And let's not forget the retarded "pinch to rotate" gesture. I don't think I've EVER met anyone, in the ENTIRE HISTORY of Google Maps for Android, who's INTENTIONALLY rotated a fucking map after the first time or two they've used it. It's one of those features that was cool to show off for 5 minutes, but actively harms the app's ultimate usability. I'm one of those people who've wished for YEARS that Google would add a preferences option to Google Maps for "disable rotate gestures"
> US East coast would be "book of Eli", at best. US West coast will probably not exist.
> any human society larger than a small town on the planet will be a miracle
Not really. Most estimates I've seen show South Florida left relatively unscathed. By ashfall, at least. SoFla's main problem would be dealing with an almost-overnight influx of ~40 million new residents (path of least resistance for people fleeing the eruption from elsewhere... in the US, Florida is theoretically reachable even on foot within a few weeks from most parts of the eastern US).
> If it had been a phone designed to be a good phone, it could have succeeded.
And more importantly, if it had been at LEAST as open (non-walled-garden) as Windows Mobile used to be. But no, Microsoft wanted to be Apple & exercise total control (and collect a share of the revenue) for everything users could run. Windows Phone had zero appeal to "Apple" type users, and even LESS appeal to Android users (many of whom WERE Windows Mobile users prior to 2009 or 2010).
Windows Mobile is a case study in corprate dementia & self-inflicted failure. WM6.5 was ugly, but the core OS functionality was good, and its abandonment set the mobile industry behind for at LEAST 2-3 years with respect to things like Bluetooth and GPS (Android Bluetooth was pretty dysfunctional until Eclair, and Android GPS in general was a nuclear-hot mess until sometime around Gingerbread or ICS).
There's one point many here have overlooked... it doesn't necessarily require a "UHD" TV to directly benefit from it.
Back around 1994, I bought my first "big" TV back when I was in college -- a 27" Daewoo. Nothing earthshaking, but it had higher-end specs... s-video input, decent stereo audio, and significantly better dot pitch than most of the other TVs at BrandsMart USA.
For the next 5 years, connected to a cable box and VCR, it mostly looked like shit. VHS tapes looked marginally less shitty at my house than at friends' houses, because I had a S-VHS VCR (mostly, so I could get S-Video output, even when watching non-SVHS tapes) -- they still had poor detail and awful color resolution, but at least the NTSC color artifacts were slightly better.
At one point, I contemplated buying a LaserDisc player. Unfortunately, by that time, LaserDisc was basically dead as a format in the US. In any case, a LaserDisc definitely wasn't something you could waltz into Blockbuster Video and rent for $5.
Then DVD arrived. Instantly, the TV's perceived video quality more than doubled. The TV's "real" physical resolution was somewhere around 360x480 under ideal conditions... but having a 720x480 video source to play with meant I could pump ever bit of signal detail into that TV that it was physically capable of handling. The improvement was dramatic.
The early 2000s arrived. My housemate bought a DLP HDTV, and we decided to switch to Voom! so we could enjoy "real" HDTV (at the time, Comcast in Miami hadn't started carrying HD channels yet, and only one or two of the local broadcasters were transmitting HD). As luck would have it, Voom! didn't screw around with "SD" boxes -- every box they gave you was capable of outputting anything from 480i60 over s-video (or composite, or component) to 720p60 and 1080i60 HD (over component or DVI). By this point, the old CRT TV from my college days was now my bedroom TV. And I quickly realized that even though it wasn't HD, Voom's channels (and OTA HDTV from the antenna diplexed into the LNB's output) looked MUCH better than anything I ever saw with cable.
Fast forward a bit more. Voom! went out of business, and my housemates and I were forced to settle for Comcast. God, it was awful. For a couple of days (before Voom shut down, after Comcast came to set us up), I was able to compare different channels from Voom and Comcast side by side. For SD channels like MTV, it was literally a night and day difference. On Voom!, MTV and MTV2 looked like DVD-quality video. On Comcast, I could barely watch it without feeling like my eyes were going to bleed from the blocky overcompression.
A couple of years later, I bought my own house, got a nice 65" DLP TV of my own, and signed up for DirecTV. I still had the same TV (from college) in the master bedroom, and decided to pay the extra $2/month to get a HD box for the master bedroom. Happy days again... DVD-quality, even from SD channels. Ditto, after switching to U-verse a couple of years later (a slight step down for HD, but more or less the same quality for 480i60 via S-video).
My point: most of the (2k, 720p60 or 1080i60) "HD" source available today via cable or streaming is very, very compressed... and Blu-Ray isn't necessarily a whole lot better (for economic reasons, most manufactured discs are single-layer, but studios feel like they need to shovel more and more stuff onto the discs, and it's usually the main video stream whose bitrate suffers the most). HOWEVER, even if it's horrifically overcompressed, "4k" UHD is STILL basically 2k HD with 4x oversampling, which means you can compress the hell out of a 4K UHD stream and STILL have it be a net improvement over what you WOULD have gotten from a real-world 720p60 or 1080i60 "2k" stream.
There's one small fly in the ointment now -- HDMI and HDCP. HDMI, because it allows the video source to ask what resolutions the TV supports instead of allowing you to forcibly set them with DIP switches or an explicit menu option. HDCP, because it means it's now a lot har
^-- Ugh. I hate posting long replies with my phone. That should read, "The area of an 8k screen where one person can perceive the added detail MIGHT BE SMALL,"
The area of an 8k screen where one person can perceive the added detail, but in the real world, you probably aren't the ONLY person watching the TV, nor is everyone necessarily watching the exact same part of the screen. The only solution is to maintain the highest meaningful resolution & achievable framerate needed to satisfy the most demanding aspects of foveal & peripheral vision across the entire scene, so it will look equally convincing REGARDLESS of which part somebody is looking at.
It's like the headphones-vs-10.2-surround debate... YOU only have 2 ears, but you aren't bolted to a single "sweet spot" in a room with flawless acoustics. Using lots of speakers in different locations enables more people in different spots to enjoy the kind of ambient surround sound that would otherwise be impossible without headphones.
It's the same with framerate. There are people who think 60fps is the limit of human motion perception. They're wrong. The thing is, for one framerate to give you that, "ooooh, smoother!" feeling compared to another, you basically need to double it.
* 100fps is unambiguously smoother than 50fps, and is still a noticeable improvement over 60fps.
* 120fps is visually-indistinguishable from 100fps **with native 120fps & 100fps content**, but 60fps content looks better at 120fps than 100fps because it's a whole multiple. Likewise, 50fps content looks smoother at 100fps than 120fps.
* For lower-framerate content to look smooth at a higher framerate that's a non-whole multiple, you'd REALLY want it to be at least 4x faster (eg, 50fps at 240fps is almost indistinguishable from 50fps at 250fps, but 50fps at 150fps looks a bit smoother than 50fps at 180fps).
* Your peripheral vision is more sensitive to framerate & flicker than your foveal vision, especially when you have high contrast (like a moving sharp white dot on a black background). This is why old computer monitors often appeared to flicker if you were reading a book in your lap... you were seeing a bright flickering white are with your peripheral vision that didn't bother you (much) when looked at directly (a least, at 72hz or 85hz... 50 & 60 progressive-scan always looked flicker-y to me even back then, and the perceived flicker from close-up interlaced video (think: 640x400 Amiga, pre-FlickerFixer) was almost painful to look at).
* With current technology, 600fps is approximately the point where diminishing returns overwhelms the improvement. To do visibly better than 600fps, you have to exceed 1000fps.
* There's still a problem, though... "The Uncanny Valley" exists in video, too. Somewhere around 300fps, video starts to seem "odd". Why? At 300fps, your foveal vision is totally happy, and your peripheral vision is "generally" content... but your brain still notices little things wrong, and interprets their existence as "danger". At low (under 150fps) framerates, there are so many things your brain sees as 'wrong', it just goes into 'this isn't real' mode & ignores them. At currently-achievable higher framerates, you fall off the "almost, but not quite, hyper-realistically-perfect" cliff into the Uncanny Valley.
