That's a disingenuous thing to say when important packages have come to depend upon it... due to redhat's influence.
Which "important packages"? The only thing I know of offhand is Gnome depending on logind (which is only one component of systemd; it's quite possible to make a system that uses logind without the rest of systemd if you want). And even there, who cares? Where is it written that Gnome needs to be the flagship DE anyway? KDE and Xcfe are better anyway, and legions of Gnome users have abandoned it for those and for MATE and Cinnamon.
The package management is part of the defining quality of the distribution, it's part of what differentiates it from other distributions already.
Um, you could say the exact same thing about the init system. In fact, this is exactly the case with Slackware, which uses a BSD-style init system.
Many people believe as you do that the descriptions of systemd's crappiness are overblown, although I can't imagine anyone familiar with Linux having that opinion, because pulseaudio.
I've been using Linux as my primary desktop since 1999. I've never had any trouble with PA. So yeah, I think the descriptions of systemd's crappiness are overblown, just like they were with PA.
What's theoretical about redhat wanting to exert more influence over Linux?
That's not a conspiracy theory, that's just normal business. RH already exerts a huge amount of influence over Linux, and has for a very long time. However, no one is forcing other distros to adopt systemd. Debian, Ubuntu, and friends haven't adopted RPM yet, and RH has been pushing that thing for almost 20 years now. Companies attempt to exert influence, that's a normal fact of life.
The conspiracy theory is that systemd is affiliated with the NSA and is being used to insert NSA-accessible backdoors in everyone's Linux box.
Slackware counts as a fringe distro, let's face it. They are more like BSD then Linux.
So? What's the problem with that? If it suits you better than the "true" Linux distros like RH/Fedora or Debian, then why don't you use it? Why are you so determined to stick with Linux, even when all the dominant distros have taken a path you clearly don't agree with? All your screaming and yelling isn't likely to make them change course now.
Oh, and the Linux Standard base calls RPM the standard package.
The LSB has done that for well over a decade, and lots of distros (including Debian and all its derivatives) have never switched to RPM. So obviously, RPM isn't the standard.
If you won't consider them simply because some people have arguments you don't like, then you're the problem.
No, I won't consider them because the anti-systemd crowd looks like a bunch of nuts, and bring it up in all kinds of discussions which are clearly way off-topic. Just because systemd involves a bunch of low-level system daemons doesn't mean screaming about how "bad" it is is warranted in a discussion about US politics, medical research, or some news about some web-based company. I never see this with the pro-systemd crowd, only calm and rational discourse rather than the conspiracy-theory-like insane ramblings that the anti-systemd crowd spouts.
The folks at Debian would like to have a word with you over rpm vs. dpkg, among many other huge differences. Slackware and Gentoo and Arch also disagree.
If you can't even compile mplayer yourself and obviously are not capable of making your own distro, then why on earth do you need to do any system-level debugging? Users like you need to just stick with an easy-to-use distro like Ubuntu.
No problem, just find an enterprise vendor who makes a Linux distro that's to your liking (assuming your boss is OK with this, if not then you need to do what your boss tells you to).
If there isn't such a vendor, then the free market has spoken!
The problem with a VM for each application is efficiency. VMs have a certain amount of overhead, and now you want to spin up a whole new OS instance for every single application, and then switch between them? That's bad enough for heavyweight applications like browsers and office programs, but what about small, simple applications like a desktop calculator?
Android's permission system doesn't have this problem because the overhead of having separate users for each app is almost nil on a Unix system.
Yes, this is a big problem with Linux distros and there's been a lot of talk about it and about solutions for it. The systemd guys even proposed an interesting solution that involves btrfs.
However, the OP seemed to be talking about having virtual machines so that different users have full control over their applications, unlike now where for the most part applications have to be installed by someone with root access.
