10 years ago, desktop computers didn't page to hard drive sectors with half-lives of 100k writes or less, and siphon the bits through a single-bit (SPI/MMC mode) or 4-bit wide (SD-mode) cocktail straw. As a few guys @ XDA learned the hard way, micro-SD wear-leveling is NOWHERE close to being as robust as what's in a desktop SSD, and onboard flash (like what's in the Nexus 4) might not have "SSD" logic *at all* (leaving its management *entirely* up to the OS to cut costs). Vigorously swap to a mostly-full microSD card, and you can *literally* push it into "hard error" land and end up with weakened cells in just one single weekend of aggressive benchmarking. Blindly swapping to internal or microSD flash desktop-style is NOT consequence-free, and is *totally* unsafe to do on any phone without microSD (at least end users can toss & replace a worn-out microSD card).
The problem isn't that Android phones have "limited ram", the problem is that Android's garbage collection sucks miserably at dealing with short-lived objects -- a problem that was fixed & mostly a non-issue with J2SE by the time 1.4 or 1.5 came out more than a DECADE ago.
10 years ago, when J2SE 1.5 was out and its garbage-collection problem was already a historical footnote, a laptop with 512mb, 32-gig hard drive, and 700-1000MHz CPU was fairly respectable. A Galaxy S3 has a gig of ram, 16 or 32 gigs of internal flash, and a 32-gig class 10 microSD card costs $20 on sale.
In some cases, strong encryption, source-tracking, and certificates will make matters *worse* by making them non-repudiatable. Right now, the NSA can judiciouscy play its "he's making *that* up" (or exaggerating) card, and might occasionally get away with it & be believed if they don't try it too often. If Snowden had not only decrypted docs (he *was* a trusted insider, after all), but ALSO had digital signatures attesting to their authenticity, the NSA would be in even *deeper* shit.
As I understand it, Lucas always intended for there to be a sequel to Star Wars, but the investors made him go through the motions of pretending Star Wars could stand on its own. The problem wasn't that they didn't think it would be popular, but rather that it wouldn't be popular enough to justify the unholy production cost of a sequel made to the same standards. They wanted to make sure that if it merely did "OK", they could keep wringing cash out of it for a few more years in second-run and international markets.
I believe The Empire Strikes Back and Return of the Jedi were kind of like the final two "Back to the Future" and "Matrix" movies... filmed mostly back-to-back, and some parts filmed simultaneously to cut costs. I think Lucas originally planned to wrap everything up in a single ~3-hour second episode, but after Star Wars turned into a license to print money, he (or his investors) decided to make it into two separate movies, and leave the door open to a 3-movie prequel that Lucas himself didn't really have any interest in making.
^^^ argh. Terrible typo I missed because I was laughing so hard while I typed this. The lion CHASING him chokes to death on the pebble, and Simba discovers that Leia the Lioness is his sister.
> The Lion King was substantially ripped off too, from Kimba the White Lion.
Twenty years from now, kids will watch the remake of the Lion King. During the climax scene, he'll be running away, and right as he's starting to get tired, he'll hear a familiar voice booming from the sky:
"Simba! Use the Horse! THE HORSE... USE THE HORSE!"
At that moment, the lion chasing him will start to laugh so hard, he won't be able to run. In fact, while skidding to a halt, a pebble will get thrown up into the air, and land in his windpipe as he rolls on the ground writhing in laughter at the cheesy attempt to tie the Lion King to Star Wars. He'll choke to death, and discover that Leia the Lioness is his sister...
Well, that's a tiny bit of an exaggeration. The harsh reality is that a 3D printer is a cool, fun, convenient way to make one-off and limited runs of plastic parts that would otherwise have to be injection-molded or extruded.
Yes, I know some systems can print starch that dissolves so you can (sort of) end up with spaces and gaps in the finished item, but in the real world, it's basically up to you to drill the precision holes, sand the rough edges, remove the burrs, and do the actual assembly yourself. We're a LONG way from "download the plans to some finished consumer good & stick it to The Man(tm) by printing yourself an unauthorized copy".
Buying a hobby-grade 3D printer today is kind of like spending $800 to buy a copy of Sculpt-Animate 4D for the Amiga 3000 20 years ago -- full of promise, totally cool, and the greatest Christmas gift someone could possibly get you... but at the end of the day, frustrating as hell.
