A bit late now, as you've already learned Spanish, but there are now translation bots available for exactly this use case. Learning how to work around the quirks of the old translator is a pretty unusual situation, I'm not surprised that doesn't work anymore:( Even so, when I translate short conversational phrases from English to German/vice-versa I've found quality to improve... in particular, with idiomatic expressions.
Interesting, what are your usage patterns? I usually translate either web pages or short speech-like phrases, and I've seen improvements in both of these.
In 2108 the "DRM war" will have resulted in a messy, inconclusive draw with neither "side" obviously winning.
Music will be completely DRM free, and mostly sold by artists themselves direct from their websites using whatever the codec-de-jure is. URLs to simple "how to pay" files will be embedded using stenography into the audio itself, imperceptible to the human ear but easily decoded by peoples web-phones whether they are heard on the radio, in clubs, live, at a friends house etc. Paying for music won't be compulsory and sharing of collections will be common, in fact it'll be almost socially required to publish your music collection via some kind of social network app.
Even so, we won't have a wasteland of amateur music. These apps will also show where you got the music from - whether you paid for it or took it from a friends collection. There'll be social stigma attached to "leeching" too much music from other people... a collection composed entirely of music you downloaded from elsewhere will be seen as somewhat akin to living your entire life off social security. There'll be exceptions to this for people with no income, like teenagers or students, but people with jobs will be expected to buy most or all of their music.
Movies and TV shows will be DRM protected, and although there'll be ways to circumvent this DRM, it'll generally be unintrusive and movies will be cheap enough that it's more hassle to break the DRM than to just buy/rent them. Once bought you gain an irrevocable right to that movie, even if the bits themselves are not stored on your local systems. This right will be protected by law. Movies will be available to buy, rent and view in cinemas simultaneously around the world, as will TV shows. A minority of people will download cracked videos but again, as movies are cheap, convenient, and simultaneously available everywhere there'll be a social stigma attached to doing that.
Desktop software won't be DRM protected, but instead will have mostly moved into well protected datacenters where the bulk of their cleverness is implemented. They'll be subscription based rather than ad-supported and only a thin layer of UI logic will be running on a local client. It'll work, and people will like it, because peoples data and the software they use with it will be available everywhere in a secure manner. Latency will be low thanks to very fast networks and local "residential" datacenters that cache content and some code.
Video games, that aren't really suited to running over a network, will have moved entirely to games consoles which are now so specialized for video games and entertainment they are hardly recognizable as regular computers. They will be protected by extremely tough DRM, as code+data will be distributed via the net rather than on discs. The hardware-implemented protection will be cracked occasionally by professional 'pirates' and due to the vast size of the video game industry (eclipsing the movie industry by orders of magnitude), such breaches will have law enforcement coming down on them with the same ferocity that drug dealers experience today.
Right, I should go to bed before I spam this thread with my daydreaming anymore:)
Well, perhaps. That's the conventional view. But in the last few years we've seen something unexpected... that when you have gigantic amounts of data, statistical approaches can work very well.
If I want an answer to a question, chances are, I can find it on Google (or another search engine). Google doesn't "understand" things, it just does very very large and complex statistical analyses of a giant dumping ground for words. Despite that, it works, most of the time.
I think you should try the new Google Translate. The old comprehension based algorithms are gone, statistical translation are new. Even now, in their very very early days, there are noticable quality improvements. Given enough time and input material to train on, I think it's entirely feasible that they'll end up doing a good job on factual/technical texts for major languages.
Answering questions given a large corpus of text, translation of texts and speech recognition can all be treated as statistical problems. Then it's just a matter of gathering enough data, and perfecting the algorithms.
Anyway, comprehension based technologies are going nowhere fast, as you say. Machines that "understand" things are still stuck in limbo, and unless Cyc changes direction sharpish, will probably stay that way.
Dude, there's no shortage of energy and I say that as a fully paid up peak oil convert.
We're surrounded by energy. Once natural oil production starts to slide (and I can believe that'll happen in 0-5 years, if not before) we can and probably will replace it with coal-to-liquids technology, which is crap for global warming but does solve the problem of powering our food trucks.
It's an open question what will happen after that. Our investment in petroleum based propulsion is gigantic. It's lockin on a far bigger scale than Windows ever was. Even relatively minor changes like ethanol have problems with things like pipeline incompatability. Bigger changes like going to all-electric cars are thrown around without thinking through the costs.
Personally, I wouldn't be surprised at all if in 2108 we're still using cars powered by petroleum. Not petroleum that we suck out of the ground with hi-tech straws of course. That'll probably end in the next 50 years. Probably, either petroleum manufactured from biomass or extracted directly from the air and water (CO from the air, H from electrolysis, CO + H == syngas, input to the fischer-trope process). Petrol is amazingly energy dense, easy to transport and we have very hundreds of millions (billions?) of vehicles deployed that use it already..... carbon-neutral renewable petroleum? What's not to like?
All of those things, with the possible exception of Google, were predicted 10 years ago. People have been talking about things like VoIP (which is really just a re-invention of digital phone networks that already exist), broadband video on demand, multi-core CPUs etc way longer than ten years ago. It was predicted by many people that clock speeds would hit a wall due to physical constraints, and would have to go multi-core, back in the 80s.
Google is a more interesting case. I think the value of a "universal answering machine" is obvious and people were dreaming about such things back in the 50s. The fact that these things are actually called "search engines" and don't work by knowing everything but by sorting and ranking documents that contain the information you need, was less well predicted. And the fact that it'd be supported by contextual advertising wasn't really predicted at all.
Well yeah, but if it's true that somebody predicted ubiquitous mobile phones in 1908 then that's damn impressive, and suggests that the future isn't as opaque as we often think it is.
In 1908, hardly anybody even had landlines, the mobile phone was still 70 years away from invention and the idea of ubiquitous portable technology hadn't really taken hold yet. Today we take miniaturization for granted but in 1908 nobody had invented integrated circuits or the microchip. To predict mobile phones that far in advance shows tremendous hindsight and ability to correctly extrapolate from current trends (phones + radios + technology gets more ubiquitous == cellphones)
In 1908 people still had the first two world wars ahead of them. I guess nobody was predicting devastating wars in the way people here are, but it didn't matter - they happened and civilisation survived, advanced anyway.
I liked some of the predictions. The eye overlay is novel, but while I'm sure the technology for that will exist in 2108 I'm not convinced augmentation will be ubiquitous. There are easier ways to get the same effect. Why have your eyeball overwrite the words on a newspaper, when the newspaper could translate itself for you? Why even have newspapers instead of websites?
