I far prefer Postgresql to MySQL, but while doing everything in the database is often "neater" it usually makes it more expensive to scale.
If you let the DB do most of the work when the DB runs out of grunt it often costs $$$$$$$ to make faster.
If you let the app servers do most of the work, when the app servers run out of grunt you add a bunch of app servers for $$$$$. Big diff.
I know the DB people will say a lot about maintaining data integrity etc.
Sure use the data integrity stuff whenever it doesn't really cost the DB anything (paid for already, or only a little cost), but once you start adding lots of stored procedures and triggers, rules etc, you better be aware that scaling up a database server often costs a lot more money.
No problem for small businesses that are going to grow to medium sized businesses - since Intel, Seagate and friends are already well on their way to taking care of your future needs. They might already have stuff for you today, that you need tomorrow.
But if you're intending to be _huge_ in the next few years, it's not going to be cheap. People can say all sorts of things about database clustering but please check the ugly details first there are often lots of limitations and gotchas.
So basically you'd want to keep the load on the DB low enough so that you can avoid the messy expensive scenarios as long as possible.
e.g. No problem if by the time you need a fault-tolerant DB "teraflop" server with 1TB RAM and 10GB/sec IO, Dell and friends are selling them for about USD2999.
Whereas if you need a fault tolerant DB that needs 10TB working mem, 100GB/sec IO and "terahurts" of performance, _TODAY_, it's going to cost you.
> Bear in mind that something exciting is only exciting in comparison to what happened before
Peaks and troughs are good.
But left4dead doesn't have you travelling for 5 boring minutes. It's just a short pause before you have lots of stuff happening again.
As for creating a contrast between boring and exciting. The really boring bits can be saved for "real life" when I'm not playing games.
For example, I don't mind boring plane flights that much. I don't want too much excitement when travelling in real life. Especially since respawning is hard for most people.
I don't know about the Concorde but in the case of planes like the airbus with the safety measures active you could do a desperate control like full hard right, and the plane will just bank right to the max of 67 degrees.
It just won't end up rolling (unless you turn off the restrictions). BUT the thing is, in most cases that's what the pilot wants anyway, even emergency cases.
You could try a 90 degree bank, but the plane might break up or crash anyway, or you could still end up killing almost the same numbers of passengers. They're not fighter jets with pilots in g suits.
Same thing goes for a "full pull up" to try to avoid hitting stuff. They don't want to stall and lose height (and make it even more likely for a crash). So the plane provides the max power available at the moment, and limits the angle of attack based on the airspeed to something that won't stall the plane.
If that's not enough, I doubt even "direct control" will help.
Sometimes the plane and people in it are just doomed and there's nothing the computers and pilots can do about it once they've gone down that path.
Only way out is to not be in that situation in the first place. Better training, screening, discipline, maintenance etc.
A significant change in latency is a fundamental enough change for me. It's like the difference between the Postal Service and the telephone. And the telephone was most certainly a significant advance.
Similarly it's a huge advance that people around the world can now watch very many _good_ lectures from various places around the world, and skim through to see which are worth watching all in just a few minutes.
You really couldn't do the same thing in the preinternet/Youtube era.
Back then you'd have to: know of the existence of the lecture, figure out who to contact, write a letter or call, figure out how to get the tape across international borders.
Now someone in India/Malaysia/etc can learn from the top lecturers around the world if they want.
"Not listening" is not the problem at all. You can definitely override the airbus computers if you have a different opinion.
The real problem is if the computer has already doomed you before you even have time to have a different opinion.
But if the computer system is as reliable or more reliable than the airframe etc or the above average pilot, then it's not likely to be a big problem in practice.
What probably happened is the plane got into a situation where the computers couldn't figure out what to do, but the situation was so bad that the pilots couldn't figure out what to do in time.
Whether the computers got the plane into that situation or the weather or something else did, I don't know. But I doubt the storm helped.
Actually, there are cases where the pilots saved the day though there were no preset procedures and they had to "wing it";).
There was no preset procedure for flying when all four 747 engines shut down due to volcanic ash, and the cockpit windows get ash-blasted so the pilots can barely see out of them.
But they still managed to make it to the airport and land without anybody dying or getting injured.
And for some reason there wasn't a preset procedure for gliding when out of fuel for Air Canada 767 pilots (they only simulated one engine landings, not zero engine landings!)- see the Gimli Glider. And even Gimli wasn't listed as a potential landing site in Air Canada's manuals. The pilot just happened to know of its existence.
