
I was going off employee resumes and some press releases. I had the impression the wifi-ish meshy thing was on the poles and the backup to it was the cellular and the link into the house was the zigbee. There certainly are several meter manufacturers that like zigbee.
If you have better sources (like, maybe you're one of the online resumes I read?) then I guess you're right, otherwise we have to compare source quality to figure which interpretation is more likely to be correct. Otherwise it'll be hard to select an interpretation.
In some area all you can get is DSL any where from 1.5 to 6 meg max. maybe 6 meg will work ok but some people can only get 1.5 or 3 meg dsl.
Cable has higher download speeds but a full block on a Node trying to use this at the same time.
satellite broadband with the FAP does not kill this the lag will.
3G 5GB cap will led to big costs for data over 5GB. And ATT's 2GB then $10 per GB will may this cost so much people will get a PC.
And if you have 2-3 systems then you may need FIOS just make it work good. And Fios is not all over.
Always fun seeing the new guys complain it could never work, even when I was doing it in the 90s with xwindows / nfs / vnc over a 14.4 modem... and liking it...
Slashdot Mirror
Maybe, just maybe, so much power over life and death shouldn't be given to for-profit organisations?
Because then you end up with crap like this.
Continuing that line of thought, you can't pay or promote individual docs for their work, or they could falsify their results. The only option available is a union payscale. Similar to military doctors whom get paid pretty much on salary.
Sadly enough, I've had a simple drive replacement tied up in meetings and other office politics for months. Write up a proposal for change, sit in meetings where various department heads without a clue discuss the potential hazards, write up the rollback process [...]
The saddest part is that a drive replacement is not a change. It's operational maintenance, or, depending on company policy, emergency response.
At one place I worked it was a capital purchase, with all that implies. After all it is a depreciating asset, and they had capital budget to burn, and after all, the drive was part of a capital purchase at one time, wasn't it? Took months. Luckily the error spewing drive didn't fail until I got the replacement, nothing as dramatic as JWSmythe's story.
I've known a lot of datacenters, where their "tech" was the security guard. Sure, I trust him to follow simple instructions (Find the machine labeled X. Reboot it.) Sometimes even those simple instructions are screwed up. There's nothing like getting a database server rebooted in the middle of the day, when you asked for some other server to be rebooted.
Boss: Hey vlm how come "eject" is marked as a mission critical / disaster recovery software component on all our linux boxes. For that matter, why do all our 1U servers have the cheapest cdrom dell ever made as an option? We could save like FIVE WHOLE DOLLARS by getting rid of those cdroms in our standard build.
vlm: Well, see, when you tell a drone which reset button to hit, you run "dish eject" which causes all still living boxes to spit out their drive tray, so you tell the guard to hit the power on the "one that isn't working". (If you don't know what dish is, it runs the same command on multiple machines).
SOME rack mount servers have a handy dandy controllable LED light on them. Personally I wish a mandatory standard part of the ATX power supply standard was a multicolor LED inside the power switch itself "OK all the boxes that are working have changed color to red, hit the only switch that is not red"
DAN appears to be a poor reinvention of flight control software for aerospace from the 70s/80s. Those whom don't know their history are doomed to poorly repeating their past.
I don't know that history, and would (sincerely) love to have you elaborate.
Well, any description as short as your DAN description would be pretty much identical, so I won't bother. But here's an example: Take your five space shuttle main computers from the late 70s, built by different people and at least two completely separate programming teams. Feed in the same positioning sensors and gyros and all that. A circuit watches the elevon (elevator/aileron) output signals. A deviation of more than X percent from the average output means power gets cut to that computer.
Works great if the failure rate is low enough that there's only one problem to avoid. Fails pretty bad if there's multiple problems on multiple processors.
There are also deadlock like problems, look at what happens if all the processors output completely random output because of a weird input situation, you'll average at 50% output (maybe) and that is meaningless for an "eject the external tank" signal.
Another fun way to combine multiple outputs, which is really old school in aerospace but would probably be new for IT, is combining multiple outputs. For example, you don't test a new autopilot design by plugging it in and praying, you hook up two old tested working pilots each with 1/3 control authority and one new mystery box also with 1/3 control authority and then run them. If the mystery box blows up, the two other boxes have 2/3 of the control authority so the plane doesn't crash. You chart recorder the ratio of the three autopilots and hopefully each always contributes about the same unless there's a really good reason not to (maybe your new one does log instead of linear interpolation). This is more fun for engine computers... You run cylinder 1, and cylinder 1 only, on the new ignition system and send the plane up.
This is all REALLY old stuff for the CS and EE-control theory guys. This is all REALLY new stuff for the IT guys.
they do not honestly consider it to be in their enlightened self-interest to contribute to the poor, caring not for societial mores or others suffering. They are strange people, but they exist by their own right.