Back then, you'd spend days, if not WEEKS, defining 3D objects, start a render at 2am before going to bed, crawl out of bed the next morning for school, be happy that you weren't greeted by 30-40 scanlines of black (indicating that it didn't like your lighting for some reason), spend the day at school praying obsessively that you'd be greeted by 2/3 of a badly-rendered image when you got home instead of a guru meditation number, and if you hit the jackpot... your preview didn't look like total shit, and vaguely resembled whatever it was you were trying to render.
A few days later, you'd go to render a raytraced preview the size of a postage stamp, then go away for the weekend, because that's about how long a 16-25MHz A3000 took to render a 80x50 thumbnail. Assuming it didn't crash, and there wasn't a thunderstorm to reboot the computer. OK, months passed, and you're about to go take a 2-week family vacation, so you launch into the Holy Grail -- a 320x200 HAM animation with 8-16 frames. You start the rendering job, go away, come home a few weeks later... and to your despair (but non-surprise), are greeted by either a guru meditation number or a rebooted computer courtesy of Florida Power & Light.
You screwed with it a few more times after that, but the magic was gone. The blue smoke evaporated. It just took too damn long to render anything meaningful, and the program had an 80% chance of crashing before it finished anyway. And when it didn't crash, it was Florida before UPSes became affordable, so 2-second power outages were almost guaranteed to nuke any multi-day rendering job before it finished even if the program DIDN'T crash. Such was life on the bleeding razor's edge of computer graphics ~20 years ago. Sigh.
> And introverts don't necessarily love the bustle of the city.
It depends... I'd argue that to introverts, "bustle" is just background noise that's easy to ignore, and maybe vaguely comforting in a passive-social-life kind of way. In contrast, the invasive social bullying you'll encounter in a small town "where everyone knows everybody" is positively soul-crushing. Being left alone when you're one in 4 million is pretty easy. Being left alone when you're one of 137 people living within 20 miles is a bit more challenging, especially if one or more of the 137 takes an invasive & annoying interest in you, or decides you don't neatly conform to their world view.
In a big city, you can buy a townhouse. It might share walls with 2 neighbors, but those walls are probably concrete and (if built within the past 50 years or so) devoid of openings & wall penetrations. If your back yard isn't surrounded by an 8-foot concrete wall, it probably COULD be done legally if you wanted one. Contrast that with typical suburbia, where you might have 12 windows and 10 feet of airspace separating you from your neighbors, and you probably aren't allowed to build anything more substantial than a 6' wall with 8' hedge before the HOA will come after you for creating an "eyesore" or blocking your neighbor's "view" (of your yard). In a city, your neighbors are closer, but you're also allowed to do more effective things to buffer yourself from them.
Narrow townhomes often have three, four, or more stories. Some people hate it, but for others, there's just something totally cool about a zig-zag floor plan where the front rooms are offset by 1/2 floor from the rear rooms & the stairs do double-duty as both stairways and interior hallways. Bonus points if you have a basement room that opens up into a walk-out excavated light court ( http://www.busyboo.com/2013/02/20/home-extension-uk-garden/ ), and official "jackpot" status if your garage has a pit-type hydraulic car stacker like this one (which allows you to independently park two cars in a single-car garage with high ceiling and pit) : http://www.youtube.com/watch?v=b3aJjCI0JSE:-D
> since NTT DoCoMo is where much of mobile innovation starts
Seriously?!? Japan is probably the only country on Earth whose mobile phone network is more locked down & arbitrarily gimped by carriers than America's. NTT even makes fsck'ing *Verizon* look like the shining light of open, interoperable Android freedom.
Wait... you're saying the HI-LINK HLK-RM04 is a complete SoC, and not just a wi-fi/serial bridge module?!? Now I'm intrigued & might need to order one;-)
You're right about not needing a GPU to open the garage door and control lights, but remember... when you're talking about ASICs, there are MASSIVE economies of scale. It's cheaper to make 5 million chips with GPU 2 million buyers don't care about than it is to make 3 million with GPU and 2 million without. Eventually, there's a point where it becomes cheaper, but for something complex like a GPU, there's also a huge area where it's still cheaper to make them all with it, test them, bin the ones with defective (or just untested) GPUs, and sell THOSE as the "GPU-less" version.
