I think zip ties might actually be more secure than TSA locks.
The T&B nylon ones are more durable than the cheap-ass ones and when wrapped tight, you pretty much have to use a wire cutter to cut them off. I've been without one when I wanted to remove some and found a pair of conventional scissors and a pocket knife inadequate. This means that opportunistic people without tools are SOL.
I've used them on luggage and I mark mine in a surreptitious way. If they're completely off, I know the bag was opened for sure, and if there's some other type of zip tie on there, I know it was opened and resealed.
It is amazing a company can make so much money doing something completely useless.
I think the social media space is full of these kinds of companies only because ad agencies are full of clueless executives who only know that they need "social media capability" and don't have it, so they farm it out to this kind of company, mark up the cost, and pass it on to their client.
Does Iran have sufficient investments to restrain their potential use of nuclear weapons, if they do in fact acquire them?
It depends on whether or not you consider them rationally motivated or religiously motivated.
If its the former, I've always considered the Iranians stupid for wasting resources on developing nuclear weapons. Unless enough time passes that they are able to match the United States in delivery capabilities (the full triad, not just a couple of land-launched ICBMs), they really can't do very much of them. And I don't see the Iranians all that close to an ICBM, let alone a fleet of bombers capable of bypassing the most powerful air force in history or a fleet of nuclear powered submarines and sea-launch ICBMs.
A nuclear strike on the US (especially) or even a close ally would almost *certainly* result in an overwhelming nuclear counter-attack, the kind of attack that would call into question the future of Iran as a nation state or even a coherent civilization. You could make a lot of arguments about why we wouldn't or shouldn't, but I think the reality is that the public would DEMAND it in a way that just couldn't be resisted. After 9/11 the public backed two dubious ground wars, one of them on outright fabricated intelligence. I might even expect a hesitant President being given the advice that unless he counterstrikes, not even the Secret Service and military would be able to guarantee the personal safety of himself and his family, such would be the public rage.
Then there's the question of using them as a threat -- well, a threat is only useful if it's realistic, and an armed Iran with nuclear ICBMs with all the right signs they might be readying to use them? It's not hard to see a preemptive strike, and even then the rational actor Iranian has to ask what's going to come of it? The glassing over of Tehran after they use them?
All in all, the Iranians would have been MUCH better off sinking all their nuclear weapons investment into something a lot less of an existential threat to their own country. I think they'd have been far better off developing long-range drones and cruise missiles and building them by the thousand lot. An Iran capable of flying hundreds of kamikaze drones thousands of miles is much scarier to me because the potential damage is small and doesn't create an opportunity for total warfare retaliation.
There was for example little or no "collateral damage" when we bombed Japan, or for that matter Dresden.
There was huge amounts of what we today call collateral damage, but back then they didn't use the term "collateral damage".
Now it's "collateral damage" because we're killing a bunch of civilians or destroying civil infrastructure we'd rather not.
Then it wasn't collateral damage because we MEANT to kill civilians and destroy civil infrastructure because we believed that breaking the enemy's ability and will to fight would aid our war effort and shorten the war.
That's why it's called total warfare (or scorched Earth warfare) -- you don't want the enemy to have ANYTHING that enables them to fight, and that includes a population able to function at any meaningful level of productivity, and they aren't very productive if they are starving, homeless and lacking any infrastructure that enables them to be productive.
This was the partial goal of the allied military and very much part of the post-war pacification of Germany, where deliberate allied policies forced the population into famine and stripped them of much of their industrial capacity. Make no mistake, there was no accidental, collateral damage to German civilians, it was a deliberate policy during and after the war to crush the German population into submission.
The county I live in actually has a garbage-to-electricity system. I think the energy side of it works fine, it's the economics of it I wonder about it.
Supposedly they're widely used in Europe where land for landfilling is more scarce.
I guess I would have assumed that GNP growth combined with all the myriad ways currency leaves general circulation in the US would have led to a general net positive demand for paper money that exceeded the recycling of worn out currency.
