I am forced to assume the articles have confused 'open relays' with 'open proxies'. Almost no one sends email using open relays anyway, they're all virus infect end users on broadband who are unwittedly operating machines that will forward connections to wherever.
The only part of the internet funded by government money is government servers and networks. Normal internet traffic does not and cannot travel over government owned wires. Even if you go to whitehouse.gov, you're over private wires the entire way, until you hit whatever government building houses it.
It sounds like I'm quibbling, but the saying 'part of the internet is public' is just wrong.
It's like talking about how part of a mall are publically funded because the USPS has rented space for a post office in it. That doesn't have anything to do with what the mall can or cannot do with restricting speech inside it, or what the stores in the mall can do, although the post office itself clearly can't restrict speech within. (In fact, it couldn't agree to a lease where the owner reserved the right to restrict speech within the post office, either.)
I sometimes sit in amazement and wonder how 'life' got to be so sacred, but apparently people can just randomly waste our time, in huge blocks, of millions of people at once, and it's okay. What is 'life' if not 'time'?
Anyone who waste my time is my enemy, they are slowly killing me. Anyone who wastes large numbers of people's time is society's enemy.
Dismissed with prejudice means 'You don't have a case, and cannot possibly ever have a case, so you can't even file again about this.'.
Whereas dismissal without prejudice means 'They cannot prove the case, but they may, indeed, have one, once they get their act together and file in the right court or under the correct statue or whatever.'
(Of course, you can appeal either of these.)
Dismissal with prejudice is what judges do to people who sue the government for alien mind control.
It has nothing to do with the name. From the website:
[netqmail] is comprised of qmail-1.03 plus a patch file, some documentation, and a shell script which prepares the files for compilation.
I.e., people are having to distribute qmail as is, and rig a script to patch it. Exactly like I said.
netqmail is just putting all that in one package.
Re:Grudgingly going back to Sendmail.
on
Postfix 2.1 Released
·
· Score: 2, Insightful
That's what I'm saying.
If you're having an excessively large volume of mail, and want to add more relay servers, the solution is to add more identical servers with round-robin DNS, and possibly a larger pipe.
Having them pass mail among themselves via a shared NFS server isn't going to help anything at all. The ideal solution would be to have most mail in and out in ten seconds.
And, if large amounts of undeliverable mail are piling up, a shared queue just hurts things, not helps them. That just means every server will check every message to see if it's time to deliver it, repeatedly. Over the network.
Now, if he wants to do something sort of like this that actually help, a solution would be what postfix calls a 'fallback relay'...a server where mail that can't get delivered in X time goes. Where you can have some extra sanity checking that will quickly delete mail that can't ever be delivered, but you don't want to run on every message.
Re:Grudgingly going back to Sendmail.
on
Postfix 2.1 Released
·
· Score: 2, Informative
That's exactly what I said. You've built a system where one system will accept a message, and then one system will attempt to deliver it, which provides no benefit at all over having one system deliver it from start to finish, except you've added race conditions and file sharing and waste all around.
As for talking about deleting things out of the queue, that's just crazy. There are commands to do that, and they run just fine remotely. (Not that running around deleting mail from a delivery queue is a normal action in the first place, and I suspect you came up with that because you know what you're talking about is silly.)
How the hell is that FUD? There's a rather large patchset already out there because the maintainer hasn't released a new version in 5 years. (In fact, there are a few posts here asserting that's a feature, that no new versions make it easy to patch, which is what normal people would call 'denial'.)
You know, just asserting things that aren't true doesn't magically make them true.
Re:Grudgingly going back to Sendmail.
on
Postfix 2.1 Released
·
· Score: 2, Informative
Because someone's a loon who's made an amazingly complicated mail system, that's why.
He's not only building relay servers that transfer mail between themselves, which there is absolutely no reason to do, (They should accept mail from X and forward to Y, not play hot potato with it. Having more than one server is fine, but they don't have anything to say to each other.) he's making them transfer mail between themselves using the mail queue instead of SMTP.
Which is rather akin to setting up a shuttle bus system between the airport and a hotel, realizing you need more than one bus to handle the load, and coming up with the 'solution' of running each bus halfway and transferring all the passengers at the midway point. Each bus driver only needs to be able to handle half the route, think of all the time and training he'll save!