Computer manufacturers already offer that. You can buy a modern computer which easily has the hardware power necessary to do all that. What they don't offer is the software for that, because computer manufacturers don't make software, they just stick Windows on there or let you put your own software on it. Unless you're talking about Apple of course, but obviously even though they control the software on their machines they can't figure out how to make all that work.
The problem with your idea is that, while most of the software components needed to do all that already exist, no one has bothered putting them all together into a package that's appliance-like and easy-to-use by regular people. In fact, regular people can't even understand the advantage of this idea of yours, so no one's going to bother putting the effort into making a shiny product which does it and marketing it to people at Walmart and Best Buy.
It does seem like it'd be a very interesting open-source project to work on.
Ah, another typical moronic anti-systemd troller, proving his utter stupidity.
Maybe if you had bothered reading my whole post, you would have seen this:
The problem is that you anti-systemd wackos can't just discuss it in places like this thread (where it's actually on-topic, a story about Linux Mint continuing to provide both systemd and upstart)
Using old versions isn't a viable option once security updates stop,
No problem, just do security updates yourself. You can backport fixes on your own, you don't need a distro to do it for you. Since you obviously know more than the distro maintainers (or else you wouldn't be disagreeing with their decision to adopt systemd), this shouldn't be a problem for you.
Slackware is not a option for a lot of scenarios (say running Oracle DB in the enterprise).
Then create your own distro.
And how many anti-systemd people are running OracleDB in an enterprise anyway?
Just like you don't need the FCC to take things off the air because you can change the channel, you don't need us to stop discussing systemd and why it's literally both bad and wrong. You can just skip the discussion.
No, we actually can't. The problem is that you anti-systemd wackos can't just discuss it in places like this thread (where it's actually on-topic, a story about Linux Mint continuing to provide both systemd and upstart). Instead, you have to jump into every Linux discussion anywhere and ramble on and on with your gripings. We can't have a discussion about, say, The GIMP without some anti-systemd nutter jumping in and complaining about systemd now.
So even if any of your complaints were actually valid, you all look like a bunch of raving loons because a bunch of you just won't shut the fuck up about it in discussions which have absolutely nothing to do with systemd. It's like the boy who cried wolf: at this point, I'm so sick of the anti-systemd nutters that I really don't take anything they say seriously, so if some of them do have any valid technical complaints, it's lost in the noise.
Just look at some university presidents and their absolutely spartan lifestyles due to working for non-profits, such as multimillion dollar homes purchased for them by the university.
While there's definitely some largess going on at many universities these days, those multimillion dollar homes are not purchased for presidents by the universities. They're owned by the universities, and the presidents are allowed to live there as long as they're president. It's free housing. When the president gets another job or retires, he has to leave the mansion and go find himself another place to live, so that the new president can live in the mansion.
Are you going to complain about the President of the US getting to live in the White House for free too?
But that is 'picking and choosing', and it's not 'simply providing a pipe'. Why should my ISP get to decide that my neighbour's Skype conversation is more important than my multiplayer, if I'm paying for the connection?
That's not what he's talking about. The problem is that the ISPs want to pick and choose between destinations, not protocols. For example, Comcast wants to give home subscribers a high-speed connection to certain video services which have paid big fees to Comcast (or better yet to Comcast's own video service), but a much slower connection to video services which have not paid them any fees.
It's one thing to prioritize video over, say, peer-to-peer sharing or FTP. It's entirely another thing to prioritize one video service over another video service.
I'm guessing it's either the GOP stance on abortion or immigration. That's usually how they keep commoners voting for them while they push legislation to benefit the 0.0001%.
Why do we design a UI by writing *text* in 2015? It should be possible to auto-generate a UI
Um, it is, and this has been possible for quite some time. Lots of IDEs auto-generate code for UIs. Qt's QtCreator comes to mind, and I'm pretty sure MS has had something like this for ages. I'm sure there's several others.
In fact, perhaps the whole packaging of Linux systems should be different. What if every user was running in a virtual environment where they can install any software they want, with the other users being isolated from those changes.
