Ahh. They weren't "systematically giving this out", and they weren't "offering" it. It was presented on request, typically when parents expressed concern about the location and safety of the student. The capacity, and the ease of obtaining the records, wasn't broadly advertised. Parents are often paying quite a lot for a student's education. and make demands based on that and on their concern for their children.
The "magic words" to most easily obtain the records was expressing concern about possible suicide. It's a very real problem on college campuses.
> there is no reason not to name names and every reason to name them.
They're already dealing with the issue. Mine was meant to be a cautionary tale: they are already dealing with the issues, including some personnel changes, so _nothing_ would be served by naming names at this point.
And no, the nature of the wrongdoing is such that very few people were aware of it, because very few people paid attention. The students and staff sign various agreements agreeing to monitoring "as deemed necessary" for security or network performance. So the university was clear, as best I could tell, for merely collecting the data. The legal and ethical difficulties were the lack of control of access to the data, and reporting of why used or provided access.
> The crowding out is being caused by humans and humans alone;
Do you count goats, rabbits, and cane toads? Goats are a core part of the expansion of the Sahara: they graze the grass too short. Australia has numerous examples of ecological disaster from humans importing other animals, especially rabbits and cane toads.
> Is it really the school's or government's responsibility to protect kids from these things?
For all students, and especially with medical issues, a school has considerable legal authority and responsibility for that child's care. They are _expected_ to support the physician's and parent's decisions about medical care: Failing to do so could kill the child, especially one with such clearly diet linked medical requirements. And failing to demonstrate that the school has done their best to cooperate with those standards is a real legal risk.
I'm not suggesting that this is genuinely wise or justified approach. I'm pointing out one of the most obvious and compelling reasons to provide such detailed monitoring.
I'm sure it will soon. There are some potential benefits: noticing that a diabetic student is buying a frosty milkshake everyday, or that a child with seafood allergies is buying fish sticks every week, or that a morbidly obese student or builimic student are buying 5 servings of ice cream every day might all be useful to the parents and the school. And the usefulness of such information can be used to justify monitoring _all_ students.
I recently encountered this sort of thing at a university where the IT department implemented extremely detailed tracking of wifi use. They would report to the parents, without notification to the student, where a student's laptop was last detected and what wifi access points they normally used at certain times of day. The nominal reason was "so the parents could contact the student". I was quite surprised, though not shocked, at their casual approach to privacy, especially since the same system monitored staff, visitors, campus police cell phones, and the personnel at the ROTC and military research facilities on their campus
Given this kind of "spin" on the problem, I'm sure the client agreements with Slashdot will include "opt-in" as the default setting on every agreement.
> Translation; Company horribly oversold the bandwidth and is too cheap to buy a bigger pipe.
Please try again. It's quite common for companies to make very reasonable assumptions that most of their customers will not be using their full bandwidth, unattended, 24x7. So they buy bandwdth from upstream and connecting to the core "tier 3" fiber optic backbones of our country. Then a modest number of Bittorrent users show up, each with 5 TB external drives, filling them 24x7 with thousands of hours of music and vides which they will never watch, simply to "complete their collecionts". Those relatively few users can take than the rest of the client base, combined. And they do not necessarily _want_ or require high bandwidth connections for this service.
A natural and reasonably fair approach is to throttle the traffic. Bittorrent is not like video streaming, or gaming or VOIP. It deals well with throttling and short losses of connection. The difficulty is that the necessary proxies and throttles to do this degrade service, somewhat, for _everyone_.
Switching to the "career of your dreams" is usually a bad idea I know several reasonably competent engineers and scientists who bankrupted themselves, and crippled their family finances, "pursuing a dream" of being a stay at home parent, pursing an artistic career and lost their engineering edge while they did it. They now regret the decision, but have no chance of recovering their engineering edge sufficiently to return to their original, much better paid fields. They'd have to start over as a 40 or 50 year old intern with obsolete skills, and there's no market for them.
If your dream is so important to you, fund it yourself as a hobby or a pasttime. I know too many reasonably competent engineers who blamed their lack of focus on their "lack of inspiration" on their lack of interest. They switched careers, and turned out to be as unfocused in their new "inspired" career. But because they were "unfocused" in a poorly funded career, they've either gone hideously broke or drained their family's finances finances supporting their career. I've known several who are literally a million dollars poorer between the loss of engineering income and with the educational costs of the career switch, for jobs they can't get because they're competing with much, much cheaper kids who are also dependent on family support to keep them fed. They spent their retirement funds and their kids' college funds on their "dream" careers, and they're pretty unhappy about it now.