But I guess I also would have expect that the overall demand for paper money would be shrinking as people pay more and more for stuff with credit cards, so maybe that has something to do with it.
I have a client who runs a waste-to-energy plant. They sort and shred trash to produce a burnable product for some power plants setup to handle it.
My contact says it's actually cheaper for haulers to landfill the trash. Currently the two counties involved in the plant require haulers to send their trash to the plant, but pay a subsidy to make up the difference.
But technology people go to work for all kinds of companies who do boring work nobody wants to hear about. Or they go to work for a company which maybe a lot of people DO want to hear about, like Apple, but they sign all kinds of secret contracts the swear them to secrecy and ruinous poverty if they reveal them.
And I think many people in technology work because the technology itself is interesting to work with, the purpose for which it is being used for, whether selling rutabagas or insurance, they really don't care about.
And the NSA's specific appeal is "so you're interested in X? How would you like to work with a football field's worth of the newest X. We have a nearly unlimited amount of money to spend on it and some problems more interesting than cutting a cube dweller's transaction time on some database screens."
This gets compared to the pitch from private industry who says "we're looking to upgrade from Windows 2008r2 to 2016 because it fits our 7 year budget cycle for server replacement, and we only do TPS reports on days ending in Y".
I have no idea if this still applies, but I've read that when pit bulls were bred for fighting they were also selected (intentionally or not) for not attacking people as "man biters" were a risk to handlers.
I think any dog breed with a strong dominance trait could potentially become a problem if it wasn't trained or socialized, and I think mostly this risk goes up with the size of the dog in question.
(Disclaimer: I have a pit bull / Great Dane mix, an he's great with people, but not so good with other dogs, partly due to being attacked when he was younger by a neighbor's unleashed lab and a golden retriever. Bad choice by those two dogs, if their owner hadn't intervened, the lab would have been dead, and that was with me putting all of my 6'1, 240 body into the leash and pinch collar to pull my dog away.)
....and probably a lot they just don't understand.
That's what this sounds like. They have metric assloads of data, so much they don't really know what to do with it all and maybe some of it they don't know what it means. But that's what correlations are for, isn't it? To manufacture causation?
You know, it's not meant to be the answer to the autonomous driving problem, just an example at how sensor packages already exist that can handle foul weather situations better than human vision.
I've driven several of those blizzards in North Dakota.
In one of them I was driving a car with radar assisted distance sensing cruise control and it was way better than I was at judging the car in front of me with limited visibility and at night.
While I'm sure there's a whole world of forum posters with their disk benchmarks in their warlording signature who go for this kind of thing because they want to be the guy with the best benchmark numbers, what's the actual performance gain in a typical kind of scenario vs. a SATA3 SSD?
These kinds of sequential benchmarks don't really tell me how much real-world time something like this will shave off booting a computer, launching an application, etc versus a more conventional solution.
And will provide you with all the power less protection you might need.
Is the threat of power loss and the data loss risk associated with it somehow a new thing that people haven't been concerned with an dealing with relatively inexpensively for a long time?
I think you could also make an argument that SSDs might be less likely to lose data vs. spinning rust, as their greater speed means that disk transactions are more likely to be completed faster, leaving the device idle more often and thus less likely to be in the middle of a transaction if the power is lost.
This was my thought. I've always wondered if there was a kind of algorithm or heuristic to service tags or if they are just kind of serially generated.
It probably wouldn't do this scam a bunch of good to use, say, tags for really obsolete models (ie, something 10+ years old, which the owner may not even still own) or for some of the non-PC equipment that Dell has sold over the years that has had otherwise similar looking service tags applied to it.
If you COULD sort out what models went with what tag ranges, it would be a lot more useful as you could pick on "home" models and maybe even prey on slightly older but not completely obsolete vintages with the idea that that class of older, home PCs models are prone to the kinds of issues that fake tech support people could use to get you to run their malware stuff on.