It's been like this for ages. When was the last time you saw a desktop Linux system where more than one person actually used it, and there were multiple accounts on it? Better yet, when was the last time you saw such a system where multiple people were using it simultaneously? Linux, just like any UNIX system, certainly has the capability built-in to have multiple simultaneous users (since it is a multi-user multi-processing time-sharing system), but in practice no one does that much any more because we use PCs now, not shared centralized machines. Servers are a little different of course, but even here people are frequently running VMs these days so they have a full Linux environment to themselves; the big exception I can think of is ultra-cheap shared web hosting, and there the capabilities available to users are limited.
Actually, if by "everyone", you mean passenger vehicles (as your sentence implies), then road costs would be largely unaffected. A single large truck does about as much damage as several thousand cars.
That's a good point which I didn't think of. However, another factor to think of is road widening: how much money is spent annually building new roads or widening and improving existing roads because of passenger car traffic? Surely if we got many of the cars off the road, we wouldn't need such wide roads and as much road work to handle the traffic volume.
You are right about the effect of trucks on the roads and their share of the costs however. Cargo simply shouldn't be hauled long-haul except maybe for express freight; it should be hauled on trains in shipping containers, and then hauled by truck the last few miles to its destination.
That's okay, they won't let you have "pleasure use only" unless you have multiple cars.
I don't see why not; what if you're a telecommuter?
How do they do that without asking or auditing you yearly? I have only ever had them ask my mileage when I switched carriers.
That's easy: the adjuster checks the odometer when you have a wreck. If your mileage if far beyond what you had claimed (when you got the policy plus the annual mileage), then they deny your claim because you committed fraud. Now you have a wrecked car and you're not getting any compensation for it, and they're dropping you.
No, it's not, but you're also leaving out all the other destinations along the NE Corridor: Baltimore, DC, and Boston namely (and a bunch of smaller stops in between). What's the distance between DC and Boston? That's probably comparable to LA to SF. On the other hand, there really isn't much between SF and LA, whereas there's a lot of traffic between all those destinations on the NE corridor.
Try telling your auto insurance company your car is for "pleasure use only" and you only put 5k miles per year on it, then use it for commuting 50 miles each way daily. See what happens when you get in a wreck and they figure out your car has far more mileage than you claimed.
Your rates change with the number of miles you put on the car yearly. Cars used for lots of commuting get higher rates than cars which spend most of their time sitting in your garage and only being driven on weekends. It's not linear, however. And a single road trip isn't going to make a difference either. But if you make a habit of driving between SF and LA regularly, versus taking the plane or train, then it will be reflected in your insurance rates if you truthfully report your annual mileage to your insurance company.
That one's irrelevant to a driver. The road is going to be there whether you use it or not, and there's no additional cost to use it (unless it's a toll road of course; this could actually add up to a lot: crossing bridges in the NY area is very expensive). It's another "prisoner's dilemma" deal: if everyone stopped driving so much and took trains more, then road costs would decrease and less of our tax dollars would be needed for that. But that just isn't going to happen. Basically, if the government really wants to get people to drive less, it needs to go ahead and build the infrastructure needed to provide a viable alternative. Build it and they will come. But this is risky: building it is expensive, and if it's shittily done (too slow, too inconvenient, too expensive), then people won't use it, proving the nay-sayers right.
This is why we need to move forward with SkyTran. It's ultra-convenient (you get your own personal car, and it comes when you order it, not according to a schedule), very cheap to deploy compared to trains and light rail ($1M per mile last I heard), reasonably fast (150mph between cities), quick to build (once the SkyTran factory is built; the components for the system are factory-built and assembled quickly on-site), it's really the answer to our transportation problems, but no one wants to do it because they think "it's impossible" or "it won't work". Most people probably thought that about smartphones back in 2000, the internet back in 1993, and personal computers in 1985 (even though there working versions of both at the time, just not scaled up and made as accessible as now).
