If you mean deadbeat after the fact (the guy moving out a few months later), then appropriate time limits should render that problem largely moot.
I was using a colloquialism to encompass any situation where the father doesn't support the child until at least the age of majority.
If you mean deadbeat to mean somebody who doesn't bother to use protection and then doesn't call, then I would argue that the blame falls equally on both parties for that decision, and that those situations aren't really different from any other case of unwanted pregnancy. Those folks are unlikely to pay child support anyway, so there's a very high probability that (no matter what the law or the courts say) the mother will end up taking full responsibility for that child's welfare. So it isn't so much letting these people off the hook as acknowledging the reality of the situation and encouraging the women in question to place those kids in a good home with a family who will be better able to take care of them. Those situations are particularly good candidates for open adoption, assuming they can find a willing family.
If you mean deadbeat to mean somebody who doesn't bother to use protection and then doesn't call, then I would argue that the blame falls equally on both parties for that decision, and that those situations aren't really different from any other case of unwanted pregnancy. Those folks are unlikely to pay child support anyway, so there's a very high probability that (no matter what the law or the courts say) the mother will end up taking full responsibility for that child's welfare. So it isn't so much letting these people off the hook as acknowledging the reality of the situation and encouraging the women in question to place those kids in a good home with a family who will be better able to take care of them.
The option of leaving the child for adoption is completely independent of this discussion and irrelevant to it. Someone who gets a woman pregnant and then cuts and runs may not be poor forever (or even at all).
Emotionally, sure. I was solely talking about the risk to the health of the person involved (infertility, fatal infections, etc.), which for drug-induced abortions are, AFAIK, fairly similar to a miscarriage in terms of the harm to the mother.
Mental health is not an irrelevant side game to physical health.
That isn't a reality, it isn't even accurate. The pregnancy is a trivial matter relative to the impact of having a child and the obligations that go with it and that impact applies to both father and mother.
Pregnancy can easily be life threatening. Raising children is just time consuming. However, bringing up child rearing completely ignores the point of discussion, which is that mothers can have abortions without the father's consent. OP was making an argument that women have "more rights" because they can't be forced to carry a child to term, not that fathers can be forced to rear a child they don't want.
Actually, there are many cases where the rights of one person can outweigh the rights of another. For instance, it is rather common for the state to force a blood draw to gather DNA or test for drugs. That is an invasive procedure with the potential complications including transmission of blood born diseases, infection, and even death.
Blood is drawn in these situations where some evidence of an offense already exists. Not the same at all.
Considering that an abortion is low risk, low impact, and has a very short term impact.
You need to talk to some women who have had abortions.
She can take birth control, the morning after pill, or simply abstain from sex so that really isn't a meaningful point at all. She has both temporary and permanent choices available to her. Further, she has no legal or practical need to inform the man of them. The man does not have these choices. Some vasectomies are reversible in theory but in practice reversal often does not work.
The man has the same high level choices as the woman to avoid pregnancy. Temporary or permanent contraception options, and abstinence.
Up until the moment of birth, yes, but that's only 9 months out of the 18.75 years in which the parents are responsible for that child.
The point of discussion was around why a woman can have an abortion without the father's consent and why describing this as women having "more rights" is moronic.
From an ethical perspective, there's actually something to be said about giving the father the right to terminate child support if the father objected to the mother keeping the baby prior to its birth. Here's why: There's a small minority of women who, when a relationship is on the rocks, stop taking birth control pills in the hopes that having a baby will fix the relationship. If the father had a legally enshrined right to say, "Give the baby up for adoption or I won't be obligated to pay any child support," those sorts of pregnancies would be significantly rarer, and the abortions that often follow a few months later would also be significantly rarer.
I don't disagree at a conceptual level. However, much like my views on capital punishment, this is where I have to let actual real-life outcomes overrule academic theorising. There's always going to be a far larger cohort of "deadbeat dads" than there are women getting pregnant for such purposes.
Not really a fair comparison. One is a procedure that ends a single pregnancy, the other is a largely permanent procedure that ends the possibility of future pregnancies.
IVF is a well-established science. In this scenario if the man is concerned he might want to have more children in the future, he can get some sperm frozen.
For that matter, non-surgical abortions (RU-486) have been legal in the U.S. for more than a decade [...]
Despite conservatives' sadistic fantasies about women being filicidal maniacs held back only by anti-abortion laws, in realty the vast, vast majority of women find an abortion to be a significant and often traumatic process. Even when it's not surgical - though the physical effects of using RU486 aren't exactly benign.
