Some "expert" whining that flash can't get any cheaper because of fabs, limitations, etc, etc.
Well, I'm not buying it. Until I start hearing something from the people who actually make the tech, I'm going to say it'll probably keep going. Supply issues are just temporary. Companies can, and are, building new fabs all over. In terms of overall cost that has been getting reduced by both process size (which doesn't seem to be stopping soon) and by advances in how data is stored. Recently we've started to have TLC flash drives, which store 3 bits per location. This comes at a cost of write/erase cycles but it turns out that you don't really write that much data in normal desktop usage, so that works out ok, and you can over provision more as you have more storage.
Eventually I'm sure we'll hit a wall of some sort, but I think there's quite a ways to go.
Also, the question isn't if they are as cheap as magnetic drives. The question is if they are cheap enough for the capacity people need and these days the answer is generally "yes". Most people don't need 4TB of storage. I don't mean that in a condescending way, I mean that they actually wouldn't use it if they had it. Hence a smaller SSD can work perfectly fine. 500GB, or less, tends to do the trick real well for most people. So it doesn't matter if you can get a big HDD, it matters if you can get a fast SSD that is cheap enough to be affordable.
For higher capacity usages, well ya, HDDs are still great and still used. We got a big ole' NAS not too long ago using magnetic drives. We needed a lot of storage and didn't want to spend tons of money since performance wasn't a big issue.
Your evidence supports that, not disproves it. The problem is the Saudi monarchy is massively corrupt, as all monarchies tend to be. That doesn't mean that a theoretical perfect one, that is unattainable with real people, wouldn't be a great system.
I mean look at a system like the US has, it's mix of democracy and republic. There is a lot of infighting in government, bureaucracy, dead weight, and so on. It is slow to respond and rather bloated. This is by necessity, and also by design. Spread the power around and create checks and balances so that nobody can abuse it. That is needed because we deal with real people, but it is inefficient.
Now imagine a system where it is a dictatorship, or other situation of absolute power, but we have a theoretical perfect being (an AI maybe) as the ruler. It is incorruptible, cares for nothing but the welfare of its citizens and nation, and makes the best choice it can, all the time. Well that would be a hell of a lot more efficient. Shit would get done. When things needed to change, they'd just change by immediate decree. No games, no pork barrel spending, no holding the budget hostage.
The problem is, we can't have that perfect ruler. Humans are imperfect and put them in a position of absolute power, they get corrupted, generally very badly. Even if you got real lucky and got one that didn't, sooner or later you'd get one that did.
Hence the need for a more complex, and inefficient, system like what we see in modern free countries. However that doesn't invalidate the theory that with a perfect ruler a system of absolute power would be more efficient.
On any issue of importance he's either agreed with them, or folded without a fight. I'd give him a "hamstrung" thing if he'd taken a number of fights to the republicans, lost each time, and has to start compromising to get anything at all done. However he hasn't done that. He's never even stood up and fought. It isn't even that he's rolled over, he's just never shown up in the first place.
This blaming the republicans is really silly. While the republican party by and large is not being helpful, they do not have any sort of control. They have a narrow majority in the house, a minority in the senate, and of course don't have the presidency. If President Obama wanted to stand up and fight on things that mattered, well he'd have a shot at least. It isn't like they could just ram legislation past him. However he hasn't, not once that I can think of.
If they have something that nobody else does, or at least that they can say nobody else does, they love to push it as the best thing EVAR. They want to convince people that the Apple product is better, faster, more capable, etc, etc.
So, since they have one of the first, if not the first, phone with a 64-bit ARM CPU (since near as I know 64-bit ARM is a very new thing) they are going to advertise and push it. They will get their zealots all in a lather as to how much better that is, without any real understanding as to what it actually means, they'll just know it is something their product has that those nasty other products don't and thus it is better.
So over the air networks are regulated by the FCC. Nudity, strong language, etc are all regulated. So they have legal limits they have to comply with. However the cable channels, they can do as they wish. Why then are only the premium ones the only ones with (more than a little) nudity? Ad money. The networks all have standards and practices divisions to work with the creators of shows to keep things such that advertisers are happy, and to work with advertisers to keep them happy. Advertisers worry about their brand being associated with certain things and so the networks have to keep them happy. I'm not saying the advertisers are being sensible, but that is how it is.
This is also why some shows can get away with more than others. If it is a big show, that lots of people watch, advertisers will be more willing to STFU and deal to have their spot played during it.
For that matter, the Internet gets similar things going on with ads. For example Fark used have "boobies" posts/threads with links to naked women. Generally pretty tame and not a large part of their content. However, advertisers kept bitching and some refused to do business, so Fark spun that content off in to a different site and now doesn't allow it on the primary domain. There was nothing legally keeping them doing this, just ad revenue.
We've already gone beyond the ability of electronics to deal with. The 144dB afforded by 24-bit recording is more than we can build converters to capture. The very best will get maybe 120-122dB dynamic range in actual operation (components have more but when implemented in a circuit that is about as good as it gets). At that point, you are dealing with the thermal noise of electrons bouncing around in the transistors, there just isn't much room for improvement with current tech.
That is not to mention microphones which have their own inherent noise. most of them it is 20dB SPL or more. Even the really low noise ones from B&K or the like are around 0dB. Your mics will introduce a noise floor there's no way to get around, unless you increase the volume of the music going in.
And those magical reel to reel tape units? Maybe 70dB SNR, if they are really good.
While it may not seem intuitive, in fact you can. If you want to see it in operation, have a look at http://www.youtube.com/watch?v=cIQ9IXSUzuM. You can see perfect reconstruction of sine waves near the Nyquist frequency, and see how things work with real analogue monitoring gear and digital signaling.
We really do understand how band-limited systems work real well, and it really does reproduce sound accurately.
Challenge them to try some ABX testing. They get a recording on their reel-to-reel machine. You then record the output of that with a good digital recorder, like an Alesis Masterlink or whatever. Sync up the playback and have an ABX switch. See if they can tell which is which.
They won't be able to, as the digital recorder will capture everything, including the flaws, of the analogue mater in all its glory. If you play the tape over and over enough to degrade it eventually there'll be an audible difference, but if you make the digital copy and then do the test, it'll sound identical.
Our technology is better than our ears these days. The challenge is in how to capture a good sounding recording, not in how to store it.
Digital goes straight down to DC with no issues. The only limit on the low end is that most devices aren't DC coupled, for various electrical reasons, and roll off on the extreme low end. However it is generally VERY low. A Benchmark ADC-1 goes down to 2 Hz and if you want, you can find DC coupled ADCs. Now try and find a microphone that can pick that up, and better yet, try and find an analogue tape that goes that low.