Interesting Google sidetrip: I don't remember the exact chip, but I know a few grad students (from Michigan, I think... maybe Arkansas, Kansas, and/or Oklahoma) have already built prototypes of low-cost short-range (~10-25 mile) weather radar arrays using basically three off-the-shelf chips (Broadcom agile baseband processor, TI DSP, and ARM-based SoC) with a 250-watt class-D amp tuned for ~5GHz to build an exciter with software-defined receiver that needs little more than a rotating high-gain antenna and power supply to be hardware-complete. Obviously there's a huge amount of software development remaining just to get it to the point where you can make the output look like current weather radar data & view it with off-the-shelf apps, but when you're taking about building something that could profitably retail for $10,000, and would currently be classified as "unbelievably cheap" with a $400,000 price tag, well... that's definitely dotcom-level "disruptive innovation";-)
Demographics. Remember, it was 1992... "two cities, sharing the same patch of ground, but otherwise existing in parallel, largely oblivious to each other's existence". Statistically, if you were a gringo from east-central Dade or South Beach, your friends and family members were overwhelmingly likely to live south of Miller Drive. North of 836, the area west of 27th avenue was like a foreign country, and the area east of 27th avenue was a riot-torn warzone. Yeah, that's an exaggeration... but not much. Miami was a VERY different place 21 years ago.
The anger came from the fact that lots of people were told that they were going to die unless they evacuated, went to stay with friends in areas that were NOT in evacuation zones, got hit directly and went through complete HELL, then came home to neighborhoods that were barely affected... and found out that government officials had known for at least a day before landfall that Andrew's eyewall wasn't going anywhere NEAR South Beach or downtown unless some truly freakish path change occurred that would have been inconceivable under every known model.
I have friends who worked for Dade County. The topic of Miami's lopsided demographics came up over and over again in private during the 48 hours before Andrew's landfall. County officials knew damn well where everyone was going to go if they evacuated, and intentionally decided it was more important to be politically correct & pretend people were going to evacuate to public shelters in neighborhoods they normally wouldn't even DRIVE THROUGH than to tell people point blank, "Look , we know your friends and family members live down south... don't go there, because they're more likely to get hit directly than YOU are. If you don't want to evacuate to Liberty City or Hialeah, go to Orlando."
The RPi ALSO has a CPU & lots of ram, so it takes the place of BOTH the serial-ethernet bridge AND 8-bit MCU. The boards you mentioned are interesting alternatives to the w5100 (though I couldn't find anything about their inherent security, or lack thereof, vs w5100), but I'm pretty sure that a RPi is probably still the most cost-effective way to get internet-connectivity that includes SSL regardless of whether or not you care about HDMI.
In the grand scheme of things, HDMI adds very little to the RPi's cost. HDMI is just a collection of fast serial ports with inline op-amps to balance the signals, and the entire implementation besides the port itself is inside the SoC anyway. A Pi board without HDMI might cost a whopping 50c less, but if you actually needed HDMI, you'd end up spending half as much as the entire Pi for a HDMI port on a 100-mil breakout board by the time you factor in the purchase price and shipping.
In case you haven't noticed, Miami is normally about 10 degrees hotter than Daytona... especially in the evening. North of Orlando, the temperature goes down a bit after sundown. In Miami, the thermometer just kind of laughs at us and stays where it is. Or worse, in the winter, we get wacky nighttime heat inversions where it'll be in the upper 70s as the sun is going down, then start going up again around 9-11pm until it's in the mid-80s by midnight. Your air conditioner runs a lot for half the day. Our air conditioners don't get as much of a break. Especially when you're trying to cool the interior down to 20 degrees or more below ambient outside temperature.
In Central Florida, a heat pump makes sense. In South Florida, you'll get more usefulness out of a dual-speed compressor that can run longer at reduced speed to wring more humidity out of the air, but kick into high gear in the mid-afternoon and keep the house cool in scenarios where a properly-sized air conditioner would actually be fighting a losing battle against the sun & your indoor temperature would be creeping up by a degree every hour or two EVEN WITH the air conditioner running constantly. Putting the whole "heat thing" into perspective, I've literally gone for entire "winters" without ever turning on the heat. Over the past year, I think we've had MAYBE 4 or 5 days when it was genuinely cool AND DRY enough outside to open the windows... at night. MAYBE two days in February or March when it was still cool and non-humid enough during the day. Even when it's in the upper 60s, we STILL have insane humidity that ruins it.
