Jump to content

webtrekker

Members
  • Posts

    3,579
  • Joined

  • Last visited

  • Days Won

    40

Everything posted by webtrekker

  1. That is so true. I think possibly the worst thing to backup onto (and many people do) is an SSD. Yes, they're wonderful: small, low energy, and fast, but you get NO warning of when an SSD is about to fail, so there is NO chance of retrieving your data.
  2. Is it an external HDD and also disconnected (ie. USB lead unplugged)? You should only connect your external backup drive when you actually need to transfer files, and make sure you are disconnected from the internet. TOTALLY. If not, you may get a ransomeware attack that will destroy your backups. Unplug your drive again as soon as your file transfers are completed. I've accumulated Terabytes of data but, as I say, the important stuff that I never want to lose is kept on a handful of DVD's and an external drive.
  3. Believe me, hard drive crashes are more common than you think. Have you ever opened up a drive and seen how femor they are? (BTW, don't do that with a working drive! They're hermetically sealed and a speck of dust can make them inoperable). I've got DVD's from over 20 years ago that still work fine. You can always copy them onto new discs every so often if you're concerned about it. Remember, I only keep IMPORTANT info on these to limit the amount I need.
  4. The thing is, you can't live your life in constant fear. I've said and posted stuff for years that would incriminate me if they so wished. I can't let that worry me, and I can't turn back time, so I just carry on with my life enjoying what I can and passing on advice to my family.
  5. When I had an important HDD crash on me a couple of years back I tried everything I knew to retrieve the data, to no avail, so I took it to a good friend who owns a computer repair shop. It was complicated because the HDD was part of a RAID system so he had to ask another mate for a RAID board that matched mine. Luckily he got one and after 3 days managed to retrieve most of the important data. I asked him the best way to keep backups to stop this happening again. Of course he mentioned the cloud and external drives etc, but said the best way was to burn DVD's. They'd probably last the rest of my life and weren't prone to virus or ransomeware attacks. So a tip is to keep all of your verey, VERY, important files on DVD's (and duplicate those too for extra protection).
  6. In fact, to be honest, I don't have a lot of data that would interest them anywa, at least not enought that THEY would go to so much effort to steal it.
  7. All of my data is on external backups. Hard drives, thumb drives and DVD's. Anything I have in the so-called 'Cloud' is backed up on physical media too so, unless THEY knock on the door and take away my backups, then there's no other way they can lay their dirty paws on them. It's got nothing to do with AI. I also have hard copies of important stuff incase of blackouts or if the internet disappears.
  8. Ok, now I'll move on with my experiments! First off, I've copied the Transcript (without timestamps) of this random youtube video. No need to watch the video, I just wanted the text of the Transcript. Here's an excerpt of the Transcript as obtained directly from youtube ... artificial intelligence is already disrupting every industry we thought it would never touch creative Industries like art music movies and video games all of which are hundreds of billions of dollar Industries or more but just a few days ago Nvidia came out with an incredible research paper focused on generative AI for video a paper that could change everything so in this episode I want to show you the state of the art in a few different areas of design and then explain how nvidia's latest AI breakthrough might be the key to bringing them all together your time is valuable so let's get right into it I want to start this video in an unusual place the body cam footage of a tactical police officer that was dispatched to a nearby abandoned warehouse this video took the World by storm last month because it's actually from a video game called on record which is built in Unreal Engine by a small French indie game Studio named drama and drama is exactly what their gameplay video started the graphics are so photorealistic that many Publications thought it was an outright scam claiming that it was just a real life video with some game elements overlaid on top but this is a real game and the photorealism is actually achieved through a few pretty clever tricks to make it feel like you're looking at real footage from a body cam this is the state of the art of computer generated Graphics today in just a couple of years small Indie Studios won't even need camera tricks to make a game or a movie feel this fully immersive but these Graphics weren't generated with AI in fact the state of video generated with AI looks very different in videos generated entirely by diffusion models the pixels from one frame help inform what should be generated in the next frame one big problem with this approach is there's no real way for them to keep track of the structure of various objects from frame to frame as a result you get this sort of psychedelic painting within a painting within a painting style of motion picture that's not very useful for telling a coherent story there are a few different ways to bring those structural elements back into this kind of video the first way is