Ooh, this is fun. I can play at this too. In 2108 TV channels as we know them will be long since dead. There will still be series and regular shows, syndicated through various online providers, but TV on demand will have completely taken over and the idea of broadcasting a stream of shows 24/7 won't have any relevance... not even in news. Actually 2108 is probably way too pessimistic for that. I wouldn't be surprised if this happens in 30 years.
In 2108 everything will be recycled, and we will absolutely be mining landfills as the article suggests. I'm not sure they'll be used as archaeological sites - the people of 2108 will have extremely good historical records of our time, far better records that we have of the world in 1908. We'll probably be mining the moon and perhaps Mars as well, as we'll have long since exhausted our supplies of some exotic metals, foolishly incinerating them and thus spreading them into the air rather than keeping them in concentrated areas until appropriate recycling/disposal technology is invented.
Children will be given iPhone-like devices at birth (practically already happening) and we won't use email addresses or phone numbers anymore, at least not directly. Instead we'll use real names, disambiguated through social networks which cater to different kinds of people, but all partake in a shared social graph protocol of some kind that is as important and ubiquitous as HTTP is today. You won't ask girls for their number, instead you can just touch two phones together for a moment to mark that person as a "recent acquaintance" (these networks will have long sinced evolved beyond binary yes/no friend relationships).
Physical books will have given way completely to eBooks, probably read on fold-out versions of iPhone like technologies. Buying a physical book will be seen as something of a historical curiousity, sort of like paying for a portrait painting would be today. Books will still consist primarily of text, but will also include movies, interactive geegaws (think flash in wikipedia) and will automatically translate themselves to whatever language the reader is most comfortable with.
On the topic of automatic translation, statistical translation engines like Googles will have been tweaked, tuned and trained to the point at which they produce practically perfect translations of technical works, and human translation is seen as an expensive luxury reserved only for emotional works of literature (actually, I'd give this one 50 years). All phone/email conversations will be automatically translated behind the scenes, if necessary.
Fluent speech recognition will finally be viable and ubiquitous after the whole process is moved into datacenters, where enormous statistical models of phrase probabilities can provide superior accuracy and speed to todays desktop-bound solutions, simply through brute force (I give that 30 years tops as well, probably sooner)
To say that UNIX is a "wheel" is garbage. UNIX (and Windows, which is based on similar concepts) is a moth eaten dirty piece of cloth. It's got giant problems. Look at malware for one. Look at how many jokes revolve around software crashes of some sort, for another.
Before claiming that UNIX is like a wheel, go read up on modern operating system research. Seeing as you have a low opinion of Microsoft, might as well start there - try reading Singularity: Rethinking the software stack from Microsoft Research. They describe an operating system that, amongst other things, operates in a single address space without using hardware memory protection. There are no traditional processes, or syscalls. Instead the basic unit of software is a "Software Isolated Process" or SIP that is statically verified and compiled to machine code at install time. SIPs cannot be arbitrarily modified after installation. The whole thing is a single address space microkernel, except without the performance problems that scuppered previous microkernel attempts (because there are no context switches). A new security model based on verifiable type systems, state-machine based messaging and pre-declared intents allow for the construction of systems that are far more resistant to malware and unstable 3rd party extensions than today.
If you did LotR without the graphics, painted scenery, costumes, epic music and special effects you'd have theatre. There's a reason cinema is way more popular than theatre - these things matter and can really take a story to the next level.
Oh, also I don't agree that if the story is bad a film can't be successful. Look at the first Star Wars movie. The plot was derivative, predictable crap. It was an amazing success because it just had that magical something to it, and awesome special effects.
No, Windows and MacOS handle it by having stable APIs that are regression tested against a wide range of apps. New features, bugfixes and performance improvements are rolled out on a regular schedule after extensive testing, in a way that avoids breaking the majority of apps.
The Linux community could do this. They don't because of that kind of thinking. Our dependency resolution is perfect! Windows is so primitive! Well, yes, in an academic sort of way it is perfect, in that when the metadata is correct it does the "right" thing. But in a practical sort of way it completely sucks for the user who just wants to upgrade Firefox or Inkscape or whatever and can't because it would uninstall his desktop environment.
I don't think it'll ever change though. The way the "platform" (I use the term loosely) is managed today is a huge albatross around the neck of desktop Linux, but nobody really wants to change it. Better to try and 'educate' users as to why wanting the latest features is wrong and naughty.
#
# Require patent applicants to outline the level of investment necessary to realize a given patent - the system was designed to protect the investments of entrepreneurs so, if little to no investment is required, then there is no need for a patent on a given idea. Also, patent suit awards could be derived from this information accordingly.
Well, a bunch of other people have pointed out problems with your proposals - but it's easy to tell people why something won't work, and harder to improve an idea constructively. So I want to contribute to making your last point a bit stronger.
The basic idea is a good one, I think, but it has a big problem - what is the level of investment required? Rather than set a fixed level ("little to none" is really vague) judge how much it cost to develop the patent (cost of workers time, materials, etc) - call it N dollars - and then create a pool of N*X dollars, where X is some fixed value akin to interest rates, decided by governments. Higher values of X encourage research/development but discourage bringing new implementations of an idea to market. Now anybody who wants to use that patent, rather than be subject to the whims of an artificial monopoly, has to pay their share of the pool.
Here's an example. Let's say I develop a Wonder Widget, and it costs me $1000 in my time (as judged by what I could have earned with my skills doing something else), materials, etc. Nailing down exactly what these costs comprise of is a jungle of random opinions and subjective decisions, but it's no worse than trying to decide if something is "obvious" like we do today, and at least it's not a binary yes/no decision.
It cost me $1000, let's say X is 1.2 (a 20% "innovation incentive"?), thus the pool is $1200 in size. Right now, it's just me selling Wonder Widgets.
Before long, another company wishes to make Wonder Widgets too. In the current patent system, I can (in theory) stop them, or put more or less any conditions I like on the patent license - even revoking it later. But in this system, the new company has to split the pool, by paying $600 to me. Now both companies are $600 out of pocket.
A third company wants to join the fun. They have to pay $400 into the pool, split between the companies already there - I get $200 and the second company to join also gets $200. Now we're all $400 out of pocket.
The process repeats, and each time, the barrier to entry for competing in this market gets lower. But of course, anybody who joins at a late stage is going into a saturated market full of experienced players.
This solves your problem, of deciding on the "right" level for a patent to be grantable, by replacing it with a scheme in which all ideas are patentable but the consequences of a cheap idea being patented is very low.