"Soap opera effect" is another manifestation of the Uncanny Valley. The problem isn't the smoothness or framerate... the problem comes from all the tiny details that are slightly wrong when the video processor tries to guess what a synthetic tween'ed frame should look like. Your brain subconsciously notices, and screams "Danger! Things are not as they appear to be!"
2. There are things that Walmart just plain doesn't sell. At least, not at their retail stores.
If you're buying something like an XBox One bundle, the console you'll get at Walmart is no different from the console you'll get from anyone else, so there's no reason to favor or avoid Walmart if you're buying one.
If you want a Denon or Yamaha home theater amp, you aren't going to be getting it from your local Walmart store AT ANY PRICE, because Walmart doesn't stock expensive niche items in retail stores at all, and is rarely price-competitive with Amazon when they have them online.
Likewise, the products sold at Walmart aren't necessarily identical in quality to seemingly-identical products sold elsewhere. For example, a TV manufacturer might have a model -- let's call it the "QZX65Y" -- that everyone at avsforum.com worships, and has 99% 5-star reviews everywhere from people who think it's the greatest TV, ever. That TV might wholesale for $520, and sell for $599 at stores like Best Buy. Now, let's scrutinize its seemingly-identical cousin, the "QZX65Y3" (sold only by Walmart). See, Walmart balked at the $520 wholesale price, and threatened to walk unless the manufacturer sold it to them for $507. But Walmart doesn't really care if their model is literally 100% identical, as long as it has the same advertised specs. So, the manufacturer grudgingly agrees to Walmart's lower wholesale price, then manufactures Walmart's run of TVs using cheaper, lower-binned backlight LEDs, and gives it a slightly different model number to avoid harming all the 5-star reviews the "real" model has gotten. Then Walmart advertises it for $589. On paper, the two look the same -- especially to a clueless consumer who doesn't even know what to look for. But if you put the two models side by side displaying uniform 50% gray, the "good" model might display a field of relatively consistent gray, while the "Walmart" model might have patches with a blue or yellow twinge, or slightly different brightness (because low-binned LEDs aren't guaranteed to be as consistent as high-binned LEDs).
Sadly, the same kind of "quality fade" can happen if, a year later, you go to some regional electronics chain (like Brandsmart in Florida) who sells mostly "last year's models, at rock-bottom prices". For the last manufacturing run or two of a given model, the manufacturer might just switch to low-binned LEDs so they can reduce the wholesale price and scrape up a few more buyers. The "good" stores won't buy the cost-reduced models anyway (they're waiting for next year's models), and by that point the manufacturer has piled up enough 5-star reviews for ANYONE who bitches about getting a late-version model with inferior low-binned LEDs to look like a wacky outlier. Companies like Linksys and Netgear do this so often, they've practically turned quality-fade into a performance art.
It happens. All the time. Every single manufacturer with models specific to Walmart does this. And it's not just TVs... a Lawn Boy or Toro lawnmower from Walmart might LOOK like the lawnmowers sold by more expensive stores, but if you scrutinize the model numbers, you'll see that they differ slightly. And if you tore down two seemingly-identical models (with slightly different model numbers) side by side, you'll find at least a few insidious differences, GUARANTEED. Maybe it's something as simple as polishing off the injection-molding burrs from the starter handle on the "good" model, while leaving them on the "Walmart" model. Or using stainless-steel screws on the "good" model, and cheaper galvanized screws on the "Walmart" model. Regardless, the differences are real, and they're there if you look hard enough for them.
One big problem SO has is reconciling old questions with "best" answers that might no longer be the best -- or even still RELEVANT.
Suppose that someone posted a message to SO in 2012 asking how to hide the mouse pointer arrow that appears if the user connects a bluetooth or USB mouse to the device when their app (say, an OpenGL ES game) is in the foreground.
Five years ago, the correct and concise answer would have been, "You can't".
Today, the proper answer would be, "You can't do it unless the device has Android N (7.0) or newer AND you target API 24+, in which case here's what you need to do..."
The problem is, if someone posted a new question like this TODAY, some mod looking for easy points would likely flag it within minutes for closure-as-duplicate, EVEN IF the older question's only answer is "you can't". Often, the mods who are the most aggressive about flagging for closure-as-duplicate aren't even subject matter experts in the platform whose questions they're moderating... they're just looking to score easy points for their "Google" skills, and don't BOTHER to actually read anything beyond the search result summaries (let alone consider whether the original questions' answers are still relevant).
This happened on a grand scale for MONTHS after Android Studio moved to Gradle-based projects. People would post questions asking how to do something specific in a Gradle-based Android Studio 2.x project, then get their question swatted down almost instantly (as "duplicate-of-older-question", usually some question about Eclipse or Android Studio 1.0) by users with high reputations, but no discernible expertise in ANDROID development (because the real Android experts KNEW that Gradle was a HUGE change that broke things up, down, left, right, and diagonally).
Even BETTER would be a Thinkpad with 12.9" 4:3 2732x2048 display (like Apple's largest iPad Pro).
Or maybe a 17-18" high-ppi widescreen that can be configured to pretend it's a side-by-side 3:4 portrait & 4:3 landscape display (aspect ratios approximate). Why "pretend"? Because Windows puts two smaller monitors to better, more-efficient use than a single widescreen having identical total resolution.
There's also the Model M13, which is basically a M (buckling springs & all) with Trackpoint. The catch is, the left & right mouse buttons (below the spacebar) are badly-designed and wear out after 5-10 years. I have two M13 keyboards in a closet with worn-out buttons. I keep hoping someone will eventually make replacement buttons for it.
Dell's Precision laptops are equally-repairable. They're practically SFF desktop workstations in laptop form.
Upgrades I've done over the years:
* added mSATA SSD
* replaced original 500-gig hybrid drive with 2TB drive, and configured mSATA drive w/64-gig cache partition for new HD.
* added 2x4 gigs of RAM
* upgraded the original non-Intel mPCIe wifi to a better (MIMO-capable) 802.11ac Intel card (the original card left one of the factory-installed antennas unused)
* replaced the original Quadro K1100m video card with thirdparty Quadro K2100m card. This one was a nail-biting gamble because there's no real standard for screw & heatsink layout, but it worked.
The only potential glitch I noted with the new video card is due to the way Dell implemented switchable graphics... to use Quadro graphics on the laptop's own panel, you HAVE to enable IntelHD graphics (basically, the internal panel is wired to IntelHD, which "passes through" the Quadro graphics).
If you don't disable switchable graphics in BIOS setup, Windows 10 stupidly tries to use IntelHD for EVERYTHING unless you explicitly configure specific apps to use Quadro graphics... even games. Since I normally run docked with two external monitors, I like to disable switchable graphics so Windows won't insidiously use IntelHD (I played WoW for months before discovering that it was using IntelHD... running it with Quadro graphics vastly improved things like reflections & antialiasing).
With the original card, it would temporarily re-enable IntelHD switchable graphics during POST & BIOS setup to ensure you always had working video at powerup... with the new card, I'm fucked if I go on a trip & forget to re-enable switchable graphics before I undock until I can hijack a TV or monitor w/HDMI input to use long enough to re-enable IntelHD (I now carry a 1.5-meter HDMI cable everywhere when traveling).
The 17" model (m6800) actually has a second 2.5" drive bay (I have a 15" m4800), and both the 15" and 17" models allow you to swap out the internal optical drive for an additional 2.5" HD. Personally, I think there could be a market for someone to make a thin, larger-diameter drive like the old Quantum Bigfoot, if they made them with a form-factor identical to the de-facto standard for laptop optical drives. IMHO, a 4-8 terabyte second internal HD would be more useful to many users than an internal optical drive, and using a wide optical bay for a 2.5" drive just seems like a waste. With how much empty space many new laptops have inside, I'm surprised the industry never came up with a standard for thin 3.5" hard drives with the same interface as 2.5" laptop drives.