With postfix, of course, he'd have to build a delivery station to offload the passengers to, but with sendmail, he apparently can transfer passengers directly from bus to bus! (Which, despite sendmail's shortcomings, I doubt was intentional.)
Gee, you really like quoting me out of context, don't you? And even fighting strawmen.
Someone said it wasn't free software, someone else asked why, and I explained. It's not free because it can't be forked, the only version that can ever be is the offical version, except for unwieldy patches. You're at the mercy of the maintainer.
I didn't say you shouldn't use it, I didn't say it was bad, I just stated facts. And you acted like I attacked your mother.
Re:Grudgingly going back to Sendmail.
on
Postfix 2.1 Released
·
· Score: 2, Insightful
And I suppose you have some explanation of how having an unreachable backup MX lowers your chances of getting email to your primary MX?
...
I thought not. SMTP doesn't work that way. A mail server tries all possible MX servers, in order.
If my primary is, for god know what reason, unreachable, then it wastes a single second of the mail sender's time to check the backup before queueing the mail. That's the entire net effect. You'll get your delayed mail a second later, in theory, which doesn't actually happen because no one runs around setting retry timers that trigger the exact second anyway.
As for relay servers...you need hundreds of those why, exactly? You're operating one of those crazy-ass overloaded networks, apparently, where mail bounces through two or even three SMTP servers both going in and out.
Those networks are brokenly designed. Sorry, but it's true. I understand why you can't fix them at this point, but it doesn't mean they were designed correctly.
There's absolutely no reason not to have a single incoming point (Or even sets of points, via NFS directories.) for each account, where a message comes in and is stored 'locally', and absolutely no reason not to have one outgoing server for each outgoing message, when then sends the message directly to the recepient. (Those can be the same system, or not.) And maybe backup MXs for collecting mail when the system is down. No one has ever explained to me why a mail system needs to be more complicated than that. Anything more complicated than that is probably just historical nonsense laying around.
I would rather a failure mode that cannot be caused on purpose via people with an agenda.
However, it's perfectly possible to reduce the error rate to wholesale tampering by the staff of an entire precinct.
You just have a computer generate a computer and human readable paper ballot, and scan those to tally the votes. No more worries about what the person meant...it's printed completely unambigiously on the paper, alone, instead of having everyone listed and trying to figure out wwhich was marked.
Some people have suggested barcodes, but I think there's too much danger the machines will 'accidently' print one name and a different barcode, and, really, printing OCRable type isn't that hard.
And as the machines would also tally who was voted for, you've gotten rid of the 'hide a few bags of ballots before they count them' that sometimes happens. The total ballots printed better equal the total votes plus the total discarded ballots.
The only possiblities for vote tampering left are a) figuring out who can vote, which we still have problems with, and b) the actions of an entire precinct, because they can go in and vote as many times as they want, or let others do that...which constrained by the fact that people can sit there and observe them. With electronic machines, though, at least we could make them do the repeated voting while they are open, and not at 4 in the morning before the polls open.
There's a reason they don't let other people into the booth with you. It's not just a 'secret ballot', in that no one can know how you voted if you don't want them too, it's supposed to be an unprovable ballot, where you can't prove how you voted even if you wanted to. If you can prove how you voted, someone can give you 10 dollars to vote a certain way...or even threat your life unless you do prove how you voted.
This is why there are concerns being raised about camera phones in voting booths.
Re:Grudgingly going back to Sendmail.
on
Postfix 2.1 Released
·
· Score: 3, Informative
Why the hell are you sharing a mail queue? It's not like more than one server can send the message at a time, or receive it. And postfix supports NFS mailboxes just fine.
And why the hell are you bouncing spam? Delete spam or reject spam, do not bounce spam.
It sounds like you don't know what you're doing, or have a really stupid setup.
And, BTW, if you're getting hammered because you're the backup MX, which spammers like to pound, it might make sense to set up a tertiary MX server that doesn't actually exist. Spammers will go after that, instead, and never hit you, as almost all spamming software is written by complete fucking morons. Whereas actual mail that failed to get your primary server will just your backup. (Or, failing to get your backup, they will then fail to get your tertiary and just queue the mail, and start back over when they retry.)