Of course it has other problems. Maybe you'd need to enforce a rule that new players can enter the pool only once a year, or something like that, to stop people applying five minutes after the last guy. Maybe 20% is too low a multiplier on the size of the initial investment (it basically caps the profit you can make purely off the research part of R&D) - perhaps 200% or 2000% is a better idea.
how difficult would it be to develop a similarly intricate for linux or OS-X if a malware author decided to target those platforms?
Here's how you could make a somewhat modern piece of malware for Linux. I'll leave out the stuff that's the same between operating systems... the control networks, etc, and just look at controlling/hiding in the system.
First question - how to get in? All the usual techniques will work. Browser exploits are still common, even after years of hardening the IE and Firefox codebases. Plugin exploits (quicktime, acrobat, etc) even more so. Emailing out virus mails that appear to come from friends is still a very effective technique - we spent years training people to not trust emails from random people, only to have that advice subverted by having the emails come from friends. There are no restrictions on sending mail on Linux, nor reading from the users address book assuming they use client-side mail. If they use webmail the same techniques will work as on Windows.
Some people might say, but Mike, it's hard to make a binary that works on all forms of Linux! In reality, it's not that hard. The basic loading/linking code and core libraries are the same across distributions. It's hard once you try to build real, interesting apps that provide GUIs and so on, but if you're willing to put in some testing (and modern malware is a professional operation, so why not) you can make the same binary work just fine on dozens of distros.
Other people might say that it's complicated to run binaries on Linux, because you have to set the +x bit. I'll ignore the fact that I think Linux isn't ever going to get 33% market share with the current way of distributing software... suffice it to say, that once you convince a user that you're legitimate and that they want your eCard (that's how this malware spreads), you can just give them a command to copy/paste into the "Run Program" dialog box.
Once you're in, you want to do a few things. You want to download the rest of the trojan... no problem with that... maybe start sending mail... again no problem... what else? Maybe you want to drain the users bank account. The easiest way to do that is to install a browser extension that waits for the user to log in, and then scripts the web app. This has already been done on Windows/IE and isn't technically difficult - although it does require testing on the banks you want to target.
What else? Stealing cookies is popular. Yep, we can do that. Maybe popping up "unkillable ads". Yes, X will let you do this.
Next, you want to hide, to make yourself hard to get rid of. This is the part where people tend to assume Linux is more robust than Windows. Is it really? Well, firstly, you can do a decent job of hiding without root. To start with, try injecting yourself into a system process... or start several copies of the same program, all of which watch each other and restart new copies if others are killed or paused. It exploits the fact that you can't send signals to groups of processes atomically. Adjust the users path in their startup scripts to let you override any binary you wish, and then use a user-mode rootkit technique to hide the fact that the file was modified. Or set yourself to startup in the KDE/GNOME config systems somehow (eg, as an invisible panel app).
What if you want to store stuff on disk, and hide those files? Doing it with a kernel rootkit is easy enough, but what about without having access to kernel space? One way to do it is ptrace every process that might be used to explore the filesystem - like shells. You can intercept the syscalls of these programs before they reach the kernel in that way, and thus make files "disappear" from the command line, from Nautilus/Konqueror, or whatever other programs you want to do. If you're worried about the ptrace
ActiveX has been "defanged" for several years. You can't install random software without asking the user anymore in IE and that's been true for a long time.
The Storm botnet has been spread by emailing out binaries that people then run, because they believe it to be something it's not. That's a hard problem to solve. It hasn't really been solved by any system yet - perhaps it can't be solved.
Any computer where you can easily add new software (and a desktop OS that doesn't let you do that is one which isn't going anywhere fast), will have this problem.
That reasoning is invalid. There are tens of millions of XBoxes in the world, all of which run a customized version of Windows, yet I'm not aware of any viruses for the XBox. I guess Windows must be entirely secure!
Or maybe desktop security and arbitrary-consumer-electronic-device security are different problems with different solutions.
The other poster is correct. There is no difference in Windows vs Linux desktop security. It's beyond trivial to phish or intercept the users root password, if you want it, which you might not bother with because there are plenty of other ways to hide in a modern operating system (google "user mode rootkit").
Of course, hiding yourself on the system and ensuring your survival could be difficult.
It's easy. Poll/proc looking for Synaptic/Ubuntu Update or whatever the GUI package manager wrapper de jour is, and when you see it being started, send SIGSTOP and open up a clone of the systems "Enter your admin password" dialog box. Wait for the user to enter the root password and away you go, rootkits ahoy. Windows has some mechanisms to try and defend against this, I've yet to see a Linux distro that does.
Assuming 11% efficiency on these cells and 25% of that urban area being able to be coated in cells, and assuming an average insolation of 200W/m^2, we get a total power production of about 180 terrawatts. Current *world* demand is only 10 terrawatts. See where I'm going with this?
Yes, but Indium supplies are expected to encounter serious trouble once world production capacity is in the 20GW range. Some say it could be even earlier than that. I looked through this whole thread to find somebody raising the alarm about Indium supplies - there are only two posts about it that I can see, both at the bottom. There are serious supply issues facing a lot of the materials we rely on today as well as materials we're thinking of relying on tomorrow. For instance, if you think electric cars will one day fully replace petroleum cars, try running the numbers on world lithium supplies. Like most attempts to figure out how much of an exotic substance we have left, you'll encounter wildly varying numbers of equal credibility, and most of them will be far too low for your liking.
I clicked through a bunch of the vulnerabilities, and a lot of them are marked as reserved for future use. What's up with that? I think whatever script the dude used to compile this table, didn't work - either that or I don't understand the CVE process being used, because I don't see any indication of which systems are affected by them.
Anyway. Such a study is ultimately pointless, we already know that MacOS X and Windows are both seriously insecure. A single vulnerability in the tangled morass of code making up modern web browsers is typically enough to compromise the entire machine (Vista being an exception to this). A single vulnerability in *any* app which talks over the network is usually enough to get your code onto the machine, and from there you have free reign to do more or less whatever you want. Requiring root is no panacea, you don't need root to do the things modern malware wants to do anyway. As that's the entire OS X desktop security system right there, we can surmise that the primary advantage it has security-wise is just obscurity. (yeah, i know 10.5 is supposed to have MAC for some basic daemons etc.... wake me up when it is properly and widely applied to desktop apps).
I don't get it. You opened port 80 on different machines, and saw different traffic, none of which managed to exploit the web server.
I'm sceptical this tells us much about anything, beyond maybe the set up of your NAT/DMZ. Otherwise you should have received exactly the same traffic on both web servers. Bots don't check the OS before sending their exploitable GET requests.
How would having it in iTunes help Linux users? BBC would still lose. Flash is the only cross-platform solution to streaming video that has some kind of DRM in it.