Then there was the case of "complete loss of hydraulic flight control systems due to being hit by a surface to air missile". I doubt there's a preset procedure in DHL's flight training manual for that;).
Fortunately the pilot there had attended a seminar given by the captain of United 232 who had to fly a plane a similar way (which was a less fortunate flight as lots of people died).
That said, in the Gimli Glider case I think a computer could actually help a lot, assuming the sensors still work. But in the case of being hit by a missile, I think a skilled human pilot who doesn't want to die, is going to be better than a programmer+computer or "preset procedure guy" at "saving the day".
Preset procedures and computer assistance are good help, especially for mediocre pilots. You don't want them to do the wrong thing.
But the procedure writer can't list all the weird stuff, way in advance. There are too many possible weird things that could happen to list down usefully for pilots to follow. And the programmer won't dare program those into the computer, because a slight difference in the assumptions could be disastrous.
When the preset procedures don't fit, what you need is pilots that have a good sense of what the plane can do, then often they can figure out what the plane can and cannot do when bits fail or get blown away.
Guild Wars is a bit like that. You still have to do stuff to get to places, but once you've been to a town, you can get back to it in a few clicks.
No monthly fee. But there are other annoyances - it's not like WAR where you can "queue up" for a PvP battle and do something else more "PvE"ish in the meantime. And they keep changing the skills every 2 weeks or so.
With Guild Wars, you have to run/walk/fight to new cities/towns first (or get someone to "run" you there - e.g. do all the hardwork while you just tag along). After that, you can teleport to that town or any other town you have been before.
It's a _chore_ having to keep running to places you've been before.
Like "same old" cutscenes you can't skip, but must keep pressing "Next" (to kill anything that gets in your way) till you finally reach the real destination (the actual battle).
Being able to teleport straight to places you've been before is a good thing. I don't care if the world feels small in that way - as long as it's diverse enough.
It's like being in a small shop with a huge variety of products, and a different product on every inch of the shelves that you can choose if you like. Compared to being in a huge hypermarket with shelves and shelves of the _same_ items, so you need to walk about a lot more to get to the stuff you want.
Guild Wars is a bit like WoW Lite in some ways. So a lot of ppl won't like it.
But the fix for SSL is not about fixing the CAs, it's getting the browsers to behave more like SSH (or better). Then at least the browser will give useful warnings for a change, that'll help people who really care about security. While it won't help the "click through" users, nothing much will help those against attackers anyway.
Then that's the way it should be - YOU decide who you want to trust, with the help of technology.
In contrast with DNSSEC, you're stuck with Verisign for.com.
_Someone_else_ decides who you HAVE to trust for some TLD, whether you like it or not.
Verisign is the CA that signed Microsoft certificates for someone who was not Microsoft- http://www.amug.org/~glguerin/opinion/revocation.html Network Solutions ( a subsidiary of Verisign that does DNS stuff) has also been known for doing dubious stuff like domain front running.
So yes, things are different in DNSSEC- you get to be stuck with one crappy company for.com and no choice other than pick some other TLD.
Different yes, but sure doesn't sound better to me at all.
p.s. FWIW, Verisign owns Geotrust, who owns RapidSSL who kept using MD5 for certs till the exploit became public.
Yeah the popular press tends to paint stuff in black/white.
Anyway it is an interesting research result (esp to salespeople/conmen;) ), maybe someone else should try to replicate it or try it with some modifications - get someone of different gender, handedness, ask for different stuff, show the request using a (not too flashy) phone with a large screen.
Well in her case maybe the connections and other normal growth related changes could be regarded as damage and repaired, just like the other health problems she had that were self-repaired.
Yes, with the SSL situation, you can pick any random "el-cheapo" SSL CA _you_ like (or can con). Your users' browsers will still trust them. There's less of a monopoly there compared to DNSSEC right?
Go to Firefox,Tools,Options,Advanced,Encryption,View Certificates,Authorities. Pick the cheapest CA there who will do what you want[1], and you're set.
But don't forget, you still end up having to pay them every year or so, it doesn't matter that they are crap, you still need to pay toll to somebody. You yourself say that the SSL CA system is crap. So it is a _parasitic_ system - takes money from people and doesn't really help them very much in return.
