Meltdown and Spectre are the two flaws. AMD is only immune to Meltdown. All are susceptible to Spectre due to the speculative nature of speculative execution. Has to do with instruction prefetching.
Very true. The slightest shift in personal paradigm opens the floodgates for ramifications of the shift. Incubation and processing time are required.
Reverse engineering assembly (which is just about all I've programmed in since 1984) is very linear. There are no abstractions beyond the conceptual design of an app or driver as a whole. The key to being good at doing it is patience, and patience comes from naturally resonating at that level of excessively fine detail. And of course you need a very good working knowledge of the operating system's internals. So it's plausible that she really does this work.
As for being transgender … if it's true, she has pulled out all the stops - adam's apple shaving, facial feminization surgery, etc. The only real supporting feature is arm length. The voice is VERY difficult to get right and she's relatively high pitched for that. The clowns would pay for it all if they had a purpose, and that purpose could be for her to woo and wow and get close to other male hackers to figure out what they know or what they're involved in. So it's plausible, but it would have been an A+ primo job of transition.
Absolutely the case. If you try to write an assembler or disassembler, things become pretty hairy because the little tricks they've employed for encoding instructions have become more and more complex and convoluted as MMX, then the progression through SSE's, and finally AVX came into being, in addition to a plethora of new instructions. The instruction encoding base is still stuck in the original 8086 arena and it's been maxed out to the point of ridiculous complexity trying to expand. They need a reset. Desperately. An entirely new system of encoding instructions that allows for a much simpler and more direct (and sensible) implementation. But regarding "intelligence," you don't have to be a rocket scientist. I look at typical equations for collision detection, plane intersect, etc. in the 3D math (trig) arena and I see all those Greek symbols and abstractions and I just close the browser tab. My brain shuts off with all that. I can process it, I just hate doing it. I like everything linear. A connects to B, no variables, no fluff, no interpretation. Which only simplifies the entire process. So it all just comes down to working well at that level of detail, which MOST people do not do. They don't like it. The ability is there; the passion for doing it isn't.
Something to keep in mind when it comes to figuring out Q's stuff.
Analysis Time
I'm posting this because it's early and we still pretty much have the night crew on.
I can write a search app for all our threads. Data can be collected from online, into offline files, processed, and saved for access. It is quite doable and it would work MIRACLES for our research. It's happened to me a thousand times: "I know I saw something about that posted earlier… was it last week? Last month? Oh forget it, I'm not going through 500 threads one at a time….." and the research doesn't get done. What if it did get done? The right tools for the job are beyond critical.
Here's the problem: it's a double-time job. And that is exactly what I would put into it. 80+ hours a week. The biggest question is, what would the longevity of such a project be? And how are we going to answer that? Ask Q? "Hey Q what is the exact date the Storm will end?" I'm going to go out on a limb and speculate that such a question would not be answered.
Back in November (the 5th) when I dove into all this, I was working on an all-assembly OS to support ONLY gaming. And a replacement for DirectX which actually eliminated that layer entirely. And of course my own games in addition, complete with dev tools. All that got shelved to do Q stuff full time.
I live with my ex-gf, now roommate; she works full time for $12.75 per hour and I get $500 per month doing an office cleaning contract. (Hopefully no doxxing here.) We scrape by. The cost of living in our rural town is so low that we're able to survive with this setup, but my car is 18 years old and could die any block. It's a never-say-die work horse and it's still hanging in there but nothing lasts forever. I just can't dive in and take on another project that could easily go on for 2+ months (before I get it all right) without some serious motivation. I have a very intense resistance to trying to capitalize on all this and commercialize Q. I would feel like a lowlife exploiter who just didn't understand or appreciate what it was all about. But the bills keep coming in and one major hit like a fridge crash (which is more likely than not hour by hour) would kill us. I've said it before; our very old house is a prime candidate for demolition and we are not in good living conditions. But we get by - barely. Taking on a project like this would dramatically increase our effectiveness; probably the biggest thing we have working against us is that 750 post limit to each thread. But it is what it is and it's what we have to work with.