Right, because ODF will never change and thus cause interop problems. Oh, wait. It'll change the moment OpenOffice release a new version, and if the standard lags behind, they'll stick the new features into another namespace and we're back to square one.
A standard like this is a way for two parties to communicate. Microsoft isn't going to switch to ODF, just like OpenOffice isn't going to switch to OOXML. The only thing that matters then is what format the next random office document you get emailed is using. It's not going to be ODF is it? It's going to either be a binary Office file, or maybe (just maybe) an OOXML file.
The KOffice people can take a "principled stand against OOXML" because hardly anyone actually uses KOffice, and the people who do aren't the type who are interacting with Windows-based organisations all day, and even if they were the KOffice team don't have any particular need to implement hard/annoying feature requests anyway.
Now I agree that dumb attempts to cheat at ISO don't help anyone, but on a purely technical level, any office suite that cares about actual interop instead of academic interop, will need to support OOXML just like they need to support the binary formats today.
I dunno for sure, but if I remember his rationale it was something like:
OOXML may specify (vaguely) tons of crappy Word bugs etc, BUT it is actually usefully large and complete, where as ODF isn't. I remember him comparing the number of spreadsheet functions defined in OOXML vs ODF. It made ODF look pretty sad. More importantly, Miguel and Jody Goldberg have actually written a real non-toy spreadsheet program (Gnumeric) that can read/write XLS files. So when Miguel says that 10 pages of documentation on spreadsheet functions aren't enough to get useful work done, I'm inclined to believe him. After all, de Icaza might have views that annoy people on Slashdot, but he does know programming and has plenty of useful code behind him to back it up.
OOXML specifies the behavior of 99.99% of the worlds actual Office documents. ODF doesn't. Thus, OOXML is more useful.
The above point has another implication - it's easier to implement for most Office programs, because they already had to be compatible with reading binary DOC/XLS/PPT files and OOXML is conceptually similar. Neither OOXML and ODF are really "neutral" - they are just XML dumps of some office suites internal state, and their structure reflects that. When writing filters for Gnumeric Jody Goldberg found it a lot easier to support OOXML than ODF. That's not really because OOXML is "better" than ODF or even that it's more specified, but rather because the work to make Gnumeric compatible with the way Office sees the world was already done.
The differences in opinion probably boil down to a pragmatic vs theoretical viewpoint. de Icaza and Novell actually work on an office suite that has to be used by, you know.... people. That means being usefully compatible with the real documents floating around. Joe Blogger or indeed Richard Stallman doesn't have to do that because they'll never write an ODF importer and find it doesn't work because the way OpenOffice implements feature XYZ isn't properly specified by the standard.
Now for my personal biases: I never wrote an Office suite, but I worked on Wine for several years. I've seen more than one armchair coder take potshots at things Microsoft has produced without knowing what the hell they're talking about. I've spent a whole ton of time figuring out the guts of Windows and whilst there's truly a lot of crap in there, too many people have the idea that feature XYZ must on Windows be worse than on Linux, just because Linux is Free.
Miguel has also spent a lot of time with both Linux and Microsoft technologies, and contrary to popular opinion that doesn't turn you into a brainwashed fanboy, it just gives you more information and experience with which to compare the two worlds. If based on his experience of writing office software, Miguel says to me that OOXML is more useful, then I believe him. Just as when I tell him why the Windows PE linkage model is better than the Linux ELF linkage model, I'd hope he'd believe me.
Canadian tar sands are bottlenecked on water supplies. The oil shale is something nobody figured out how to extract economically. Last time I looked at that, they were thinking of building large numbers of nuclear reactors on sites to finish off the process - seems unlikely and very expensive to me.
The current problem with rapidly escalating oil prices is not related to US refinery capacity. If the bottleneck was at the refinery level, crude would be cheap and gasoline would be expensive. In fact, both are expensive.
A statement like "more oil is currently known to exist than any other time in human history" is absurd, and I wonder how you arrived at it. We don't know how much oil exists, partly because some of it is - as you say - "undiscovered", ie we think it might be out there but we don't know for sure - but mostly because OPEC lie about the size of their remaining reserves. Data quality in the oil industry is poor to non-existant. We don't even have accurate figures for how much we pump out of the ground each day, let alone how much we have left.
You are confused about the pricing of sour crudes. Yes, they are cheaper, but not significantly so. The spot price of Mexican Maya on the 16th was $79, only about $10-$12 less than the price of the high quality stuff. Given that oil used to cost $10-$12 the fact that sour has risen to slightly less than sweet is really of no consequence.
You say there's no factual information to indicate that we're at peak. But world production has been flat since the summer of 2004, despite progressively increasing prices (due to increased demand from Asia) providing every incentive to pump more. This behavior has not been seen before and strongly suggests that world production capacity is maxed out - there are huge wins to be had by any company or country that can significantly boost production, but doing an analysis of an oil major like ExxonMobil, will show that their existing fields decline as fast as they can replace them. To me this is a pretty good sign that we're at peak - inability to raise production despite huge demand.
The whole "it's an oil industry conspiracy" won't wash, sorry. This isn't like the computer industry where one or two companies can dominate the landscape - oil is a commodity, and the price is not set by the oil companies but by supply and demand. It's the simplest market you can get. Anybody who is sitting on top of a giant oil field right now would be an idiot to leave it for tomorrow, because there's no guarantee we'll want that oil tomorrow - maybe there's a recession and oil demand is reduced. Maybe we discover better ways to power our cars.
Right now there's a lead-in time of at least 5 years from discovering a field to first commercial oil, sometimes longer. Even if you start today, there is risk. If you leave it longer, the risk gets even bigger. At least for private oil companies, there are huge financial incentives to boost production and thus get a leg up over your competitors in stock price and profits. To claim that the entire industry is in a cartel to deliberately hold back production is to reveal your lack of knowledge around discovery trends, skills shortages and the impact on depletion rates of modern production techniques like horizontal wells/waterflooding.
The video is sad but not for the reasons you give.
Sure, it's a shame that nearly all of them confuse the word "suffrage" with "suffering".
But seriously, I couldn't care less about that. All the guy proves is that English is a stupid language, and we all knew that already. If he had phrased it in a way that wasn't deliberately confusing he would have got sensible answers and no video.
What's actually sad about is that the girls all sign a petition not knowing what it's for. Even if the dude hadn't been playing word games, a petition to "end womens suffering" is completely meaningless - what's it supposed to achieve? How would he achieve it? What, exactly, are they putting their name to?