I still remember joyfully watching a huge cold front making its way south a few months ago (January?), I think it even snowed in Jacksonville, then... DAMMIT! The cold front NEVER MADE IT SOUTH OF WEST PALM BEACH. It's like it just ran into our wall of heat & humidity & fizzled out after dumping 3 feet of snow on the entire eastern seaboard and midwest.
For the morbidly-curious, here's a book that might give you somewhat of an idea of what USED to be involved with interfacing a microcontroller with a network over Ethernet pre-Wiznet w5100, and give the benefit of context to understand why that module (and its descendants) have been so wildly popular among embedded developers working with 8-bit microcontrollers.
The Microchip ENC28J60 falls somewhere between the older chips written about in that book and a "plug & play" module like the W5100. With the older chips, you were lucky to hack together your own personal networking protocol that (barely) managed to coexist on the same wire as NETBIOS, TCP/IP, and IPX/SPX. The ENC28J60 does for networking kind of what the ATI Rage Theater chipset did for MPEG-2 video compression... it accelerates and automates some of the grunt work of interacting with signals on the cable so you can pay attention to bigger details, like your actual protocol. I've never personally used it, but from what I've read, ENC28J60 TCP/IP is "do-able, but with a few cautions and limits". By comparison, the W5100 is pure black magic... to your embedded app, it turns the Internet and/or your local LAN into a big virtual serial cable.
When the w5100 came out ~5-6 years ago, embedded developers were LITERALLY dancing in the streets, because it was dirt cheap and "just worked". Security wasn't even a CONSIDERATION until 2-3 years later, when the consequences of exposing the serial ports of devices with no security besides physical access to the port started to really sink in... and the devices themselves had almost no serial-port security, because pre-Wiznet, an ethernet-serial adapter cost somewhere between $250 and $400... at RESELLER prices. Pre-w5100, serial ports just plain didn't get exposed to the internet, because the adapters to do it were too expensive to even contemplate.
Because they'll cut your AC precisely on the hottest days of the year when you need it the most. But wait, it gets worse. If you cut the compressor, but allow the fan to run, you're effectively running dry air over a pool of water in the evaporator pan. If your compressor can only run for 20 minutes per hour, it's 100+ degrees outside, and your thermostat is set to 74, the interior temperature is probably going to go up more in 40 minutes than it can be cooled in the remaining 20, which means your blower is going to run nonstop for hour after hour -- raising the humidity in your house, allowing the temperature to creep up, AND using almost half as much power running the blower alone for an hour that you WOULD have used if they'd allowed the compressor + blower to run normally for 10-15 minutes two or three times per hour.
Years ago, I rented an apartment that had a FPL on-demand box installed. In return for a piddling discount of something like $5/month off an electric bill that normally exceeded $180-220, they left me completely MISERABLE several dozen times per year (2-3 times per month during the month or two we call "winter", AT LEAST 3 or 4 times PER WEEK during the remainder of the year).
Even worse, I had to fight with them for MONTHS to get it removed, and finally had to get the Florida PSC involved to force them to remove it. They apparently figured that since I was renting, if they just stonewalled me long enough, eventually I'd move out with the box still in place.
Any "savings" were completely neutralized by the fact that I had to cool the house down to 70 before 11am and keep it there all afternoon as a defensive measure against having it disabled when I needed it the most. The irony is that my bills went DOWN by ~$20-30/month (even after the discount was removed), because I no longer had to supercool the house daily before noon to subvert them.
Oh, there will be PLENTY of such designs coming over the next year or two.
The RPi is fundamentally game-changing now that they've finally gotten the shipping charges down to non-anal-rape levels, because it means you can finally get an ARM with enough ram to do SSL via software for less than the cost of an Atmel Atmega 2560 + W5100.
In the past, it was the need for external RAM that killed everyone who tried to homebrew ARM boards with it. RAM looks deceptively easy on a schematic, but it's a BIATCH to actually hand-solder a.5mm TQFP MCU to a board with 30MHz+ bus and make it work reliably. A $30 ARM9 with a few megs of ram, glue chips, and voltage regulator would be a hell of a deal, even if the board literally did nothing else besides break out the remaining pins on the MCU to 100-mil headers.
> For criminy's sake. TLS is *there*. It's *free*. Why the hell aren't these guys using it??