So what makes you so confident that the situation with DNSSEC will be so much better for the users and site owners? How will the DNSSEC situation be more competitive than the SSL CA situation? How is it better?
In contrast, the DNSCurve situation is definitely very competitive - it would cost users and sites about as much to use dnscurve, as it does to make an ssh connection to someplace. I can certainly see why CAs would prefer DNSSEC to DNSCurve;).
[1] Has Mozilla ever removed a CA for being sloppy yet? I know there have been CAs who have signed stuff they shouldn't have.
Don't be silly, the first step is getting effective and practical artificial gravity and radiation shielding.
We don't even have that, and we talk about going to the moon and mars. We've already got stuff to those planets. Why waste billions of dollars and time doing the same crap over and over again and still not be able to stay in space long term? It's not as if Mars is a much friendlier environment than "Space", and you have to spend months in "Space" anyway getting to Mars.
One of the subsequent steps is seeing if a bunch of humans can stay in the "next gen" space station (with artificial gravity and shielding) for months AND maintain good physical and mental health.
You probably need decent power sources. Solar panels would be fine for places not too far from the Sun - so you may have to figure out how to manufacture those in space.
Once you have all that, then you can do some tests on mining and construction in space. After the initial tests, you could send a probe to the asteroid belt to test some of the mining tech out to see if there are gotchas before sending the "real thing".
But if you're impatient you could actually take a risk and send the space station with humans all the way there without sending the robot miner mission first. After all if the mining is unsuccessful, it doesn't necessarily mean the humans will die- since by then you would have a space station that will work long term, and would not be a piece of primitive crap like the ISS or an unreliable suicide canister like the space shuttle.
When you have something that can mine asteroids and build new space stations and other stuff, you can get one into a parking orbit over Mars and do all the "fun" planet stuff.
And by that time maybe the humans on the Earth no longer need to pay that much for such space missions - since more of the resources are coming from the asteroids and the space colonies.
If this sort of stuff is really infeasible, we might as well forget about this humans in space thing. Big waste of time and money.
1) Plenty of asteroids about in the asteroid belt to mine. 2) Yes using resources from planets may eventually be necessary, BUT my point is you need to get the space station tech right first - radiation shielding, artificial gravity etc. Then you can travel to planets and colonize them. It's easier to colonize and mine a planet when you already have a colony in orbit around the planet for you to start from.
Otherwise what do you propose to do? Send all the necessary resources all the way from Earth to the planet? That sounds like a bigger waste of resources to me. Sure it might work (at a very high cost), but it does not scale and is not a long term solution.
If we want humans in space sustainably, we need to develop practical tech to do so. Sending people to Mars/Moon with our current tech is just "Space Theatre/Circuses". Not worth the bang for buck.
All the tech from Apollo can't even get us space station where humans can live on fairly permanently without their bones going soft, and bring up future generations of humans.
And how much did the Apollo project cost? A LOT. And here are NASA and friends trying to do the same thing again.
Sounds like a big waste of time and resources to me.
> what kind of shielding it would take to protect someone from a full blown solar flare? We don't have good tech there either.
You're making my point for me. If you're going to take months to travel to Mars or wherever, you are going to need that shielding anyway. Unless you like suicide missions.
Might as well throw in the artificial gravity then.
Slashdot Mirror
I far prefer Postgresql to MySQL, but while doing everything in the database is often "neater" it usually makes it more expensive to scale.
If you let the DB do most of the work when the DB runs out of grunt it often costs $$$$$$$ to make faster.
If you let the app servers do most of the work, when the app servers run out of grunt you add a bunch of app servers for $$$$$. Big diff.
I know the DB people will say a lot about maintaining data integrity etc.
Sure use the data integrity stuff whenever it doesn't really cost the DB anything (paid for already, or only a little cost), but once you start adding lots of stored procedures and triggers, rules etc, you better be aware that scaling up a database server often costs a lot more money.
No problem for small businesses that are going to grow to medium sized businesses - since Intel, Seagate and friends are already well on their way to taking care of your future needs. They might already have stuff for you today, that you need tomorrow.
But if you're intending to be _huge_ in the next few years, it's not going to be cheap. People can say all sorts of things about database clustering but please check the ugly details first there are often lots of limitations and gotchas.