So, I think the dark haired girl at the start who says that it's mostly ended is smarter than the rest (with the exception of the one who refused to sign it:p). She actually thought "what the hell is this guy claiming?" - even though she let the pressure of being on camera sucker her into signing it, that's still better than 99% of them.
How is that a tax? The payments don't go to the government. They'd go to the people who were affected by the impact of climate change, probably via a chain of insurance companies.
Slashdot Mirror
A bit late now, as you've already learned Spanish, but there are now translation bots available for exactly this use case. Learning how to work around the quirks of the old translator is a pretty unusual situation, I'm not surprised that doesn't work anymore :( Even so, when I translate short conversational phrases from English to German/vice-versa I've found quality to improve ... in particular, with idiomatic expressions.
Interesting, what are your usage patterns? I usually translate either web pages or short speech-like phrases, and I've seen improvements in both of these.
In 2108 the "DRM war" will have resulted in a messy, inconclusive draw with neither "side" obviously winning.
Music will be completely DRM free, and mostly sold by artists themselves direct from their websites using whatever the codec-de-jure is. URLs to simple "how to pay" files will be embedded using stenography into the audio itself, imperceptible to the human ear but easily decoded by peoples web-phones whether they are heard on the radio, in clubs, live, at a friends house etc. Paying for music won't be compulsory and sharing of collections will be common, in fact it'll be almost socially required to publish your music collection via some kind of social network app.
Even so, we won't have a wasteland of amateur music. These apps will also show where you got the music from - whether you paid for it or took it from a friends collection. There'll be social stigma attached to "leeching" too much music from other people ... a collection composed entirely of music you downloaded from elsewhere will be seen as somewhat akin to living your entire life off social security. There'll be exceptions to this for people with no income, like teenagers or students, but people with jobs will be expected to buy most or all of their music.
Movies and TV shows will be DRM protected, and although there'll be ways to circumvent this DRM, it'll generally be unintrusive and movies will be cheap enough that it's more hassle to break the DRM than to just buy/rent them. Once bought you gain an irrevocable right to that movie, even if the bits themselves are not stored on your local systems. This right will be protected by law. Movies will be available to buy, rent and view in cinemas simultaneously around the world, as will TV shows. A minority of people will download cracked videos but again, as movies are cheap, convenient, and simultaneously available everywhere there'll be a social stigma attached to doing that.
Desktop software won't be DRM protected, but instead will have mostly moved into well protected datacenters where the bulk of their cleverness is implemented. They'll be subscription based rather than ad-supported and only a thin layer of UI logic will be running on a local client. It'll work, and people will like it, because peoples data and the software they use with it will be available everywhere in a secure manner. Latency will be low thanks to very fast networks and local "residential" datacenters that cache content and some code.
Video games, that aren't really suited to running over a network, will have moved entirely to games consoles which are now so specialized for video games and entertainment they are hardly recognizable as regular computers. They will be protected by extremely tough DRM, as code+data will be distributed via the net rather than on discs. The hardware-implemented protection will be cracked occasionally by professional 'pirates' and due to the vast size of the video game industry (eclipsing the movie industry by orders of magnitude), such breaches will have law enforcement coming down on them with the same ferocity that drug dealers experience today.
Right, I should go to bed before I spam this thread with my daydreaming anymore :)
Well, perhaps. That's the conventional view. But in the last few years we've seen something unexpected ... that when you have gigantic amounts of data, statistical approaches can work very well.
If I want an answer to a question, chances are, I can find it on Google (or another search engine). Google doesn't "understand" things, it just does very very large and complex statistical analyses of a giant dumping ground for words. Despite that, it works, most of the time.
I think you should try the new Google Translate. The old comprehension based algorithms are gone, statistical translation are new. Even now, in their very very early days, there are noticable quality improvements. Given enough time and input material to train on, I think it's entirely feasible that they'll end up doing a good job on factual/technical texts for major languages.
Answering questions given a large corpus of text, translation of texts and speech recognition can all be treated as statistical problems. Then it's just a matter of gathering enough data, and perfecting the algorithms.
Anyway, comprehension based technologies are going nowhere fast, as you say. Machines that "understand" things are still stuck in limbo, and unless Cyc changes direction sharpish, will probably stay that way.
Dude, there's no shortage of energy and I say that as a fully paid up peak oil convert.
We're surrounded by energy. Once natural oil production starts to slide (and I can believe that'll happen in 0-5 years, if not before) we can and probably will replace it with coal-to-liquids technology, which is crap for global warming but does solve the problem of powering our food trucks.
It's an open question what will happen after that. Our investment in petroleum based propulsion is gigantic. It's lockin on a far bigger scale than Windows ever was. Even relatively minor changes like ethanol have problems with things like pipeline incompatability. Bigger changes like going to all-electric cars are thrown around without thinking through the costs.
Personally, I wouldn't be surprised at all if in 2108 we're still using cars powered by petroleum. Not petroleum that we suck out of the ground with hi-tech straws of course. That'll probably end in the next 50 years. Probably, either petroleum manufactured from biomass or extracted directly from the air and water (CO from the air, H from electrolysis, CO + H == syngas, input to the fischer-trope process). Petrol is amazingly energy dense, easy to transport and we have very hundreds of millions (billions?) of vehicles deployed that use it already ..... carbon-neutral renewable petroleum? What's not to like?
All of those things, with the possible exception of Google, were predicted 10 years ago. People have been talking about things like VoIP (which is really just a re-invention of digital phone networks that already exist), broadband video on demand, multi-core CPUs etc way longer than ten years ago. It was predicted by many people that clock speeds would hit a wall due to physical constraints, and would have to go multi-core, back in the 80s.
Google is a more interesting case. I think the value of a "universal answering machine" is obvious and people were dreaming about such things back in the 50s. The fact that these things are actually called "search engines" and don't work by knowing everything but by sorting and ranking documents that contain the information you need, was less well predicted. And the fact that it'd be supported by contextual advertising wasn't really predicted at all.
Well yeah, but if it's true that somebody predicted ubiquitous mobile phones in 1908 then that's damn impressive, and suggests that the future isn't as opaque as we often think it is.
In 1908, hardly anybody even had landlines, the mobile phone was still 70 years away from invention and the idea of ubiquitous portable technology hadn't really taken hold yet. Today we take miniaturization for granted but in 1908 nobody had invented integrated circuits or the microchip. To predict mobile phones that far in advance shows tremendous hindsight and ability to correctly extrapolate from current trends (phones + radios + technology gets more ubiquitous == cellphones)
In 1908 people still had the first two world wars ahead of them. I guess nobody was predicting devastating wars in the way people here are, but it didn't matter - they happened and civilisation survived, advanced anyway.