Quite a few embedded home automation devices are built around 8-bit MCUs like the Atmel AVR family. You'd be massively challenged to get even a minimal subset of TCP/IP working with a chip like the Microchip ENC28J60 ethernet controller and an Atmel Atmega 128. SSL/TLS? ROFLMAO. It's not happening. You could probably kludge something with more chips and sram, but by that point, you'd be better off throwing in the towel and buying a RPi board.
Pre-RPi, ARM boards with additional RAM were pretty expensive (at least $80-150), so a $10 AVR plus $15 Wiznet board represented a huge cost savings. Now that you can get a RPi for $30, it's kind of stupid to keep building controllers with 8-bit MCUs and ethernet-serial bridge boards... but a year ago, the RPi basically didn't exist, and even 6 months ago, it was pretty expensive once you factored in rape-level shipping charges to the US. Genuinely cheap ARM chips with external RAM are game-changing for anything that involves communication over the internet.
In almost every case, the big, impressive waves arrive LONG before the storm itself, and the weather deteriorates & becomes unpleasant long before it actually becomes *dangerous* to be within sight of the waves. A hurricane isn't an event, it's a process. A *tornado* is an event... the storm quickly rolls in, the funnel drops from the sky (while the condensation cloud forms and makes it visible from the ground up), and it's *there*. Hurricanes? Hours, if not DAYS, of getting bitchslapped by rainbands that generally get worse and worse, until the eyewall FINALLY makes landfall and passes nearby (or directly above). Then, you get hit with a tornado-like windfield, have a few minutes of calm, another hit by the windfield (in the opposite direction), followed by a few more hours/days of the process in reverse.
Now... the authorities might certainly *prefer* that everyone just leave instead of making their jobs more complicated by having to manage tourists and traffic as a hurricane is approaching, but it's not like the beach suddenly becomes a lethal place to be just because a hurricane is approaching and will be there in a few hours.
The local road network matters, too. In Southwest Florida, yeah... there's a ~25 mile stretch of beach between Bonita Springs and Fort Myers with a 2-lane bridge at each end, another 10-15 miles of traffic just to get to I-75, and literally no other route to the mainland in between. Getting everyone "off the islands" is a major exercise in logistics. Contrast that with South Beach, where there's a 6-lane freeway and multiple other roads connecting it to the mainland, and a quarter of the population drives ashore every weekday over the span of 2-3 hours just getting to work. If anything, at the "Miami" end, the problem isn't evacuating Miami Beach to the mainland... it's the fact that people who live in western Dade County -- many of whom vividly remember being lied to by the government(*) -- trying to evacuate to Orlando (or worse, get frustrated when northbound I-75, Turnpike, and I-95 turn into parking lots, and head WEST across I-75 towards Naples, not realizing that they're going to encounter even WORSE northbound traffic on 75 long before they even make it across the county line).
(*)When Andrew hit, the county authorities were ADAMANT that it was going to directly hit South Beach, because they wanted to scare people who lived there into evacuating. The National Hurricane Center wasn't happy about it, but grudgingly went along with it because the county's rationale sounded reasonable. The problem was, it was a total lie from the first press release, and they knew it. As a result, people who lived in places like northern Coral Gables, Coconut Grove, Brickell, and downtown Miami fled SOUTH to the homes of friends and family members... and ended up evacuating INTO the area that was the hardest-hit. NOBODY who remembers Andrew trusts the authorities anymore. It's 80% of the reason why I learned how to run GFS myself and got into stormchasing in the first place... so I could independently run my own sims and do my own fact-checking instead of being forced to take the authorities' word at face value and hope they weren't lying again.
Connecting HA gear to the internet in a way that's both secure and works (especially with the manufacturer's own Android/iOS/ActiveX software) is actually pretty hard to do with real-world equipment, mainly because the overwhelming majority of stuff that's affordable (and shocking amounts of stuff that's supposedly top of the line) gets implemented with little more than symbolic security that's the equivalent of a TSA lock.
One of the most common ways embedded hardware gets connected to the Internet is via Wiznet w5100 modules (and variants). Basically, the w5100 is a bridge between UDP and a serial port. Data arrives on some port, and gets blindly sent to the serial port. Data comes in through the serial port, and gets sprayed out via UDP. There's some minimal logic that implements a half-assed 8-character "password" that -- at best -- might be equivalent to a 64-bit random number IF you hacked their setup utility to accept arbitrary byte values instead of just letters and digits (effectively turning it into more like a 40-bit random value). Keep in mind that there's no rate-limiting or lockout, so the only limiting factor at which someone can try to bruteforce you is your internet connection.