I, personally, set up a 'backup MX' record to point at one of my IPs that didn't actually run a mail server, and cut my daily spam attempts by 30%.
While I currently use postfix and think it's great, I will point out there's another mail server out there without the issues of sendmail or qmail, named exim, that's pretty good also.
Re:this SMTP server vs Qmail and Sendmail
on
Postfix 2.1 Released
·
· Score: 5, Informative
qmail isn't free software because it's non-forkable.
You can freely redistribute the source and binaries compiled from clean source. And you can distribute patches to it.
However, the point is, the qmail maintainer is the only person who can release new versions of qmail. And hence it's not free software.
There are two very large dangers with qmail...that it will go off in a random direction no one agrees with, and you'll either have to follow along or go that way, and that the qmail maintainer will just stop releasing new versions. With free software, if enough people use it, they will simply make a fork...but they can't do that with qmail. Technically they could grab a random version and keep building patches off that, but that becomes unmaintainable real fast.
In other words, qmail is basically 'freeware', not 'free software', although it does come in source form, and you have been granted the ability to modify it and even share the modifications. But not the end result.
Nope, censorship is the wrong word, as the government is nothing to do with this. They're just protecting themselves from time-wasters and law suits related to copyright infringment.
I don't know what universe you live in, but mine, universities are often owned and operated by, gasp, the government.
That's what I said! Sci-fi involves scientific or technological changes to the world, and fantasy involves impossible changes to the world, and then they both deal with said changes. However, there are possible changes that do not fit under either sci-fi or fantasy, like alternate histories, and 1984, which is a sort of 'alternate future'.
And Divine Comedy is fantasy. It's the special brand of fantasy that no one seems to recognize, religion fantasy, like Touched by an Angel or the Left Behind series. It's 'What if this religion were correct?'. I suspect people don't like to recognize it because, to them, it's already true. But just because someone thinks they are a vampire doesn't mean we should stop classifying Buffy as fantasy, and just because someone believes in angels doesn't make Touched by an Angel not fantasy.
And Animal Farm is not really SF/F, because it's just an allegory. Animal Farm doesn't explore what would actually happen if animals became sentient, started talking, and took over, it's an allegory of 'workers' taking over. If it was real sci-fi or fantasy we'd have the authorities showing up to cart all these amazing talking animals off to be tested on.
That's not to say that sci-fi can't be an allegory, almost all good sci-fi is, but saying 'animals can talk' and then not dealing with that concept is not exploring how that new world works. If they had treated that concept, with the animals having no rights, and comparing that to slavery, sure, they used a fantasy alteration to reality and explore the world, and thus it's fantasy.
But Animal Farm, while it does have a single fantastic concept, doesn't actually explore the effects of that concept on society, instead chosing to explore a worker's revolt. Which is all well and good, I like the book, but it's not fantasy anymore than Mickey Mouse cartoons are fantasy because they have a talking mouse.
Science fiction, and fantasy, postulate a world that functions differently than ours, be it by a little, or a lot, and explore how that world works. It's not about the future at all.
Now, you can argue that there's not enough scientific change in 1984 to make it sci-fi, but that's never been what sci-fi is about. Even if you don't want to call it sci-fi, you have to admit it's on the same spectrum as sci-fi and fantasy, it's just that fantasy postulates impossible changes in society due to 'magic', science fiction postulates possible changes in science due to technology, and 1984 postulates possible changes in society, not due to anything stated.
Which is, BTW, the same reason that alternate histories get grouped under sci-fi and fantasy....they belong there. It's 'what would happen if...', that's all sci-fi and fantasy is, it's just set in the past.
Some people use the term 'speculative fiction', and, yes, that is what 1984 is, just like it is what Gulliver's Travels is, or what a Conneticut Yankee in King Arthur's Court is.
I don't have any idea how what I described is a cause or an effect...I described an experiment. One that can be done, and in fact has been done, repeatedly, at various intervals, over the last 100 years, in attempt to measure the movement of the planet 'through' space with light.
Not the slightest trace of this movement has ever been detected. There are only the two options that I said...either we, for some unknown reason, are not moving through space, which makes the entire universe spin around us in random patterns, or our movement through space somehow has no effect on how fast light moves past us, that unlike every other speed in existence, the speed of light is not additive with other velocities, aka, it's 'constant'.