I live in Florida. My house sits on several feet of crushed limestone, with quite a few underground globs of waste cement (presumably whatever the construction workers dumped at the end of the day whenever they were doing concrete work). I dug up a bunch of the limestone boulders and concrete globs earlier this year while redoing my front yard's landscaping. After ~35 years underground, exposed to groundwater, plant roots, and everything else, the two can be pretty damn hard to tell apart... they've both been eroded by water, and many of the concrete globs have limestone with visible fossils embedded in them (presumably because the concrete was dumped onto a pile of rocks). In most cases, it's not at all obvious where the "real" rock ends and the hardened concrete glob begins.
It's also noteworthy that the researchers pointed out that the concrete they believe the ancient Egyptians made would NOT have been like modern concrete -- it would have looked more like "natural" stone, because it would have contained much larger chunks of natural stone than modern concrete aggregate.
As for lasting 4,000 years, remember... it would have been more like Roman concrete than modern portland cement-based concrete. Modern reinforced concrete depends upon embedded steel for tensile strength. The steel rusts, and the concrete eventually comes crashing down because it can't support its own weight without it. The Romans used concrete the same way they used cut stone (arches, vaults, etc). If the Romans had built hundred-foot spans of steel-reinforced concrete, they would have probably fallen apart long before the empire itself did.
At the moment, both possibilities are still very much topics ripe for further research. Pretty much everyone agrees that at least some of the blocks were quarried, and there seems to be pretty good evidence that at least some of the blocks were cast on site.
Is the theory a bit of a reach? Sure. So would have been telling a group of paleontologists in the 1950s that birds aren't distant just DESCENDANTS of dinosaurs... birds ARE dinosaurs. And T-rex had feathers.
> Except it is known the blocks used are quarried limestone and granite, not concrete.
No, they've always been ASSUMED to have been quarried limestone and granite. About 10 years ago, someone analyzed a chunk of "stone" from one of the pyramids & discovered the same kind of bubbles you'd find in manmade cast stone.
The conclusion of the above: the pyramids are a combination of cast and quarried stone... basically the lower stones were quarried, and the upper stones were cast... basically, they used quarried stone up to the point where it became more difficult/expensive to transport and lift the blocks into place, and used cast stone for the rest (because cast stone would have been too expensive to use for everything, so they only used it where they HAD to).
Personally, I think the most obvious theory is likely to be the closest to the correct explanation: that the blocks WEREN'T quarried, but are some form of manmade cast stone made from ancient concrete.
If they were cast instead of directly quarried, the builders could have just built the whole thing like a modern freeway embankment... build the perimeter, backfill the inside with sifted & graded crushed rock & sand. Maybe put down an occasional layer of cloth to stabilize it horizontally (knowing the cloth will eventually decay, but only really NEEDING it for stabilization during construction). Cast the next row of stones, move them horizontally into place, and backfill the interior up to the next level. Stir, rinse, and repeat until you're done. Modern retained-earth construction obviously goes quite a bit further, (like using steel cables to pull the retaining walls inward so they can be vertical instead of sloped, and using precast wall segments instead of casting them on-site), but the basic idea is the same.
Moving big, heavy things HORIZONTALLY is fairly straightforward. So is moving crushed-rock cementious slurry up a hill in small buckets. If they're cast stone, the pyramids' construction basically just becomes a matter of having lots of money, immense HR management resources, and good supply-chain management.
From what I've read, Egypt's antiquities ministry is part of the reason why relatively little is known about the "nuts and bolts" construction details of the pyramids. It WANTS to preserve the aura of mystery, because the official narrative drives tourism and brings enormous amounts of money into Egypt. From their point of view, the absolute WORST thing that could happen is if someone were to demonstrate that the pyramids were no more special than a random freeway embankment.
The problem is, in real-world use, so-called "HD Radio" sounds like over-compressed total SHIT compared to even mediocre analog FM. In terms of audio fidelity, digital radio is a HUGE step backwards. To me, at least, everything on "HD Radio" sounds like a bit-starved mp3 file (no channel separation, no sense of "spatiality", general "dead" sound... the higher-order audio artifacts that stick out like a sore thumb once you know what to listen for).
The problem with TELEVISION in American phones & tablets isn't power, it's the fact that ATSC 1.x is damn-near IMPOSSIBLE to tune at all in a moving vehicle (somehow, the ATSC committee totally forgot about doppler-shift as a "something to care about" when evaluating 8VSB), and even the doppler-shift induced by an antenna swaying in hurricane-force winds is enough to render ATSC 1.x channels unwatchable.
Continuing on the "power" theme, I think you'll find that right now, people in Puerto Rico fall into basically two categories:
* People who don't own a generator, and for whom watching TV is probably the least of their real-world concerns right now.
* People who own a generator. If the generator is running, the amount of extra power required by a modern LCD TV (or a wireless access point, HD-Homerun, and one or more phones or tablets running Silicon Dust's TV app) is minimal. If the generator isn't running, they're temporarily screwed anyway & probably too miserable in the heat & humidity to care about TV tonight.
As far as I'm concerned, solar power doesn't even matter in this equation. Compared to a cheap gas generator, solar power is EXPENSIVE. Solar is for people who already own a generator, and have enough money to invest in a last-ditch backup power source to use when the generator isn't available.
On the other hand, Ku-band satellite internet isn't all that hard for someone who knows how to aim the dish to set up, and is DIRT CHEAP by comparison.
Ku band satellite is only expensive if it has to be "portable", in the sense of "drive around in a van & have it working in 5 minutes". If you set the bar a lot lower (3-4 hour setup time, dish mounted to pole in 5-gallon Home Depot bucket half-filled with concrete), it's really cheap.
Scenario 1: ham spends 16 hours & relays a few hundred "We're alive" messages.
Scenario 2: ham bolts a VSAT satellite dish onto a pole w/southern view, fires up a small generator, aims the dish, configures the hughes.net (or WildBlue, or some other service) satellite modem, sets up an 802.11ac access point, then tells 2,000 people "it's working!", so they can post THEIR OWN messages instead of relying on one ham to do everything.
IMHO, it's a no-brainer which scenario would put the ham's skills to best use & benefit the most people directly (scenario #2, obviously).
And the gear isn't even all that expensive... maybe $3k/site, if FEMA or the American Red Cross gets totally raped & (over-)pays full list price for everything. More like $2k/site if they even TRIED to get a good deal. Add maybe $500-2000/month per site for satellite internet service and generator gas, max. To FEMA or the Red Cross, this is literally pocket change... they probably spend more money transporting a single truckload of bottled water.
The problem with 1200-baud packet is that the protocol is "all or nothing", with NO forward error correction. It just keeps blindly retransmitting after pseudorandom delays until a packet manages to get through error-free.
As others have noted, packet has ALWAYS had major problems with scalability & dealing with congestion. It takes very little to saturate a packet radio link into unusability due to retransmissions & collisions.
We already have enormously better protocols, but the prohibition against using encryption on ham bands seriously limits the ability of hams to build emergency adhoc internet backhaul (to places that DO have working internet connectivity) that others can seamlessly use directly.
Puerto Rican hams HAVE put together some mesh networks over the past week, but they've generally done it in ways that don't require a ham license (ie, fixed point to point 802.11), because a mesh network operating legally within the constraints of FCC ham regulations would be mostly useless to the public for emergency internet access since they'd have to block https & anything else that involves encrypted communication & risk exposing the real internet credentials of people to anyone running WireShark.
Update: apparently, if a Nexus 6P with N2G47W radio modem firmware is without working LTE, HSPA(+), and GPRS for several days, it stops even TRYING to use them until the next hard shutdown & full reboot.
I've done numerous "hot" reboots over the past two weeks (restarting Android, without going all the way and rebooting everything), toggled wi-fi off and on, and tried using T-mobile away from my home's wi-fi multiple times over the past 12 days, to no avail. Zero data.