I liked some of the predictions. The eye overlay is novel, but while I'm sure the technology for that will exist in 2108 I'm not convinced augmentation will be ubiquitous. There are easier ways to get the same effect. Why have your eyeball overwrite the words on a newspaper, when the newspaper could translate itself for you? Why even have newspapers instead of websites?
Ooh, this is fun. I can play at this too. In 2108 TV channels as we know them will be long since dead. There will still be series and regular shows, syndicated through various online providers, but TV on demand will have completely taken over and the idea of broadcasting a stream of shows 24/7 won't have any relevance ... not even in news. Actually 2108 is probably way too pessimistic for that. I wouldn't be surprised if this happens in 30 years.
In 2108 everything will be recycled, and we will absolutely be mining landfills as the article suggests. I'm not sure they'll be used as archaeological sites - the people of 2108 will have extremely good historical records of our time, far better records that we have of the world in 1908. We'll probably be mining the moon and perhaps Mars as well, as we'll have long since exhausted our supplies of some exotic metals, foolishly incinerating them and thus spreading them into the air rather than keeping them in concentrated areas until appropriate recycling/disposal technology is invented.
Children will be given iPhone-like devices at birth (practically already happening) and we won't use email addresses or phone numbers anymore, at least not directly. Instead we'll use real names, disambiguated through social networks which cater to different kinds of people, but all partake in a shared social graph protocol of some kind that is as important and ubiquitous as HTTP is today. You won't ask girls for their number, instead you can just touch two phones together for a moment to mark that person as a "recent acquaintance" (these networks will have long sinced evolved beyond binary yes/no friend relationships).
Physical books will have given way completely to eBooks, probably read on fold-out versions of iPhone like technologies. Buying a physical book will be seen as something of a historical curiousity, sort of like paying for a portrait painting would be today. Books will still consist primarily of text, but will also include movies, interactive geegaws (think flash in wikipedia) and will automatically translate themselves to whatever language the reader is most comfortable with.
On the topic of automatic translation, statistical translation engines like Googles will have been tweaked, tuned and trained to the point at which they produce practically perfect translations of technical works, and human translation is seen as an expensive luxury reserved only for emotional works of literature (actually, I'd give this one 50 years). All phone/email conversations will be automatically translated behind the scenes, if necessary.
Fluent speech recognition will finally be viable and ubiquitous after the whole process is moved into datacenters, where enormous statistical models of phrase probabilities can provide superior accuracy and speed to todays desktop-bound solutions, simply through brute force (I give that 30 years tops as well, probably sooner)
Yeah, I'll stop there ...
You just made the authors point for him, bravo.
To say that UNIX is a "wheel" is garbage. UNIX (and Windows, which is based on similar concepts) is a moth eaten dirty piece of cloth. It's got giant problems. Look at malware for one. Look at how many jokes revolve around software crashes of some sort, for another.
Before claiming that UNIX is like a wheel, go read up on modern operating system research. Seeing as you have a low opinion of Microsoft, might as well start there - try reading Singularity: Rethinking the software stack from Microsoft Research. They describe an operating system that, amongst other things, operates in a single address space without using hardware memory protection. There are no traditional processes, or syscalls. Instead the basic unit of software is a "Software Isolated Process" or SIP that is statically verified and compiled to machine code at install time. SIPs cannot be arbitrarily modified after installation. The whole thing is a single address space microkernel, except without the performance problems that scuppered previous microkernel attempts (because there are no context switches). A new security model based on verifiable type systems, state-machine based messaging and pre-declared intents allow for the construction of systems that are far more resistant to malware and unstable 3rd party extensions than today.
And they only just got started.
If you did LotR without the graphics, painted scenery, costumes, epic music and special effects you'd have theatre. There's a reason cinema is way more popular than theatre - these things matter and can really take a story to the next level.
Oh, also I don't agree that if the story is bad a film can't be successful. Look at the first Star Wars movie. The plot was derivative, predictable crap. It was an amazing success because it just had that magical something to it, and awesome special effects.
No, Windows and MacOS handle it by having stable APIs that are regression tested against a wide range of apps. New features, bugfixes and performance improvements are rolled out on a regular schedule after extensive testing, in a way that avoids breaking the majority of apps.
The Linux community could do this. They don't because of that kind of thinking. Our dependency resolution is perfect! Windows is so primitive! Well, yes, in an academic sort of way it is perfect, in that when the metadata is correct it does the "right" thing. But in a practical sort of way it completely sucks for the user who just wants to upgrade Firefox or Inkscape or whatever and can't because it would uninstall his desktop environment.
I don't think it'll ever change though. The way the "platform" (I use the term loosely) is managed today is a huge albatross around the neck of desktop Linux, but nobody really wants to change it. Better to try and 'educate' users as to why wanting the latest features is wrong and naughty.
Well, a bunch of other people have pointed out problems with your proposals - but it's easy to tell people why something won't work, and harder to improve an idea constructively. So I want to contribute to making your last point a bit stronger.
The basic idea is a good one, I think, but it has a big problem - what is the level of investment required? Rather than set a fixed level ("little to none" is really vague) judge how much it cost to develop the patent (cost of workers time, materials, etc) - call it N dollars - and then create a pool of N*X dollars, where X is some fixed value akin to interest rates, decided by governments. Higher values of X encourage research/development but discourage bringing new implementations of an idea to market. Now anybody who wants to use that patent, rather than be subject to the whims of an artificial monopoly, has to pay their share of the pool.
Here's an example. Let's say I develop a Wonder Widget, and it costs me $1000 in my time (as judged by what I could have earned with my skills doing something else), materials, etc. Nailing down exactly what these costs comprise of is a jungle of random opinions and subjective decisions, but it's no worse than trying to decide if something is "obvious" like we do today, and at least it's not a binary yes/no decision.
It cost me $1000, let's say X is 1.2 (a 20% "innovation incentive"?), thus the pool is $1200 in size. Right now, it's just me selling Wonder Widgets.
Before long, another company wishes to make Wonder Widgets too. In the current patent system, I can (in theory) stop them, or put more or less any conditions I like on the patent license - even revoking it later. But in this system, the new company has to split the pool, by paying $600 to me. Now both companies are $600 out of pocket.
A third company wants to join the fun. They have to pay $400 into the pool, split between the companies already there - I get $200 and the second company to join also gets $200. Now we're all $400 out of pocket.
The process repeats, and each time, the barrier to entry for competing in this market gets lower. But of course, anybody who joins at a late stage is going into a saturated market full of experienced players.
This solves your problem, of deciding on the "right" level for a patent to be grantable, by replacing it with a scheme in which all ideas are patentable but the consequences of a cheap idea being patented is very low.