A few months ago, I estimated that an attacker who knows you have something specific behind a Wiznet interface that responds to a known command with a known response would take about 1-3 months to bruteforce if they kept the rate low enough to not noticeably affect your internet access or attract undue attention, and less than a weekend if they just all-out hammered you as fast as they could, trying only 8-character alphanumeric values and starting with those that begin with digits & plaintext English words.
That itself isn't necessarily the problem per se... there's nothing that says you can't encrypt the data being sent via UDP and in response, and implement stronger authentication and authorization checks on your own... except nothing actually DOES.
99.9% of the time, you have a circuit with almost no real network-level security that was developed with the assumption that someone with physical access to the serial port has already demonstrated some level of authorization, connected to a serial-ethernet bridge whose "security" is almost a complete sham, with predictable results: disaster.
Short answer: if you want to connect consumer gear over the internet, buy a Raspberry Pi, and use it as a middleware gateway device that accepts incoming connections via https, enforces its own strong authentication, passes no raw commands directly between the internet and embedded device, or at least requires that any raw value be signed with a pinned certificate. Then connect THAT to the embedded device through the Wiznet serial-ethernet adapter. Never, ever, EVER expose a serial port directly to the internet through a serial-ethernet adapter... I can almost guarantee that any such adapter that ISN'T built around a RPi and costs less than $200 is inherently insecure and a hack attack waiting to happen.
If you absolutely MUST expose some consumer-grade device with insecure ethernet-serial interface over the internet, at least hide it behind a router running OpenWRT/Tomato/dd-WRT and use something like port knocking and IP range-blocking to temporarily unblock access to your mobile device's current IP address for short periods of time when you intentionally enable it (keeping in mind that with many wireless providers, switching between HSPA/EVDO and LTE will radically change your IP address, and your address might change from request to request ANYWAY.
>That's the same reason smartphones don't have replaceable batteries much anymore: > by the time the battery wears out, 99% of users are going to be looking for a new phone anyways
Within ~4 hours of purchase? That's about how long a stock battery *might* last for me... if I don't use it much.
User-replaceable batteries mean you can toss the wimpy 1800mAH toy into a drawer & snap in a nice, big, beefy 4,000-8,000mAH extended battery that *might* last until at least midnight. Or swap the stock battery for a spare every few hours as they die.
And unfortunately, eBook reading is STILL a completely *dreadful* experience for anything besides cover-to-cover novels and magazines.
E-ink takes an eternity to flip pages, tablets are eye-searing, low-contrast, low-res, or some combination... and both totally suck for random-access reference-type "flip back and forth between a few sections of the book" reading. I'm as guilty as anyone of buying eBooks when I need some book RIGHT NOW, but damn, I wish somebody could make ebooks not suck so miserably.
> Let's say patents were strong and eternal. This could easily lead to land grabs - there's a rush to discover new areas > that people aren't even thinking about right now and bring them to market so that they can be patented. The value of a > patent portfolio is high, and so companies may innovate to create such portfolios. Cross-licensing would be the order > of the day in that environment, but it would hardly mean innovation would be dead.
No, it means we'd be in eternal stagnant IP feudalism, where only companies the size of Sony, Samsung, Apple, and Microsoft were allowed to innovate anything meaningful commercially. Everyone else would be -- at best -- IP sharecroppers forced to kneel before them and grant their patents for a pittance in return for the right to make commercial use of them at all, because just about any conceivable implementation would otherwise infringe upon two dozen other patents.
Think about it. When's the last time a huge corporation held by risk-averse institutional shareholders concerned with maximizing the next quarter's profits truly innovated *anything* that could be described as "disruptive"? The infant Microsoft would have been thrown into the well and drowned by IBM, DEC, Burroughs, Wang, and all the other companies that used to dominate the American computer industry back in the 1970s/80s. Woz would have gotten served a C&D 3 days after showing off his new prototype to the computer club. Commodore would have never existed, because its whole economic foundation was a semiconductor firm who built what was practically a 6800 clone. Fairchild *and* Shockley would have been eternal vassals of Bell Labs.