[...] so it's more on the same scale as forcing a man to take Viagra.
No, it's not. At all. That comparison is beyond bad and into absurd.
Another is that a man has no right to a child until after it is born. The mother does not have to inform the father that she is going to have an abortion. One more place where the right of the male is less than the right of a female.
No, that's recognition of the reality that a woman has a lot more at stake in a pregnancy than a man.
And a male cannot force a woman to have an abortion even if he is the one who has to pay for it for the next 18 years.
Indeed. One person cannot force another to undergo an invasive and traumatic medical procedure. Truly they are being oppressed ! On the flipside, a woman who does not want children cannot force her husband to have a vasectomy or be castrated, so it seems that balances out.
I won't comment on your moral compass since you a merely a victim of a society that teaches that property ownership is right and moral. That being said, there is nothing natural about property.
That's drawing a rather long bow. Even "dumb" animals like dogs have a concept of property.
Without the threat of violence people would be able to use whatever it is they can find as long as it requires no violence to use. There is no harm done to others if I use something no one else is, regardless of who thinks they have a right to control that resource.
The problem is - unless you've developed some psychic powers - you never know when someone might want to use their stuff. Just because they don't want to use it now, doesn't mean they won't want to five minutes from now.
This sounds good in theory, but I think it has the potential for unintended consequences, such as companies exchanging all low-paygrade employees for external contractors.
I'm sure there would be unintended consequences, there always are.
That one, however, is so obvious it shouldn't be hard to build something into any legislation to defeat it. "FTE equivalence", for example.
Don't underestimate the risk of having to walk away with a $5M golden parachute and look for another job while your peers are making $10M-$20M at the same time. $5M would let you live a quite comfortable life for 40 years ($125k/yr) - and that's assuming all you did with it was draw down the capital.
It's actually kind of hard to underestimate that risk. It's basically nonexistant.
Maybe the law should state that you get absolutely nothing if the company tanks. Then we'll know that high salaries are given for performance.
I'd rather see a law that said the highest-remunerated employee in the company could not be given more than 10x that of the lowest-remunerated employee in the company. That way the people who deserve the windfall of a successful and profitable business are the ones receiving it.
A government deciding it has the right to control the deal is a government deciding that there IS no property.
So fraud doesn't exist in your worldview ? Curious.
Using that metric, all property belongs to the community.
This is what's called a non-sequitur.
It's funny to hear people say "equal" with respect to socialism. Ideas so good, they have to be mandatory and enforced by people with guns.
Some ideas - property rights, for example - have to be mandatory and enforced by people with guns because of the handful of psychopathic arseholes who ruin it for the rest of us.
I'm not sure why you think it's so difficult to make a quiet machine full of disks. You don't need a great deal of airflow to keep 5400rpm drives at a safe temperature (30ish C). A couple of drive carriers with low-RPM fans in the back and a case with a single big low-RPM fan at the top, and it's done.
I'm sure there are people out there who can hear a fly fart that would consider this thing incredibly loud. For normal people, however, in a normal situation where there are things like ceiling fans, air conditioners, music/TV, refrigerators, dishwashers, and the like, it's utterly unnoticeable.
Hosting a database in a VM, especially a shared VM, is a bad idea. Unless you are absolutely positive it will be used *very* sparingly. Databases want lots of memory all to themselves. A stale cache causing disk IO kills performance pretty fast - and VPSes tend to restrict guest OS RAM. There are many other scenarios that work better on bare metal - even medium-volume Email runs much better outside a VM. Ditto anything directory related (which is very rough on IO once you hit a certain workload saturation point.)
Utter rubbish. A VM will deliver high-90s % of native performance in all but a few pathological corner cases, unless grossly misconfigured or mismanaged.
There's no reason not to host databases, mailservers, or whatever else you want on a VM.
Any computer that doesn't stand 2 ft. tall and sound like a damned jet isn't going to have enough internal storage for a decent-sized media library, and so will be using external storage. The 2 ft tall part isn't usually a problem, but that damned jet-noise just doesn't cut it in a media application.
My fileserver has sixteen drives in it. You can't even tell it's on in a room with normal levels of ambient noise.
Simple when you look at the bottom line the amount they make as a consumer electronics company is insane, all of X86 sales at Apple I doubt would even add up to what they make on the iPod, much less iTunes or iPad. Finally you have the fact that with ARM Apple has complete control of the pipe, something that not only Apple has consistently shown a preference for but its current CEO spent over a year flying around the world making deals JUST to insure their supply chain would be the way they desired? it don't take a psychic to see what the future seems to be holding.