Of course it has other problems. Maybe you'd need to enforce a rule that new players can enter the pool only once a year, or something like that, to stop people applying five minutes after the last guy. Maybe 20% is too low a multiplier on the size of the initial investment (it basically caps the profit you can make purely off the research part of R&D) - perhaps 200% or 2000% is a better idea.
Here's how you could make a somewhat modern piece of malware for Linux. I'll leave out the stuff that's the same between operating systems ... the control networks, etc, and just look at controlling/hiding in the system.
First question - how to get in? All the usual techniques will work. Browser exploits are still common, even after years of hardening the IE and Firefox codebases. Plugin exploits (quicktime, acrobat, etc) even more so. Emailing out virus mails that appear to come from friends is still a very effective technique - we spent years training people to not trust emails from random people, only to have that advice subverted by having the emails come from friends. There are no restrictions on sending mail on Linux, nor reading from the users address book assuming they use client-side mail. If they use webmail the same techniques will work as on Windows.
Some people might say, but Mike, it's hard to make a binary that works on all forms of Linux! In reality, it's not that hard. The basic loading/linking code and core libraries are the same across distributions. It's hard once you try to build real, interesting apps that provide GUIs and so on, but if you're willing to put in some testing (and modern malware is a professional operation, so why not) you can make the same binary work just fine on dozens of distros.
Other people might say that it's complicated to run binaries on Linux, because you have to set the +x bit. I'll ignore the fact that I think Linux isn't ever going to get 33% market share with the current way of distributing software ... suffice it to say, that once you convince a user that you're legitimate and that they want your eCard (that's how this malware spreads), you can just give them a command to copy/paste into the "Run Program" dialog box.
Once you're in, you want to do a few things. You want to download the rest of the trojan ... no problem with that ... maybe start sending mail ... again no problem ... what else? Maybe you want to drain the users bank account. The easiest way to do that is to install a browser extension that waits for the user to log in, and then scripts the web app. This has already been done on Windows/IE and isn't technically difficult - although it does require testing on the banks you want to target.
What else? Stealing cookies is popular. Yep, we can do that. Maybe popping up "unkillable ads". Yes, X will let you do this.
Next, you want to hide, to make yourself hard to get rid of. This is the part where people tend to assume Linux is more robust than Windows. Is it really? Well, firstly, you can do a decent job of hiding without root. To start with, try injecting yourself into a system process ... or start several copies of the same program, all of which watch each other and restart new copies if others are killed or paused. It exploits the fact that you can't send signals to groups of processes atomically. Adjust the users path in their startup scripts to let you override any binary you wish, and then use a user-mode rootkit technique to hide the fact that the file was modified. Or set yourself to startup in the KDE/GNOME config systems somehow (eg, as an invisible panel app).
What if you want to store stuff on disk, and hide those files? Doing it with a kernel rootkit is easy enough, but what about without having access to kernel space? One way to do it is ptrace every process that might be used to explore the filesystem - like shells. You can intercept the syscalls of these programs before they reach the kernel in that way, and thus make files "disappear" from the command line, from Nautilus/Konqueror, or whatever other programs you want to do. If you're worried about the ptrace
ActiveX has been "defanged" for several years. You can't install random software without asking the user anymore in IE and that's been true for a long time.
The Storm botnet has been spread by emailing out binaries that people then run, because they believe it to be something it's not. That's a hard problem to solve. It hasn't really been solved by any system yet - perhaps it can't be solved.
Any computer where you can easily add new software (and a desktop OS that doesn't let you do that is one which isn't going anywhere fast), will have this problem.
That reasoning is invalid. There are tens of millions of XBoxes in the world, all of which run a customized version of Windows, yet I'm not aware of any viruses for the XBox. I guess Windows must be entirely secure!
Or maybe desktop security and arbitrary-consumer-electronic-device security are different problems with different solutions.
The other poster is correct. There is no difference in Windows vs Linux desktop security. It's beyond trivial to phish or intercept the users root password, if you want it, which you might not bother with because there are plenty of other ways to hide in a modern operating system (google "user mode rootkit").
It's easy. Poll /proc looking for Synaptic/Ubuntu Update or whatever the GUI package manager wrapper de jour is, and when you see it being started, send SIGSTOP and open up a clone of the systems "Enter your admin password" dialog box. Wait for the user to enter the root password and away you go, rootkits ahoy. Windows has some mechanisms to try and defend against this, I've yet to see a Linux distro that does.
Yes, but Indium supplies are expected to encounter serious trouble once world production capacity is in the 20GW range. Some say it could be even earlier than that. I looked through this whole thread to find somebody raising the alarm about Indium supplies - there are only two posts about it that I can see, both at the bottom. There are serious supply issues facing a lot of the materials we rely on today as well as materials we're thinking of relying on tomorrow. For instance, if you think electric cars will one day fully replace petroleum cars, try running the numbers on world lithium supplies. Like most attempts to figure out how much of an exotic substance we have left, you'll encounter wildly varying numbers of equal credibility, and most of them will be far too low for your liking.
I clicked through a bunch of the vulnerabilities, and a lot of them are marked as reserved for future use. What's up with that? I think whatever script the dude used to compile this table, didn't work - either that or I don't understand the CVE process being used, because I don't see any indication of which systems are affected by them.
Anyway. Such a study is ultimately pointless, we already know that MacOS X and Windows are both seriously insecure. A single vulnerability in the tangled morass of code making up modern web browsers is typically enough to compromise the entire machine (Vista being an exception to this). A single vulnerability in *any* app which talks over the network is usually enough to get your code onto the machine, and from there you have free reign to do more or less whatever you want. Requiring root is no panacea, you don't need root to do the things modern malware wants to do anyway. As that's the entire OS X desktop security system right there, we can surmise that the primary advantage it has security-wise is just obscurity. (yeah, i know 10.5 is supposed to have MAC for some basic daemons etc .... wake me up when it is properly and widely applied to desktop apps).
I don't get it. You opened port 80 on different machines, and saw different traffic, none of which managed to exploit the web server.
I'm sceptical this tells us much about anything, beyond maybe the set up of your NAT/DMZ. Otherwise you should have received exactly the same traffic on both web servers. Bots don't check the OS before sending their exploitable GET requests.
How would having it in iTunes help Linux users? BBC would still lose. Flash is the only cross-platform solution to streaming video that has some kind of DRM in it.
Right, because ODF will never change and thus cause interop problems. Oh, wait. It'll change the moment OpenOffice release a new version, and if the standard lags behind, they'll stick the new features into another namespace and we're back to square one.