Digital is king of bass and that's part of the reason we've seen an increase in the use of heavy bass in music, is that digital allows for it. You couldn't do it very well with analogue, and not at all with many of the systems (like phonograph).
In terms of high frequencies, well you can record those with digital if you want, we just generally don't bother because there is no point. You can't hear it, you can't feel it, etc. Really, this is something that is pretty easy to test, and indeed has been tested. Our ears don't do that high, we can't hear it.
However it CAN be a bad thing, as what you can get is aliasing. Your gear may not be able to handle those high frequencies properly and your speakers or amp may generate aliased subharmonics that you can hear. So you end up getting distortion, sound that wasn't there in the recording that your setup is playing because it can't handle the ultrasonic content. Hence band limiting can be quite useful.
So digital sound is generally recorded at 48kHz (or 44.1kHz) because there is no reason to waste space recording the higher frequency content, which there is very little of anyhow (look at a spectrograph of it some time if you are interested), cannot be heard, and can cause problems on lower end gear, and even some high end gear.
Problems with recording and reproduction vs reality do not have to do with the digital storage medium, rather they have to do with the analogue recording and playback components (mics and speakers) as well as dealing with placement issues. If you want something that sounds quite real, get yourself some binaural recordings, and some good headphones to listen to them with.
All you have to do is refresh and convert it. So if a new standard is becoming the thing that everyone uses and the old standard is going away, well there will be quite a bit of time when stuff can deal with both. During that time, you convert. You can keep doing this, as often as is required. Same deal with the actual storage medium. In addition to backups, you transfer it to new media periodically.
This isn't theoretical shit I'm talking about either, we do this at work. We have data on our NAS and tapes that originated two decades ago. The original storage devices are long dead but it doesn't matter, it was transferred, and can be transferred again.
So ya, if you have a digital format and you let it sit in a vault for decades, never touching it, then it may degrade and be useless. Of course that is true of anything. Go have a look at the Constitution some day and see the extremes they have to go to to protect it, since age has not been kind to it.
However digital has the benefit of being dead fucking simple to transfer to new media with perfection any time you like. Just do that, and you are good. In fact standards these days are made to make that real easy. Like LTO, any drive must, by the standard, be able to read tapes from two versions prior. So if you buy a new LTO-6 drive, you can load up your LTO-4 tapes, and transfer the data. Then the same deal later when you move to LTO-8. That aside you can go to a completely new/different device just by having them talk over another standard like Ethernet or FC.
I would bet on the long term permanence of digital over analogue any day. Both can be fragile, if you take no steps to protect it, but digital can be perfectly replicated as often as you like, and that gives it a survivability analogue doesn't have.
The ISP's I've deal with USWest/Qwest/Centurylink, Speakeasy, and Cox, have all been happy to sell a business account for the asking. You need to call their business division, which is separate from consumer, but that is all. They'll put the account in your name (in the case of Cox a business name is required so they just use my first and last name) and life is good. They are happy to get the money. The account costs more, and you can spend money on addon services like multiple static IPs.
It isn't like they try and keep them away from people to be mean, they are just targeted at businesses because they cost more and most people don't want to pay. In my case I pay about the same for a 30/5 as a consumer would for a 150/20 account. However, I have no bandwith cap, no prohibition on running servers (which I do) and 4 static IPs. The consumer plan has a 400GB cap, and they disallow running servers (and block some common inbound server ports, like SMTP).
Maybe some ISPs are jerks, but try it first. Most you just need to look up the number for their business services division and call them. I've never had one say no. Money is money after all.
So long as people play nice. Companies do the same thing. You'll never find a workplace where every desktop has dedicated bandwidth out to the Internet, or even to the central server. However, so long as people play nice and don't slam it with torrents 24/7, it actually works really well.
Like at work we have a Dell Equallogic storage unit that is where people can put important data. Highly reliable NAS, we back it up to tape, all that jazz. However, it only has 2 gbps to its disks on the SAN. We have one array, and it has two gig uplinks. The NAS head itself only has 8 gbps out to the network so even if we get more arrays, it doesn't have a ton of bandwidth. Each and every desktop has a gig link, and we have a several hundred that might want to access that NAS. So we are massively oversubscribed.
However, it isn't a problem. Performance is good. Why? Because we all play nice and share. People access data when they need it, and let it sit idle otherwise. In fact, since most of the data is archival, fairly static, it really has more capacity than we need overall.
Same deal with our overall network. I work on a campus with tens of thousands of computers and only a gig or two of total Internet bandwidth (and 10 gbps of Internet 2). However things are always fast. Downloads fly, pages load instantly, etc. Reason is we all share and play nice. We get what we need and let the connection sit idle otherwise. If someone did slam the connection, they'd get a phone call from network operations.
I will say that bandwidth caps aren't the best solution to making people share and play nice, but other solutions are a good deal harder to implement, and for customers to understand. Bandwidth caps are something that is pretty easy for people to comprehend, and that will give them a reason to keep usage in check.
ISPs seem to really like money and business class liens cost more money. I've had business class service for about 13 years from 3 ISPs over that time, and all were plenty happy to sell it to me. Cox is who I'm with currently and they had no problem giving me a business account in my own name. Their software doesn't understand the concept, the "business name" field has my first and last name in it, but that doesn't hurt anything, and I get my nice business class Internet.
With wireless, well ya, you are going to have issues with that because there are hard bandwidth limits enforced by power output, frequency availability, and atmospheric noise (look up Shannon's Law) that you just can't get around. Wireless bandwidth will always have limits, particularly when you are talking wide area wireless.
Nothing is all benefit, or all cost. Nothing is all good or all bad. Everything has tradeoffs, everything has a cost. The trick then is weighing the costs and benefits to decide if something is good or bad overall, and what course of action to take.
Any time something is presented as completely one-sided, someone is pushing an agenda. You can see this with advertising all the time. They talk up how awesome whatever their product is, all the cool shit it does for you. They never talk about any limitations, downsides, etc, unless it is legally required (like the drug advertising). The reason is, of course, they are trying to sell you on it. They want you to think it is awesome so it is all benefits, all the time.
You are correct, it is something to watch for. When people present a very one-sided stance it is often because they are pushing a given point, not because there is no counter-point.