Where's the business case in going to ARM Macs ? What benefits does it deliver ? How are the losses and additional costs recovered ?
"Control" is a means, not an end. It's only important in so much as it can deliver an advantage or eliminate a weakness. It's a struggle to see any advantage delivered by ARM, and similarly difficult to see any weakness from continuing to use Intel CPUs.
What happened to Xserves? Everybody switched to Linux. If running racks of Mac Minis is a profitable business, clearly there's still lots of people out there who want OS X.
Clearly there is a demand for 'affordable' OSX servers, and the fact that companies are rack mounting minis in custom enclosures i think speaks more to just how poor value a mac pro is than anything else.
No, it doesn't say anything about the value of a Mac Pro. A Mac Pro is an even worse option for co-location. If you want to co-locate a Mac, the decision between a Mac Mini and a Mac Pro is clearly the Mac Mini.
What it does speak volumes to is how much people still want Xserves or a virtualisation-friendly OS X license.
No, Apple are just completely disinterested in any market except consumer devices.
Which does make me wonder why they don't try and make some easy cash with a "Virtualised OS X" license at a few $thousand per socket for unlimited instances (like Windows Datacentre). It's not like they'd have to do anything more than write a new EULA.
But when that single, oversold, 5U rack "monster server" (which has been oversold because you have to make that $24k back somehow!) falls over, you'd better have ANOTHER $24k server to swap it out, pronto; because now FIFTY clients are breathing down your neck while you spend a couple of hours getting everything reloaded and working again (that is, assuming you even HAVE that spare $24k server just lying around)...
Except you wouldn't have a single 5U server, you'd have 2x 2U servers or 4x1U servers and a virtualisation cluster. So a single hardware outage would mean a bunch of VMs just restarted automatically (and nearly instantly) on another host. Or if you were really fancy and were using VMware's Fault Tolerance, they'd never have gone down at all.
The question is, what's the processing or storage density of a bunch of Mac Minis vs a racked configuration?
That's only a small part of the question. Then you have things like IO options, redundant connectivity, redundant power, manageability, servicability and longevity.
A HP C7000 blade chassis with BL460c blades will give you 16 blades, equalling 256 HT cores, 8TB RAM and 32 spindles in 10U. That's 25.6 cores, 800GB of RAM and 3.2 spindles for every RU. Other vendors deliver similar numbers.
According to this, Mac Mini Vault gets 140 Mac Minis in a 48U enclosure. 140 Mac Minis would provide 560 HT cores, 2.2T RAM and 280 spindles. So, 11.67 cores, 46.67GB RAM and 5.83 spindles per RU.
From this it's clear blades provide about twice as much processing density and more than an order of magnitude more RAM. They lose out with spindles, but an option if you need more disk is either replacing every second compute blade with a 12-spindle storage blade, delivering 12.8 cores, 400GB RAM and 11.2 spindles per RU, or b) a SAN.
Additionally with blades, you get redundant power and connectivity, pretty much any IO option that matters and huge amounts of bandwidth, remote management, excellent servicability and form factors that stand for 10-15 years. If you're buying a lot of blades and chassis, HP (or IBM or Dell) will probably give you a 50-80% discount from list price. If you're buying a lot of Mac Minis, Apple might give you 20-30% off list, if you're lucky.
Finally, there's the need for custom racks, and substantially higher network and power cabling infrastructure within each rack for the Mac Minis. In fairness jamming four chassis into a single rack presents a heat and power challenge, but I'd be quite surprised if it were even as expensive as Mac Mini racks. Plus you only need two chassis in a rack (easy) and you've got the same density as Mac Minis but with all the other superiorities of blades.
Good on Mac Mini Vault for making money off it, but if they didn't have a specialist product, it's hard to see how they'd have a business model. If Apple were to allow virtualisation of OS X on non-Apple hardware, even at a license price of $thousands/socket, their current model would be screwed. They're lucky Apple has basically zero interest outside of the consumer market.
Yes, of course they change things all the time, but the article was referring to the external form factor for the mini, which hasn't really changed much in years even if the ports and guts have changed quite a bit. The last significant revision was 2010. Unusual for an Apple machine, maybe, but pretty normal for big-vendor, business-oriented PCs.
The problem I was attempting to point out with your analogy is that a drive failure in a 2-drive mirror is the exact same as running a non-raid box with one drive. No, it's not.