A standard like this is a way for two parties to communicate. Microsoft isn't going to switch to ODF, just like OpenOffice isn't going to switch to OOXML. The only thing that matters then is what format the next random office document you get emailed is using. It's not going to be ODF is it? It's going to either be a binary Office file, or maybe (just maybe) an OOXML file.
The KOffice people can take a "principled stand against OOXML" because hardly anyone actually uses KOffice, and the people who do aren't the type who are interacting with Windows-based organisations all day, and even if they were the KOffice team don't have any particular need to implement hard/annoying feature requests anyway.
Now I agree that dumb attempts to cheat at ISO don't help anyone, but on a purely technical level, any office suite that cares about actual interop instead of academic interop, will need to support OOXML just like they need to support the binary formats today.
OOXML may specify (vaguely) tons of crappy Word bugs etc, BUT it is actually usefully large and complete, where as ODF isn't. I remember him comparing the number of spreadsheet functions defined in OOXML vs ODF. It made ODF look pretty sad. More importantly, Miguel and Jody Goldberg have actually written a real non-toy spreadsheet program (Gnumeric) that can read/write XLS files. So when Miguel says that 10 pages of documentation on spreadsheet functions aren't enough to get useful work done, I'm inclined to believe him. After all, de Icaza might have views that annoy people on Slashdot, but he does know programming and has plenty of useful code behind him to back it up.
OOXML specifies the behavior of 99.99% of the worlds actual Office documents. ODF doesn't. Thus, OOXML is more useful.
The above point has another implication - it's easier to implement for most Office programs, because they already had to be compatible with reading binary DOC/XLS/PPT files and OOXML is conceptually similar. Neither OOXML and ODF are really "neutral" - they are just XML dumps of some office suites internal state, and their structure reflects that. When writing filters for Gnumeric Jody Goldberg found it a lot easier to support OOXML than ODF. That's not really because OOXML is "better" than ODF or even that it's more specified, but rather because the work to make Gnumeric compatible with the way Office sees the world was already done.
The differences in opinion probably boil down to a pragmatic vs theoretical viewpoint. de Icaza and Novell actually work on an office suite that has to be used by, you know .... people. That means being usefully compatible with the real documents floating around. Joe Blogger or indeed Richard Stallman doesn't have to do that because they'll never write an ODF importer and find it doesn't work because the way OpenOffice implements feature XYZ isn't properly specified by the standard.
Now for my personal biases: I never wrote an Office suite, but I worked on Wine for several years. I've seen more than one armchair coder take potshots at things Microsoft has produced without knowing what the hell they're talking about. I've spent a whole ton of time figuring out the guts of Windows and whilst there's truly a lot of crap in there, too many people have the idea that feature XYZ must on Windows be worse than on Linux, just because Linux is Free.
Miguel has also spent a lot of time with both Linux and Microsoft technologies, and contrary to popular opinion that doesn't turn you into a brainwashed fanboy, it just gives you more information and experience with which to compare the two worlds. If based on his experience of writing office software, Miguel says to me that OOXML is more useful, then I believe him. Just as when I tell him why the Windows PE linkage model is better than the Linux ELF linkage model, I'd hope he'd believe me.
Canadian tar sands are bottlenecked on water supplies. The oil shale is something nobody figured out how to extract economically. Last time I looked at that, they were thinking of building large numbers of nuclear reactors on sites to finish off the process - seems unlikely and very expensive to me.
The current problem with rapidly escalating oil prices is not related to US refinery capacity. If the bottleneck was at the refinery level, crude would be cheap and gasoline would be expensive. In fact, both are expensive.
A statement like "more oil is currently known to exist than any other time in human history" is absurd, and I wonder how you arrived at it. We don't know how much oil exists, partly because some of it is - as you say - "undiscovered", ie we think it might be out there but we don't know for sure - but mostly because OPEC lie about the size of their remaining reserves. Data quality in the oil industry is poor to non-existant. We don't even have accurate figures for how much we pump out of the ground each day, let alone how much we have left.
You are confused about the pricing of sour crudes. Yes, they are cheaper, but not significantly so. The spot price of Mexican Maya on the 16th was $79, only about $10-$12 less than the price of the high quality stuff. Given that oil used to cost $10-$12 the fact that sour has risen to slightly less than sweet is really of no consequence.
You say there's no factual information to indicate that we're at peak. But world production has been flat since the summer of 2004, despite progressively increasing prices (due to increased demand from Asia) providing every incentive to pump more. This behavior has not been seen before and strongly suggests that world production capacity is maxed out - there are huge wins to be had by any company or country that can significantly boost production, but doing an analysis of an oil major like ExxonMobil, will show that their existing fields decline as fast as they can replace them. To me this is a pretty good sign that we're at peak - inability to raise production despite huge demand.
The whole "it's an oil industry conspiracy" won't wash, sorry. This isn't like the computer industry where one or two companies can dominate the landscape - oil is a commodity, and the price is not set by the oil companies but by supply and demand. It's the simplest market you can get. Anybody who is sitting on top of a giant oil field right now would be an idiot to leave it for tomorrow, because there's no guarantee we'll want that oil tomorrow - maybe there's a recession and oil demand is reduced. Maybe we discover better ways to power our cars.
Right now there's a lead-in time of at least 5 years from discovering a field to first commercial oil, sometimes longer. Even if you start today, there is risk. If you leave it longer, the risk gets even bigger. At least for private oil companies, there are huge financial incentives to boost production and thus get a leg up over your competitors in stock price and profits. To claim that the entire industry is in a cartel to deliberately hold back production is to reveal your lack of knowledge around discovery trends, skills shortages and the impact on depletion rates of modern production techniques like horizontal wells/waterflooding.
The video is sad but not for the reasons you give.
Sure, it's a shame that nearly all of them confuse the word "suffrage" with "suffering".
But seriously, I couldn't care less about that. All the guy proves is that English is a stupid language, and we all knew that already. If he had phrased it in a way that wasn't deliberately confusing he would have got sensible answers and no video.
What's actually sad about is that the girls all sign a petition not knowing what it's for. Even if the dude hadn't been playing word games, a petition to "end womens suffering" is completely meaningless - what's it supposed to achieve? How would he achieve it? What, exactly, are they putting their name to?
So, I think the dark haired girl at the start who says that it's mostly ended is smarter than the rest (with the exception of the one who refused to sign it :p). She actually thought "what the hell is this guy claiming?" - even though she let the pressure of being on camera sucker her into signing it, that's still better than 99% of them.
How is that a tax? The payments don't go to the government. They'd go to the people who were affected by the impact of climate change, probably via a chain of insurance companies.