2% of the console's price is pretty exorbitant for open standards patents. The whole deal with (F)RAND stuff is "Reasonable and Nondiscriminatory". Now you don't have to license your stuff under that model, but that's how open standards like MPEG-4 and 802.11 are done. Companies pool their patents and set up a standard, and the licenses are fixed. The idea is that anyone can license it for the same amount, and that amount is fair and reasonable.
The reason companies do that is to get their patents used and licensed. I mean if I develop some cool new video compression, but I won't set licensing terms, everyone has to come to me and I decide if you get a license, and if so what it costs, well that will hamper adoption. Many companies will give that a miss since they don't know why it'll cost them. However if it is all out in the open, then it is much more likely to get used and licensed.
Also a lot of standards agencies require it. If you want your IP to be part of whatever standard they make, you have to disclose it, and license it under RAND terms. You don't want to, then it is excluded from the standard.
Well, if you decide to do that, you can't then go and decide to try and stick it to a company you don't like. You can't say "Yes, all our stuff is available under this fair license for all to use, oh except for you, we don't like you so you pay more." Sorry, you gave up that ability when you decided to do the open standards thing and RAND licensing.
Hence, the court decision. Google wanted to play hardball with MS, but they were doing it with patents they'd said they wouldn't do that with. So they got slapped down.
So ya, exorbitant demands. Particularly in context of what we are talking about. Remember Google doesn't own H.264 or 802.11. They only have a small number of the patents on it. So if their share was like 2%, then total cost could easily be 10-20%. If that was the kind of money demanded for those standards, they'd not be used. Google just wanted to screw MS.
Windows 7 does nothing at all with Blu-ray content. It doesn't understand how to play it. All it does in relation to any of this is provide a method for programs to inquire to drivers if everything is (supposedly) secure. A Blu-ray player can inquire as to the encryption status of the links and make sure things aren't being captured and so on. For that matter, so can other programs. It isn't Blu-ray specific, however only the media companies give a shit so that's all that really does it. Games don't mind at all if their output is being captured.
Doesn't matter the interface. DVI, HDMI, and DP can all encrypt the signal. There's nothing special, on a computer at least, about HDMI.
It is then up to the software how it acts on that. However, due to licensing requirements, the software has to disable the video out if everything isn't encrypted. If it doesn't they won't be able to get a license for the keys to decode the media.
Same deal on any platform. It isn't like Windows is special in this way. If your chosen platform doesn't support the necessary "protection" then there won't be any licensed Blu-ray playback software.
This is a media industry thing, not an OS thing. The OS provides the ability to have verified driver paths, but it does nothing at related to changing anything. That is up to the software, and that is dictated by licensing.
How, precisely, would you propose to build something backward compatible with the current spec that can push that kind of bandwidth, and be built for a reasonable cost?
The reason for these limits aren't arbitrary. It gets rather difficult and expensive to generate these real high bandwidth signals. Same reason why 10 gig ethernet costs so much more than gigE and needs better cabling to boot.
It isn't magic, as technology advances (particularly smaller lithography) it becomes possible to do higher clock rates at a lower cost and thus increase the bandwidth going over the cables. However it isn't something where we could just make it as fast as we wanted, easily and cheaply. If it were, well we'd have a lot higher interconnect speed.
So if you know some engineering voodoo that nobody else does that will allow for a 2-4x increase in bandwidth while still keeping cost low, well then off to the patent office with you You'll be able to make a mint. However if you are just whining that you can't have everything, without any actual understanding, then please stop.
HDMI is video and audio transport. Closed captioning works fine over it, since it comes from the video source. Be it your cable, DVD, Blu-ray, whatever, the CC information is processed on the relevant device, and then sent out as part of the video.
Asking HDMI to do closed captioning is like asking Ethernet to do packet filtering: You are looking at the wrong area.
When you are talking shared wires like cable modems or PON there is always the solution of just add more frequencies. With DOCSIS 3, you can have an arbitrary number of channels devoted to cable modems. So if what you have isn't enough, you can add more. Now that means taking them from something else, of course, however there's likely to be plenty available. When analogue service is discontinued, well that's 100 channels you've got right there (on many cable networks the first 600MHz is analogue cable service, the next 400MHz is all the digital stuff).
With PON it is even easier: Just shoot another laser down the fibre. That is already how it works: You have one wavelength for downstream, another for upstream, and another for video service. Well WDM doesn't stop at 3 wavelengths, you can have more. So as needed, more wavelengths can be added, giving more throughput.
In terms of the sharing issues, well it is easy to see what wins on a busy campus, like where I work. We all share bandwidth, of course, the campus has a gig or two of total Internet speed, not a gig for each desktop. However for all that, wired access is fast all the time. We have enough. Wireless, well that's another story. You go to the student union at lunch time and you find it is really, really bad. Things load slow, if they load at all. You have to reload pages plenty and so on. There are just too many people fighting for what's there. It isn't for lack of APs either, the university puts hundreds (literally) in a building, but when people are all in the same space, all hitting the same APs, well it gets overloaded.
There is this false idea that wireless is better than wired, that we will all move over, everything will be wireless all the time and life will be grand.
Nope. You can always get more bandwidth, quite a bit more, out of wires than wireless. That pesky Shannonâ"Hartley Coding Theorem just keeps cropping up and getting in the way. If you want more bits per second, you either need more bandwidth (meaning more spectrum) or a better signal to noise ratio. When you are talking wireless the only thing you can do about SNR is to up transmission power, which is not without its own issues, and there is just only so much bandwidth you can have, particularly with given properties.
See part of the problem is that as you move up the spectrum to higher frequencies, it gets easier to have more bandwidth, of course. However your signal gets more and more directional, and has less and less penetrating power. VHF and UHF are really good for transmissions. They are pretty non-directional and can penetrate most buildings without a whole lot of issues. However if you are operating on, say a 700MHz carrier your bandwidth is going to be limited, particularly when you have multiple services that want to use it. Indeed in the US you find that it is partitioned up in to 6, 10, 12, and 22MHz blocks.
Now if you go way up in frequency, this isn't a problem. Go up to a few hundred THz, instead of MHz. Now bandwidth isn't a big issue. If you have a carrier of 700THz then you can have a few THz of bandwidth, no problem and thus tons of information... Only one issue. 700THz might be more popularly called "blue". You are up in the light range now, and of course light can't penetrate for shit. Even a piece of paper would be sufficient to disrupt the signal. It is also highly, highly directional.
Finally there's a big issue which is that everyone has to share wireless. Anyone on a given segment, node, access point, etc is sharing whatever bandwidth there is. You don't each get your own bandwidth, you all have to share. So the more users, the less there is for each and there's really no way around that.