The server will keep functioning, sure (though performance may be impaired). But a twin-engined plane on one engine will keep flying as well.
However, the SLA defining your server's purpose has been compromised, because there is now a greater risk of outage and data loss. Similarly, a multi-engine plane with a failed engine is in an "emergency condition" - you don't want to keep flying it around as if nothing were wrong, but it's not going to come instantly crashing down out of the sky.
But going back to the first point, statistically speaking, the probability that you will lose that 80 to 90 percent climb performance in a twin engine plane is greater than the probability you will lose 100% of your climb performance in a single engine plane.
But the _consequences_ are not identical. That's the point. You are implicitly arguing they are because you're making an argument about overall safety.
I didn't mean this to be a general and detailed analysis. I'm not doing any sort of "detailed" analysis, I'm pointing out that your conclusion is based on a false equivalence a child could identify.
I'm pretty relaxed about it, as well.
If you're not going to be doing that, you'd rather just minimize the chances of entering any emergency condition. I'd rather minimise my chances of finding myself in an involuntary glider.
If any actual pilots want to chime in an explain why it's safer to be in a twin-engined plane running on one engine than a single-engined plane running on no engines, I'm happy to listen. However, from a passenger perspective, I'd much rather be in the former situation than the latter.
A twin engine plane running on a single engine is an emergency condition, it's not like running a RAID system in which everything keeps working just fine until you lose more drives. Any competent sysadmin will treat a drive failure in a 2-drive mirror as an emergency condition.
Especially if it happens at the most likely time, during takeoff, where you can lose 80 to 90 percent of climb performance [faasafety.gov]. How is that a worse situation than losing 100% of climb performance ?
The comparison is well-intended but not congruent. A better comparison would be comparing a twin-engined plane to a quad-engined plane. I don't know if it's still true, but it was true that most twin-engine planes couldn't even cruise on two engines. However, most quad-engined planes can cruise on three engines, and that long has been true. Of course, engines are not tires and so there's never going to be better than false congruence here, anyway.
I'm fairly sure all multi-engined planes must be able to remain airborn with only a single engine operational.
Slashdot Mirror
I was using a colloquialism to encompass any situation where the father doesn't support the child until at least the age of majority.
If you mean deadbeat to mean somebody who doesn't bother to use protection and then doesn't call, then I would argue that the blame falls equally on both parties for that decision, and that those situations aren't really different from any other case of unwanted pregnancy. Those folks are unlikely to pay child support anyway, so there's a very high probability that (no matter what the law or the courts say) the mother will end up taking full responsibility for that child's welfare. So it isn't so much letting these people off the hook as acknowledging the reality of the situation and encouraging the women in question to place those kids in a good home with a family who will be better able to take care of them. Those situations are particularly good candidates for open adoption, assuming they can find a willing family.
The option of leaving the child for adoption is completely independent of this discussion and irrelevant to it.
Someone who gets a woman pregnant and then cuts and runs may not be poor forever (or even at all).
Mental health is not an irrelevant side game to physical health.
Pregnancy can easily be life threatening. Raising children is just time consuming.
However, bringing up child rearing completely ignores the point of discussion, which is that mothers can have abortions without the father's consent. OP was making an argument that women have "more rights" because they can't be forced to carry a child to term, not that fathers can be forced to rear a child they don't want.
Blood is drawn in these situations where some evidence of an offense already exists. Not the same at all.
You need to talk to some women who have had abortions.
The man has the same high level choices as the woman to avoid pregnancy. Temporary or permanent contraception options, and abstinence.
The point of discussion was around why a woman can have an abortion without the father's consent and why describing this as women having "more rights" is moronic.
I don't disagree at a conceptual level. However, much like my views on capital punishment, this is where I have to let actual real-life outcomes overrule academic theorising. There's always going to be a far larger cohort of "deadbeat dads" than there are women getting pregnant for such purposes.
IVF is a well-established science. In this scenario if the man is concerned he might want to have more children in the future, he can get some sperm frozen.
Despite conservatives' sadistic fantasies about women being filicidal maniacs held back only by anti-abortion laws, in realty the vast, vast majority of women find an abortion to be a significant and often traumatic process. Even when it's not surgical - though the physical effects of using RU486 aren't exactly benign.
No, it's not. At all. That comparison is beyond bad and into absurd.
No, that's recognition of the reality that a woman has a lot more at stake in a pregnancy than a man.