Or not -- maybe you'd pick on NEW models, with the idea that if you were building malware/identity theft empire newer computers would represent more affluent people (more money to steal), would generally be less likely to have other malware/rot problems and perhaps even have access to better networking connection (ie, rich-guy 100 meg cable versus less-rich-guy shared wireless or something).
Either way, being able to decode tags for models BEFORE you exploited tag lookup online would be beneficial. Maybe they just had a lot of time on their hands and access to enough platforms that they could guess ranges.
I get that the old console strategy has its limitations, especially if you're older, want to get into online content, etc. You gotta pay to play, at least if you want to play at the current release.
What's kind of funny is that he has a fairly modern PC (well, maybe not gaming wise, it's a top of the line NUC) and if given the choice for screen time, about half the time he will play stupid flash games from some web site they have access to at school.
It's kind of like the old story about Christmas day, where the kid has more fun playing with empty boxes than the new toys they came in.
The local Microsoft store generously donated an XBox 360 to our school's charity raffle, probably 6 months after the XBox One was released.
We didn't own any gaming system at all, and my son immediately griped that there would be "no games because its old". The day after we got the console, we went to a local pawn shop and bought 5 games for $30, all of which played just fine. I think we might be up to about 15 games now, and I'd doubt that even with the 2 games my son has bought new, we're out more than $150 on games.
I think with consoles, this is the way to do it -- there are so many used games for the previous model that if you stay just slightly behind whatever's current, you have an endless supply of cheap games.
Maybe this is a problem for someone who's really into gaming, wants the latest and greatest, but honestly, for an 11 year old boy (and a 49 year old man...) I have a hard time understanding what you're missing on a brand new console for the extra money.
Was your "psychological harm" so great that you could demonstrate financial damage from it?
Did you lose wages from work, have to obtain counseling or incur any other monetary costs associated with this?
If not, I don't think anyone would buy into your psychological harm because you really can't demonstrate any actual consequences from it. I don't think transitory emotional states without any demonstrable consequences count as psychological damage.
Why do you think a video of a guy sitting in front of his computer adds anything worth the headache of watching a video?
Is there some kind of Skype-type intimacy that clearly doesn't register with me that I seem to be missing?
I can almost get a normal TV-type interview where the camera films the subject actually engaging with the interviewer and we catch some kind of something from his human interaction, but here it's just some guy in front of a computer.
How can they possibly be expected to keep up with the latest trends in user interface graphic design if they spend their resources on complex and time-consuming security reviews? We need to rev this sucker every 9-12 months with new GUI designs to keep it fresh and relevant. UI experts are finding new and improved ways to do familiar tasks based on the latest trends in all-white backgrounds, small type, big buttons and the latest in little spinning widgets that show you how hard your phone is working.
If we devoted efforts on security reviews, we'd have to keep pushing out a familiar, old-fashioned user interfaces that would look out of place at the newest yoga studios, craft breweries and locally sourced, cruelty free, organic beard grooming parlors.
If I could go back in time with a bunch of money, one thing I think would have been interesting to do would be to have forked/extended IMAPv4 in the 1990s to include a bunch of extensions for calendaring, contact management and mailbox management and coupled them with a local delivery agent and very lightweight directory on the back end. I think you would have had to address the local mail storage issue as well -- mbox vs. maildir, or do you go full-on database storage.
This would have (mostly) given you a single piece of software which would have enabled Exchange-like functionality in the Unix/open software space. I think the linking of calendaring, directory and mail in one bit of software might have significantly staved off the hegemony of Exchange/Outlook. It would have at least enabled developers creating GUI clients a single protocol and server to work against that combined all the basic elements people used -- email, calendar, contacts, making client development simpler (one protocol, one server).