Frankly, I see the same thing played out regularly for people who have doubts about their lovers or their spouses. They abandon decent, workable relationships in favor of their "soul mate" or someone else tempting who is "the one". If the alternative pastime, or alternative partner, is so ideally suited to you, let them work for it. Don't abandon your current working life or your current working relationship in favor of an unlikely dream. There are far too many broken careers, and broken hearts, from such switches.
Most simply put, I'll offer the advice that so many agents and editors give to their dreaming clients. "Don't quit your day job". If you turn out to be that good at your hobby, you'll find the opportunity to turn to it later as a full-time career. It's much less heart and wallet braking to work at your primary job and your primary relatonships. Just don't _lie_ about it, and over commit.
Taking in a critical router during daylight hours, when Comcast's brick and mortar installations are open, can be awkward. Ensuring that you have failover capacity that can actually take your network load while that equipment is serviced and replaced, and restoring the configuration without interruption, can be nighmarish. This is especially the case in small shops that are trying to grow, shops where the failover capacity has been billed as being enabled but has never been tested or ever actually existed.That is the kind of embarrassment that causes middle managers to get IT personnel fired, especially to avoid the blame themselves.
Yes, that's the important word right there. "Intent" does not protect developers from the consequences of failing to learn and use the existing standards. A great deal of my professional life is spent reverse engineering discarded standards, and bringing previously working software into compliance with new standards, while leaving the software compatible with _every single_ old standard for compatibility reason. The amount of work maintaining compatibility often overwhelms the benefits of the new standard quite quickly.
In this case one the main difficulties of the new "universal standard" are that it broke critical features: screen sharing, in particular, is quite useful during conferences. I don't know anyone using Skype in the technical world who hasn't taken advantage of that feature.
Except for the many, many complaints about systemd that are not FUD, such as the overwhelming need by systemd developers to integrate non-init fucntions into the systemd suite such as logging, DHCP, and mounting attached media, the casual reworking of the file-system-hierarchy by ignoring all standards for "/run" and "/media", and insane inconsistencies of "using a symlink for/etc/resolv.conf into the DHCP substructure, except not resetting the symlink if the user puts a flat text file in/etc/resolv.conf and never updating it again"
One of the announced goals of systemd is to "create a stateless Linux", to essentially clear away or prevent any modifications to "/etc" and "/var" and set all local system configuration withing systemd itself. This _breaks_ decades of system configuration work, including every Linux compatible system management tool such as puppit, CFengine, chef, BladeManager, and even installation systems such as Debian'a.dev system, RPM's packaging, and commercial software installation tools./etc/resolv.conf is merely an example of breaking such tools, but it's a fundamental example of the disdain for non-systemd based system components. It's an example of the chaos being engendered is from such broad goals that have_nothing to do_ with an init system, and that erratically and unpredictably break decades of well-written software that followed the published file system hierarchy standards in favor of underdocumented, non-portable, rapidly evolving and functionally unnecessary switches to a new layout.
Those are only a few of the well documented issues: so overall, there's a lot to like about systemd. But the tendency of its proponents to characterize concerns about it as "FUD" is disingenuous.
> though, it might finally lead to ads being delivered from the same host as the rest of the content on the page
This cannot be done. much of graphical and especially streaming content is from more robust "content delivery networks", such as Akamai, that host much larger proxies closer to the web browser's "final mile". Even modest icon or graphical content on a web page will overwhelm many corporate core web servers without these third-party hosted proxies, and it's especially true for ad content. Slashdot itself benefits from its extremely simple format, and can thus handle its load quite easily with quite modest resources. But if they tried to host the advertising content all on the same systems, I'm quite certain it would collapse the servers if not the firewalls themselves.
> Who you fuck is not a matter for the state to be prosecuting or otherwise interfering with.
Except for rape, child molestation, incest, trading sex for promotions, demanding sex for promotions, sex with infected partners who refuse to warn others, and shall i go on?
The "state" has always regulated sex, often though not always for compelling reason.
Technology for culturing muscle _cells_ is available and has been for some time. Growing enervated, intact muscle with working neural structures, reflexes, and viable attachments to tendon requires far, far more complex support and has never been achieved without a living host, at least not in any reliable literature that I can find.