And thus the problem. You can't "just get more" bandwidth when you are talking wireless. You run in to physical limits. Your SNR is limited by power considerations (and distance) and the atmosphere, your bandwidth is limited by what is useful, and not used by other things, and so on.
With wired, not such an issue. You can go way up in frequency, particularly when you are talking fiber optic. However the real thing is that you can just lay more wires. You don't have to send a signal down one pair, you can have multiple. Ethernet is a good example of that. Gig and 10 gig use all 4 pairs, two to send, two to receive. Need more bandwidth on the same tech? Just lay another bundle. 8 pairs, as in two Cat-6a cables, will get you 20gbps, 12 pairs 30gpbs, and so on.
That's all dedicated (and full duplex) too. Only the endpoints use that. You can have stacks of cables running right next to each other, connecting different devices, and none of them trod on each other, they all have separate bandwidth.
So while wireless is cool, and useful, if we want fast speeds, if we want the ability to transfer lots of data all the time, we need wires. Wireless won't cut it. For that matter to the extent we can make it work well, it needs to be short range. You can use higher frequencies, have better SNR, and have less people per segment if you build the segments out. However that means lots of access points all over the place and those need to have backhaul, and that is going to be wired.
I get amused when people do this. Your "we/I" thing. I say something you don't like, so you extrapolate my views on the situation. You assume that I don't believe that humans are causing climate change. You'll notice I said nothing on the issue. I do that on purpose. My views as to the veracity of the science really aren't relevant. I am talking about the situation of what is being claimed and what to do about it.
However you take this somewhat religious point of view in that I argued something you didn't like, so I'm a heretic, an unbelieve, clearly I reject everything, I'm anti-science! You are assuming facts not in evidence.
I'm just pointing out one (of a number) of issues with getting widespread action on the issue: If we are past a tipping point, where reductions in emissions will not do much (and there are models that show this), then the solutions that are based around that are not useful.
I have a pretty good understanding of economics and one of the things I quite well understand is that setting up a system where rich countries give more money to poor countries, which would be the big effect of cap and trade, isn't going to help reduce emissions. If anything, it'll raise them since the poor countries will be able to develop their industry and then start using more fossil fuels themselves. You would likely find that many who thought the agreement was good when they were getting money would decide to withdraw later when it suited them.
If cleaner energy is the desired result and economics are the method you wish to use, then subsidies for R&D are the way to go. Taxes on coal to support nuclear, taxes on fossil fuels to support biofuels, that kind of thing. You pour money in to developing better biofuel technology or the like, and money in to getting the process up to the point where it is cheap, that will do something. Just shuffling money around country to country will accomplish nothing.
In terms of climate models ok, if you don't like that assertion then which models do you like, and why are those ones correct? What you have to realize is that there are a lot of climate models out there and their predictions are not consistent. They do not agree on how much warming, how fast, and what effect changes will have. So, if there's a correct one, then let's hear it, and evidence as to why that one is correct. If not, well that was relating back to my original point.
You seem to think that this is an argument claiming against climate change. It's not, I try to not interject my personal opinion at all in to these things. Rather this is talking about one of the problems relating taking any wide scale action. The actions that largely seem to be proposed will not do much to actually deal with the issue and, if we are indeed past a tipping point, be worse than useless (since they'd take up resources that should be spent in other ways).
I think a big problem is that people confuse and oversimplify the climate change argument. They think "There is a consensus, the science is settled, thus there is only one course of action we can take!" Nope, rather you can break it down in to four rough levels:
1) The fact of global warming, that the average surface temperature is rising outside of known cycles. This is an observation, a measurement. While it is a complex one, it is not really arguable unless you can find a flaw in the measurements or calculations.
2) The theory of global warming, in particular that the prime or exclusive cause of the warming is an increase in atmospheric CO2 (which is quite easy to measure) cause by human emissions. It provides an explanation for the causality of the observations we've made. As with all theories, you can debate this if you can find evidence that it is incomplete, or contradictory evidence or the like. Also that is a fairly general overview, there are researchers working on much more specific, predictive, versions that would explain more precisely the change in temperature with a given change in atmospheric CO2.
3) The assertion or judgement that this is a net bad thing for humanity. This is based on other theories and models as to what will happen if the temperature increase continues. You can then evaluate those and decide if it is a good or bad thing overall (everything has costs and benefits). For that matter two people could agree on the outcome, but disagree on the judgement of if it is good.
4) The politics or policy of what to do about it. This is the kind of thing there can be a lot of disagreement on, even if you agree on all other points. There's no "right" here. It is all about what seems to be the most beneficial to spend resources on.
The problem is that people seem to do sufficient reading to be convinced that #1 is true, or that #1 and #2 are true, and then decide that means you have to go all the way to the end, and that whatever their given source advocates for #4 is the One True Way(tm) and you are anti-science if you don't buy in.
Is that it is just one of shuffling around who produces what. Now this may well work, if all we need to do is maintain our level of emissions or decrease them. However if we are past a tipping point, where nothing short of a massive reduction (and perhaps not even that) in emissions will stop the warming then it does no good.
What I was talking about with proposals and so on was things outside of "emit less CO2" or "here are way to try and emit less CO2". The reason is this is assuming that we are indeed past a tipping point, as some climate researchers claim. In that case, the issue of emissions is not one to concentrate on, but rather on how to either prevent the change via other means, prepare ourselves to deal with the change, or some of both.
We should all be happy to go back to the pre-industrial ages. Sure it means the vast majority of humans will have to die off, and the ones that live will have much shorter, harder, lives but hey, it would be good for the planet (depending on how you define good)! As such all of us should be happy, no honoured to do that. Excepting for professors, of course. They advance knowledge so they clearly need to be allowed to keep all of their modern conveniences. But the rest of us, back to the dark ages!
That is what always amuses me about the "industrial society is bad!" types is I've never seen any of them practice what they preach. None of them go and live in the wilderness, off the land, eschewing all modern technology except for the rare times they come to give a talk on it. Heck none of them even go back to Amish/Mennonite levels of technology. They live modern lives, enjoying all the conveniences, and then say others shouldn't.
How about, try it first, then see if maybe there's a reason we like all this new stuff?
That means we'd be down to around 45% of current levels by 2050. That's a big reduction. Now probably doable and worth trying, but you'd want to be fairly certain that it would, indeed, fix the problem if you are going to make the tradeoffs necessary to do so. You wouldn't want to spend a bunch on a big change to make all this happen only to find out no, sorry, but that isn't in fact going to help.