For the hell of it, I decided to try totally shutting down the phone & starting it back up again. Sonofabitch, my phone now appears to have working T-mobile data service again.
I don't know who to be angrier at... T-mo for being down for days when other carriers weren't, Qualcomm for allowing their radio modem to just give up even TRYING to connect to data, or Huawei for making an Android 7.1.2 rom that allows Qualcomm's radio modem to throw in the towel and give up without so much as a "hey, your phone has totally given up trying to find working data & won't bother to try again until your next complete shutdown" notification (or a, "hey, your phone's radio modem appears to have crashed... do a full shutdown to try and resolve it." notification).
Moral: if your phone's data is nonworking for a long time, it might STAY nonworking for a long time after the underlying problem itself gets fixed. Ugh.
> The reputation is unfair: T-Mobile is superb right now
Unless you're in South Florida, where a big chunk of Dade & Broward counties STILL have no working T-mobile data service almost TWO WEEKS post-Irma (Verizon and AT&T had few outages that were mostly resolved within a day or two). Somehow, they're STILL having backhaul problems for data service.
I mean, fuck, even SPRINT had data service mostly restored a few days ago. T-Mo must have somehow found shittiest, least-resilient south Broward backhaul provider in the history of mobile phone service.
Disclaimer: current T-mo customer who's extremely likely to switch to Verizon this afternoon if T-mobile doesn't get my goddamn data working before then (I found out last night that a Nexus 6P *is* Verizon-compatible, eliminating the last thing that kept me from switching).
A lithium battery has a finite lifespan. It's guaranteed to eventually need replacement, and unless Apple is literally molding the lithium gel around the circuit board in a way that makes its replacement physically impossible, they should be required to sell replacement batteries at a fair price (or if they don't want to, then they shouldn't be allowed to prevent anyone ELSE from making compatible replacements).
A mechanical button or poorly-attached microUSB likewise has a finite life... a life that might very well be less than the reasonably-expected life of the phone.
Screens break and crack. It's just something that happens to glass when dropped.
A good step forward would be for the FTC to require manufactures to explicitly disclose the repair cost & minimum availability for key components at purchase. Ex:
Battery: 1 year free warranty, guaranteed availability for a maximum charge of $n until yyyy/MM/dd(*)
Screen: 1 year free warranty for defects, guaranteed availability for a maximum charge of $x until yyyy/MM/dd, and $y until YyYy/Mm/Dd
Electrolytic Capacitors: 10 year warranty, or replacement with newer model (not necessarily the newEST model) that's at least as good.
Buttons & connectors: 1 year warranty, 5 year guaranteed availability of repair with maximum charge of $k(*).
(*) And if they can't satisfy the repairability guarantee, they'll have to refund some fair fraction of the original purchase price based on age (say, 100% up to 18 months, 80% after 18-24 months, decreasing by another 10% per 6 months thereafter).
I'd even allow them to be assholes & enforce absurdly-short (or outrageously expensive) terms... as long as they were required to accurately & effectively communicate those terms to buyers at purchase, and really bend over backwards to make sure consumers know about terms the FTC deemed 'unreasonable'
concrete example: Consumer buys new iPhone 17 (or Samsung Galaxy S16, or Google VoxelQ). The manufacturer guarantees battery-availability only until 8/31/22, which is less than 60 months from the date of first sale. After opening the box, the phone is in another sealed box with prominent federally-mandated warning (in English & Spanish for US phones) that says something like, "Warning, the manufacturer of this phone has chosen to not guarantee the availability of replacement batteries after 8/31/22. The FTC has determined that 80% of batteries will have less than 50% of their original capacity after 8/31/21. By law, you have an absolute right to return this phone to the seller for 100% refund and full cancellation of all contractual obligations and shipping charges arising from its purchase without charge as long as this seal is not broken. This right can not be waived or limited, regardless of seller policy or conditions of sale."
ie, the Feds couldn't *stop* Apple (& others) from being assholes, but they could ensure that consumers KNOW Apple (& others) are being assholes & protect them from being victims unless they're absolutely HELLBENT on being victims. As long as the FTC's threshold for requiring the warning was reasonable enough for most vendors to avoid, its presence would be effective & would subject the mfr. to criticism & ridicule from magazines, reviewers, and probably late-night comedians. The key is to make sure it doesn't turn into a stupid, pointless warning, like "Warning: may contain peanuts" on a jar of peanuts or peanut butter.
Consumer protection laws with teeth can and do work. Just ask anyone who was given a hard time by underlings or junior staff about getting a warranty repair for a jailbroken iphone or rooted Android phone until they summoned the manager & invoked the magic phrase, "Magnuson-Moss Warranty Act"... at which point the manager, if he had any sense & valued his job, apologized *profusely* for the "misunderstanding" and *personally* made sure the phone got fixed, because anything else would risk a $50,000 per incident fine and mountain of subs
My point was, HDCP-encrypted HDMI inflates the bitrate to several gigabytes per second (for HD video, at least). You just plain *can't* sustain capturing of THAT MUCH bulk data to persistent storage (SSD, network, whatever) without some seriously exotic hardware. It's like trying to fill martini glasses with a firehose (without spilling any bits).
The size-inflation that makes the bitstream impossibly huge to capture for offline decryption is a major element *of* HDCP.
The only reason why HDMI-over-cat5e extenders for HD video are even POSSIBLE is because they decrypt the HDCP-inflated HDMI, trivially compress it down to something sane(*), re-encrypt it using an ephemeral AES-128 key that changes frequently(**), and transmit THAT... then decrypt the AES, re-inflate it, re-encrypt it with HDCP, and output it via HDMI at the other end.
(*) HDCP takes something like an 8-bit sample & pads it with additional random bits to make the number that gets encrypted bigger. The receiver knows the bottom N-8 bits of each sample are junk, so it just lops them off and ignores them.
(**) At least, this (or something comparably-robust) is what HDCP licensing officially requires. Quite a few of the single-cable HDMI extenders from China just strip the HDCP and leave it as unencrypted DVI. Or some later ones just re-pad the samples with '0' bits prior to HDCP encryption, which makes bruteforcing the key a lot easier (this attack vector became moot once the HDCP master key itself was leaked).
Now that the master key is known, implementing HDCP decryption in hardware (like a FPGA) is borderline-trivial. The hardest part is getting a pair of hdmi ports soldered to the FPGA, which is why the HDMI people used to aggressively go after anyone selling $5 HDMI breakout boards or cheap FPGA dev boards with two HDMI ports... it's technically prohibited by HDMI licensing, and made it harder for freshman EE students to build a DIY hdcp stripper as an afternoon project. Now that the industry is focused on HDCP 2.2, they seem to have mostly given up on trying to swat down everyone on eBay since it was always a hopeless game of 'whack-a-mole' anyway. They could fight against $100 dev boards with two ports because an injunction against (or customs-seizure of) a thousand would have been a crushing loss for a seller, but $5 boards that cost 20 cents to make are another battle entirely.
Sure it is. With h.264 and AES implemented in silicon and the keys known only to the video chip, it's fairly straightforward for the chip to decompress a h.264 video stream, encrypt it with AES, and write it to the frame buffer... then at output time, read it back from ram, decrypt it, re-encrypt it with HDCP, and output it to the display. The CPU itself never has direct access to unencrypted video frames.
Yes, HDCP's master key has been compromised, but decrypting HDCP on the fly still requires dedicated hardware (ASIC or FPGA). Decrypting it in realtime using CPU PIO and bus-snooping just isn't practical, and HDCP intentionally inflates the bitrate to make it impractical to capture the raw encrypted bitstream for offline decryption... not even a SSD can sustain writing THAT MUCH data for extended amounts of time without either falling behind or running out of drive space.