Indeed. One person cannot force another to undergo an invasive and traumatic medical procedure. Truly they are being oppressed ! On the flipside, a woman who does not want children cannot force her husband to have a vasectomy or be castrated, so it seems that balances out.
Indeed. This is precisely why high-end storage arrays have such atrocious performance.
I won't comment on your moral compass since you a merely a victim of a society that teaches that property ownership is right and moral. That being said, there is nothing natural about property.
That's drawing a rather long bow. Even "dumb" animals like dogs have a concept of property.
Without the threat of violence people would be able to use whatever it is they can find as long as it requires no violence to use. There is no harm done to others if I use something no one else is, regardless of who thinks they have a right to control that resource.
The problem is - unless you've developed some psychic powers - you never know when someone might want to use their stuff. Just because they don't want to use it now, doesn't mean they won't want to five minutes from now.
This sounds good in theory, but I think it has the potential for unintended consequences, such as companies exchanging all low-paygrade employees for external contractors.
I'm sure there would be unintended consequences, there always are.
That one, however, is so obvious it shouldn't be hard to build something into any legislation to defeat it. "FTE equivalence", for example.
Don't underestimate the risk of having to walk away with a $5M golden parachute and look for another job while your peers are making $10M-$20M at the same time.
$5M would let you live a quite comfortable life for 40 years ($125k/yr) - and that's assuming all you did with it was draw down the capital.
It's actually kind of hard to underestimate that risk. It's basically nonexistant.
Maybe the law should state that you get absolutely nothing if the company tanks. Then we'll know that high salaries are given for performance.
I'd rather see a law that said the highest-remunerated employee in the company could not be given more than 10x that of the lowest-remunerated employee in the company. That way the people who deserve the windfall of a successful and profitable business are the ones receiving it.
If the reason you don't take and use other people's stuff is because you're afraid of being hurt, you've got a pretty fucked up moral compass.
So fraud doesn't exist in your worldview ? Curious.
This is what's called a non-sequitur.
Some ideas - property rights, for example - have to be mandatory and enforced by people with guns because of the handful of psychopathic arseholes who ruin it for the rest of us.
I'm not sure why you think it's so difficult to make a quiet machine full of disks. You don't need a great deal of airflow to keep 5400rpm drives at a safe temperature (30ish C). A couple of drive carriers with low-RPM fans in the back and a case with a single big low-RPM fan at the top, and it's done.
I'm sure there are people out there who can hear a fly fart that would consider this thing incredibly loud. For normal people, however, in a normal situation where there are things like ceiling fans, air conditioners, music/TV, refrigerators, dishwashers, and the like, it's utterly unnoticeable.
Utter rubbish. A VM will deliver high-90s % of native performance in all but a few pathological corner cases, unless grossly misconfigured or mismanaged.
There's no reason not to host databases, mailservers, or whatever else you want on a VM.
My fileserver has sixteen drives in it. You can't even tell it's on in a room with normal levels of ambient noise.
Where's the business case in going to ARM Macs ? What benefits does it deliver ? How are the losses and additional costs recovered ?
"Control" is a means, not an end. It's only important in so much as it can deliver an advantage or eliminate a weakness. It's a struggle to see any advantage delivered by ARM, and similarly difficult to see any weakness from continuing to use Intel CPUs.
What happened to Xserves? Everybody switched to Linux.
If running racks of Mac Minis is a profitable business, clearly there's still lots of people out there who want OS X.
No, it doesn't say anything about the value of a Mac Pro. A Mac Pro is an even worse option for co-location. If you want to co-locate a Mac, the decision between a Mac Mini and a Mac Pro is clearly the Mac Mini.
What it does speak volumes to is how much people still want Xserves or a virtualisation-friendly OS X license.
Indeed. A VM is superior in pretty much every measurable way to a physical machine.
No, Apple are just completely disinterested in any market except consumer devices.
Which does make me wonder why they don't try and make some easy cash with a "Virtualised OS X" license at a few $thousand per socket for unlimited instances (like Windows Datacentre). It's not like they'd have to do anything more than write a new EULA.
Except you wouldn't have a single 5U server, you'd have 2x 2U servers or 4x1U servers and a virtualisation cluster. So a single hardware outage would mean a bunch of VMs just restarted automatically (and nearly instantly) on another host. Or if you were really fancy and were using VMware's Fault Tolerance, they'd never have gone down at all.
Based on personal experience, my guess would be "all of it".