Currently, there's no TSA checkpoints at Amtrak stations. TSA actually tried pulling that a while back somewhere and the Amtrak Police forcibly escorted them off the premises. With Amtrak trains, you can just walk right on, ticket or not (they check the tickets later, after the train is in motion).
Now, how long this will remain this way is anyone's guess.
But for traveling along the northeast corridor, Amtrak is actually pretty convenient and reasonably fast compared to airlines, since there's no security and it doesn't stop at traffic lights like cars do (and between Baltimore and Philly, and Philly and NYC, it can get up to 125mph). The biggest downside is that the train only runs at certain times (usually once or twice a day for a given route I think). As for what to do on the other end, that depends on where you're going. If your destination is Manhattan, then a car is actually a huge liability and expense (where do you park it? Parking in Manhattan is horrifically expensive.), so the train works out great since it takes you right to the city and then you can walk or take the subway wherever you're going. In Philly, if you're going someplace downtown the same deal probably applies; same with DC and Boston. If you're going somewhere in the outskirts, then you're looking at the rental car thing. As for cost savings, there's no loss of cost savings if you take mass transit in NYC; a one-way subway fare is $2.25 I believe. Compared to the train ticket (or an airline ticket), that's nothing.
Finally, one thing people are missing in these cost analyses is the true cost to drive a car; it's more than gas, it's also tolls (at least in the NE), and maintenance and upkeep, especially tires. Tires are expensive and they wear proportionally with mileage driven. There's also insurance costs, but a single 500-mile trip probably isn't going to affect that, but if you make a lot of trips, your insurance company does charge you more unless you're underreporting your annual mileage (which can cause them to reject a claim, so that's a bad idea).
Slashdot Mirror
That's a disingenuous thing to say when important packages have come to depend upon it... due to redhat's influence.
Which "important packages"? The only thing I know of offhand is Gnome depending on logind (which is only one component of systemd; it's quite possible to make a system that uses logind without the rest of systemd if you want). And even there, who cares? Where is it written that Gnome needs to be the flagship DE anyway? KDE and Xcfe are better anyway, and legions of Gnome users have abandoned it for those and for MATE and Cinnamon.
The package management is part of the defining quality of the distribution, it's part of what differentiates it from other distributions already.
Um, you could say the exact same thing about the init system. In fact, this is exactly the case with Slackware, which uses a BSD-style init system.
Many people believe as you do that the descriptions of systemd's crappiness are overblown, although I can't imagine anyone familiar with Linux having that opinion, because pulseaudio.
I've been using Linux as my primary desktop since 1999. I've never had any trouble with PA. So yeah, I think the descriptions of systemd's crappiness are overblown, just like they were with PA.
What's theoretical about redhat wanting to exert more influence over Linux?
That's not a conspiracy theory, that's just normal business. RH already exerts a huge amount of influence over Linux, and has for a very long time. However, no one is forcing other distros to adopt systemd. Debian, Ubuntu, and friends haven't adopted RPM yet, and RH has been pushing that thing for almost 20 years now. Companies attempt to exert influence, that's a normal fact of life.
The conspiracy theory is that systemd is affiliated with the NSA and is being used to insert NSA-accessible backdoors in everyone's Linux box.
Slackware counts as a fringe distro, let's face it. They are more like BSD then Linux.
So? What's the problem with that? If it suits you better than the "true" Linux distros like RH/Fedora or Debian, then why don't you use it? Why are you so determined to stick with Linux, even when all the dominant distros have taken a path you clearly don't agree with? All your screaming and yelling isn't likely to make them change course now.
Oh, and the Linux Standard base calls RPM the standard package.
The LSB has done that for well over a decade, and lots of distros (including Debian and all its derivatives) have never switched to RPM. So obviously, RPM isn't the standard.
If you won't consider them simply because some people have arguments you don't like, then you're the problem.