I'm afraid I didn't say "standing, walking and talking are reflexes", and I'm sorry if I gave that impression. But all of them rely on extensive muscular reflexes, such as the plantar and babinski reflexes in the feet, the patellar reflex in the knee, numerous tongue and throat reflexes for speech, the "arthrokinetic" reflexes for hip movement, etc.
I' urge you to meet people who grew up physically disabled and had some function restored medically later in life. There's a great deal of development of limbs and physiology, of capability, that depends on ongoing stimulation, and which is profoundly distorted by the lack of normal stimulation or activity. And there are profound reasons that dancers and gymnasts start training _early_. The strength and flexibility they require cannot simple be overlaid later. Now take that to an extreme, with a body that hasn't engaged in regular exercise _at all_, until adulthood and understand the useless morass of a functional body that would arise.
Unless there is something like an auto-exercise machine that can provide a full range of human activity to a brainless corpse, the process would be pointless. It seems much safer and potentially more effective to simply raise a clone child and simply confiscate the body when the transplantee is ready.
Yes the idea is a long standing one. Most proposals that have made more physical sense involve a mountain based launcher reaching the highest altitude feasible to reduce atmospheric drag as much as possible. There are big trade-offs of lighter mass of spacecraft with less fuel, and what can feasibly and safely be accelerated along the longest available launch tube. And the more fuel aboard the spacecraft, the more destructive the _inevitable_, eventual launch failure.
> There is no reason why a jar-grown clone would need to be anything other than completely brain dead.
There's not currently any technology, or hope of technology, that can produce muscle tone and muscular reflexes to stand, walk, speak, or handle tools without actually doing the activities. So there are many reasons to give the clones basic human activity. A sodden mas of undeveloped muscle tissue, wrapped around inexperienced or developed organs and undeveloped bone, would require ongoing medical issues as extensive as being extremely old and infirm. "Jar-grown" does not mean mature, viable organ systems.
It requires a vacuum equipped rail gun of quite an incredible length, designed to tolerate spacecraft zipping down it and out of the end of the tube about Mach 25. We don't have anything built or even tested at Mach 1 that can even approach those launch velocities, or be cacuum sealed effectively along its entire launch length. And when it leaves the launch tube at Mach 1, the sonic boom is likely to _destroy_ the end of any launching tube as it reverbates back.
The idea of "new bodies for old" is an old one in fiction and in science. This was explored in some dramatic detail in Lois McMaster Bujold's stories of "Miles Vorkosigan". Old people would have a clone made, the clone raised in foster care until mature enough to support a full grown brain, and then the brain transplanted. Raising the clone required raising them to at least puberty, to support the brain and mature nervous system of the transplant candidate, and they were certainly sentient beings being sacrificed.
> Speaking as an accountant, get/use something compatible with your accounting system.
Oh, my, yes. This. Getting buy-in from the people who actually do the work, so that they know why you selected certain features, is incredibly helpful. This applies to accounting, VOIP, trouble ticket systems, Wikis, email, and any other IT supported system. And the people on the ground may know of critical features or flaws that their own supervisors are simply unaware of or which get "summarized" out of the critical feature list at the several dozens of meetings before such a project is actually implemented.
Facebook is an interesting case I'd not dealt with. But I'm afraid their external services are rooted in IPv4. I cannot find a single exposed service for them that is only IPv6, and some of their services are IPv4 only. For example, their exposed MX records point to 'msgin.vvv.facebook.com', which is IPv4 only. And this sort of claim that an environment is "dial stack" is generally mistaken. A few exposed services are both, but most of the exposed network services remain IPv4. And until their customer base _stops_ using IPv4-only routers, firewalls, local proxies, and local computer configurations, they _must_ continue to support IPv4.
The key to being able to continue to support IPv4 has been the switch to NAT. When you say they are "IPv6 only internally", they don't have to expose that network space, and most companies do not _want_ to expose that network space. So the internal IPv6 becomes irrelevant to most of their external communications.
Slashdot Mirror
Ahh. They weren't "systematically giving this out", and they weren't "offering" it. It was presented on request, typically when parents expressed concern about the location and safety of the student. The capacity, and the ease of obtaining the records, wasn't broadly advertised. Parents are often paying quite a lot for a student's education. and make demands based on that and on their concern for their children.