Slashdot Mirror
And let's not forget the retarded "pinch to rotate" gesture. I don't think I've EVER met anyone, in the ENTIRE HISTORY of Google Maps for Android, who's INTENTIONALLY rotated a fucking map after the first time or two they've used it. It's one of those features that was cool to show off for 5 minutes, but actively harms the app's ultimate usability. I'm one of those people who've wished for YEARS that Google would add a preferences option to Google Maps for "disable rotate gestures"
> US East coast would be "book of Eli", at best. US West coast will probably not exist.
> any human society larger than a small town on the planet will be a miracle
Not really. Most estimates I've seen show South Florida left relatively unscathed. By ashfall, at least. SoFla's main problem would be dealing with an almost-overnight influx of ~40 million new residents (path of least resistance for people fleeing the eruption from elsewhere... in the US, Florida is theoretically reachable even on foot within a few weeks from most parts of the eastern US).
> If it had been a phone designed to be a good phone, it could have succeeded.
And more importantly, if it had been at LEAST as open (non-walled-garden) as Windows Mobile used to be. But no, Microsoft wanted to be Apple & exercise total control (and collect a share of the revenue) for everything users could run. Windows Phone had zero appeal to "Apple" type users, and even LESS appeal to Android users (many of whom WERE Windows Mobile users prior to 2009 or 2010).
Windows Mobile is a case study in corprate dementia & self-inflicted failure. WM6.5 was ugly, but the core OS functionality was good, and its abandonment set the mobile industry behind for at LEAST 2-3 years with respect to things like Bluetooth and GPS (Android Bluetooth was pretty dysfunctional until Eclair, and Android GPS in general was a nuclear-hot mess until sometime around Gingerbread or ICS).
There's one point many here have overlooked... it doesn't necessarily require a "UHD" TV to directly benefit from it.
Back around 1994, I bought my first "big" TV back when I was in college -- a 27" Daewoo. Nothing earthshaking, but it had higher-end specs... s-video input, decent stereo audio, and significantly better dot pitch than most of the other TVs at BrandsMart USA.
For the next 5 years, connected to a cable box and VCR, it mostly looked like shit. VHS tapes looked marginally less shitty at my house than at friends' houses, because I had a S-VHS VCR (mostly, so I could get S-Video output, even when watching non-SVHS tapes) -- they still had poor detail and awful color resolution, but at least the NTSC color artifacts were slightly better.
At one point, I contemplated buying a LaserDisc player. Unfortunately, by that time, LaserDisc was basically dead as a format in the US. In any case, a LaserDisc definitely wasn't something you could waltz into Blockbuster Video and rent for $5.
Then DVD arrived. Instantly, the TV's perceived video quality more than doubled. The TV's "real" physical resolution was somewhere around 360x480 under ideal conditions... but having a 720x480 video source to play with meant I could pump ever bit of signal detail into that TV that it was physically capable of handling. The improvement was dramatic.
The early 2000s arrived. My housemate bought a DLP HDTV, and we decided to switch to Voom! so we could enjoy "real" HDTV (at the time, Comcast in Miami hadn't started carrying HD channels yet, and only one or two of the local broadcasters were transmitting HD). As luck would have it, Voom! didn't screw around with "SD" boxes -- every box they gave you was capable of outputting anything from 480i60 over s-video (or composite, or component) to 720p60 and 1080i60 HD (over component or DVI). By this point, the old CRT TV from my college days was now my bedroom TV. And I quickly realized that even though it wasn't HD, Voom's channels (and OTA HDTV from the antenna diplexed into the LNB's output) looked MUCH better than anything I ever saw with cable.
Fast forward a bit more. Voom! went out of business, and my housemates and I were forced to settle for Comcast. God, it was awful. For a couple of days (before Voom shut down, after Comcast came to set us up), I was able to compare different channels from Voom and Comcast side by side. For SD channels like MTV, it was literally a night and day difference. On Voom!, MTV and MTV2 looked like DVD-quality video. On Comcast, I could barely watch it without feeling like my eyes were going to bleed from the blocky overcompression.
A couple of years later, I bought my own house, got a nice 65" DLP TV of my own, and signed up for DirecTV. I still had the same TV (from college) in the master bedroom, and decided to pay the extra $2/month to get a HD box for the master bedroom. Happy days again... DVD-quality, even from SD channels. Ditto, after switching to U-verse a couple of years later (a slight step down for HD, but more or less the same quality for 480i60 via S-video).
My point: most of the (2k, 720p60 or 1080i60) "HD" source available today via cable or streaming is very, very compressed... and Blu-Ray isn't necessarily a whole lot better (for economic reasons, most manufactured discs are single-layer, but studios feel like they need to shovel more and more stuff onto the discs, and it's usually the main video stream whose bitrate suffers the most). HOWEVER, even if it's horrifically overcompressed, "4k" UHD is STILL basically 2k HD with 4x oversampling, which means you can compress the hell out of a 4K UHD stream and STILL have it be a net improvement over what you WOULD have gotten from a real-world 720p60 or 1080i60 "2k" stream.
There's one small fly in the ointment now -- HDMI and HDCP. HDMI, because it allows the video source to ask what resolutions the TV supports instead of allowing you to forcibly set them with DIP switches or an explicit menu option. HDCP, because it means it's now a lot har
^-- Ugh. I hate posting long replies with my phone. That should read, "The area of an 8k screen where one person can perceive the added detail MIGHT BE SMALL,"
The area of an 8k screen where one person can perceive the added detail, but in the real world, you probably aren't the ONLY person watching the TV, nor is everyone necessarily watching the exact same part of the screen. The only solution is to maintain the highest meaningful resolution & achievable framerate needed to satisfy the most demanding aspects of foveal & peripheral vision across the entire scene, so it will look equally convincing REGARDLESS of which part somebody is looking at.
It's like the headphones-vs-10.2-surround debate... YOU only have 2 ears, but you aren't bolted to a single "sweet spot" in a room with flawless acoustics. Using lots of speakers in different locations enables more people in different spots to enjoy the kind of ambient surround sound that would otherwise be impossible without headphones.
It's the same with framerate. There are people who think 60fps is the limit of human motion perception. They're wrong. The thing is, for one framerate to give you that, "ooooh, smoother!" feeling compared to another, you basically need to double it.
* 100fps is unambiguously smoother than 50fps, and is still a noticeable improvement over 60fps.
* 120fps is visually-indistinguishable from 100fps **with native 120fps & 100fps content**, but 60fps content looks better at 120fps than 100fps because it's a whole multiple. Likewise, 50fps content looks smoother at 100fps than 120fps.
* For lower-framerate content to look smooth at a higher framerate that's a non-whole multiple, you'd REALLY want it to be at least 4x faster (eg, 50fps at 240fps is almost indistinguishable from 50fps at 250fps, but 50fps at 150fps looks a bit smoother than 50fps at 180fps).
* Your peripheral vision is more sensitive to framerate & flicker than your foveal vision, especially when you have high contrast (like a moving sharp white dot on a black background). This is why old computer monitors often appeared to flicker if you were reading a book in your lap... you were seeing a bright flickering white are with your peripheral vision that didn't bother you (much) when looked at directly (a least, at 72hz or 85hz... 50 & 60 progressive-scan always looked flicker-y to me even back then, and the perceived flicker from close-up interlaced video (think: 640x400 Amiga, pre-FlickerFixer) was almost painful to look at).
* With current technology, 600fps is approximately the point where diminishing returns overwhelms the improvement. To do visibly better than 600fps, you have to exceed 1000fps.
* There's still a problem, though... "The Uncanny Valley" exists in video, too. Somewhere around 300fps, video starts to seem "odd". Why? At 300fps, your foveal vision is totally happy, and your peripheral vision is "generally" content... but your brain still notices little things wrong, and interprets their existence as "danger". At low (under 150fps) framerates, there are so many things your brain sees as 'wrong', it just goes into 'this isn't real' mode & ignores them. At currently-achievable higher framerates, you fall off the "almost, but not quite, hyper-realistically-perfect" cliff into the Uncanny Valley.