No, I won't consider them because the anti-systemd crowd looks like a bunch of nuts, and bring it up in all kinds of discussions which are clearly way off-topic. Just because systemd involves a bunch of low-level system daemons doesn't mean screaming about how "bad" it is is warranted in a discussion about US politics, medical research, or some news about some web-based company. I never see this with the pro-systemd crowd, only calm and rational discourse rather than the conspiracy-theory-like insane ramblings that the anti-systemd crowd spouts.
Get ready for a bunch of nasty posts telling you to go move to Europe. This is why everything is so backwards in America: NIH.
The folks at Debian would like to have a word with you over rpm vs. dpkg, among many other huge differences. Slackware and Gentoo and Arch also disagree.
If you can't even compile mplayer yourself and obviously are not capable of making your own distro, then why on earth do you need to do any system-level debugging? Users like you need to just stick with an easy-to-use distro like Ubuntu.
No problem, just find an enterprise vendor who makes a Linux distro that's to your liking (assuming your boss is OK with this, if not then you need to do what your boss tells you to).
If there isn't such a vendor, then the free market has spoken!
The problem with a VM for each application is efficiency. VMs have a certain amount of overhead, and now you want to spin up a whole new OS instance for every single application, and then switch between them? That's bad enough for heavyweight applications like browsers and office programs, but what about small, simple applications like a desktop calculator?
Android's permission system doesn't have this problem because the overhead of having separate users for each app is almost nil on a Unix system.
Yes, this is a big problem with Linux distros and there's been a lot of talk about it and about solutions for it. The systemd guys even proposed an interesting solution that involves btrfs.
However, the OP seemed to be talking about having virtual machines so that different users have full control over their applications, unlike now where for the most part applications have to be installed by someone with root access.
Computer manufacturers already offer that. You can buy a modern computer which easily has the hardware power necessary to do all that. What they don't offer is the software for that, because computer manufacturers don't make software, they just stick Windows on there or let you put your own software on it. Unless you're talking about Apple of course, but obviously even though they control the software on their machines they can't figure out how to make all that work.
The problem with your idea is that, while most of the software components needed to do all that already exist, no one has bothered putting them all together into a package that's appliance-like and easy-to-use by regular people. In fact, regular people can't even understand the advantage of this idea of yours, so no one's going to bother putting the effort into making a shiny product which does it and marketing it to people at Walmart and Best Buy.
It does seem like it'd be a very interesting open-source project to work on.
Ah, another typical moronic anti-systemd troller, proving his utter stupidity.
Maybe if you had bothered reading my whole post, you would have seen this:
The problem is that you anti-systemd wackos can't just discuss it in places like this thread (where it's actually on-topic, a story about Linux Mint continuing to provide both systemd and upstart)
Or are you too stupid to understand this?
Using old versions isn't a viable option once security updates stop,
No problem, just do security updates yourself. You can backport fixes on your own, you don't need a distro to do it for you. Since you obviously know more than the distro maintainers (or else you wouldn't be disagreeing with their decision to adopt systemd), this shouldn't be a problem for you.
Slackware is not a option for a lot of scenarios (say running Oracle DB in the enterprise).
Then create your own distro.
And how many anti-systemd people are running OracleDB in an enterprise anyway?
Just like you don't need the FCC to take things off the air because you can change the channel, you don't need us to stop discussing systemd and why it's literally both bad and wrong. You can just skip the discussion.
No, we actually can't. The problem is that you anti-systemd wackos can't just discuss it in places like this thread (where it's actually on-topic, a story about Linux Mint continuing to provide both systemd and upstart). Instead, you have to jump into every Linux discussion anywhere and ramble on and on with your gripings. We can't have a discussion about, say, The GIMP without some anti-systemd nutter jumping in and complaining about systemd now.