No, they exist because someone created an unfree market by getting rid of the guillotine. In the old days, they had fun for awhile until it was off with their heads. They didn't breed too successfully. The libertarian solution would be to allow the guillotine thus periodically cleansing the population of sociopaths.
Also I don't think it terribly enlightened to position yourself as the wealthy well fed dude with lots of food, surrounded by starving poor. That strikes me as fairly stupid. For example, I do not vacation in Haiti or Somalia. A stable socialized support system isn't there to benefit the poor, its the most effective way ever invented to prevent the rich (the guys in charge) from getting the guillotine.
Why does the radiology equiment need to be given access to the internet anyway such that it would ever get infected by conficker?
So they can update the bugs in the software, as required by the FDA. Even if there are no known bugs, what if they later discovered the billionth patient would get irradiated to a crisp, they need to prove to the FDA they could theoretically deploy a patch.
Also some very expensive embedded hardware (not exclusively medical) phones diagnostic data home for troubleshooting. You're not going to print out a one million line trace file, are you?
That would cause the hospital to loose the medicare/caid funding and consequently have to turn away or eject patients that would be a huge cost to them that would otherwise get treatment.
This assumes that medicare/medicaid patients are, and always will be, your most profitable. If your assumption ever turns out to be wrong...
What's your rate of infections by viruses? If you're running a MSWind network, that might be a fair test. If it's zero, you may be doing pretty well.
This wouldn't work on a Linux or Unix network, as there basically aren't any viruses to probe the network, but on MSWind they might do a fair job of testing you.
Its just an anecdote, but a casual acquaintance I met at HOPE 2006 in NYC or something, worked at a hospital and their solution to preventing LAN/network based infections was to create the semi-mythical one PC vlan.
So the linux side spoke dot1q and had a zillion interfaces and spoke smb via samba or whatever the heck the biomed device used. There was also some confusing discussion of mac address filtering, he was guite proud that any ole sysadmin could do iptables at OSI 3 and up, but he was doing all manner of layer 2 MAC filtering. For example, all the "whatchamacallits" had the same first three bytes of the MAC address, the OUI, and they were only allowed traffic from THE mac address of some remote monitor thingy.
He had a lot of semi-automated script and claimed if anyone swapped the server LAN card his switches would all autoupdate or something based on a vast pile of perl.
Some of the most sensitive "LAN" devices used a "LAN" of a crossover cable instead of running thru wifi and etherswitches. Obviously a little hard to monitor. So he used a hub (not a switch) and some dual ported RMON probe thingy to watch the sensitive "net".
He also did this mac address lockout thing such that only certain macs work in certain ports...
The point of this rambling recollection is you put a regular Fortune 500 IT dude in a hospital IT dept and you get security holes galore, a hospital IT guy merely needs some new outlook, some new ways of thinking.
I sure hope they're talking about a random soldier rank pv2 E1 whom happens to be named "Obama" not the much more famous guy at the tippy top peak of the military chain of command of the worlds largest most powerful military. You'd think the C-in-C might have enough power to take care of this without begging other countries for help?
I'm not trying to debate right or wrong here, trying to focus on an utter lack of efficiency and competence. We're not talking about rewriting the worlds biggest most corrupt banking system or the worlds biggest most corrupt medical-industrial complex. Just a tiny freaking prison in the middle of nowhere.
An article by the NY Times analyzed cables released which indicated the US is having difficulties in fulfilling Obama's promise to close the Guantánamo Bay detention camp and is now considering incentives in return for other countries accepting detainees, including a one-on-one meeting with Obama or assistance with the IMF. '
and how many people are always changing minor things without change control because they feel this is their baby and they can do anything?
That's because the development server IS the production server, for whatever reasons. Its not a maintenance procedure problem, but a design problem way upstream of scheduled maintenance.
The other scenario is when you're breaking individual (or world wide) new ground. It works when a huge team can spend months debating the route and design for some new railroad tracks, however an operating engineer needs full and instant discretion of how and when to work the throttle and brake levers.
There's some things that aren't well defined. You grow your OSPF areas until they blow up, then you make multiple areas. How do you know when they'll blow up? Well, pretty much you don't, unless you estimate so low that you go bankrupt on hardware costs...
Works, although excruciatingly slowly for planned work.
The collision of excruciatingly slow proactive planned work, and reactive trouble tickets, always is a source of utter hilarity. Usually the end result is you only do planned proactive paper shuffling for meaningless stuff "lets change the background color to be 0.001% darker" and ram thru development as part of a trouble ticket with no oversight at all (well, to make our big customer happy, we've decided to completely redo our database schema and stored procedures this afternoon as part of the ticket).