"Soap opera effect" is another manifestation of the Uncanny Valley. The problem isn't the smoothness or framerate... the problem comes from all the tiny details that are slightly wrong when the video processor tries to guess what a synthetic tween'ed frame should look like. Your brain subconsciously notices, and screams "Danger! Things are not as they appear to be!"
1. Walmart isn't always the cheapest.
2. There are things that Walmart just plain doesn't sell. At least, not at their retail stores.
If you're buying something like an XBox One bundle, the console you'll get at Walmart is no different from the console you'll get from anyone else, so there's no reason to favor or avoid Walmart if you're buying one.
If you want a Denon or Yamaha home theater amp, you aren't going to be getting it from your local Walmart store AT ANY PRICE, because Walmart doesn't stock expensive niche items in retail stores at all, and is rarely price-competitive with Amazon when they have them online.
Likewise, the products sold at Walmart aren't necessarily identical in quality to seemingly-identical products sold elsewhere. For example, a TV manufacturer might have a model -- let's call it the "QZX65Y" -- that everyone at avsforum.com worships, and has 99% 5-star reviews everywhere from people who think it's the greatest TV, ever. That TV might wholesale for $520, and sell for $599 at stores like Best Buy. Now, let's scrutinize its seemingly-identical cousin, the "QZX65Y3" (sold only by Walmart). See, Walmart balked at the $520 wholesale price, and threatened to walk unless the manufacturer sold it to them for $507. But Walmart doesn't really care if their model is literally 100% identical, as long as it has the same advertised specs. So, the manufacturer grudgingly agrees to Walmart's lower wholesale price, then manufactures Walmart's run of TVs using cheaper, lower-binned backlight LEDs, and gives it a slightly different model number to avoid harming all the 5-star reviews the "real" model has gotten. Then Walmart advertises it for $589. On paper, the two look the same -- especially to a clueless consumer who doesn't even know what to look for. But if you put the two models side by side displaying uniform 50% gray, the "good" model might display a field of relatively consistent gray, while the "Walmart" model might have patches with a blue or yellow twinge, or slightly different brightness (because low-binned LEDs aren't guaranteed to be as consistent as high-binned LEDs).
Sadly, the same kind of "quality fade" can happen if, a year later, you go to some regional electronics chain (like Brandsmart in Florida) who sells mostly "last year's models, at rock-bottom prices". For the last manufacturing run or two of a given model, the manufacturer might just switch to low-binned LEDs so they can reduce the wholesale price and scrape up a few more buyers. The "good" stores won't buy the cost-reduced models anyway (they're waiting for next year's models), and by that point the manufacturer has piled up enough 5-star reviews for ANYONE who bitches about getting a late-version model with inferior low-binned LEDs to look like a wacky outlier. Companies like Linksys and Netgear do this so often, they've practically turned quality-fade into a performance art.
It happens. All the time. Every single manufacturer with models specific to Walmart does this. And it's not just TVs... a Lawn Boy or Toro lawnmower from Walmart might LOOK like the lawnmowers sold by more expensive stores, but if you scrutinize the model numbers, you'll see that they differ slightly. And if you tore down two seemingly-identical models (with slightly different model numbers) side by side, you'll find at least a few insidious differences, GUARANTEED. Maybe it's something as simple as polishing off the injection-molding burrs from the starter handle on the "good" model, while leaving them on the "Walmart" model. Or using stainless-steel screws on the "good" model, and cheaper galvanized screws on the "Walmart" model. Regardless, the differences are real, and they're there if you look hard enough for them.
One big problem SO has is reconciling old questions with "best" answers that might no longer be the best -- or even still RELEVANT.
Suppose that someone posted a message to SO in 2012 asking how to hide the mouse pointer arrow that appears if the user connects a bluetooth or USB mouse to the device when their app (say, an OpenGL ES game) is in the foreground.
Five years ago, the correct and concise answer would have been, "You can't".
Today, the proper answer would be, "You can't do it unless the device has Android N (7.0) or newer AND you target API 24+, in which case here's what you need to do..."
The problem is, if someone posted a new question like this TODAY, some mod looking for easy points would likely flag it within minutes for closure-as-duplicate, EVEN IF the older question's only answer is "you can't". Often, the mods who are the most aggressive about flagging for closure-as-duplicate aren't even subject matter experts in the platform whose questions they're moderating... they're just looking to score easy points for their "Google" skills, and don't BOTHER to actually read anything beyond the search result summaries (let alone consider whether the original questions' answers are still relevant).
This happened on a grand scale for MONTHS after Android Studio moved to Gradle-based projects. People would post questions asking how to do something specific in a Gradle-based Android Studio 2.x project, then get their question swatted down almost instantly (as "duplicate-of-older-question", usually some question about Eclipse or Android Studio 1.0) by users with high reputations, but no discernible expertise in ANDROID development (because the real Android experts KNEW that Gradle was a HUGE change that broke things up, down, left, right, and diagonally).
Even BETTER would be a Thinkpad with 12.9" 4:3 2732x2048 display (like Apple's largest iPad Pro).
Or maybe a 17-18" high-ppi widescreen that can be configured to pretend it's a side-by-side 3:4 portrait & 4:3 landscape display (aspect ratios approximate). Why "pretend"? Because Windows puts two smaller monitors to better, more-efficient use than a single widescreen having identical total resolution.
There's also the Model M13, which is basically a M (buckling springs & all) with Trackpoint. The catch is, the left & right mouse buttons (below the spacebar) are badly-designed and wear out after 5-10 years. I have two M13 keyboards in a closet with worn-out buttons. I keep hoping someone will eventually make replacement buttons for it.
Dell's Precision laptops are equally-repairable. They're practically SFF desktop workstations in laptop form.
Upgrades I've done over the years:
* added mSATA SSD
* replaced original 500-gig hybrid drive with 2TB drive, and configured mSATA drive w/64-gig cache partition for new HD.
* added 2x4 gigs of RAM
* upgraded the original non-Intel mPCIe wifi to a better (MIMO-capable) 802.11ac Intel card (the original card left one of the factory-installed antennas unused)
* replaced the original Quadro K1100m video card with thirdparty Quadro K2100m card. This one was a nail-biting gamble because there's no real standard for screw & heatsink layout, but it worked.
The only potential glitch I noted with the new video card is due to the way Dell implemented switchable graphics... to use Quadro graphics on the laptop's own panel, you HAVE to enable IntelHD graphics (basically, the internal panel is wired to IntelHD, which "passes through" the Quadro graphics).
If you don't disable switchable graphics in BIOS setup, Windows 10 stupidly tries to use IntelHD for EVERYTHING unless you explicitly configure specific apps to use Quadro graphics... even games. Since I normally run docked with two external monitors, I like to disable switchable graphics so Windows won't insidiously use IntelHD (I played WoW for months before discovering that it was using IntelHD... running it with Quadro graphics vastly improved things like reflections & antialiasing).
With the original card, it would temporarily re-enable IntelHD switchable graphics during POST & BIOS setup to ensure you always had working video at powerup... with the new card, I'm fucked if I go on a trip & forget to re-enable switchable graphics before I undock until I can hijack a TV or monitor w/HDMI input to use long enough to re-enable IntelHD (I now carry a 1.5-meter HDMI cable everywhere when traveling).
The 17" model (m6800) actually has a second 2.5" drive bay (I have a 15" m4800), and both the 15" and 17" models allow you to swap out the internal optical drive for an additional 2.5" HD. Personally, I think there could be a market for someone to make a thin, larger-diameter drive like the old Quantum Bigfoot, if they made them with a form-factor identical to the de-facto standard for laptop optical drives. IMHO, a 4-8 terabyte second internal HD would be more useful to many users than an internal optical drive, and using a wide optical bay for a 2.5" drive just seems like a waste. With how much empty space many new laptops have inside, I'm surprised the industry never came up with a standard for thin 3.5" hard drives with the same interface as 2.5" laptop drives.
The fundamental problem is that the hacking victims aren't Equifax's CUSTOMERS, they're Equifax's PRODUCT.