The "magic words" to most easily obtain the records was expressing concern about possible suicide. It's a very real problem on college campuses.
> , it would be easy for anyone to get it
Not in bulk, already digitized.
> there is no reason not to name names and every reason to name them.
They're already dealing with the issue. Mine was meant to be a cautionary tale: they are already dealing with the issues, including some personnel changes, so _nothing_ would be served by naming names at this point.
And no, the nature of the wrongdoing is such that very few people were aware of it, because very few people paid attention. The students and staff sign various agreements agreeing to monitoring "as deemed necessary" for security or network performance. So the university was clear, as best I could tell, for merely collecting the data. The legal and ethical difficulties were the lack of control of access to the data, and reporting of why used or provided access.
> The crowding out is being caused by humans and humans alone;
Do you count goats, rabbits, and cane toads? Goats are a core part of the expansion of the Sahara: they graze the grass too short. Australia has numerous examples of ecological disaster from humans importing other animals, especially rabbits and cane toads.
> Is it really the school's or government's responsibility to protect kids from these things?
For all students, and especially with medical issues, a school has considerable legal authority and responsibility for that child's care. They are _expected_ to support the physician's and parent's decisions about medical care: Failing to do so could kill the child, especially one with such clearly diet linked medical requirements. And failing to demonstrate that the school has done their best to cooperate with those standards is a real legal risk.
I'm not suggesting that this is genuinely wise or justified approach. I'm pointing out one of the most obvious and compelling reasons to provide such detailed monitoring.
I'm sure it will soon. There are some potential benefits: noticing that a diabetic student is buying a frosty milkshake everyday, or that a child with seafood allergies is buying fish sticks every week, or that a morbidly obese student or builimic student are buying 5 servings of ice cream every day might all be useful to the parents and the school. And the usefulness of such information can be used to justify monitoring _all_ students.
I recently encountered this sort of thing at a university where the IT department implemented extremely detailed tracking of wifi use. They would report to the parents, without notification to the student, where a student's laptop was last detected and what wifi access points they normally used at certain times of day. The nominal reason was "so the parents could contact the student". I was quite surprised, though not shocked, at their casual approach to privacy, especially since the same system monitored staff, visitors, campus police cell phones, and the personnel at the ROTC and military research facilities on their campus
Given this kind of "spin" on the problem, I'm sure the client agreements with Slashdot will include "opt-in" as the default setting on every agreement.
> Translation; Company horribly oversold the bandwidth and is too cheap to buy a bigger pipe.
Please try again. It's quite common for companies to make very reasonable assumptions that most of their customers will not be using their full bandwidth, unattended, 24x7. So they buy bandwdth from upstream and connecting to the core "tier 3" fiber optic backbones of our country. Then a modest number of Bittorrent users show up, each with 5 TB external drives, filling them 24x7 with thousands of hours of music and vides which they will never watch, simply to "complete their collecionts". Those relatively few users can take than the rest of the client base, combined. And they do not necessarily _want_ or require high bandwidth connections for this service.
A natural and reasonably fair approach is to throttle the traffic. Bittorrent is not like video streaming, or gaming or VOIP. It deals well with throttling and short losses of connection. The difficulty is that the necessary proxies and throttles to do this degrade service, somewhat, for _everyone_.
Switching to the "career of your dreams" is usually a bad idea I know several reasonably competent engineers and scientists who bankrupted themselves, and crippled their family finances, "pursuing a dream" of being a stay at home parent, pursing an artistic career and lost their engineering edge while they did it. They now regret the decision, but have no chance of recovering their engineering edge sufficiently to return to their original, much better paid fields. They'd have to start over as a 40 or 50 year old intern with obsolete skills, and there's no market for them.
If your dream is so important to you, fund it yourself as a hobby or a pasttime. I know too many reasonably competent engineers who blamed their lack of focus on their "lack of inspiration" on their lack of interest. They switched careers, and turned out to be as unfocused in their new "inspired" career. But because they were "unfocused" in a poorly funded career, they've either gone hideously broke or drained their family's finances finances supporting their career. I've known several who are literally a million dollars poorer between the loss of engineering income and with the educational costs of the career switch, for jobs they can't get because they're competing with much, much cheaper kids who are also dependent on family support to keep them fed. They spent their retirement funds and their kids' college funds on their "dream" careers, and they're pretty unhappy about it now.