Slashdot Mirror
Some "expert" whining that flash can't get any cheaper because of fabs, limitations, etc, etc.
Well, I'm not buying it. Until I start hearing something from the people who actually make the tech, I'm going to say it'll probably keep going. Supply issues are just temporary. Companies can, and are, building new fabs all over. In terms of overall cost that has been getting reduced by both process size (which doesn't seem to be stopping soon) and by advances in how data is stored. Recently we've started to have TLC flash drives, which store 3 bits per location. This comes at a cost of write/erase cycles but it turns out that you don't really write that much data in normal desktop usage, so that works out ok, and you can over provision more as you have more storage.
Eventually I'm sure we'll hit a wall of some sort, but I think there's quite a ways to go.
Also, the question isn't if they are as cheap as magnetic drives. The question is if they are cheap enough for the capacity people need and these days the answer is generally "yes". Most people don't need 4TB of storage. I don't mean that in a condescending way, I mean that they actually wouldn't use it if they had it. Hence a smaller SSD can work perfectly fine. 500GB, or less, tends to do the trick real well for most people. So it doesn't matter if you can get a big HDD, it matters if you can get a fast SSD that is cheap enough to be affordable.
For higher capacity usages, well ya, HDDs are still great and still used. We got a big ole' NAS not too long ago using magnetic drives. We needed a lot of storage and didn't want to spend tons of money since performance wasn't a big issue.
Your evidence supports that, not disproves it. The problem is the Saudi monarchy is massively corrupt, as all monarchies tend to be. That doesn't mean that a theoretical perfect one, that is unattainable with real people, wouldn't be a great system.
I mean look at a system like the US has, it's mix of democracy and republic. There is a lot of infighting in government, bureaucracy, dead weight, and so on. It is slow to respond and rather bloated. This is by necessity, and also by design. Spread the power around and create checks and balances so that nobody can abuse it. That is needed because we deal with real people, but it is inefficient.
Now imagine a system where it is a dictatorship, or other situation of absolute power, but we have a theoretical perfect being (an AI maybe) as the ruler. It is incorruptible, cares for nothing but the welfare of its citizens and nation, and makes the best choice it can, all the time. Well that would be a hell of a lot more efficient. Shit would get done. When things needed to change, they'd just change by immediate decree. No games, no pork barrel spending, no holding the budget hostage.
The problem is, we can't have that perfect ruler. Humans are imperfect and put them in a position of absolute power, they get corrupted, generally very badly. Even if you got real lucky and got one that didn't, sooner or later you'd get one that did.
Hence the need for a more complex, and inefficient, system like what we see in modern free countries. However that doesn't invalidate the theory that with a perfect ruler a system of absolute power would be more efficient.
On any issue of importance he's either agreed with them, or folded without a fight. I'd give him a "hamstrung" thing if he'd taken a number of fights to the republicans, lost each time, and has to start compromising to get anything at all done. However he hasn't done that. He's never even stood up and fought. It isn't even that he's rolled over, he's just never shown up in the first place.
This blaming the republicans is really silly. While the republican party by and large is not being helpful, they do not have any sort of control. They have a narrow majority in the house, a minority in the senate, and of course don't have the presidency. If President Obama wanted to stand up and fight on things that mattered, well he'd have a shot at least. It isn't like they could just ram legislation past him. However he hasn't, not once that I can think of.
That's the problem.
If they have something that nobody else does, or at least that they can say nobody else does, they love to push it as the best thing EVAR. They want to convince people that the Apple product is better, faster, more capable, etc, etc.
So, since they have one of the first, if not the first, phone with a 64-bit ARM CPU (since near as I know 64-bit ARM is a very new thing) they are going to advertise and push it. They will get their zealots all in a lather as to how much better that is, without any real understanding as to what it actually means, they'll just know it is something their product has that those nasty other products don't and thus it is better.
So over the air networks are regulated by the FCC. Nudity, strong language, etc are all regulated. So they have legal limits they have to comply with. However the cable channels, they can do as they wish. Why then are only the premium ones the only ones with (more than a little) nudity? Ad money. The networks all have standards and practices divisions to work with the creators of shows to keep things such that advertisers are happy, and to work with advertisers to keep them happy. Advertisers worry about their brand being associated with certain things and so the networks have to keep them happy. I'm not saying the advertisers are being sensible, but that is how it is.
This is also why some shows can get away with more than others. If it is a big show, that lots of people watch, advertisers will be more willing to STFU and deal to have their spot played during it.
For that matter, the Internet gets similar things going on with ads. For example Fark used have "boobies" posts/threads with links to naked women. Generally pretty tame and not a large part of their content. However, advertisers kept bitching and some refused to do business, so Fark spun that content off in to a different site and now doesn't allow it on the primary domain. There was nothing legally keeping them doing this, just ad revenue.
We've already gone beyond the ability of electronics to deal with. The 144dB afforded by 24-bit recording is more than we can build converters to capture. The very best will get maybe 120-122dB dynamic range in actual operation (components have more but when implemented in a circuit that is about as good as it gets). At that point, you are dealing with the thermal noise of electrons bouncing around in the transistors, there just isn't much room for improvement with current tech.
That is not to mention microphones which have their own inherent noise. most of them it is 20dB SPL or more. Even the really low noise ones from B&K or the like are around 0dB. Your mics will introduce a noise floor there's no way to get around, unless you increase the volume of the music going in.
And those magical reel to reel tape units? Maybe 70dB SNR, if they are really good.
While it may not seem intuitive, in fact you can. If you want to see it in operation, have a look at http://www.youtube.com/watch?v=cIQ9IXSUzuM. You can see perfect reconstruction of sine waves near the Nyquist frequency, and see how things work with real analogue monitoring gear and digital signaling.
We really do understand how band-limited systems work real well, and it really does reproduce sound accurately.
Challenge them to try some ABX testing. They get a recording on their reel-to-reel machine. You then record the output of that with a good digital recorder, like an Alesis Masterlink or whatever. Sync up the playback and have an ABX switch. See if they can tell which is which.
They won't be able to, as the digital recorder will capture everything, including the flaws, of the analogue mater in all its glory. If you play the tape over and over enough to degrade it eventually there'll be an audible difference, but if you make the digital copy and then do the test, it'll sound identical.
Our technology is better than our ears these days. The challenge is in how to capture a good sounding recording, not in how to store it.
Digital goes straight down to DC with no issues. The only limit on the low end is that most devices aren't DC coupled, for various electrical reasons, and roll off on the extreme low end. However it is generally VERY low. A Benchmark ADC-1 goes down to 2 Hz and if you want, you can find DC coupled ADCs. Now try and find a microphone that can pick that up, and better yet, try and find an analogue tape that goes that low.