Another example, if it takes a month and endless meetings to replace a failing drive during scheduled maint, and a half hour to replace a failed drive at any time, this simply eliminates all proactive maintenance. Much easier / cheaper to burn the power supply out, have a nice long outage, and then replace the whole device, than to get permission to blow dust out of the air filter.
The end result is usually much worse than it was at the beginning.
would be enormously cumbersome if they needed a consensus of administrators to execute.
Thats why you leave changes to the 24x7 onsite operations team not one lone admin doin' his thing in the cube. They're the ones monitoring the systems, seems most sensible if they "push the buttons" on the things they watch. Ideally you have one team that does nothing but watch and one team that does nothing but do, and theoretically they cooperate.
And besides, DAN appears still to be vaporware.
DAN appears to be a poor reinvention of flight control software for aerospace from the 70s/80s. Those whom don't know their history are doomed to poorly repeating their past.
Next up, we'll reinvent the concept of the security office from AS/400, or maybe the idea of hard realtime control.
Maybe someone out there could could reinvent the concept of the watchdog timer so the "DAN" cluster doesn't go into deadlock? Naah, we'll let them "discover" it themselves, the hard way.
Spoken like someone who has never had to live on a fixed income, but almost certainly will someday, IF he is lucky.
My grandmother gets a COLA pay raise from SS almost every year. I haven't kept up with inflation since 2001 in other words I've been poorer every year relative to inflation.
Statistically my anecdote is pretty much the normal situation. SS recipients have a fixed income, relative to inflation, and the median income of wage earners relative to inflation has been decreasing since the 70s.
I cam only dream of moving up to a fixed income instead of a declining income.
if SS is anything like the civil war era northern army pension system, this would still be a substantial cost savings over doing it manually,
Yeah, the Social Security Administration records are maintained with quills and scrolls. They need $500 million to upgrade to computers otherwise they will have to go back to using stone tablets.
Folks disabled during the war of northern aggression generally were documented very early with fountain pen on paper (say, heart disease due to typhus or getting shot), but the old age pensioners generally got typewritten notes. I don't believe any of them survived to even the early electromechanical punchcard era.
I was thinking more in terms of workflow, apply, provide proof, approve/disapprove, appeal, pay out till they die at which point they worked remarkably fast to close the account. There can be quite a remarkable stack of paperwork, especially if multiple appeals are involved due to a contested disability, address changes, in and out of nursing home/hospitals, etc.
Can you give an example of a chemical anomaly? Not trolling, just interested.
Basically stuff we can't reliably engineer yet due to materials science mysteries and the failure of mechanical engineers. Look into things like controlled and predictable supersaturation, superheating of liquids above their boiling point, supercooling, etc. Beyond a pretty obvious "its gotta be clean stuff in a clean smooth container" no one has any solid engineering data on exactly how to reliably make a beaker superheat pure distilled H2O at STP to exactly 104.1 C before explosive boiling sets in.
Also some weird stuff happens in flames and generally speaking non-equilibrium reactions. Thats where the fullerenes aka buckballs were found after centuries of burning stuff and saying, eh, its just as bunch of ash.
Basically anywhere working as a chemist would be a huge pain due to uncontrolled / hard to reproduce conditions. Those are the long term answers that no one has solved for decades.
The short term answer is go visit your local PHD candidate or researcher and ask them what they're having trouble figuring out today. Why is the yield on catalyst bed number 7 so low unless its purged with (something) even though technically that should have no effect? Why won't this blasted grignard reagent form when theoretically its supposed to "just work"? This (insert swearwords) organic compound just won't freaking crystallize.
1) Data acquisition (who has incrementally paid into the system) - this is B.S. accounting because it's all going to come from the general tax fund soon anyway, so why the charade?
They gather that data to verify eligibility and calculated benefit payout. Don't you get an annual statement listing what you earned for the past X years and if you were to become disabled you would get Y dollars per month, etc? I just got my annual statement on Friday.
They send this out annually because if the food store that I worked at in 1991 forgot to credit me with that income, its a heck of a lot easier to set the record straight in 1992 rather than waiting for me to retire or become disabled decades later and ooops I haven't paid in enough "fully qualified" years for whatever benefit.
Also if someone steals your SS number and works under it, you can trivially figure out how much money they earn, which is interesting.
$500M is about $1.40 per US citizen.
From experience doing genealogical research, if SS is anything like the civil war era northern army pension system, this would still be a substantial cost savings over doing it manually, crazy as that might sound.
Also from having been involved in major data conversion projects over the decades, a cost of $2 per account converted would have been an incredible daydream. Just having a final step of a semi-intelligent semi-trained human being review the converted account will cost more. The only thing saving the SS is probably huge scale and frankly all the accounts are pretty much the same story other than personally identifiable information.