Unfortunately the unix/open software world's lack of appetite for integration and use of separate and disparate packages for functionality turned Exchange/Outlook like email into a clusterfuck of sendmail/postfix/qmail, various LDAP implementations (which were almost always more complicated than necessary for email), and then the various schemes for scaling up mbox/maildir/database (the latter often with an additional layer of separate database software) and all the dozens of local delivery options which stubbornly remained unintegrated with the MTA.
Anyway, in my time-travel fantasy all of this results in a robust does-it-all backend that remains compatible with existing mail-only IMAP clients and enables client developers a consolidated target to work against. I'm sure there's reasons this never happened, but it always struck me that IMAPv4 came so close to providing a back end that could be adapted to handle more than email.
I'd argue that Microsoft wants it that way. I've installed and run Exchange since 2000 and by 2010 Microsoft mostly hit the sweet spot in terms of useful management interface and pretty damn good reliability and performance, especially for the large feature set it employed.
But in 2010, they killed off the GUI management for a web interface that maybe does half of what even small organizations need in terms of admin, shunting the rest of their management intrerface to the overly verbose and Byzantine PowerShell. It's like running Postfix but using a line editor to manage the config file.
Worse, they made it less reliable and resilient and I've had Microsoft SEs tell me the same thing.
My (conspiratorial) belief is that MS made it worse to manage in a deliberate attempt to drive SMBs to O365 and jack up cash flow through a more expensive and continuous billing cycle. We've priced out O365 vs. on premise installations and head counts beyond about 20 people are cheaper over 3-5 years for on premise than O365. Exactly none of the clients who have gone O365 have dropped IT staff from what I've seen.
And it's not like there aren't outages and downtime associated with O365, probably statistically more than 2010 or 2013 installs I've seen.
I'm sure that some large organizations have found the increased PowerShell aspect of Exchange management useful, but these are the kind of IT shops with relatively large groups of dedicated Exchange teams anyway -- they have the time and headcount to make all that scripting worthwhile.
Slashdot Mirror
I think zip ties might actually be more secure than TSA locks.
The T&B nylon ones are more durable than the cheap-ass ones and when wrapped tight, you pretty much have to use a wire cutter to cut them off. I've been without one when I wanted to remove some and found a pair of conventional scissors and a pocket knife inadequate. This means that opportunistic people without tools are SOL.
I've used them on luggage and I mark mine in a surreptitious way. If they're completely off, I know the bag was opened for sure, and if there's some other type of zip tie on there, I know it was opened and resealed.
It is amazing a company can make so much money doing something completely useless.
I think the social media space is full of these kinds of companies only because ad agencies are full of clueless executives who only know that they need "social media capability" and don't have it, so they farm it out to this kind of company, mark up the cost, and pass it on to their client.
Does Iran have sufficient investments to restrain their potential use of nuclear weapons, if they do in fact acquire them?
It depends on whether or not you consider them rationally motivated or religiously motivated.
If its the former, I've always considered the Iranians stupid for wasting resources on developing nuclear weapons. Unless enough time passes that they are able to match the United States in delivery capabilities (the full triad, not just a couple of land-launched ICBMs), they really can't do very much of them. And I don't see the Iranians all that close to an ICBM, let alone a fleet of bombers capable of bypassing the most powerful air force in history or a fleet of nuclear powered submarines and sea-launch ICBMs.
A nuclear strike on the US (especially) or even a close ally would almost *certainly* result in an overwhelming nuclear counter-attack, the kind of attack that would call into question the future of Iran as a nation state or even a coherent civilization. You could make a lot of arguments about why we wouldn't or shouldn't, but I think the reality is that the public would DEMAND it in a way that just couldn't be resisted. After 9/11 the public backed two dubious ground wars, one of them on outright fabricated intelligence. I might even expect a hesitant President being given the advice that unless he counterstrikes, not even the Secret Service and military would be able to guarantee the personal safety of himself and his family, such would be the public rage.
Then there's the question of using them as a threat -- well, a threat is only useful if it's realistic, and an armed Iran with nuclear ICBMs with all the right signs they might be readying to use them? It's not hard to see a preemptive strike, and even then the rational actor Iranian has to ask what's going to come of it? The glassing over of Tehran after they use them?