Digital is king of bass and that's part of the reason we've seen an increase in the use of heavy bass in music, is that digital allows for it. You couldn't do it very well with analogue, and not at all with many of the systems (like phonograph).
In terms of high frequencies, well you can record those with digital if you want, we just generally don't bother because there is no point. You can't hear it, you can't feel it, etc. Really, this is something that is pretty easy to test, and indeed has been tested. Our ears don't do that high, we can't hear it.
However it CAN be a bad thing, as what you can get is aliasing. Your gear may not be able to handle those high frequencies properly and your speakers or amp may generate aliased subharmonics that you can hear. So you end up getting distortion, sound that wasn't there in the recording that your setup is playing because it can't handle the ultrasonic content. Hence band limiting can be quite useful.
So digital sound is generally recorded at 48kHz (or 44.1kHz) because there is no reason to waste space recording the higher frequency content, which there is very little of anyhow (look at a spectrograph of it some time if you are interested), cannot be heard, and can cause problems on lower end gear, and even some high end gear.
Problems with recording and reproduction vs reality do not have to do with the digital storage medium, rather they have to do with the analogue recording and playback components (mics and speakers) as well as dealing with placement issues. If you want something that sounds quite real, get yourself some binaural recordings, and some good headphones to listen to them with.
All you have to do is refresh and convert it. So if a new standard is becoming the thing that everyone uses and the old standard is going away, well there will be quite a bit of time when stuff can deal with both. During that time, you convert. You can keep doing this, as often as is required. Same deal with the actual storage medium. In addition to backups, you transfer it to new media periodically.
This isn't theoretical shit I'm talking about either, we do this at work. We have data on our NAS and tapes that originated two decades ago. The original storage devices are long dead but it doesn't matter, it was transferred, and can be transferred again.
So ya, if you have a digital format and you let it sit in a vault for decades, never touching it, then it may degrade and be useless. Of course that is true of anything. Go have a look at the Constitution some day and see the extremes they have to go to to protect it, since age has not been kind to it.
However digital has the benefit of being dead fucking simple to transfer to new media with perfection any time you like. Just do that, and you are good. In fact standards these days are made to make that real easy. Like LTO, any drive must, by the standard, be able to read tapes from two versions prior. So if you buy a new LTO-6 drive, you can load up your LTO-4 tapes, and transfer the data. Then the same deal later when you move to LTO-8. That aside you can go to a completely new/different device just by having them talk over another standard like Ethernet or FC.
I would bet on the long term permanence of digital over analogue any day. Both can be fragile, if you take no steps to protect it, but digital can be perfectly replicated as often as you like, and that gives it a survivability analogue doesn't have.
The ISP's I've deal with USWest/Qwest/Centurylink, Speakeasy, and Cox, have all been happy to sell a business account for the asking. You need to call their business division, which is separate from consumer, but that is all. They'll put the account in your name (in the case of Cox a business name is required so they just use my first and last name) and life is good. They are happy to get the money. The account costs more, and you can spend money on addon services like multiple static IPs.
It isn't like they try and keep them away from people to be mean, they are just targeted at businesses because they cost more and most people don't want to pay. In my case I pay about the same for a 30/5 as a consumer would for a 150/20 account. However, I have no bandwith cap, no prohibition on running servers (which I do) and 4 static IPs. The consumer plan has a 400GB cap, and they disallow running servers (and block some common inbound server ports, like SMTP).
Maybe some ISPs are jerks, but try it first. Most you just need to look up the number for their business services division and call them. I've never had one say no. Money is money after all.
So long as people play nice. Companies do the same thing. You'll never find a workplace where every desktop has dedicated bandwidth out to the Internet, or even to the central server. However, so long as people play nice and don't slam it with torrents 24/7, it actually works really well.
Like at work we have a Dell Equallogic storage unit that is where people can put important data. Highly reliable NAS, we back it up to tape, all that jazz. However, it only has 2 gbps to its disks on the SAN. We have one array, and it has two gig uplinks. The NAS head itself only has 8 gbps out to the network so even if we get more arrays, it doesn't have a ton of bandwidth. Each and every desktop has a gig link, and we have a several hundred that might want to access that NAS. So we are massively oversubscribed.
However, it isn't a problem. Performance is good. Why? Because we all play nice and share. People access data when they need it, and let it sit idle otherwise. In fact, since most of the data is archival, fairly static, it really has more capacity than we need overall.
Same deal with our overall network. I work on a campus with tens of thousands of computers and only a gig or two of total Internet bandwidth (and 10 gbps of Internet 2). However things are always fast. Downloads fly, pages load instantly, etc. Reason is we all share and play nice. We get what we need and let the connection sit idle otherwise. If someone did slam the connection, they'd get a phone call from network operations.
I will say that bandwidth caps aren't the best solution to making people share and play nice, but other solutions are a good deal harder to implement, and for customers to understand. Bandwidth caps are something that is pretty easy for people to comprehend, and that will give them a reason to keep usage in check.
ISPs seem to really like money and business class liens cost more money. I've had business class service for about 13 years from 3 ISPs over that time, and all were plenty happy to sell it to me. Cox is who I'm with currently and they had no problem giving me a business account in my own name. Their software doesn't understand the concept, the "business name" field has my first and last name in it, but that doesn't hurt anything, and I get my nice business class Internet.
With wireless, well ya, you are going to have issues with that because there are hard bandwidth limits enforced by power output, frequency availability, and atmospheric noise (look up Shannon's Law) that you just can't get around. Wireless bandwidth will always have limits, particularly when you are talking wide area wireless.
Nothing is all benefit, or all cost. Nothing is all good or all bad. Everything has tradeoffs, everything has a cost. The trick then is weighing the costs and benefits to decide if something is good or bad overall, and what course of action to take.
Any time something is presented as completely one-sided, someone is pushing an agenda. You can see this with advertising all the time. They talk up how awesome whatever their product is, all the cool shit it does for you. They never talk about any limitations, downsides, etc, unless it is legally required (like the drug advertising). The reason is, of course, they are trying to sell you on it. They want you to think it is awesome so it is all benefits, all the time.
You are correct, it is something to watch for. When people present a very one-sided stance it is often because they are pushing a given point, not because there is no counter-point.