with video to video Runway is a company that co-created the AI image generator stable diffusion they have a wide set of generative tools for image and video editing their gen 1 video generator takes a video as an input and then lets you edit it using text prompts and image overlays the result keeps the structure of the input video but switches out the look and feel based on the added prompts likewise Wonder studio is an AI toolkit that automatically places computer generated characters into live action scenes it automatically identifies individual actors and can overlay any CG character including all relevant animations and lighting effects based on the camera's position and motion so there are two parts to the process here creating the actual structure and composition of the scene and then adding in the correct skins textures and visual effects on top of that let me quickly highlight a few big Innovations in each of these air areas you probably already know how fast and cheap it is to generate an image with AI tools like stable diffusion or mid-journey but Runway has another tool called infinite image where you can take an existing image and expand it forever using text prompts so now you don't even need to create the whole overlay image or Texture Pack by yourself you can create a part of it and then generate the rest with AI Not very legible is it! So I pasted the excerpt into ChatGPT-4 with this preceeding it in the prompt ... 'Correct this youtube video transcript with correct capialization and punctuation but keep the words as they are originally :' After running the prompt, this is what I got ... Artificial intelligence is already disrupting every industry we thought it would never touch: creative industries like art, music, movies, and video games, all of which are hundreds of billions of dollar industries or more. But just a few days ago, Nvidia came out with an incredible research paper focused on generative AI for video, a paper that could change everything. So in this episode, I want to show you the state of the art in a few different areas of design and then explain how Nvidia's latest AI breakthrough might be the key to bringing them all together. Your time is valuable, so let's get right into it. I want to start this video in an unusual place: the body cam footage of a tactical police officer that was dispatched to a nearby abandoned warehouse. This video took the world by storm last month because it's actually from a video game called "On Record," which is built in Unreal Engine by a small French indie game studio named Drama. And drama is exactly what their gameplay video started. The graphics are so photorealistic that many publications thought it was an outright scam, claiming that it was just a real-life video with some game elements overlaid on top. But this is a real game, and the photorealism is actually achieved through a few pretty clever tricks to make it feel like you're looking at real footage from a body cam. This is the state of the art of computer-generated graphics today. In just a couple of years, small indie studios won't even need camera tricks to make a game or a movie feel this fully immersive. But these graphics weren't generated with AI. In fact, the state of video generated with AI looks very different. In videos generated entirely by diffusion models, the pixels from one frame help inform what should be generated in the next frame. One big problem with this approach is there's no real way for them to keep track of the structure of various objects from frame to frame. As a result, you get this sort of psychedelic painting within a painting within a painting style of motion picture that's not very useful for telling a coherent story. There are a few different ways to bring those structural elements back into this kind of video. The first way is with video-to-video. Runway is a company that co-created the AI image generator Stable Diffusion. They have a wide set of generative tools for image and video editing. Their Gen 1 video generator takes a video as an input and then lets you edit it using text prompts and image overlays. The result keeps the structure of the input video but switches out the look and feel based on the added prompts. Likewise, Wonder Studio is an AI toolkit that automatically places computer-generated characters into live-action scenes. It automatically identifies individual actors and can overlay any CG character, including all relevant animations and lighting effects, based on the camera's position and motion. So there are two parts to the process here: creating the actual structure and composition of the scene and then adding in the correct skins, textures, and visual effects on top of that. Let me quickly highlight a few big innovations in each of these areas. You probably already know how fast and cheap it is to generate an image with AI tools like Stable Diffusion or Mid-Journey. But Runway has another tool called Infinite Image where you can take an existing image and expand it forever using text prompts. So now you don't even need to create the whole overlay image or texture pack by yourself. You can create a part of it and then generate the rest with AI. 100% more readable I think you'll agree! All nicely capitalized and puntuated.
  9. I've already mentioned a couple of these in the Cornavirus thread so I'll copy those into here first ... Please read those two posts first as they contain good tips.
  10. webtrekker

    Ww3?