All in all, the Iranians would have been MUCH better off sinking all their nuclear weapons investment into something a lot less of an existential threat to their own country. I think they'd have been far better off developing long-range drones and cruise missiles and building them by the thousand lot. An Iran capable of flying hundreds of kamikaze drones thousands of miles is much scarier to me because the potential damage is small and doesn't create an opportunity for total warfare retaliation.
There was for example little or no "collateral damage" when we bombed Japan, or for that matter Dresden.
There was huge amounts of what we today call collateral damage, but back then they didn't use the term "collateral damage".
Now it's "collateral damage" because we're killing a bunch of civilians or destroying civil infrastructure we'd rather not.
Then it wasn't collateral damage because we MEANT to kill civilians and destroy civil infrastructure because we believed that breaking the enemy's ability and will to fight would aid our war effort and shorten the war.
That's why it's called total warfare (or scorched Earth warfare) -- you don't want the enemy to have ANYTHING that enables them to fight, and that includes a population able to function at any meaningful level of productivity, and they aren't very productive if they are starving, homeless and lacking any infrastructure that enables them to be productive.
This was the partial goal of the allied military and very much part of the post-war pacification of Germany, where deliberate allied policies forced the population into famine and stripped them of much of their industrial capacity. Make no mistake, there was no accidental, collateral damage to German civilians, it was a deliberate policy during and after the war to crush the German population into submission.
The county I live in actually has a garbage-to-electricity system. I think the energy side of it works fine, it's the economics of it I wonder about it.
Supposedly they're widely used in Europe where land for landfilling is more scarce.
I guess I would have assumed that GNP growth combined with all the myriad ways currency leaves general circulation in the US would have led to a general net positive demand for paper money that exceeded the recycling of worn out currency.
But I guess I also would have expect that the overall demand for paper money would be shrinking as people pay more and more for stuff with credit cards, so maybe that has something to do with it.
I have a client who runs a waste-to-energy plant. They sort and shred trash to produce a burnable product for some power plants setup to handle it.
My contact says it's actually cheaper for haulers to landfill the trash. Currently the two counties involved in the plant require haulers to send their trash to the plant, but pay a subsidy to make up the difference.
But technology people go to work for all kinds of companies who do boring work nobody wants to hear about. Or they go to work for a company which maybe a lot of people DO want to hear about, like Apple, but they sign all kinds of secret contracts the swear them to secrecy and ruinous poverty if they reveal them.
And I think many people in technology work because the technology itself is interesting to work with, the purpose for which it is being used for, whether selling rutabagas or insurance, they really don't care about.
And the NSA's specific appeal is "so you're interested in X? How would you like to work with a football field's worth of the newest X. We have a nearly unlimited amount of money to spend on it and some problems more interesting than cutting a cube dweller's transaction time on some database screens."
This gets compared to the pitch from private industry who says "we're looking to upgrade from Windows 2008r2 to 2016 because it fits our 7 year budget cycle for server replacement, and we only do TPS reports on days ending in Y".
I thought the Treasury sold currency to the Fed at face value, and the profit (value of currency minus cost to make it) was called seigniorage.
I would think that on balance the seigniorage profit from paper money was way more than the loss from small coins.
I have no idea if this still applies, but I've read that when pit bulls were bred for fighting they were also selected (intentionally or not) for not attacking people as "man biters" were a risk to handlers.
I think any dog breed with a strong dominance trait could potentially become a problem if it wasn't trained or socialized, and I think mostly this risk goes up with the size of the dog in question.
(Disclaimer: I have a pit bull / Great Dane mix, an he's great with people, but not so good with other dogs, partly due to being attacked when he was younger by a neighbor's unleashed lab and a golden retriever. Bad choice by those two dogs, if their owner hadn't intervened, the lab would have been dead, and that was with me putting all of my 6'1, 240 body into the leash and pinch collar to pull my dog away.)