Frankly, I see the same thing played out regularly for people who have doubts about their lovers or their spouses. They abandon decent, workable relationships in favor of their "soul mate" or someone else tempting who is "the one". If the alternative pastime, or alternative partner, is so ideally suited to you, let them work for it. Don't abandon your current working life or your current working relationship in favor of an unlikely dream. There are far too many broken careers, and broken hearts, from such switches.
Most simply put, I'll offer the advice that so many agents and editors give to their dreaming clients. "Don't quit your day job". If you turn out to be that good at your hobby, you'll find the opportunity to turn to it later as a full-time career. It's much less heart and wallet braking to work at your primary job and your primary relatonships. Just don't _lie_ about it, and over commit.
Taking in a critical router during daylight hours, when Comcast's brick and mortar installations are open, can be awkward. Ensuring that you have failover capacity that can actually take your network load while that equipment is serviced and replaced, and restoring the configuration without interruption, can be nighmarish. This is especially the case in small shops that are trying to grow, shops where the failover capacity has been billed as being enabled but has never been tested or ever actually existed.That is the kind of embarrassment that causes middle managers to get IT personnel fired, especially to avoid the blame themselves.
> the intent
Yes, that's the important word right there. "Intent" does not protect developers from the consequences of failing to learn and use the existing standards. A great deal of my professional life is spent reverse engineering discarded standards, and bringing previously working software into compliance with new standards, while leaving the software compatible with _every single_ old standard for compatibility reason. The amount of work maintaining compatibility often overwhelms the benefits of the new standard quite quickly.
In this case one the main difficulties of the new "universal standard" are that it broke critical features: screen sharing, in particular, is quite useful during conferences. I don't know anyone using Skype in the technical world who hasn't taken advantage of that feature.
Except for the many, many complaints about systemd that are not FUD, such as the overwhelming need by systemd developers to integrate non-init fucntions into the systemd suite such as logging, DHCP, and mounting attached media, the casual reworking of the file-system-hierarchy by ignoring all standards for "/run" and "/media", and insane inconsistencies of "using a symlink for /etc/resolv.conf into the DHCP substructure, except not resetting the symlink if the user puts a flat text file in /etc/resolv.conf and never updating it again"
One of the announced goals of systemd is to "create a stateless Linux", to essentially clear away or prevent any modifications to "/etc" and "/var" and set all local system configuration withing systemd itself. This _breaks_ decades of system configuration work, including every Linux compatible system management tool such as puppit, CFengine, chef, BladeManager, and even installation systems such as Debian'a .dev system, RPM's packaging, and commercial software installation tools. /etc/resolv.conf is merely an example of breaking such tools, but it's a fundamental example of the disdain for non-systemd based system components. It's an example of the chaos being engendered is from such broad goals that have_nothing to do_ with an init system, and that erratically and unpredictably break decades of well-written software that followed the published file system hierarchy standards in favor of underdocumented, non-portable, rapidly evolving and functionally unnecessary switches to a new layout.
Those are only a few of the well documented issues: so overall, there's a lot to like about systemd. But the tendency of its proponents to characterize concerns about it as "FUD" is disingenuous.
It's more an example of the "create a new universal standard" approach to programming: The obligatory XKCD cartoon is:
https://xkcd.com/927/
> though, it might finally lead to ads being delivered from the same host as the rest of the content on the page
This cannot be done. much of graphical and especially streaming content is from more robust "content delivery networks", such as Akamai, that host much larger proxies closer to the web browser's "final mile". Even modest icon or graphical content on a web page will overwhelm many corporate core web servers without these third-party hosted proxies, and it's especially true for ad content. Slashdot itself benefits from its extremely simple format, and can thus handle its load quite easily with quite modest resources. But if they tried to host the advertising content all on the same systems, I'm quite certain it would collapse the servers if not the firewalls themselves.
LibreSSL for drop-in compatibility? Or gnutls?
> Who you fuck is not a matter for the state to be prosecuting or otherwise interfering with.
Except for rape, child molestation, incest, trading sex for promotions, demanding sex for promotions, sex with infected partners who refuse to warn others, and shall i go on?
The "state" has always regulated sex, often though not always for compelling reason.
And probably put the man in a book.
Niven, Pournelle? You've put a few friends in wonderful stories. If you collaborate again, this man deserves it.