    Never mind. Just keep thinking about it and one day it will hit you like a bom....................
  11. Yep, just the same as THEY now have a copy of the post you've just made. ChatGPT is shortly being allowed to use the entire internet (indeed, it may already have) as its training dataset, so there'll be no escape anyway unless you air-gap your entire life I'm afraid. Life is very short (something I'm all too aware of now that I'm 71!) and I personally find it very helpful to be able to summarize articles that would normally consume a lot of my precious time. If the summay looks interesting then I will explore the article in more depth. In this way I can gain vastly more knowledge in the time available to me. Even just being able to list the main points of an article as bullet points is amazing and of great use to me.
  12. If you need a more powerful example of how useful this can be, consider reducing this article into a concise, bulleted summary ... (Yes, I know that screenshot is long and unreadable, but it was just to show the actual ChatGPT-4 page and how long the prompt was) So this is the final summary produced by ChatGPT-4 ... Temperature change always precedes CO2 change globally, suggesting that CO2 change cannot be the cause of temperature change. Data from Mauna Loa Observatory in Hawaii shows that variations in the rate of change of CO2 concentration reflect the lunar synodic periods, indicating that changes in CO2 are not causing the moon to rotate around the Earth. The monthly CO2 concentration has steadily increased over time, with an average rate of increase accelerating almost four times over 64 years. The seasonal variation in CO2 concentration is out of phase with the claim that increased CO2 causes temperature increase. Empirical data does not show a significant relationship between the satellite lower troposphere temperature and CO2 concentration at Mauna Loa Observatory or other locations, refuting the UN IPCC's claim. The rate of change of CO2 concentration is influenced by the temperature level, but temperature change does not result from CO2 concentration change. The correlation between temperature and CO2 concentration is not statistically significant, and CO2 changes lag temperature changes by eleven months. Analysis of data from various locations supports the conclusion that there is no statistically significant correlation between CO2 concentration and temperature. During the major 1997-98 El Nino event, the temperature change preceded the CO2 concentration change by five months, further indicating that CO2 change cannot cause temperature change. Cumulative temperature closely matches the trend of cumulative CO2 concentration, supporting the hypothesis that temperature drives the rate of CO2 generation. Ice core data and modern CO2 and global data show that CO2 change lags temperature change by several centuries to more than a millennium. Cross correlation analysis confirms that changes in CO2 concentration lag temperature changes, reinforcing the conclusion that CO2 does not cause temperature change. HOWZAT! A huge article condensed into 12 bullet points!
  13. Ok, I know that post is a bit long-winded and many will just skip over it, so here's a great suggestion ... Whenever you have a fairly long document to digest, copy it and post it into ChatGPT-4 asking for a concise summary, then you'll get something like this ... Now that's much easier on the eyes (and mind!) and could serve as a 'hook' to prompt people into actually reading the article, or as a quick summary for readers with little time. I'd suggest posting the summary above the longer article.
  14. webtrekker

    Ww3?

    Tell that to the folk from Hiroshima and Nagasaki, Alexa!
  15. webtrekker

    Ww3?