If you, as a consumer, get harmed by Equifax's negligence, they aren't going to care until regulators MAKE them care.
I live in Florida. My house sits on several feet of crushed limestone, with quite a few underground globs of waste cement (presumably whatever the construction workers dumped at the end of the day whenever they were doing concrete work). I dug up a bunch of the limestone boulders and concrete globs earlier this year while redoing my front yard's landscaping. After ~35 years underground, exposed to groundwater, plant roots, and everything else, the two can be pretty damn hard to tell apart... they've both been eroded by water, and many of the concrete globs have limestone with visible fossils embedded in them (presumably because the concrete was dumped onto a pile of rocks). In most cases, it's not at all obvious where the "real" rock ends and the hardened concrete glob begins.
It's also noteworthy that the researchers pointed out that the concrete they believe the ancient Egyptians made would NOT have been like modern concrete -- it would have looked more like "natural" stone, because it would have contained much larger chunks of natural stone than modern concrete aggregate.
As for lasting 4,000 years, remember... it would have been more like Roman concrete than modern portland cement-based concrete. Modern reinforced concrete depends upon embedded steel for tensile strength. The steel rusts, and the concrete eventually comes crashing down because it can't support its own weight without it. The Romans used concrete the same way they used cut stone (arches, vaults, etc). If the Romans had built hundred-foot spans of steel-reinforced concrete, they would have probably fallen apart long before the empire itself did.
At the moment, both possibilities are still very much topics ripe for further research. Pretty much everyone agrees that at least some of the blocks were quarried, and there seems to be pretty good evidence that at least some of the blocks were cast on site.
Is the theory a bit of a reach? Sure. So would have been telling a group of paleontologists in the 1950s that birds aren't distant just DESCENDANTS of dinosaurs... birds ARE dinosaurs. And T-rex had feathers.
> Except it is known the blocks used are quarried limestone and granite, not concrete.
No, they've always been ASSUMED to have been quarried limestone and granite. About 10 years ago, someone analyzed a chunk of "stone" from one of the pyramids & discovered the same kind of bubbles you'd find in manmade cast stone.
http://www.materials.drexel.ed...
The conclusion of the above: the pyramids are a combination of cast and quarried stone... basically the lower stones were quarried, and the upper stones were cast... basically, they used quarried stone up to the point where it became more difficult/expensive to transport and lift the blocks into place, and used cast stone for the rest (because cast stone would have been too expensive to use for everything, so they only used it where they HAD to).
Personally, I think the most obvious theory is likely to be the closest to the correct explanation: that the blocks WEREN'T quarried, but are some form of manmade cast stone made from ancient concrete.
If they were cast instead of directly quarried, the builders could have just built the whole thing like a modern freeway embankment... build the perimeter, backfill the inside with sifted & graded crushed rock & sand. Maybe put down an occasional layer of cloth to stabilize it horizontally (knowing the cloth will eventually decay, but only really NEEDING it for stabilization during construction). Cast the next row of stones, move them horizontally into place, and backfill the interior up to the next level. Stir, rinse, and repeat until you're done. Modern retained-earth construction obviously goes quite a bit further, (like using steel cables to pull the retaining walls inward so they can be vertical instead of sloped, and using precast wall segments instead of casting them on-site), but the basic idea is the same.
Moving big, heavy things HORIZONTALLY is fairly straightforward. So is moving crushed-rock cementious slurry up a hill in small buckets. If they're cast stone, the pyramids' construction basically just becomes a matter of having lots of money, immense HR management resources, and good supply-chain management.
From what I've read, Egypt's antiquities ministry is part of the reason why relatively little is known about the "nuts and bolts" construction details of the pyramids. It WANTS to preserve the aura of mystery, because the official narrative drives tourism and brings enormous amounts of money into Egypt. From their point of view, the absolute WORST thing that could happen is if someone were to demonstrate that the pyramids were no more special than a random freeway embankment.
The problem is, in real-world use, so-called "HD Radio" sounds like over-compressed total SHIT compared to even mediocre analog FM. In terms of audio fidelity, digital radio is a HUGE step backwards. To me, at least, everything on "HD Radio" sounds like a bit-starved mp3 file (no channel separation, no sense of "spatiality", general "dead" sound... the higher-order audio artifacts that stick out like a sore thumb once you know what to listen for).
The problem with TELEVISION in American phones & tablets isn't power, it's the fact that ATSC 1.x is damn-near IMPOSSIBLE to tune at all in a moving vehicle (somehow, the ATSC committee totally forgot about doppler-shift as a "something to care about" when evaluating 8VSB), and even the doppler-shift induced by an antenna swaying in hurricane-force winds is enough to render ATSC 1.x channels unwatchable.
Continuing on the "power" theme, I think you'll find that right now, people in Puerto Rico fall into basically two categories:
* People who don't own a generator, and for whom watching TV is probably the least of their real-world concerns right now.
* People who own a generator. If the generator is running, the amount of extra power required by a modern LCD TV (or a wireless access point, HD-Homerun, and one or more phones or tablets running Silicon Dust's TV app) is minimal. If the generator isn't running, they're temporarily screwed anyway & probably too miserable in the heat & humidity to care about TV tonight.
As far as I'm concerned, solar power doesn't even matter in this equation. Compared to a cheap gas generator, solar power is EXPENSIVE. Solar is for people who already own a generator, and have enough money to invest in a last-ditch backup power source to use when the generator isn't available.
Satellite PHONES are cost-prohibitive to use.
On the other hand, Ku-band satellite internet isn't all that hard for someone who knows how to aim the dish to set up, and is DIRT CHEAP by comparison.
Ku band satellite is only expensive if it has to be "portable", in the sense of "drive around in a van & have it working in 5 minutes". If you set the bar a lot lower (3-4 hour setup time, dish mounted to pole in 5-gallon Home Depot bucket half-filled with concrete), it's really cheap.
I agree totally.
Imagine two scenarios:
Scenario 1: ham spends 16 hours & relays a few hundred "We're alive" messages.
Scenario 2: ham bolts a VSAT satellite dish onto a pole w/southern view, fires up a small generator, aims the dish, configures the hughes.net (or WildBlue, or some other service) satellite modem, sets up an 802.11ac access point, then tells 2,000 people "it's working!", so they can post THEIR OWN messages instead of relying on one ham to do everything.
IMHO, it's a no-brainer which scenario would put the ham's skills to best use & benefit the most people directly (scenario #2, obviously).
And the gear isn't even all that expensive... maybe $3k/site, if FEMA or the American Red Cross gets totally raped & (over-)pays full list price for everything. More like $2k/site if they even TRIED to get a good deal. Add maybe $500-2000/month per site for satellite internet service and generator gas, max. To FEMA or the Red Cross, this is literally pocket change... they probably spend more money transporting a single truckload of bottled water.
The problem with 1200-baud packet is that the protocol is "all or nothing", with NO forward error correction. It just keeps blindly retransmitting after pseudorandom delays until a packet manages to get through error-free.
As others have noted, packet has ALWAYS had major problems with scalability & dealing with congestion. It takes very little to saturate a packet radio link into unusability due to retransmissions & collisions.
We already have enormously better protocols, but the prohibition against using encryption on ham bands seriously limits the ability of hams to build emergency adhoc internet backhaul (to places that DO have working internet connectivity) that others can seamlessly use directly.
Puerto Rican hams HAVE put together some mesh networks over the past week, but they've generally done it in ways that don't require a ham license (ie, fixed point to point 802.11), because a mesh network operating legally within the constraints of FCC ham regulations would be mostly useless to the public for emergency internet access since they'd have to block https & anything else that involves encrypted communication & risk exposing the real internet credentials of people to anyone running WireShark.
Update: apparently, if a Nexus 6P with N2G47W radio modem firmware is without working LTE, HSPA(+), and GPRS for several days, it stops even TRYING to use them until the next hard shutdown & full reboot.