....and probably a lot they just don't understand.
That's what this sounds like. They have metric assloads of data, so much they don't really know what to do with it all and maybe some of it they don't know what it means. But that's what correlations are for, isn't it? To manufacture causation?
Didn't Bowie strike an unusual deal a few years ago where he sold all his past AND future music rights?
You know, it's not meant to be the answer to the autonomous driving problem, just an example at how sensor packages already exist that can handle foul weather situations better than human vision.
I've driven several of those blizzards in North Dakota.
In one of them I was driving a car with radar assisted distance sensing cruise control and it was way better than I was at judging the car in front of me with limited visibility and at night.
While I'm sure there's a whole world of forum posters with their disk benchmarks in their warlording signature who go for this kind of thing because they want to be the guy with the best benchmark numbers, what's the actual performance gain in a typical kind of scenario vs. a SATA3 SSD?
These kinds of sequential benchmarks don't really tell me how much real-world time something like this will shave off booting a computer, launching an application, etc versus a more conventional solution.
And will provide you with all the power less protection you might need.
Is the threat of power loss and the data loss risk associated with it somehow a new thing that people haven't been concerned with an dealing with relatively inexpensively for a long time?
I think you could also make an argument that SSDs might be less likely to lose data vs. spinning rust, as their greater speed means that disk transactions are more likely to be completed faster, leaving the device idle more often and thus less likely to be in the middle of a transaction if the power is lost.
This was my thought. I've always wondered if there was a kind of algorithm or heuristic to service tags or if they are just kind of serially generated.
It probably wouldn't do this scam a bunch of good to use, say, tags for really obsolete models (ie, something 10+ years old, which the owner may not even still own) or for some of the non-PC equipment that Dell has sold over the years that has had otherwise similar looking service tags applied to it.
If you COULD sort out what models went with what tag ranges, it would be a lot more useful as you could pick on "home" models and maybe even prey on slightly older but not completely obsolete vintages with the idea that that class of older, home PCs models are prone to the kinds of issues that fake tech support people could use to get you to run their malware stuff on.
Or not -- maybe you'd pick on NEW models, with the idea that if you were building malware/identity theft empire newer computers would represent more affluent people (more money to steal), would generally be less likely to have other malware/rot problems and perhaps even have access to better networking connection (ie, rich-guy 100 meg cable versus less-rich-guy shared wireless or something).
Either way, being able to decode tags for models BEFORE you exploited tag lookup online would be beneficial. Maybe they just had a lot of time on their hands and access to enough platforms that they could guess ranges.
I get that the old console strategy has its limitations, especially if you're older, want to get into online content, etc. You gotta pay to play, at least if you want to play at the current release.
What's kind of funny is that he has a fairly modern PC (well, maybe not gaming wise, it's a top of the line NUC) and if given the choice for screen time, about half the time he will play stupid flash games from some web site they have access to at school.
It's kind of like the old story about Christmas day, where the kid has more fun playing with empty boxes than the new toys they came in.
The local Microsoft store generously donated an XBox 360 to our school's charity raffle, probably 6 months after the XBox One was released.
We didn't own any gaming system at all, and my son immediately griped that there would be "no games because its old". The day after we got the console, we went to a local pawn shop and bought 5 games for $30, all of which played just fine. I think we might be up to about 15 games now, and I'd doubt that even with the 2 games my son has bought new, we're out more than $150 on games.
I think with consoles, this is the way to do it -- there are so many used games for the previous model that if you stay just slightly behind whatever's current, you have an endless supply of cheap games.
Maybe this is a problem for someone who's really into gaming, wants the latest and greatest, but honestly, for an 11 year old boy (and a 49 year old man...) I have a hard time understanding what you're missing on a brand new console for the extra money.
Was your "psychological harm" so great that you could demonstrate financial damage from it?
Did you lose wages from work, have to obtain counseling or incur any other monetary costs associated with this?