So basically you'd want to keep the load on the DB low enough so that you can avoid the messy expensive scenarios as long as possible.
e.g. No problem if by the time you need a fault-tolerant DB "teraflop" server with 1TB RAM and 10GB/sec IO, Dell and friends are selling them for about USD2999.
Whereas if you need a fault tolerant DB that needs 10TB working mem, 100GB/sec IO and "terahurts" of performance, _TODAY_, it's going to cost you.
Yeah and it's DNA marked too I bet.
> Where is my right to bare poultry?
On aisle 4, next to the salad dressing.
Thing is the Buddhists and many Hindus consider it a penalty not a bonus.
Yeah to me "distribution to the public" = even strangers, people you don't know can easily get the stuff.
Since only about 24% of the people in the world have access to the Internet is the Internet a "public" network or a limited network?
> Bear in mind that something exciting is only exciting in comparison to what happened before
Peaks and troughs are good.
But left4dead doesn't have you travelling for 5 boring minutes. It's just a short pause before you have lots of stuff happening again.
As for creating a contrast between boring and exciting. The really boring bits can be saved for "real life" when I'm not playing games.
For example, I don't mind boring plane flights that much. I don't want too much excitement when travelling in real life. Especially since respawning is hard for most people.
I don't know about the Concorde but in the case of planes like the airbus with the safety measures active you could do a desperate control like full hard right, and the plane will just bank right to the max of 67 degrees.
It just won't end up rolling (unless you turn off the restrictions). BUT the thing is, in most cases that's what the pilot wants anyway, even emergency cases.
You could try a 90 degree bank, but the plane might break up or crash anyway, or you could still end up killing almost the same numbers of passengers. They're not fighter jets with pilots in g suits.
Same thing goes for a "full pull up" to try to avoid hitting stuff. They don't want to stall and lose height (and make it even more likely for a crash). So the plane provides the max power available at the moment, and limits the angle of attack based on the airspeed to something that won't stall the plane.
If that's not enough, I doubt even "direct control" will help.
Sometimes the plane and people in it are just doomed and there's nothing the computers and pilots can do about it once they've gone down that path.
Only way out is to not be in that situation in the first place. Better training, screening, discipline, maintenance etc.
A significant change in latency is a fundamental enough change for me. It's like the difference between the Postal Service and the telephone. And the telephone was most certainly a significant advance.
Similarly it's a huge advance that people around the world can now watch very many _good_ lectures from various places around the world, and skim through to see which are worth watching all in just a few minutes.
You really couldn't do the same thing in the preinternet/Youtube era.
Back then you'd have to: know of the existence of the lecture, figure out who to contact, write a letter or call, figure out how to get the tape across international borders.
Now someone in India/Malaysia/etc can learn from the top lecturers around the world if they want.
> The "Nobel Prize winner" myth: Every school child will have access to a Nobel Prize winner
In some ways yes:
http://www.youtube.com/watch?v=qn8PNMTSlwo
Plenty of other lectures/talks from MIT, Stanford, and other universities around the world are available online.
> it is doubtful that Nobel Prize winners will look forward to getting a few thousand e-mail messages a day.
I'm sure Feynman isn't too worried about that :).
FWIW, you can learn a lot from people without sending email to them, or communicating with them.
"Not listening" is not the problem at all. You can definitely override the airbus computers if you have a different opinion.
The real problem is if the computer has already doomed you before you even have time to have a different opinion.
But if the computer system is as reliable or more reliable than the airframe etc or the above average pilot, then it's not likely to be a big problem in practice.
What probably happened is the plane got into a situation where the computers couldn't figure out what to do, but the situation was so bad that the pilots couldn't figure out what to do in time.
Whether the computers got the plane into that situation or the weather or something else did, I don't know. But I doubt the storm helped.
Actually, there are cases where the pilots saved the day though there were no preset procedures and they had to "wing it" ;).
There was no preset procedure for flying when all four 747 engines shut down due to volcanic ash, and the cockpit windows get ash-blasted so the pilots can barely see out of them.
See: http://en.wikipedia.org/wiki/British_Airways_Flight_9
But they still managed to make it to the airport and land without anybody dying or getting injured.
And for some reason there wasn't a preset procedure for gliding when out of fuel for Air Canada 767 pilots (they only simulated one engine landings, not zero engine landings!)- see the Gimli Glider. And even Gimli wasn't listed as a potential landing site in Air Canada's manuals. The pilot just happened to know of its existence.