2% of the console's price is pretty exorbitant for open standards patents. The whole deal with (F)RAND stuff is "Reasonable and Nondiscriminatory". Now you don't have to license your stuff under that model, but that's how open standards like MPEG-4 and 802.11 are done. Companies pool their patents and set up a standard, and the licenses are fixed. The idea is that anyone can license it for the same amount, and that amount is fair and reasonable.
The reason companies do that is to get their patents used and licensed. I mean if I develop some cool new video compression, but I won't set licensing terms, everyone has to come to me and I decide if you get a license, and if so what it costs, well that will hamper adoption. Many companies will give that a miss since they don't know why it'll cost them. However if it is all out in the open, then it is much more likely to get used and licensed.
Also a lot of standards agencies require it. If you want your IP to be part of whatever standard they make, you have to disclose it, and license it under RAND terms. You don't want to, then it is excluded from the standard.
Well, if you decide to do that, you can't then go and decide to try and stick it to a company you don't like. You can't say "Yes, all our stuff is available under this fair license for all to use, oh except for you, we don't like you so you pay more." Sorry, you gave up that ability when you decided to do the open standards thing and RAND licensing.
Hence, the court decision. Google wanted to play hardball with MS, but they were doing it with patents they'd said they wouldn't do that with. So they got slapped down.
So ya, exorbitant demands. Particularly in context of what we are talking about. Remember Google doesn't own H.264 or 802.11. They only have a small number of the patents on it. So if their share was like 2%, then total cost could easily be 10-20%. If that was the kind of money demanded for those standards, they'd not be used. Google just wanted to screw MS.
Windows 7 does nothing at all with Blu-ray content. It doesn't understand how to play it. All it does in relation to any of this is provide a method for programs to inquire to drivers if everything is (supposedly) secure. A Blu-ray player can inquire as to the encryption status of the links and make sure things aren't being captured and so on. For that matter, so can other programs. It isn't Blu-ray specific, however only the media companies give a shit so that's all that really does it. Games don't mind at all if their output is being captured.
Doesn't matter the interface. DVI, HDMI, and DP can all encrypt the signal. There's nothing special, on a computer at least, about HDMI.
It is then up to the software how it acts on that. However, due to licensing requirements, the software has to disable the video out if everything isn't encrypted. If it doesn't they won't be able to get a license for the keys to decode the media.
Same deal on any platform. It isn't like Windows is special in this way. If your chosen platform doesn't support the necessary "protection" then there won't be any licensed Blu-ray playback software.
This is a media industry thing, not an OS thing. The OS provides the ability to have verified driver paths, but it does nothing at related to changing anything. That is up to the software, and that is dictated by licensing.
How, precisely, would you propose to build something backward compatible with the current spec that can push that kind of bandwidth, and be built for a reasonable cost?
The reason for these limits aren't arbitrary. It gets rather difficult and expensive to generate these real high bandwidth signals. Same reason why 10 gig ethernet costs so much more than gigE and needs better cabling to boot.
It isn't magic, as technology advances (particularly smaller lithography) it becomes possible to do higher clock rates at a lower cost and thus increase the bandwidth going over the cables. However it isn't something where we could just make it as fast as we wanted, easily and cheaply. If it were, well we'd have a lot higher interconnect speed.
So if you know some engineering voodoo that nobody else does that will allow for a 2-4x increase in bandwidth while still keeping cost low, well then off to the patent office with you You'll be able to make a mint. However if you are just whining that you can't have everything, without any actual understanding, then please stop.
HDMI is video and audio transport. Closed captioning works fine over it, since it comes from the video source. Be it your cable, DVD, Blu-ray, whatever, the CC information is processed on the relevant device, and then sent out as part of the video.
Asking HDMI to do closed captioning is like asking Ethernet to do packet filtering: You are looking at the wrong area.
When you are talking shared wires like cable modems or PON there is always the solution of just add more frequencies. With DOCSIS 3, you can have an arbitrary number of channels devoted to cable modems. So if what you have isn't enough, you can add more. Now that means taking them from something else, of course, however there's likely to be plenty available. When analogue service is discontinued, well that's 100 channels you've got right there (on many cable networks the first 600MHz is analogue cable service, the next 400MHz is all the digital stuff).
With PON it is even easier: Just shoot another laser down the fibre. That is already how it works: You have one wavelength for downstream, another for upstream, and another for video service. Well WDM doesn't stop at 3 wavelengths, you can have more. So as needed, more wavelengths can be added, giving more throughput.
In terms of the sharing issues, well it is easy to see what wins on a busy campus, like where I work. We all share bandwidth, of course, the campus has a gig or two of total Internet speed, not a gig for each desktop. However for all that, wired access is fast all the time. We have enough. Wireless, well that's another story. You go to the student union at lunch time and you find it is really, really bad. Things load slow, if they load at all. You have to reload pages plenty and so on. There are just too many people fighting for what's there. It isn't for lack of APs either, the university puts hundreds (literally) in a building, but when people are all in the same space, all hitting the same APs, well it gets overloaded.
There is this false idea that wireless is better than wired, that we will all move over, everything will be wireless all the time and life will be grand.
Nope. You can always get more bandwidth, quite a bit more, out of wires than wireless. That pesky Shannonâ"Hartley Coding Theorem just keeps cropping up and getting in the way. If you want more bits per second, you either need more bandwidth (meaning more spectrum) or a better signal to noise ratio. When you are talking wireless the only thing you can do about SNR is to up transmission power, which is not without its own issues, and there is just only so much bandwidth you can have, particularly with given properties.
See part of the problem is that as you move up the spectrum to higher frequencies, it gets easier to have more bandwidth, of course. However your signal gets more and more directional, and has less and less penetrating power. VHF and UHF are really good for transmissions. They are pretty non-directional and can penetrate most buildings without a whole lot of issues. However if you are operating on, say a 700MHz carrier your bandwidth is going to be limited, particularly when you have multiple services that want to use it. Indeed in the US you find that it is partitioned up in to 6, 10, 12, and 22MHz blocks.
Now if you go way up in frequency, this isn't a problem. Go up to a few hundred THz, instead of MHz. Now bandwidth isn't a big issue. If you have a carrier of 700THz then you can have a few THz of bandwidth, no problem and thus tons of information... Only one issue. 700THz might be more popularly called "blue". You are up in the light range now, and of course light can't penetrate for shit. Even a piece of paper would be sufficient to disrupt the signal. It is also highly, highly directional.
Finally there's a big issue which is that everyone has to share wireless. Anyone on a given segment, node, access point, etc is sharing whatever bandwidth there is. You don't each get your own bandwidth, you all have to share. So the more users, the less there is for each and there's really no way around that.