There simply are no other possibilities to explain this oddity. As velocity is just space divided by time, you obviously need to postulate some some sort of time and/or space change to explain this velocity inconsistancy. Einstein postulated both, that they are in fact the same thing and converted back and forth, which keeps us from having to explain where space came from or where the extra time went after it was used.
And just in case you're wondering, there are other experiments that rule out the first possiblity, although I can't think of them off the top of my head. You'd probably object to them because, obviously, if we're checking the speed of light with anything except vs the earth's speed, at some point the 'clock' is going to moving. Interesting how all those moving clocks fail in exactly the correct way to make us think relativity exists, but whatever.
Now, if you want to use another set of math to explain the universe, feel free...99.999% of the time, people use Newtonic physics anyway. We don't need to worry about time dilation when everyone is going within a close fraction of the same speed in the same direction.
You don't need a moving clock to show the speed of light is constant, because the planet itself is moving. You just take a beam of light, and split it in half with a mirror, send it along two perpendicular paths, bounce it back, and see if it arrives in sync. As the earth is moving around the sun, it's trivially easy to make one direction that's 'forward' along the motion of the sun, and one direction that's 'sideways'.
Ah, but there's nothing proving the mirrors are the same distance, or possibly the combined motion of the earth and the sun and everything cancelled out...but this is the clever bit: You wait 12 hours, when the entire arrangment has been rotated so it faces the opposite direction with regard to the movement of the sun, thanks to the rotation of the planet, and run it again. In fact, run it at six hour intervals, just for the hell of it.
Oh, and here's the other gag: You don't need any clocks. You don't need to measure the time, the two beams of light coming together will create a noticable interference pattern. If the beams at any point took even slightly longer, the pattern would change.
And, just so you don't actually have to run that experiment, I'll tell you the results...no matter in what direction you aim the arrangement in, in relation to the movement of the planet, light always takes the same amount of time to run the same distance. Yes, this experiment has been run plenty of times, with more than enough accuracy to measure the movement of the planet.
So you have two options: Either the planet isn't moving, with the rest of the universe spinning wildly around it, or the speed of light is constant no matter how fast you're moving.
Do you honestly think the scientific community accepted such a crazy theory as relativity without having inconvenient facts that required such a theory?
Build a giant hollow sphere in space. Blow it up. A second after the structure comes apart, attach plumb bobs to the pieces flying in different directions. (Good luck with that.)
Tada. Non-parallel plumb bobs due to accelleration.
Likewise, buld an infinitely large plane with enough mass to generate 1 g. Attach plumb bobs above it.
Tada. Parallel plumb bobs due to gravity. (Well, technically, they'd lean slightly toward each other due to their own gravity, but so would the ones in a rocket. You could measure with one, and just move it, if you cared.)
Yes, my two examples are physically difficult. But they certainly can exist in theory.
Now, if you're just taking issue with the 'we can't prove if we're accelerating or falling' that's commonly claimed...we all know that's a lie. Otherwise, we wouldn't know, on this planet, which we clearly do. The standard example uses an elevator/small room specifically to keep you from being able to measure variations in direction.
And, no gravity at the surface of the earth isn't uniformly curved, anyway. It's entirely possible to end up with two points ten feet apart that do have an exactly parallel 'downs', or even 'downs' which point away from each other, due to the composition of matter in the earth's crust.
Stick you in a room over one of those places, and you'd cleverly prove that gravity there was not only from acceleration, but from centripical acceleration.
Slashdot Mirror
I am forced to assume the articles have confused 'open relays' with 'open proxies'. Almost no one sends email using open relays anyway, they're all virus infect end users on broadband who are unwittedly operating machines that will forward connections to wherever.
The only part of the internet funded by government money is government servers and networks. Normal internet traffic does not and cannot travel over government owned wires. Even if you go to whitehouse.gov, you're over private wires the entire way, until you hit whatever government building houses it.
It sounds like I'm quibbling, but the saying 'part of the internet is public' is just wrong.
It's like talking about how part of a mall are publically funded because the USPS has rented space for a post office in it. That doesn't have anything to do with what the mall can or cannot do with restricting speech inside it, or what the stores in the mall can do, although the post office itself clearly can't restrict speech within. (In fact, it couldn't agree to a lease where the owner reserved the right to restrict speech within the post office, either.)