    My native language is English. I've called it KIev all of my life, as have everyone I know, and I'll not change now, same as I don't call Moscow Mockba.
  16. Oh yeah! LifeLog was 'coincidentally' cancelled on the exact same day that Zuckererg launched facebook. In fact, now it's been mentioned, here is a great Quora post about LifeLog/Facebook ... Don Romeo Conscript, Hellenic ArmyAuthor has 205 answers and 115.3K answer views1y There is no doubt for an intelligent, rationality-based person that Facebook is probably the largest part of the LifeLog system. Before getting into detail, consider this: The US Defense Department will arrest you, a private citizen, if you find a prime number - a string of digits from 1 to 9 - that is long enough to be of interest, and you dont report it to them. That is now a crime, and they will imprison you for posessing that mere string of digits. In contrast, Facebook is the worlds most powerful-ever intelligence network. No one in America out-spies the NSA, but Facebook does. Its quite hard to see how Facebook, apparently run by a couple of college nerds, would be able to run a competitor to the US intelligence community, on US soil, without them coming in and taking over. In fact, the notion of Mark Zuckerberg physically fighting off a stream of soldiers and special agents to maintain Facebooks autonomy is quite frankly ridiculous. There is no way - literally, I mean NONE - that the Pentagon would allow an independent company to not give them full access and control of a US military asset of such enormity - when they will punish the mere finding of a prime number. Some things are absolutely impossible, without a single, most infintessimal possibility of being true - and Facebook running the worlds largest ever spy network in California without being taken over by the Pentagon, is one of those things. Hopefully anyone who believes that this could ever be possible has now been quietly and unembarrassingly corrected, and can now pretend that they would never be so ignorant as to harbour such a fantastical delusion. So now lets start with the detail. After 9/11 (as all recent major problems originate) DARPA, in close collaboration with the US intelligence community (specifically the CIA), began developing a “precrime” approach to combatting terrorism known as Total Information Awareness or TIA. The purpose of TIA was to develop an “all-seeing” military-surveillance apparatus. The official excuse behind TIA was that invasive surveillance of the entire US population was necessary to prevent 'terrorist attacks' and disease outbreaks. The leader and designer was John Poindexter, previously Ronald Reagan's national security advisor. The TIA program met with considerable citizen outrage after it was revealed to the public in early 2003. The American Civil Liberties Union claimed that the surveillance effort would “kill privacy in America” because “every aspect of our lives would be catalogued,” while several mainstream media outlets warned that TIA was “fighting terror by terrifying US citizens.” As a result of the pressure, DARPA changed the program’s name to Terrorist Information Awareness. After considerable controversy and criticism, in late 2003, TIA was shut down and defunded by Congress, just months after it was launched. It was only later revealed that that TIA was never actually shut down, with its various programs having been covertly divided up among the web of military and intelligence agencies that make up the US national-security state. Some of it was privatized. So DARPA just moved these to classified portfolios of the Pentagon and US Intelligence Community, out of sight. A close friend of Poindexter, DARPA's program manager Douglas Gage, created LifeLog, which sought to “build a database tracking a person’s entire existence” that included peoples relationships, communications, thoughts, media consumption habits, purchases, behaviour, and much more, in order to build a digital record of “everything an individual says, sees, or does.” This was the first phase, and the data entry method was self-reporting (essentially getting people to spy on themselves). The second phase was take this unstructured data and organise it into "discreet episodes" and use it for "mapping out relationships, memories, events and experiences". LifeLog creates a permanent and searchable electronic diary of a person’s entire life. It seems AI was to be applied to this data, developed by Howard Shrobe and others. While DARPA publicly denied clandestine surveillance, DARPA’s own documentation on LifeLog noted that the project “will be able . . . to infer the user’s routines, habits and relationships with other people, organizations, places and objects, and to exploit these patterns to ease its task,” which acknowledged its potential use as a tool of mass surveillance. The application of these two steps is to completely model and predict human behaviour. And for network modelling, enabling an unimaginably powerful and broad set of possibilities for control of populations on every conceivable level. Among critics, Lee Tien of the Electronic Frontier Foundation told VICE at the time