Finally from having been around for awhile I know that shock stories like this are based on rolling everything they possibly can into that figure, rounding up, passing along to the next guy whom adds some more (maybe even the same stuff) and rounds up again, repeat until a scary enough figure is generated. So this is probably like three annual department budgets plus training budget plus a lifetime supply of backup tapes plus a couple weeks salary for all front end personnel (assuming they're paid during training).
Can you create an Apt repository that exists for pay for software? As in, software the user has to pay for to use?
Easily, yes. As far as I know, no one has ever tried.
apt-get lives on the client side. On the server side its just a plain ole http or ftp server that has a very precise layout of certain generated files. The apache server or proftpd server doesn't really "know" its acting as an apt repository, its just slinging files.
So your question simplifies into "can I set up a FTP or HTTP server that someone has to pay to access?" Several solutions seem to involve embedding an identifier into the URL, although there would be nothing wrong with securing it at a VPN layer, or very simplistically playing games with ip access lists.
the iPhone started out at $600, ... eventually they got to $250 or $200.
Where in the world do you live?
Here in the USA, the box plus some accessories and apps will probably set you back a couple hundred bucks, at which point you own a ipod touch. To turn it into an iphone, you need to sign a two year contract at over $100/month, lets round down to $2500.
In the US I bought an ipod touch for a couple hundred, but there is no way you can buy an iphone for much less than $3000.
Has the iphone dropped in price by a factor of 10 since I last researched it?
And then you get into straight out stupid shit like electric carving knives, wtf? A vibrating knife to cut chicken?
Another application error. You cut leather-like ham with that, not "fork tender" barbecue chicken.
You also missed some economic issues... like the kcup thing, at work if your labor is worth a buck a minute, homemade coffee is terribly expensive compared to practically any other non-drive-thru solution, unless you're making coffee as a hobby.
I'm not that savvy on the melting point of various metals, but how would the rails handle the time spent exposed to the sun?
Float them on pools of liquid lead... "strong enough" once it freezes, mushy enough at temperature to prevent kinks. This was discussed in the sci fi story, or online, don't remember. I always thought the story made more sense if you ran the train thru a tunnel with greenhouses on the roof of the tunnel, so you travel along as you eat your food. Cooling would be a problem, but you'd have nearly infinite solar power, so ...
This imaginary Mrecury sounds like a great place to live. What happens when, lets say, the rail needs maintenance or is damaged and needs to be repaired? The whole population gets cooked and then new people can move in once the rail is repaired and get free fried chicken? Sounds awesome.
Park in the nearest tunnel. Sucks to spend that much time in the dark instead of perfect twilight, but it beats death. I don't remember if that was actually in the sci fi story, or some online discussion.
The other thing is that Mercury rotates kind of slowly compared to the earth. Anything other than a major viaduct / bridge and a crane is trivially fast enough to haul a train past the damaged track, or just wait until the track is fixed...
I would have to do the math to make sure, but at some latitude a human being could quite easily outwalk the sunrise... Heck on the equator I think a person could outwalk the sunrise, if you assume it rotates a hundred times slower and the circumference is quite smaller. On the earth you need to maintain a couple thousand miles per hour at the equator, but a factor of a hundred slower rotation plus a factor for smaller circumference, its doable. One trivial solution is not one really long train but many that shuffle back and forth and could theoretically be walked past in a disaster.
One proposed method of colonizing Mercury is to build a city on rails that circles the slowly-spinning planet, always staying in the shady site where temperatures are cool enough for human habitats.
With this story, such an idea doesn't seem nearly so far-fetched.
Charles Stross story? I read some sci fi story along those lines in the 80s.
OK very interesting.
I was going off employee resumes and some press releases. I had the impression the wifi-ish meshy thing was on the poles and the backup to it was the cellular and the link into the house was the zigbee. There certainly are several meter manufacturers that like zigbee.
If you have better sources (like, maybe you're one of the online resumes I read?) then I guess you're right, otherwise we have to compare source quality to figure which interpretation is more likely to be correct. Otherwise it'll be hard to select an interpretation.
ISP caps and slow down speeds will NOT work this.
A 1920 x 1200 desktop at 32 bits a lot of data.
In some area all you can get is DSL any where from 1.5 to 6 meg max. maybe 6 meg will work ok but some people can only get 1.5 or 3 meg dsl.
Cable has higher download speeds but a full block on a Node trying to use this at the same time.
satellite broadband with the FAP does not kill this the lag will.
3G 5GB cap will led to big costs for data over 5GB. And ATT's 2GB then $10 per GB will may this cost so much people will get a PC.
And if you have 2-3 systems then you may need FIOS just make it work good. And Fios is not all over.
Always fun seeing the new guys complain it could never work, even when I was doing it in the 90s with xwindows / nfs / vnc over a 14.4 modem ... and liking it ...