I've done numerous "hot" reboots over the past two weeks (restarting Android, without going all the way and rebooting everything), toggled wi-fi off and on, and tried using T-mobile away from my home's wi-fi multiple times over the past 12 days, to no avail. Zero data.
For the hell of it, I decided to try totally shutting down the phone & starting it back up again. Sonofabitch, my phone now appears to have working T-mobile data service again.
I don't know who to be angrier at... T-mo for being down for days when other carriers weren't, Qualcomm for allowing their radio modem to just give up even TRYING to connect to data, or Huawei for making an Android 7.1.2 rom that allows Qualcomm's radio modem to throw in the towel and give up without so much as a "hey, your phone has totally given up trying to find working data & won't bother to try again until your next complete shutdown" notification (or a, "hey, your phone's radio modem appears to have crashed... do a full shutdown to try and resolve it." notification).
Moral: if your phone's data is nonworking for a long time, it might STAY nonworking for a long time after the underlying problem itself gets fixed. Ugh.
> The reputation is unfair: T-Mobile is superb right now
Unless you're in South Florida, where a big chunk of Dade & Broward counties STILL have no working T-mobile data service almost TWO WEEKS post-Irma (Verizon and AT&T had few outages that were mostly resolved within a day or two). Somehow, they're STILL having backhaul problems for data service.
I mean, fuck, even SPRINT had data service mostly restored a few days ago. T-Mo must have somehow found shittiest, least-resilient south Broward backhaul provider in the history of mobile phone service.
Disclaimer: current T-mo customer who's extremely likely to switch to Verizon this afternoon if T-mobile doesn't get my goddamn data working before then (I found out last night that a Nexus 6P *is* Verizon-compatible, eliminating the last thing that kept me from switching).
A lithium battery has a finite lifespan. It's guaranteed to eventually need replacement, and unless Apple is literally molding the lithium gel around the circuit board in a way that makes its replacement physically impossible, they should be required to sell replacement batteries at a fair price (or if they don't want to, then they shouldn't be allowed to prevent anyone ELSE from making compatible replacements).
A mechanical button or poorly-attached microUSB likewise has a finite life... a life that might very well be less than the reasonably-expected life of the phone.
Screens break and crack. It's just something that happens to glass when dropped.
A good step forward would be for the FTC to require manufactures to explicitly disclose the repair cost & minimum availability for key components at purchase. Ex:
Battery: 1 year free warranty, guaranteed availability for a maximum charge of $n until yyyy/MM/dd(*)
Screen: 1 year free warranty for defects, guaranteed availability for a maximum charge of $x until yyyy/MM/dd, and $y until YyYy/Mm/Dd
Electrolytic Capacitors: 10 year warranty, or replacement with newer model (not necessarily the newEST model) that's at least as good.
Buttons & connectors: 1 year warranty, 5 year guaranteed availability of repair with maximum charge of $k(*).
(*) And if they can't satisfy the repairability guarantee, they'll have to refund some fair fraction of the original purchase price based on age (say, 100% up to 18 months, 80% after 18-24 months, decreasing by another 10% per 6 months thereafter).
I'd even allow them to be assholes & enforce absurdly-short (or outrageously expensive) terms... as long as they were required to accurately & effectively communicate those terms to buyers at purchase, and really bend over backwards to make sure consumers know about terms the FTC deemed 'unreasonable'
concrete example: Consumer buys new iPhone 17 (or Samsung Galaxy S16, or Google VoxelQ). The manufacturer guarantees battery-availability only until 8/31/22, which is less than 60 months from the date of first sale. After opening the box, the phone is in another sealed box with prominent federally-mandated warning (in English & Spanish for US phones) that says something like, "Warning, the manufacturer of this phone has chosen to not guarantee the availability of replacement batteries after 8/31/22. The FTC has determined that 80% of batteries will have less than 50% of their original capacity after 8/31/21. By law, you have an absolute right to return this phone to the seller for 100% refund and full cancellation of all contractual obligations and shipping charges arising from its purchase without charge as long as this seal is not broken. This right can not be waived or limited, regardless of seller policy or conditions of sale."
ie, the Feds couldn't *stop* Apple (& others) from being assholes, but they could ensure that consumers KNOW Apple (& others) are being assholes & protect them from being victims unless they're absolutely HELLBENT on being victims. As long as the FTC's threshold for requiring the warning was reasonable enough for most vendors to avoid, its presence would be effective & would subject the mfr. to criticism & ridicule from magazines, reviewers, and probably late-night comedians. The key is to make sure it doesn't turn into a stupid, pointless warning, like "Warning: may contain peanuts" on a jar of peanuts or peanut butter.
Consumer protection laws with teeth can and do work. Just ask anyone who was given a hard time by underlings or junior staff about getting a warranty repair for a jailbroken iphone or rooted Android phone until they summoned the manager & invoked the magic phrase, "Magnuson-Moss Warranty Act"... at which point the manager, if he had any sense & valued his job, apologized *profusely* for the "misunderstanding" and *personally* made sure the phone got fixed, because anything else would risk a $50,000 per incident fine and mountain of subs
My point was, HDCP-encrypted HDMI inflates the bitrate to several gigabytes per second (for HD video, at least). You just plain *can't* sustain capturing of THAT MUCH bulk data to persistent storage (SSD, network, whatever) without some seriously exotic hardware. It's like trying to fill martini glasses with a firehose (without spilling any bits).
The size-inflation that makes the bitstream impossibly huge to capture for offline decryption is a major element *of* HDCP.
The only reason why HDMI-over-cat5e extenders for HD video are even POSSIBLE is because they decrypt the HDCP-inflated HDMI, trivially compress it down to something sane(*), re-encrypt it using an ephemeral AES-128 key that changes frequently(**), and transmit THAT... then decrypt the AES, re-inflate it, re-encrypt it with HDCP, and output it via HDMI at the other end.
(*) HDCP takes something like an 8-bit sample & pads it with additional random bits to make the number that gets encrypted bigger. The receiver knows the bottom N-8 bits of each sample are junk, so it just lops them off and ignores them.
(**) At least, this (or something comparably-robust) is what HDCP licensing officially requires. Quite a few of the single-cable HDMI extenders from China just strip the HDCP and leave it as unencrypted DVI. Or some later ones just re-pad the samples with '0' bits prior to HDCP encryption, which makes bruteforcing the key a lot easier (this attack vector became moot once the HDCP master key itself was leaked).
Now that the master key is known, implementing HDCP decryption in hardware (like a FPGA) is borderline-trivial. The hardest part is getting a pair of hdmi ports soldered to the FPGA, which is why the HDMI people used to aggressively go after anyone selling $5 HDMI breakout boards or cheap FPGA dev boards with two HDMI ports... it's technically prohibited by HDMI licensing, and made it harder for freshman EE students to build a DIY hdcp stripper as an afternoon project. Now that the industry is focused on HDCP 2.2, they seem to have mostly given up on trying to swat down everyone on eBay since it was always a hopeless game of 'whack-a-mole' anyway. They could fight against $100 dev boards with two ports because an injunction against (or customs-seizure of) a thousand would have been a crushing loss for a seller, but $5 boards that cost 20 cents to make are another battle entirely.
Sure it is. With h.264 and AES implemented in silicon and the keys known only to the video chip, it's fairly straightforward for the chip to decompress a h.264 video stream, encrypt it with AES, and write it to the frame buffer... then at output time, read it back from ram, decrypt it, re-encrypt it with HDCP, and output it to the display. The CPU itself never has direct access to unencrypted video frames.
Yes, HDCP's master key has been compromised, but decrypting HDCP on the fly still requires dedicated hardware (ASIC or FPGA). Decrypting it in realtime using CPU PIO and bus-snooping just isn't practical, and HDCP intentionally inflates the bitrate to make it impractical to capture the raw encrypted bitstream for offline decryption... not even a SSD can sustain writing THAT MUCH data for extended amounts of time without either falling behind or running out of drive space.