That's only a small part of the question. Then you have things like IO options, redundant connectivity, redundant power, manageability, servicability and longevity.
A HP C7000 blade chassis with BL460c blades will give you 16 blades, equalling 256 HT cores, 8TB RAM and 32 spindles in 10U. That's 25.6 cores, 800GB of RAM and 3.2 spindles for every RU. Other vendors deliver similar numbers.
According to this, Mac Mini Vault gets 140 Mac Minis in a 48U enclosure. 140 Mac Minis would provide 560 HT cores, 2.2T RAM and 280 spindles. So, 11.67 cores, 46.67GB RAM and 5.83 spindles per RU.
From this it's clear blades provide about twice as much processing density and more than an order of magnitude more RAM. They lose out with spindles, but an option if you need more disk is either replacing every second compute blade with a 12-spindle storage blade, delivering 12.8 cores, 400GB RAM and 11.2 spindles per RU, or b) a SAN.
Additionally with blades, you get redundant power and connectivity, pretty much any IO option that matters and huge amounts of bandwidth, remote management, excellent servicability and form factors that stand for 10-15 years. If you're buying a lot of blades and chassis, HP (or IBM or Dell) will probably give you a 50-80% discount from list price. If you're buying a lot of Mac Minis, Apple might give you 20-30% off list, if you're lucky.
Finally, there's the need for custom racks, and substantially higher network and power cabling infrastructure within each rack for the Mac Minis. In fairness jamming four chassis into a single rack presents a heat and power challenge, but I'd be quite surprised if it were even as expensive as Mac Mini racks. Plus you only need two chassis in a rack (easy) and you've got the same density as Mac Minis but with all the other superiorities of blades.
Good on Mac Mini Vault for making money off it, but if they didn't have a specialist product, it's hard to see how they'd have a business model. If Apple were to allow virtualisation of OS X on non-Apple hardware, even at a license price of $thousands/socket, their current model would be screwed. They're lucky Apple has basically zero interest outside of the consumer market.
Yes, of course they change things all the time, but the article was referring to the external form factor for the mini, which hasn't really changed much in years even if the ports and guts have changed quite a bit. The last significant revision was 2010.
Unusual for an Apple machine, maybe, but pretty normal for big-vendor, business-oriented PCs.
The problem I was attempting to point out with your analogy is that a drive failure in a 2-drive mirror is the exact same as running a non-raid box with one drive.
No, it's not.
The server will keep functioning, sure (though performance may be impaired). But a twin-engined plane on one engine will keep flying as well.
However, the SLA defining your server's purpose has been compromised, because there is now a greater risk of outage and data loss. Similarly, a multi-engine plane with a failed engine is in an "emergency condition" - you don't want to keep flying it around as if nothing were wrong, but it's not going to come instantly crashing down out of the sky.
But going back to the first point, statistically speaking, the probability that you will lose that 80 to 90 percent climb performance in a twin engine plane is greater than the probability you will lose 100% of your climb performance in a single engine plane.
But the _consequences_ are not identical. That's the point. You are implicitly arguing they are because you're making an argument about overall safety.
I didn't mean this to be a general and detailed analysis.
I'm not doing any sort of "detailed" analysis, I'm pointing out that your conclusion is based on a false equivalence a child could identify.
I'm pretty relaxed about it, as well.
If you're not going to be doing that, you'd rather just minimize the chances of entering any emergency condition.
I'd rather minimise my chances of finding myself in an involuntary glider.
If any actual pilots want to chime in an explain why it's safer to be in a twin-engined plane running on one engine than a single-engined plane running on no engines, I'm happy to listen. However, from a passenger perspective, I'd much rather be in the former situation than the latter.
A twin engine plane running on a single engine is an emergency condition, it's not like running a RAID system in which everything keeps working just fine until you lose more drives.
Any competent sysadmin will treat a drive failure in a 2-drive mirror as an emergency condition.
Especially if it happens at the most likely time, during takeoff, where you can lose 80 to 90 percent of climb performance [faasafety.gov].
How is that a worse situation than losing 100% of climb performance ?
The comparison is well-intended but not congruent. A better comparison would be comparing a twin-engined plane to a quad-engined plane. I don't know if it's still true, but it was true that most twin-engine planes couldn't even cruise on two engines. However, most quad-engined planes can cruise on three engines, and that long has been true. Of course, engines are not tires and so there's never going to be better than false congruence here, anyway.
I'm fairly sure all multi-engined planes must be able to remain airborn with only a single engine operational.