of LifeLog’s cancellation, “It would not surprise me to learn that the government continued to fund research that pushed this area forward without calling it LifeLog.” MIT’s David Karger was also certain that the DARPA project would continue in a repackaged form. He told Wired that “I am sure such research will continue to be funded under some other title . . . I can’t imagine DARPA dropping out of a such a key research area.” LifeLog was officially closed on February 4th, 2004. DARPA never provided an explanation for its quiet move to shutter LifeLog, with a spokesperson stating only that it was related to “a change in priorities” for the agency. On February 4th, 2004 - the exact same day - Facebook was officially launched. I hope its obvious that this, if an accident, would be an extraordinary coincidence. Facebook performs the same tasks as LifeLog in gathering everything about peoples social connections, activities and behaviour. The first point at which I became deeply skeptical of Facebook was, in a signing-up experiment, when it demanded by personal information. Having been on platforms like Twitter, Myspace and others before (Quora is another), I knew that online anonymity is a part of the deal, with no expectation to use real names, passport photos, institutions attended, birthplace, real date of birth, or anything else that Facebook was now coercing me to give them. Passport information is something that only governments demand (employers too but this wasnt a job application). The notion that Facebook might be government-run, asking as it was for my passport information like a police officer, was already at that point a possibility. At an oldly candid moment in 2015, Gage, DARPAs program manager, himself told VICE that “Facebook is the real face of pseudo-LifeLog at this point.” It is also on record that figures in the Facebook-Instagram-WhatsApp (etc) group, now renamed 'Meta', are figures in or related to the FBI, NSA, CIA, and DARPA itself. Examples of US Intelligence Community links: Max Kelly - FBI and NSA, also US Digital Service in the Executive Office of the US President. His movements in NSA after - official - departure from Facebook, were kept secret for three years. Chief Security Officer of Facebook. Sean Parker - CIA. Recruited by them at 16. First President of Facebook. Considered by many to be the person who brought to Facebook Peter Thiel. Thiel was then in coordination with the CIA (Parker's employer), and was actively trying to resurrect controversial DARPA programs that had been officially dismantled the previous year. Thiel had been developing the core panopticon software that was the aim of TIA, with his shady company Palantir. Richard Perle, the Reagan-Bush neoconservative and architect of the Invasion of Iraq, called TIA's Poindexter to introduce him to Thiel and associate Alex Karp, now Palantir's CEO. Thus another link to George Bushes 9/11-led government, as well as the link from TIA to CIA, and from CIA to Facebook. During their first meeting, Thiel and Karp sought “to pick the brain of the man now widely viewed as the godfather of modern surveillance.” Regina Dugan - DARPA. The group where all of this started. Dugan worked for DARPA from 1996, and became non other than their Director. In other words - Dugan led DARPA. There, she also led 'strategic initiatives in the field of social media'. After that she worked at Google, and then Facebook, where she headed Facebooks mysterious 'Building 8', which is considered some sort of skunk-works secret projects department. Marne Levine - daughter-in-law of 9/11-involved John Deutch, former Director of the CIA. First COO of Instagram. Previously worked at US Treasury Department. Joel Kaplan - in the White House Chiefs of Staff for George Bush, to whom he was a special assistant as policy advisor. At Facebook, he succeeded Levine as vice president of global public policy. Others of such a sensitive nature likely exist in influential roles at Facebook, now 'Meta'. My own experience: Being coerced by society to again sign up for a Facebook account (albeit only for emailing someone who gave no other contact information) I decided to make an 'anonymous' one - without real name, date of birth, passport photo, or other personal information that Facebook uniquely demands. It wasnt possible. It was not possible, and the account was blocked. Two highlights stick out to me: 1. When it locked me out, it demanded that I provide a photo, and either a copy of my passport or my drivers license. Have a think about that for a minute. A presumably civilian 'socialising' site is demanding your actual passport to make an account. On no other website in my life has this demand been levelled at me. This is clearly not a civillian operation. 