Install CygWin and use wget, instead. The CygWin installer fits easily on a USB stick.
Technology for culturing muscle _cells_ is available and has been for some time. Growing enervated, intact muscle with working neural structures, reflexes, and viable attachments to tendon requires far, far more complex support and has never been achieved without a living host, at least not in any reliable literature that I can find.
I'm afraid I didn't say "standing, walking and talking are reflexes", and I'm sorry if I gave that impression. But all of them rely on extensive muscular reflexes, such as the plantar and babinski reflexes in the feet, the patellar reflex in the knee, numerous tongue and throat reflexes for speech, the "arthrokinetic" reflexes for hip movement, etc.
I' urge you to meet people who grew up physically disabled and had some function restored medically later in life. There's a great deal of development of limbs and physiology, of capability, that depends on ongoing stimulation, and which is profoundly distorted by the lack of normal stimulation or activity. And there are profound reasons that dancers and gymnasts start training _early_. The strength and flexibility they require cannot simple be overlaid later. Now take that to an extreme, with a body that hasn't engaged in regular exercise _at all_, until adulthood and understand the useless morass of a functional body that would arise.
Unless there is something like an auto-exercise machine that can provide a full range of human activity to a brainless corpse, the process would be pointless. It seems much safer and potentially more effective to simply raise a clone child and simply confiscate the body when the transplantee is ready.
Yes the idea is a long standing one. Most proposals that have made more physical sense involve a mountain based launcher reaching the highest altitude feasible to reduce atmospheric drag as much as possible. There are big trade-offs of lighter mass of spacecraft with less fuel, and what can feasibly and safely be accelerated along the longest available launch tube. And the more fuel aboard the spacecraft, the more destructive the _inevitable_, eventual launch failure.
> There is no reason why a jar-grown clone would need to be anything other than completely brain dead.
There's not currently any technology, or hope of technology, that can produce muscle tone and muscular reflexes to stand, walk, speak, or handle tools without actually doing the activities. So there are many reasons to give the clones basic human activity. A sodden mas of undeveloped muscle tissue, wrapped around inexperienced or developed organs and undeveloped bone, would require ongoing medical issues as extensive as being extremely old and infirm. "Jar-grown" does not mean mature, viable organ systems.
It requires a vacuum equipped rail gun of quite an incredible length, designed to tolerate spacecraft zipping down it and out of the end of the tube about Mach 25. We don't have anything built or even tested at Mach 1 that can even approach those launch velocities, or be cacuum sealed effectively along its entire launch length. And when it leaves the launch tube at Mach 1, the sonic boom is likely to _destroy_ the end of any launching tube as it reverbates back.
The idea of "new bodies for old" is an old one in fiction and in science. This was explored in some dramatic detail in Lois McMaster Bujold's stories of "Miles Vorkosigan". Old people would have a clone made, the clone raised in foster care until mature enough to support a full grown brain, and then the brain transplanted. Raising the clone required raising them to at least puberty, to support the brain and mature nervous system of the transplant candidate, and they were certainly sentient beings being sacrificed.
> Speaking as an accountant, get/use something compatible with your accounting system.
Oh, my, yes. This. Getting buy-in from the people who actually do the work, so that they know why you selected certain features, is incredibly helpful. This applies to accounting, VOIP, trouble ticket systems, Wikis, email, and any other IT supported system. And the people on the ground may know of critical features or flaws that their own supervisors are simply unaware of or which get "summarized" out of the critical feature list at the several dozens of meetings before such a project is actually implemented.
> facebook.com are IPv6 only internally
Facebook is an interesting case I'd not dealt with. But I'm afraid their external services are rooted in IPv4. I cannot find a single exposed service for them that is only IPv6, and some of their services are IPv4 only. For example, their exposed MX records point to 'msgin.vvv.facebook.com', which is IPv4 only. And this sort of claim that an environment is "dial stack" is generally mistaken. A few exposed services are both, but most of the exposed network services remain IPv4. And until their customer base _stops_ using IPv4-only routers, firewalls, local proxies, and local computer configurations, they _must_ continue to support IPv4.
The key to being able to continue to support IPv4 has been the switch to NAT. When you say they are "IPv6 only internally", they don't have to expose that network space, and most companies do not _want_ to expose that network space. So the internal IPv6 becomes irrelevant to most of their external communications.