So even if any of your complaints were actually valid, you all look like a bunch of raving loons because a bunch of you just won't shut the fuck up about it in discussions which have absolutely nothing to do with systemd. It's like the boy who cried wolf: at this point, I'm so sick of the anti-systemd nutters that I really don't take anything they say seriously, so if some of them do have any valid technical complaints, it's lost in the noise.
There's a simple solution for you: don't upgrade.
For Gnome, it's really easy: just switch to MATE. It's Gnome2 by another name.
For Windows, it's easy: just stick with 7.
For systemd, it's easy: just switch to Slackware, or use an old version of your preferred distro.
Just look at some university presidents and their absolutely spartan lifestyles due to working for non-profits, such as multimillion dollar homes purchased for them by the university.
While there's definitely some largess going on at many universities these days, those multimillion dollar homes are not purchased for presidents by the universities. They're owned by the universities, and the presidents are allowed to live there as long as they're president. It's free housing. When the president gets another job or retires, he has to leave the mansion and go find himself another place to live, so that the new president can live in the mansion.
Are you going to complain about the President of the US getting to live in the White House for free too?
But that is 'picking and choosing', and it's not 'simply providing a pipe'. Why should my ISP get to decide that my neighbour's Skype conversation is more important than my multiplayer, if I'm paying for the connection?
That's not what he's talking about. The problem is that the ISPs want to pick and choose between destinations, not protocols. For example, Comcast wants to give home subscribers a high-speed connection to certain video services which have paid big fees to Comcast (or better yet to Comcast's own video service), but a much slower connection to video services which have not paid them any fees.
It's one thing to prioritize video over, say, peer-to-peer sharing or FTP. It's entirely another thing to prioritize one video service over another video service.
I'm guessing it's either the GOP stance on abortion or immigration. That's usually how they keep commoners voting for them while they push legislation to benefit the 0.0001%.
Why do we design a UI by writing *text* in 2015? It should be possible to auto-generate a UI
Um, it is, and this has been possible for quite some time. Lots of IDEs auto-generate code for UIs. Qt's QtCreator comes to mind, and I'm pretty sure MS has had something like this for ages. I'm sure there's several others.
In fact, perhaps the whole packaging of Linux systems should be different. What if every user was running in a virtual environment where they can install any software they want, with the other users being isolated from those changes.
It's been like this for ages. When was the last time you saw a desktop Linux system where more than one person actually used it, and there were multiple accounts on it? Better yet, when was the last time you saw such a system where multiple people were using it simultaneously? Linux, just like any UNIX system, certainly has the capability built-in to have multiple simultaneous users (since it is a multi-user multi-processing time-sharing system), but in practice no one does that much any more because we use PCs now, not shared centralized machines. Servers are a little different of course, but even here people are frequently running VMs these days so they have a full Linux environment to themselves; the big exception I can think of is ultra-cheap shared web hosting, and there the capabilities available to users are limited.
Actually, if by "everyone", you mean passenger vehicles (as your sentence implies), then road costs would be largely unaffected. A single large truck does about as much damage as several thousand cars.
That's a good point which I didn't think of. However, another factor to think of is road widening: how much money is spent annually building new roads or widening and improving existing roads because of passenger car traffic? Surely if we got many of the cars off the road, we wouldn't need such wide roads and as much road work to handle the traffic volume.
You are right about the effect of trucks on the roads and their share of the costs however. Cargo simply shouldn't be hauled long-haul except maybe for express freight; it should be hauled on trains in shipping containers, and then hauled by truck the last few miles to its destination.
That's okay, they won't let you have "pleasure use only" unless you have multiple cars.
I don't see why not; what if you're a telecommuter?
How do they do that without asking or auditing you yearly? I have only ever had them ask my mileage when I switched carriers.
That's easy: the adjuster checks the odometer when you have a wreck. If your mileage if far beyond what you had claimed (when you got the policy plus the annual mileage), then they deny your claim because you committed fraud. Now you have a wrecked car and you're not getting any compensation for it, and they're dropping you.
CA is not the northeast.