And thus the problem. You can't "just get more" bandwidth when you are talking wireless. You run in to physical limits. Your SNR is limited by power considerations (and distance) and the atmosphere, your bandwidth is limited by what is useful, and not used by other things, and so on.
With wired, not such an issue. You can go way up in frequency, particularly when you are talking fiber optic. However the real thing is that you can just lay more wires. You don't have to send a signal down one pair, you can have multiple. Ethernet is a good example of that. Gig and 10 gig use all 4 pairs, two to send, two to receive. Need more bandwidth on the same tech? Just lay another bundle. 8 pairs, as in two Cat-6a cables, will get you 20gbps, 12 pairs 30gpbs, and so on.
That's all dedicated (and full duplex) too. Only the endpoints use that. You can have stacks of cables running right next to each other, connecting different devices, and none of them trod on each other, they all have separate bandwidth.
So while wireless is cool, and useful, if we want fast speeds, if we want the ability to transfer lots of data all the time, we need wires. Wireless won't cut it. For that matter to the extent we can make it work well, it needs to be short range. You can use higher frequencies, have better SNR, and have less people per segment if you build the segments out. However that means lots of access points all over the place and those need to have backhaul, and that is going to be wired.
I get amused when people do this. Your "we/I" thing. I say something you don't like, so you extrapolate my views on the situation. You assume that I don't believe that humans are causing climate change. You'll notice I said nothing on the issue. I do that on purpose. My views as to the veracity of the science really aren't relevant. I am talking about the situation of what is being claimed and what to do about it.
However you take this somewhat religious point of view in that I argued something you didn't like, so I'm a heretic, an unbelieve, clearly I reject everything, I'm anti-science! You are assuming facts not in evidence.
I'm just pointing out one (of a number) of issues with getting widespread action on the issue: If we are past a tipping point, where reductions in emissions will not do much (and there are models that show this), then the solutions that are based around that are not useful.
I have a pretty good understanding of economics and one of the things I quite well understand is that setting up a system where rich countries give more money to poor countries, which would be the big effect of cap and trade, isn't going to help reduce emissions. If anything, it'll raise them since the poor countries will be able to develop their industry and then start using more fossil fuels themselves. You would likely find that many who thought the agreement was good when they were getting money would decide to withdraw later when it suited them.
If cleaner energy is the desired result and economics are the method you wish to use, then subsidies for R&D are the way to go. Taxes on coal to support nuclear, taxes on fossil fuels to support biofuels, that kind of thing. You pour money in to developing better biofuel technology or the like, and money in to getting the process up to the point where it is cheap, that will do something. Just shuffling money around country to country will accomplish nothing.
In terms of climate models ok, if you don't like that assertion then which models do you like, and why are those ones correct? What you have to realize is that there are a lot of climate models out there and their predictions are not consistent. They do not agree on how much warming, how fast, and what effect changes will have. So, if there's a correct one, then let's hear it, and evidence as to why that one is correct. If not, well that was relating back to my original point.
You seem to think that this is an argument claiming against climate change. It's not, I try to not interject my personal opinion at all in to these things. Rather this is talking about one of the problems relating taking any wide scale action. The actions that largely seem to be proposed will not do much to actually deal with the issue and, if we are indeed past a tipping point, be worse than useless (since they'd take up resources that should be spent in other ways).
I think a big problem is that people confuse and oversimplify the climate change argument. They think "There is a consensus, the science is settled, thus there is only one course of action we can take!" Nope, rather you can break it down in to four rough levels:
1) The fact of global warming, that the average surface temperature is rising outside of known cycles. This is an observation, a measurement. While it is a complex one, it is not really arguable unless you can find a flaw in the measurements or calculations.
2) The theory of global warming, in particular that the prime or exclusive cause of the warming is an increase in atmospheric CO2 (which is quite easy to measure) cause by human emissions. It provides an explanation for the causality of the observations we've made. As with all theories, you can debate this if you can find evidence that it is incomplete, or contradictory evidence or the like. Also that is a fairly general overview, there are researchers working on much more specific, predictive, versions that would explain more precisely the change in temperature with a given change in atmospheric CO2.
3) The assertion or judgement that this is a net bad thing for humanity. This is based on other theories and models as to what will happen if the temperature increase continues. You can then evaluate those and decide if it is a good or bad thing overall (everything has costs and benefits). For that matter two people could agree on the outcome, but disagree on the judgement of if it is good.
4) The politics or policy of what to do about it. This is the kind of thing there can be a lot of disagreement on, even if you agree on all other points. There's no "right" here. It is all about what seems to be the most beneficial to spend resources on.
The problem is that people seem to do sufficient reading to be convinced that #1 is true, or that #1 and #2 are true, and then decide that means you have to go all the way to the end, and that whatever their given source advocates for #4 is the One True Way(tm) and you are anti-science if you don't buy in.
That isn't how it works.
Is that it is just one of shuffling around who produces what. Now this may well work, if all we need to do is maintain our level of emissions or decrease them. However if we are past a tipping point, where nothing short of a massive reduction (and perhaps not even that) in emissions will stop the warming then it does no good.
What I was talking about with proposals and so on was things outside of "emit less CO2" or "here are way to try and emit less CO2". The reason is this is assuming that we are indeed past a tipping point, as some climate researchers claim. In that case, the issue of emissions is not one to concentrate on, but rather on how to either prevent the change via other means, prepare ourselves to deal with the change, or some of both.
We should all be happy to go back to the pre-industrial ages. Sure it means the vast majority of humans will have to die off, and the ones that live will have much shorter, harder, lives but hey, it would be good for the planet (depending on how you define good)! As such all of us should be happy, no honoured to do that. Excepting for professors, of course. They advance knowledge so they clearly need to be allowed to keep all of their modern conveniences. But the rest of us, back to the dark ages!
That is what always amuses me about the "industrial society is bad!" types is I've never seen any of them practice what they preach. None of them go and live in the wilderness, off the land, eschewing all modern technology except for the rare times they come to give a talk on it. Heck none of them even go back to Amish/Mennonite levels of technology. They live modern lives, enjoying all the conveniences, and then say others shouldn't.
How about, try it first, then see if maybe there's a reason we like all this new stuff?
That means we'd be down to around 45% of current levels by 2050. That's a big reduction. Now probably doable and worth trying, but you'd want to be fairly certain that it would, indeed, fix the problem if you are going to make the tradeoffs necessary to do so. You wouldn't want to spend a bunch on a big change to make all this happen only to find out no, sorry, but that isn't in fact going to help.