Anyone who waste my time is my enemy, they are slowly killing me. Anyone who wastes large numbers of people's time is society's enemy.
With apologizes to Dogbert. ;)
Whereas dismissal without prejudice means 'They cannot prove the case, but they may, indeed, have one, once they get their act together and file in the right court or under the correct statue or whatever.'
(Of course, you can appeal either of these.)
Dismissal with prejudice is what judges do to people who sue the government for alien mind control.
[netqmail] is comprised of qmail-1.03 plus a patch file, some documentation, and a shell script which prepares the files for compilation.
I.e., people are having to distribute qmail as is, and rig a script to patch it. Exactly like I said.
netqmail is just putting all that in one package.
If you're having an excessively large volume of mail, and want to add more relay servers, the solution is to add more identical servers with round-robin DNS, and possibly a larger pipe.
Having them pass mail among themselves via a shared NFS server isn't going to help anything at all. The ideal solution would be to have most mail in and out in ten seconds.
And, if large amounts of undeliverable mail are piling up, a shared queue just hurts things, not helps them. That just means every server will check every message to see if it's time to deliver it, repeatedly. Over the network.
Now, if he wants to do something sort of like this that actually help, a solution would be what postfix calls a 'fallback relay'...a server where mail that can't get delivered in X time goes. Where you can have some extra sanity checking that will quickly delete mail that can't ever be delivered, but you don't want to run on every message.
As for talking about deleting things out of the queue, that's just crazy. There are commands to do that, and they run just fine remotely. (Not that running around deleting mail from a delivery queue is a normal action in the first place, and I suspect you came up with that because you know what you're talking about is silly.)
You know, just asserting things that aren't true doesn't magically make them true.
He's not only building relay servers that transfer mail between themselves, which there is absolutely no reason to do, (They should accept mail from X and forward to Y, not play hot potato with it. Having more than one server is fine, but they don't have anything to say to each other.) he's making them transfer mail between themselves using the mail queue instead of SMTP.
Which is rather akin to setting up a shuttle bus system between the airport and a hotel, realizing you need more than one bus to handle the load, and coming up with the 'solution' of running each bus halfway and transferring all the passengers at the midway point. Each bus driver only needs to be able to handle half the route, think of all the time and training he'll save!
With postfix, of course, he'd have to build a delivery station to offload the passengers to, but with sendmail, he apparently can transfer passengers directly from bus to bus! (Which, despite sendmail's shortcomings, I doubt was intentional.)
Someone said it wasn't free software, someone else asked why, and I explained. It's not free because it can't be forked, the only version that can ever be is the offical version, except for unwieldy patches. You're at the mercy of the maintainer.
I didn't say you shouldn't use it, I didn't say it was bad, I just stated facts. And you acted like I attacked your mother.
I thought not. SMTP doesn't work that way. A mail server tries all possible MX servers, in order.
If my primary is, for god know what reason, unreachable, then it wastes a single second of the mail sender's time to check the backup before queueing the mail. That's the entire net effect. You'll get your delayed mail a second later, in theory, which doesn't actually happen because no one runs around setting retry timers that trigger the exact second anyway.
As for relay servers...you need hundreds of those why, exactly? You're operating one of those crazy-ass overloaded networks, apparently, where mail bounces through two or even three SMTP servers both going in and out.
Those networks are brokenly designed. Sorry, but it's true. I understand why you can't fix them at this point, but it doesn't mean they were designed correctly.
There's absolutely no reason not to have a single incoming point (Or even sets of points, via NFS directories.) for each account, where a message comes in and is stored 'locally', and absolutely no reason not to have one outgoing server for each outgoing message, when then sends the message directly to the recepient. (Those can be the same system, or not.) And maybe backup MXs for collecting mail when the system is down. No one has ever explained to me why a mail system needs to be more complicated than that. Anything more complicated than that is probably just historical nonsense laying around.
Whereas the Republicans think the solution is to not count the votes at all, and just have a court decide it.
However, it's perfectly possible to reduce the error rate to wholesale tampering by the staff of an entire precinct.