I can write code to automate the process of collecting all the bread data off the web; save it locally and even distribute it into blindingly high-speed B-tree database file format. It's all very doable. But it takes time.
We can't hold off fate forever. The car is going to go one day. The fridge is and the washer is and I have to get income. Working a standard job isn't going to happen. My age, my male gender, my White race, none are helping me. And trust me, every boss I have goes nuts, implodes, and fires me. I'm the most conscientious, polite, respectful, responsible, diligent, and capable worker there ever was. I don't care what it is I'm doing. But bosses implode when they're over me. They melt down in very short order and send me packing, never really having a clue what I did that was considered "fireable." Because I didn't do anything. I performed so well that I should have been promoted to CEO. But egos are egos and in the real world, that's the way it happens. So money has to come from somewhere else.
I don't even want to address the fact that I am in NO WAY soliciting money. NEVER. I don't do things that way. I'm just outlining the situation and the factors that are involved in deciding "do I want to do this."
Am I missing options? What would YOU do?
… what makes this so viable is the fact that once a thread fills up and closes out, it's locked in and it will not change. It's set in stone.
The support is seriously appreciated. Really.
The biggest consideration is how long it will be good for? Nobody can answer that.
Yesterday I lectured my roommate about how carefully life is controlled from a level we rarely if ever comprehend. Nothing is left to chance. If logic worked in mapping out a good life, we'd all have zero problems. It isn't about logic and planning - although those have their place; mostly in control over how we use the resources we do have. But fortune, income, lack of it, it all happens to us. Nobody controls that. We like to think we do, or that we can gain that control, but in reality, it doesn't ever work. Not really.
We need this app. We really do. I just keep thinking of how much more we would figure out than we're figuring out now. How many loose ends are untied.
I wonder if CodeMonkey would let me have the algorithm for how user ID's are changed when they change. I wouldn't need to extract the IP address from it; I probably couldn't. Just some way of telling what a given user was ID'd as las thread vs. this thread. Some people show up here who are very cryptic or have a lot of pull while being very mysterious and it would be really nice to be able to track them across threads. But that's probably flying in the face of the very anonymity we exist for.
Retrieve the catalog page. From that, identify the thread #'s for all the threads not yet archived. The ones that have been archived need to be retrieved another way (I'd work it out). Then I could step through the threads, retrieve the entire data block, and parse it out. Just as I do with Q messages, build db files with all identifying data: date, message #, etc.
There isn't a viable way to do that with message content. It would have to be searched brute force. But per local machine (all the data would be downloaded to each user's computer, which could go a long way toward preserving it) it would still be a tolerable time frame. The stuff I write always runs 10x to 100x faster than stuff written in any other language. I've been doing assembly apps for 34 years and I'm current with today. I known what the numbers are.
I'm going to have to figure out what to do. I'll start by mapping out the project. A firm decision will happen to me.
People died for what they did. Families endured endless suffering for their greed. For many, many years. A momentary change of heart does not undo all that. I wouldn't presume to begin plotting revenge for all people, but at the same time, the benefit of anonymity (and the attendant lack of accountability that goes with it) is not something they are owed no matter what their stand today.
Yes. In short, CPU's derive about 30% of their performance from reading ahead in the code stream and trying to predict what will happen. This prediction has to occur within some kind of work space on the CPU. That work space is readable, indirectly if not directly and any language can do it. The flaw is that even if the code "up ahead" is executing blatantly illegal instructions, that is not taken into account for the "what if" phase of speculation. Hence the flaw.
Others can set that up, and will probably need to. I only do Windows programming and a lot of people use other platforms. The benefit of my work will be the speed of execution. Absolutely blinding. Far beyond what anybody thinks possible today.
Well, they're hex VALUES, which can be interpreted as colors since each RGB channel goes from 0 to FF. But it's much more likely just a unique representation of each user.