No, it's not, but you're also leaving out all the other destinations along the NE Corridor: Baltimore, DC, and Boston namely (and a bunch of smaller stops in between). What's the distance between DC and Boston? That's probably comparable to LA to SF. On the other hand, there really isn't much between SF and LA, whereas there's a lot of traffic between all those destinations on the NE corridor.
No, it's mileage-based.
Try telling your auto insurance company your car is for "pleasure use only" and you only put 5k miles per year on it, then use it for commuting 50 miles each way daily. See what happens when you get in a wreck and they figure out your car has far more mileage than you claimed.
Your rates change with the number of miles you put on the car yearly. Cars used for lots of commuting get higher rates than cars which spend most of their time sitting in your garage and only being driven on weekends. It's not linear, however. And a single road trip isn't going to make a difference either. But if you make a habit of driving between SF and LA regularly, versus taking the plane or train, then it will be reflected in your insurance rates if you truthfully report your annual mileage to your insurance company.
That one's irrelevant to a driver. The road is going to be there whether you use it or not, and there's no additional cost to use it (unless it's a toll road of course; this could actually add up to a lot: crossing bridges in the NY area is very expensive). It's another "prisoner's dilemma" deal: if everyone stopped driving so much and took trains more, then road costs would decrease and less of our tax dollars would be needed for that. But that just isn't going to happen. Basically, if the government really wants to get people to drive less, it needs to go ahead and build the infrastructure needed to provide a viable alternative. Build it and they will come. But this is risky: building it is expensive, and if it's shittily done (too slow, too inconvenient, too expensive), then people won't use it, proving the nay-sayers right.
This is why we need to move forward with SkyTran. It's ultra-convenient (you get your own personal car, and it comes when you order it, not according to a schedule), very cheap to deploy compared to trains and light rail ($1M per mile last I heard), reasonably fast (150mph between cities), quick to build (once the SkyTran factory is built; the components for the system are factory-built and assembled quickly on-site), it's really the answer to our transportation problems, but no one wants to do it because they think "it's impossible" or "it won't work". Most people probably thought that about smartphones back in 2000, the internet back in 1993, and personal computers in 1985 (even though there working versions of both at the time, just not scaled up and made as accessible as now).
Currently, there's no TSA checkpoints at Amtrak stations. TSA actually tried pulling that a while back somewhere and the Amtrak Police forcibly escorted them off the premises. With Amtrak trains, you can just walk right on, ticket or not (they check the tickets later, after the train is in motion).
Now, how long this will remain this way is anyone's guess.
But for traveling along the northeast corridor, Amtrak is actually pretty convenient and reasonably fast compared to airlines, since there's no security and it doesn't stop at traffic lights like cars do (and between Baltimore and Philly, and Philly and NYC, it can get up to 125mph). The biggest downside is that the train only runs at certain times (usually once or twice a day for a given route I think). As for what to do on the other end, that depends on where you're going. If your destination is Manhattan, then a car is actually a huge liability and expense (where do you park it? Parking in Manhattan is horrifically expensive.), so the train works out great since it takes you right to the city and then you can walk or take the subway wherever you're going. In Philly, if you're going someplace downtown the same deal probably applies; same with DC and Boston. If you're going somewhere in the outskirts, then you're looking at the rental car thing. As for cost savings, there's no loss of cost savings if you take mass transit in NYC; a one-way subway fare is $2.25 I believe. Compared to the train ticket (or an airline ticket), that's nothing.
Finally, one thing people are missing in these cost analyses is the true cost to drive a car; it's more than gas, it's also tolls (at least in the NE), and maintenance and upkeep, especially tires. Tires are expensive and they wear proportionally with mileage driven. There's also insurance costs, but a single 500-mile trip probably isn't going to affect that, but if you make a lot of trips, your insurance company does charge you more unless you're underreporting your annual mileage (which can cause them to reject a claim, so that's a bad idea).