You just have a computer generate a computer and human readable paper ballot, and scan those to tally the votes. No more worries about what the person meant...it's printed completely unambigiously on the paper, alone, instead of having everyone listed and trying to figure out wwhich was marked.
Some people have suggested barcodes, but I think there's too much danger the machines will 'accidently' print one name and a different barcode, and, really, printing OCRable type isn't that hard.
And as the machines would also tally who was voted for, you've gotten rid of the 'hide a few bags of ballots before they count them' that sometimes happens. The total ballots printed better equal the total votes plus the total discarded ballots.
The only possiblities for vote tampering left are a) figuring out who can vote, which we still have problems with, and b) the actions of an entire precinct, because they can go in and vote as many times as they want, or let others do that...which constrained by the fact that people can sit there and observe them. With electronic machines, though, at least we could make them do the repeated voting while they are open, and not at 4 in the morning before the polls open.
There's a reason they don't let other people into the booth with you. It's not just a 'secret ballot', in that no one can know how you voted if you don't want them too, it's supposed to be an unprovable ballot, where you can't prove how you voted even if you wanted to. If you can prove how you voted, someone can give you 10 dollars to vote a certain way...or even threat your life unless you do prove how you voted.
This is why there are concerns being raised about camera phones in voting booths.
It's not gratuitous.
And why the hell are you bouncing spam? Delete spam or reject spam, do not bounce spam.
It sounds like you don't know what you're doing, or have a really stupid setup.
And, BTW, if you're getting hammered because you're the backup MX, which spammers like to pound, it might make sense to set up a tertiary MX server that doesn't actually exist. Spammers will go after that, instead, and never hit you, as almost all spamming software is written by complete fucking morons. Whereas actual mail that failed to get your primary server will just your backup. (Or, failing to get your backup, they will then fail to get your tertiary and just queue the mail, and start back over when they retry.)
I, personally, set up a 'backup MX' record to point at one of my IPs that didn't actually run a mail server, and cut my daily spam attempts by 30%.
While I currently use postfix and think it's great, I will point out there's another mail server out there without the issues of sendmail or qmail, named exim, that's pretty good also.
You can freely redistribute the source and binaries compiled from clean source. And you can distribute patches to it.
However, the point is, the qmail maintainer is the only person who can release new versions of qmail. And hence it's not free software.
There are two very large dangers with qmail...that it will go off in a random direction no one agrees with, and you'll either have to follow along or go that way, and that the qmail maintainer will just stop releasing new versions. With free software, if enough people use it, they will simply make a fork...but they can't do that with qmail. Technically they could grab a random version and keep building patches off that, but that becomes unmaintainable real fast.
In other words, qmail is basically 'freeware', not 'free software', although it does come in source form, and you have been granted the ability to modify it and even share the modifications. But not the end result.
I don't know what universe you live in, but mine, universities are often owned and operated by, gasp, the government.
And Divine Comedy is fantasy. It's the special brand of fantasy that no one seems to recognize, religion fantasy, like Touched by an Angel or the Left Behind series. It's 'What if this religion were correct?'. I suspect people don't like to recognize it because, to them, it's already true. But just because someone thinks they are a vampire doesn't mean we should stop classifying Buffy as fantasy, and just because someone believes in angels doesn't make Touched by an Angel not fantasy.
And Animal Farm is not really SF/F, because it's just an allegory. Animal Farm doesn't explore what would actually happen if animals became sentient, started talking, and took over, it's an allegory of 'workers' taking over. If it was real sci-fi or fantasy we'd have the authorities showing up to cart all these amazing talking animals off to be tested on.
That's not to say that sci-fi can't be an allegory, almost all good sci-fi is, but saying 'animals can talk' and then not dealing with that concept is not exploring how that new world works. If they had treated that concept, with the animals having no rights, and comparing that to slavery, sure, they used a fantasy alteration to reality and explore the world, and thus it's fantasy.
But Animal Farm, while it does have a single fantastic concept, doesn't actually explore the effects of that concept on society, instead chosing to explore a worker's revolt. Which is all well and good, I like the book, but it's not fantasy anymore than Mickey Mouse cartoons are fantasy because they have a talking mouse.