2. I searched for an informal photo of some random person of my gender and roughly my age, to get the account unlocked. Once I uploaded this picture, I think a notice came saying something like 'our staff will check the validity of this picture as a representation of you'. I was now beyond all reasonable doubt. Lets agree on something: NO BUSINESS will spend resources on something like the accuracy of a free account-users photograph. Certainly not a genuine 'social network'. This does not increase their profit (there is no money in it) and it doesnt improve the user experience by increasing customer loyalty. However, it DOES cost money - you have to pay people wages to check all the tens or hundreds of thousands of profiles being made. It does cost time and resources. So if this was a corporate entity, they are wasting time and money on something that has no benefit to them. In fact, this demand actually risks LOSING customers through its overstepping of privacy rights. The only non-work entity that demands a passport photo is the government. The only entity of any kind that would actually RUN CHECKS on a submitted photo is the government. No business would waste resources to check its validity. Therefor, Facebook has to be a government agency. Sure enough, the photo was rejected. How did they know?! Well, they made it their business to find out. Not their business to do marketing, or improve the website, or make me feel welcome - but to check that my photo could be used to identify me. That they could put passport records to any future social and network activity. A large part of the gravity of the LifeLog problem is that is does what no major entity did before it. Generally, governments demanded your formal information. But your social life was largely unknown to them. And also, websites where you divulge your thoughts, your interests, beliefs, and social network, did not demand your formal birth information. There was a degree of separation between the two worlds. What LifeLog has done is gather your psychosocial/behavioural information, and then pinned that to your government information. So they know everything about you - and can use that information however they want, whenever they want, whyever they want. This just describes the outline of what LifeLog is. The actual possibilities with this system are, and I dont use this word lightly, unimaginable. I dont think theres a person on this planet who can picture in their mind the scope for abuse that Meta has for their mass and nature of information. 2-3 Billion humans, some of whom have been giving the Pentagon their personal, interactional, psychological and lifestyle data, daily, for two decades. Its the most powerful dataset that has ever been possessed - and in relation to the 'next most', by a multiple that itself is hard to conceptualise. What we can be almost sure of, is that they know what we are going to do or think, years before we do it. You just cant deny the predictive might of a trillion human datapoints. Most unforgivable, is that it has all been given *voluntarily*, it has been proactively *offered* to the Pentagon, by Facebook and Instagram users. By people *spying on themselves* - without even asking for a DoD salary. Together, the people on the Meta data-gathering platform have done something that has never been achieved in the wildest dreams of even the most genocidal, oppressive fascist: they have given the government the information to predict, and control, every minute detail of what it is to be human, on any population level they wish, in any conceivable situation, for as long as this data can be preserved. The psychological, behavioural, relational data can now be fed into a quantum computer, to calculate any (and probably every) eventuality. What a government can do with that informational power is beyond any conceivable limit.
  17. On Svalbard, there also just happens to be the Arctic World Archive described as ... A safe zero-emission repository on the safest natural location on earth The Arctic World Archive (AWA) is a unhackable, secure data and file storage vault located deep inside an Arctic Mountain on the isolated archipelago, Svalbard. Here we secure the world's digital data, memory, history, treasures and ensure that it is made available and accessible for future generations without data loss. Right next to the Global Seed Vault! How convenient!
  18. Too true. I used to think that DUMB's were just bolt-holes for the cabal if things turned nasty, but they're obviously much more than that. They've created a whole alternative existence for themselves while they annihilate most of the global population. Then there's the Svalbard Global Seed Vault ... ... which just happens to have its own airfield! I mean, WTF?!!! Interestingly, Norwegian law has prohibited the storing of genetically modified seeds at the vault! That's obviously for THEIR benefit while they poison the rest of us with GMO crops. Fucking turds!
  19. Consumer cryptocurrency trading should be regulated as gambling, Treasury Committee says in new report 17 May 2023 The Treasury Committee today calls for consumer trading in unbacked crypto to be regulated as gambling. In a new report, the cross-party Committee of MPs highlights that cryptocurrencies such as Bitcoin have no intrinsic value and serve no useful social purpose, while consuming large amounts of energy and being used by criminals in scams, fraud and money laundering. More here. That's great news! The proceeds from Gambling are Tax Free, aren't they? ₿
×
×
  • Create New...