Then there was the case of "complete loss of hydraulic flight control systems due to being hit by a surface to air missile". I doubt there's a preset procedure in DHL's flight training manual for that ;).
http://en.wikipedia.org/wiki/2003_Baghdad_DHL_attempted_shootdown_incident
Fortunately the pilot there had attended a seminar given by the captain of United 232 who had to fly a plane a similar way (which was a less fortunate flight as lots of people died).
See: http://en.wikipedia.org/wiki/United_Airlines_Flight_232
That said, in the Gimli Glider case I think a computer could actually help a lot, assuming the sensors still work. But in the case of being hit by a missile, I think a skilled human pilot who doesn't want to die, is going to be better than a programmer+computer or "preset procedure guy" at "saving the day".
Preset procedures and computer assistance are good help, especially for mediocre pilots. You don't want them to do the wrong thing.
But the procedure writer can't list all the weird stuff, way in advance. There are too many possible weird things that could happen to list down usefully for pilots to follow.
And the programmer won't dare program those into the computer, because a slight difference in the assumptions could be disastrous.
When the preset procedures don't fit, what you need is pilots that have a good sense of what the plane can do, then often they can figure out what the plane can and cannot do when bits fail or get blown away.
Guild Wars is a bit like that. You still have to do stuff to get to places, but once you've been to a town, you can get back to it in a few clicks.
No monthly fee. But there are other annoyances - it's not like WAR where you can "queue up" for a PvP battle and do something else more "PvE"ish in the meantime. And they keep changing the skills every 2 weeks or so.
Small is not a bad thing.
With Guild Wars, you have to run/walk/fight to new cities/towns first (or get someone to "run" you there - e.g. do all the hardwork while you just tag along). After that, you can teleport to that town or any other town you have been before.
It's a _chore_ having to keep running to places you've been before.
Like "same old" cutscenes you can't skip, but must keep pressing "Next" (to kill anything that gets in your way) till you finally reach the real destination (the actual battle).
Being able to teleport straight to places you've been before is a good thing. I don't care if the world feels small in that way - as long as it's diverse enough.
It's like being in a small shop with a huge variety of products, and a different product on every inch of the shelves that you can choose if you like. Compared to being in a huge hypermarket with shelves and shelves of the _same_ items, so you need to walk about a lot more to get to the stuff you want.
Guild Wars is a bit like WoW Lite in some ways. So a lot of ppl won't like it.
> Where will your food come from? Do you have any idea how much land area it takes to feed your ass, and other people?
We'll have that same problem even if staying on the Moon and Mars. I doubt you can grow food in those places without any "sealed bubble".
So if we're thinking about space in the long term we have to solve that as well.
But the fix for SSL is not about fixing the CAs, it's getting the browsers to behave more like SSH (or better). Then at least the browser will give useful warnings for a change, that'll help people who really care about security. While it won't help the "click through" users, nothing much will help those against attackers anyway.
Then that's the way it should be - YOU decide who you want to trust, with the help of technology.
In contrast with DNSSEC, you're stuck with Verisign for .com.
_Someone_else_ decides who you HAVE to trust for some TLD, whether you like it or not.
Verisign is the CA that signed Microsoft certificates for someone who was not Microsoft- http://www.amug.org/~glguerin/opinion/revocation.html
Network Solutions ( a subsidiary of Verisign that does DNS stuff) has also been known for doing dubious stuff like domain front running.
So yes, things are different in DNSSEC- you get to be stuck with one crappy company for .com and no choice other than pick some other TLD.
Different yes, but sure doesn't sound better to me at all.
p.s. FWIW, Verisign owns Geotrust, who owns RapidSSL who kept using MD5 for certs till the exploit became public.
Switzerland, good social freedoms in the "big picture". And has cities in the "best cities to live in" list.
:).
It seems there are lots of social rules in the "small picture" though - when you can do your laundry, how untidy your garden can be, etc
I use Gandi but they're french...
;).
I'm just hoping the potential for communication and jurisdiction problems will cause the assholes to give up before Gandi surrenders
FWIW, Gandi's T&C are better than most, and I wouldn't use godaddy.
Let me guess, you used to work for Verizon, and now work for AT&T?
:).
I wonder what would happen if you started working for the Tax dept, then again maybe AIG or Citibank needs your skills more
Yeah the popular press tends to paint stuff in black/white.