Big companies owned by institutional investors are almost genetically incapable of disruptive innovation, because it involves risk, and risk means the next quarter's profits might be disrupted. When alliances of similarly risk-averse companies have de-facto veto power over everyone else by virtue of strong patent protection, disruptive innovation stagnates.
In a way, the former Soviet Union is a perfect indirect example. In the Soviet Union, risk-averse bureaucrats had de-facto (if not de-jure) veto power over everything. The same way a company like Sony can unleash its lawyers to stop disruptive media technologies in their tracks in the holy name of IP law, the Soviet government could unleash its army of law enforcement officers and bureaucrats to prevent a Soviet citizen with a disruptive invention from ever doing anything with it. There was actually quite a bit of underground innovation in the Soviet Union... it's just that it was all basically masturbation, because their efforts were stymied at every level (by buraucratic indifference if they were lucky, by aggressive political backlash if they were perceived as presenting a risk to someone in a position of power, no matter how petty or meaningless).
Going back further to the dark ages, how much real innovation occurred when everyone was forced (by law, if not practical necessity) into being vassals under feudalism that subordinated everyone and everything to landed nobility and the Church? Yes, we had people like Kepler, Newton, and Galileo... and their contemporary influence was almost nonexistent. Their discoveries were talked about privately, behind closed doors, and most of their effort was spent trying to avoid getting crushed by those who cared mainly about preserving the status quo. They're revered today, centuries after their deaths, but in their day, they were basically viewed as godless heretics who deserved to burn for their sins, or at least spend a few decades repenting them in a dark prison cell somewhere.
People forget that the US itself was founded as a reaction to the old world order. Pre-Berne, American IP law was almost at a 90-degree angle to European IP law. European IP law viewed IP as a moral right, like being granted land by a sovereign king acting through the grace of God. American IP law, in stark contrast, was scandalously utilitarian. Up until t
> I wish there were some type of test suite that phone makers could offer to emulate their products,
There is... kind of... the pile of old Android phones most Android developers have accumulated by now;-)
My pile: HTC HeroC, Samsung Epic4G, Motorola Photon, Galaxy S3 (current phone). Of the phones in the pile, only the Photon sits (battery removed) unloved and hated in a drawer, awaiting the day someone cracks that poor gimped phone's Motorola-permalocked 2.3.5 bootloader, breaks its chains & shackles, and lets it play Martin Luther King's "Free at Last!" exclamation while booting into Cyanogen.
> So if Google announced a major new version tomorrow how long would it be before I > could exclusively target that version?
Depends upon whether your new app needs immediate sales, and whether you're talking about an expensive app that needs best-of-breed hardware to run acceptably well, anyway. If your app will suck on anything less than a GNex/S3/Note2/Nexus4/OneX, you won't lose many real sales by requiring 4.1, and probably won't lose many more by aiming for 4.2.
Just don't make the mistake of setting the hardware/version bar high, then releasing an app that sucks. Aim high, knock everybody's socks off, and you'll do OK. Aim low & avoid sucking, and you'll probably make some cash anyway. Aim high & suck, and you'll get buried under bad reviews & refunds.
Slashdot Mirror
10 years ago, desktop computers didn't page to hard drive sectors with half-lives of 100k writes or less, and siphon the bits through a single-bit (SPI/MMC mode) or 4-bit wide (SD-mode) cocktail straw. As a few guys @ XDA learned the hard way, micro-SD wear-leveling is NOWHERE close to being as robust as what's in a desktop SSD, and onboard flash (like what's in the Nexus 4) might not have "SSD" logic *at all* (leaving its management *entirely* up to the OS to cut costs). Vigorously swap to a mostly-full microSD card, and you can *literally* push it into "hard error" land and end up with weakened cells in just one single weekend of aggressive benchmarking. Blindly swapping to internal or microSD flash desktop-style is NOT consequence-free, and is *totally* unsafe to do on any phone without microSD (at least end users can toss & replace a worn-out microSD card).
The problem isn't that Android phones have "limited ram", the problem is that Android's garbage collection sucks miserably at dealing with short-lived objects -- a problem that was fixed & mostly a non-issue with J2SE by the time 1.4 or 1.5 came out more than a DECADE ago.