Now, you can argue that there's not enough scientific change in 1984 to make it sci-fi, but that's never been what sci-fi is about. Even if you don't want to call it sci-fi, you have to admit it's on the same spectrum as sci-fi and fantasy, it's just that fantasy postulates impossible changes in society due to 'magic', science fiction postulates possible changes in science due to technology, and 1984 postulates possible changes in society, not due to anything stated.
Which is, BTW, the same reason that alternate histories get grouped under sci-fi and fantasy....they belong there. It's 'what would happen if...', that's all sci-fi and fantasy is, it's just set in the past.
Some people use the term 'speculative fiction', and, yes, that is what 1984 is, just like it is what Gulliver's Travels is, or what a Conneticut Yankee in King Arthur's Court is.
Not the slightest trace of this movement has ever been detected. There are only the two options that I said...either we, for some unknown reason, are not moving through space, which makes the entire universe spin around us in random patterns, or our movement through space somehow has no effect on how fast light moves past us, that unlike every other speed in existence, the speed of light is not additive with other velocities, aka, it's 'constant'.
There simply are no other possibilities to explain this oddity. As velocity is just space divided by time, you obviously need to postulate some some sort of time and/or space change to explain this velocity inconsistancy. Einstein postulated both, that they are in fact the same thing and converted back and forth, which keeps us from having to explain where space came from or where the extra time went after it was used.
And just in case you're wondering, there are other experiments that rule out the first possiblity, although I can't think of them off the top of my head. You'd probably object to them because, obviously, if we're checking the speed of light with anything except vs the earth's speed, at some point the 'clock' is going to moving. Interesting how all those moving clocks fail in exactly the correct way to make us think relativity exists, but whatever.
Now, if you want to use another set of math to explain the universe, feel free...99.999% of the time, people use Newtonic physics anyway. We don't need to worry about time dilation when everyone is going within a close fraction of the same speed in the same direction.
You don't need a moving clock to show the speed of light is constant, because the planet itself is moving. You just take a beam of light, and split it in half with a mirror, send it along two perpendicular paths, bounce it back, and see if it arrives in sync. As the earth is moving around the sun, it's trivially easy to make one direction that's 'forward' along the motion of the sun, and one direction that's 'sideways'.
Ah, but there's nothing proving the mirrors are the same distance, or possibly the combined motion of the earth and the sun and everything cancelled out...but this is the clever bit: You wait 12 hours, when the entire arrangment has been rotated so it faces the opposite direction with regard to the movement of the sun, thanks to the rotation of the planet, and run it again. In fact, run it at six hour intervals, just for the hell of it.
Oh, and here's the other gag: You don't need any clocks. You don't need to measure the time, the two beams of light coming together will create a noticable interference pattern. If the beams at any point took even slightly longer, the pattern would change.
And, just so you don't actually have to run that experiment, I'll tell you the results...no matter in what direction you aim the arrangement in, in relation to the movement of the planet, light always takes the same amount of time to run the same distance. Yes, this experiment has been run plenty of times, with more than enough accuracy to measure the movement of the planet.
So you have two options: Either the planet isn't moving, with the rest of the universe spinning wildly around it, or the speed of light is constant no matter how fast you're moving.
Do you honestly think the scientific community accepted such a crazy theory as relativity without having inconvenient facts that required such a theory?
Tada. Non-parallel plumb bobs due to accelleration.
Likewise, buld an infinitely large plane with enough mass to generate 1 g. Attach plumb bobs above it.
Tada. Parallel plumb bobs due to gravity. (Well, technically, they'd lean slightly toward each other due to their own gravity, but so would the ones in a rocket. You could measure with one, and just move it, if you cared.)
Yes, my two examples are physically difficult. But they certainly can exist in theory.
Now, if you're just taking issue with the 'we can't prove if we're accelerating or falling' that's commonly claimed...we all know that's a lie. Otherwise, we wouldn't know, on this planet, which we clearly do. The standard example uses an elevator/small room specifically to keep you from being able to measure variations in direction.
And, no gravity at the surface of the earth isn't uniformly curved, anyway. It's entirely possible to end up with two points ten feet apart that do have an exactly parallel 'downs', or even 'downs' which point away from each other, due to the composition of matter in the earth's crust.
Stick you in a room over one of those places, and you'd cleverly prove that gravity there was not only from acceleration, but from centripical acceleration.