;) ), maybe someone else should try to replicate it or try it with some modifications - get someone of different gender, handedness, ask for different stuff, show the request using a (not too flashy) phone with a large screen.
Anyway it is an interesting research result (esp to salespeople/conmen
Well in her case maybe the connections and other normal growth related changes could be regarded as damage and repaired, just like the other health problems she had that were self-repaired.
> So she's Wolverine?
But maybe with her "normal" growth is considered damage and thus "fixed".
Just like the brain tumour that appeared and vanished, any parts that try to go "next stage" get repaired.
Yes, with the SSL situation, you can pick any random "el-cheapo" SSL CA _you_ like (or can con). Your users' browsers will still trust them. There's less of a monopoly there compared to DNSSEC right?
;).
Go to Firefox,Tools,Options,Advanced,Encryption,View Certificates,Authorities. Pick the cheapest CA there who will do what you want[1], and you're set.
But don't forget, you still end up having to pay them every year or so, it doesn't matter that they are crap, you still need to pay toll to somebody. You yourself say that the SSL CA system is crap. So it is a _parasitic_ system - takes money from people and doesn't really help them very much in return.
So what makes you so confident that the situation with DNSSEC will be so much better for the users and site owners? How will the DNSSEC situation be more competitive than the SSL CA situation? How is it better?
In contrast, the DNSCurve situation is definitely very competitive - it would cost users and sites about as much to use dnscurve, as it does to make an ssh connection to someplace. I can certainly see why CAs would prefer DNSSEC to DNSCurve
[1] Has Mozilla ever removed a CA for being sloppy yet? I know there have been CAs who have signed stuff they shouldn't have.
Don't be silly, the first step is getting effective and practical artificial gravity and radiation shielding.
We don't even have that, and we talk about going to the moon and mars. We've already got stuff to those planets. Why waste billions of dollars and time doing the same crap over and over again and still not be able to stay in space long term? It's not as if Mars is a much friendlier environment than "Space", and you have to spend months in "Space" anyway getting to Mars.
One of the subsequent steps is seeing if a bunch of humans can stay in the "next gen" space station (with artificial gravity and shielding) for months AND maintain good physical and mental health.
You probably need decent power sources. Solar panels would be fine for places not too far from the Sun - so you may have to figure out how to manufacture those in space.
Once you have all that, then you can do some tests on mining and construction in space. After the initial tests, you could send a probe to the asteroid belt to test some of the mining tech out to see if there are gotchas before sending the "real thing".
But if you're impatient you could actually take a risk and send the space station with humans all the way there without sending the robot miner mission first. After all if the mining is unsuccessful, it doesn't necessarily mean the humans will die- since by then you would have a space station that will work long term, and would not be a piece of primitive crap like the ISS or an unreliable suicide canister like the space shuttle.
When you have something that can mine asteroids and build new space stations and other stuff, you can get one into a parking orbit over Mars and do all the "fun" planet stuff.
And by that time maybe the humans on the Earth no longer need to pay that much for such space missions - since more of the resources are coming from the asteroids and the space colonies.
If this sort of stuff is really infeasible, we might as well forget about this humans in space thing. Big waste of time and money.
1) Plenty of asteroids about in the asteroid belt to mine.
2) Yes using resources from planets may eventually be necessary, BUT my point is you need to get the space station tech right first - radiation shielding, artificial gravity etc. Then you can travel to planets and colonize them. It's easier to colonize and mine a planet when you already have a colony in orbit around the planet for you to start from.
Otherwise what do you propose to do? Send all the necessary resources all the way from Earth to the planet? That sounds like a bigger waste of resources to me. Sure it might work (at a very high cost), but it does not scale and is not a long term solution.
If we want humans in space sustainably, we need to develop practical tech to do so. Sending people to Mars/Moon with our current tech is just "Space Theatre/Circuses". Not worth the bang for buck.
All the tech from Apollo can't even get us space station where humans can live on fairly permanently without their bones going soft, and bring up future generations of humans.
And how much did the Apollo project cost? A LOT. And here are NASA and friends trying to do the same thing again.
Sounds like a big waste of time and resources to me.
> what kind of shielding it would take to protect someone from a full blown solar flare? We don't have good tech there either.
You're making my point for me. If you're going to take months to travel to Mars or wherever, you are going to need that shielding anyway. Unless you like suicide missions.
Might as well throw in the artificial gravity then.