If not, I don't think anyone would buy into your psychological harm because you really can't demonstrate any actual consequences from it. I don't think transitory emotional states without any demonstrable consequences count as psychological damage.
Why do you think a video of a guy sitting in front of his computer adds anything worth the headache of watching a video?
Is there some kind of Skype-type intimacy that clearly doesn't register with me that I seem to be missing?
I can almost get a normal TV-type interview where the camera films the subject actually engaging with the interviewer and we catch some kind of something from his human interaction, but here it's just some guy in front of a computer.
20 loads? Why don't they just say about 3 days worth.
How can they possibly be expected to keep up with the latest trends in user interface graphic design if they spend their resources on complex and time-consuming security reviews? We need to rev this sucker every 9-12 months with new GUI designs to keep it fresh and relevant. UI experts are finding new and improved ways to do familiar tasks based on the latest trends in all-white backgrounds, small type, big buttons and the latest in little spinning widgets that show you how hard your phone is working.
If we devoted efforts on security reviews, we'd have to keep pushing out a familiar, old-fashioned user interfaces that would look out of place at the newest yoga studios, craft breweries and locally sourced, cruelty free, organic beard grooming parlors.
If I could go back in time with a bunch of money, one thing I think would have been interesting to do would be to have forked/extended IMAPv4 in the 1990s to include a bunch of extensions for calendaring, contact management and mailbox management and coupled them with a local delivery agent and very lightweight directory on the back end. I think you would have had to address the local mail storage issue as well -- mbox vs. maildir, or do you go full-on database storage.
This would have (mostly) given you a single piece of software which would have enabled Exchange-like functionality in the Unix/open software space. I think the linking of calendaring, directory and mail in one bit of software might have significantly staved off the hegemony of Exchange/Outlook. It would have at least enabled developers creating GUI clients a single protocol and server to work against that combined all the basic elements people used -- email, calendar, contacts, making client development simpler (one protocol, one server).
Unfortunately the unix/open software world's lack of appetite for integration and use of separate and disparate packages for functionality turned Exchange/Outlook like email into a clusterfuck of sendmail/postfix/qmail, various LDAP implementations (which were almost always more complicated than necessary for email), and then the various schemes for scaling up mbox/maildir/database (the latter often with an additional layer of separate database software) and all the dozens of local delivery options which stubbornly remained unintegrated with the MTA.
Anyway, in my time-travel fantasy all of this results in a robust does-it-all backend that remains compatible with existing mail-only IMAP clients and enables client developers a consolidated target to work against. I'm sure there's reasons this never happened, but it always struck me that IMAPv4 came so close to providing a back end that could be adapted to handle more than email.
I'd argue that Microsoft wants it that way. I've installed and run Exchange since 2000 and by 2010 Microsoft mostly hit the sweet spot in terms of useful management interface and pretty damn good reliability and performance, especially for the large feature set it employed.
But in 2010, they killed off the GUI management for a web interface that maybe does half of what even small organizations need in terms of admin, shunting the rest of their management intrerface to the overly verbose and Byzantine PowerShell. It's like running Postfix but using a line editor to manage the config file.
Worse, they made it less reliable and resilient and I've had Microsoft SEs tell me the same thing.
My (conspiratorial) belief is that MS made it worse to manage in a deliberate attempt to drive SMBs to O365 and jack up cash flow through a more expensive and continuous billing cycle. We've priced out O365 vs. on premise installations and head counts beyond about 20 people are cheaper over 3-5 years for on premise than O365. Exactly none of the clients who have gone O365 have dropped IT staff from what I've seen.
And it's not like there aren't outages and downtime associated with O365, probably statistically more than 2010 or 2013 installs I've seen.
I'm sure that some large organizations have found the increased PowerShell aspect of Exchange management useful, but these are the kind of IT shops with relatively large groups of dedicated Exchange teams anyway -- they have the time and headcount to make all that scripting worthwhile.