10 years ago, when J2SE 1.5 was out and its garbage-collection problem was already a historical footnote, a laptop with 512mb, 32-gig hard drive, and 700-1000MHz CPU was fairly respectable. A Galaxy S3 has a gig of ram, 16 or 32 gigs of internal flash, and a 32-gig class 10 microSD card costs $20 on sale.
In some cases, strong encryption, source-tracking, and certificates will make matters *worse* by making them non-repudiatable. Right now, the NSA can judiciouscy play its "he's making *that* up" (or exaggerating) card, and might occasionally get away with it & be believed if they don't try it too often. If Snowden had not only decrypted docs (he *was* a trusted insider, after all), but ALSO had digital signatures attesting to their authenticity, the NSA would be in even *deeper* shit.
As I understand it, Lucas always intended for there to be a sequel to Star Wars, but the investors made him go through the motions of pretending Star Wars could stand on its own. The problem wasn't that they didn't think it would be popular, but rather that it wouldn't be popular enough to justify the unholy production cost of a sequel made to the same standards. They wanted to make sure that if it merely did "OK", they could keep wringing cash out of it for a few more years in second-run and international markets.
I believe The Empire Strikes Back and Return of the Jedi were kind of like the final two "Back to the Future" and "Matrix" movies... filmed mostly back-to-back, and some parts filmed simultaneously to cut costs. I think Lucas originally planned to wrap everything up in a single ~3-hour second episode, but after Star Wars turned into a license to print money, he (or his investors) decided to make it into two separate movies, and leave the door open to a 3-movie prequel that Lucas himself didn't really have any interest in making.
^^^ argh. Terrible typo I missed because I was laughing so hard while I typed this. The lion CHASING him chokes to death on the pebble, and Simba discovers that Leia the Lioness is his sister.
> The Lion King was substantially ripped off too, from Kimba the White Lion.
Twenty years from now, kids will watch the remake of the Lion King. During the climax scene, he'll be running away, and right as he's starting to get tired, he'll hear a familiar voice booming from the sky:
"Simba! Use the Horse! THE HORSE... USE THE HORSE!"
At that moment, the lion chasing him will start to laugh so hard, he won't be able to run. In fact, while skidding to a halt, a pebble will get thrown up into the air, and land in his windpipe as he rolls on the ground writhing in laughter at the cheesy attempt to tie the Lion King to Star Wars. He'll choke to death, and discover that Leia the Lioness is his sister...
Well, that's a tiny bit of an exaggeration. The harsh reality is that a 3D printer is a cool, fun, convenient way to make one-off and limited runs of plastic parts that would otherwise have to be injection-molded or extruded.
Yes, I know some systems can print starch that dissolves so you can (sort of) end up with spaces and gaps in the finished item, but in the real world, it's basically up to you to drill the precision holes, sand the rough edges, remove the burrs, and do the actual assembly yourself. We're a LONG way from "download the plans to some finished consumer good & stick it to The Man(tm) by printing yourself an unauthorized copy".
Buying a hobby-grade 3D printer today is kind of like spending $800 to buy a copy of Sculpt-Animate 4D for the Amiga 3000 20 years ago -- full of promise, totally cool, and the greatest Christmas gift someone could possibly get you... but at the end of the day, frustrating as hell.
Back then, you'd spend days, if not WEEKS, defining 3D objects, start a render at 2am before going to bed, crawl out of bed the next morning for school, be happy that you weren't greeted by 30-40 scanlines of black (indicating that it didn't like your lighting for some reason), spend the day at school praying obsessively that you'd be greeted by 2/3 of a badly-rendered image when you got home instead of a guru meditation number, and if you hit the jackpot... your preview didn't look like total shit, and vaguely resembled whatever it was you were trying to render.
A few days later, you'd go to render a raytraced preview the size of a postage stamp, then go away for the weekend, because that's about how long a 16-25MHz A3000 took to render a 80x50 thumbnail. Assuming it didn't crash, and there wasn't a thunderstorm to reboot the computer. OK, months passed, and you're about to go take a 2-week family vacation, so you launch into the Holy Grail -- a 320x200 HAM animation with 8-16 frames. You start the rendering job, go away, come home a few weeks later... and to your despair (but non-surprise), are greeted by either a guru meditation number or a rebooted computer courtesy of Florida Power & Light.
You screwed with it a few more times after that, but the magic was gone. The blue smoke evaporated. It just took too damn long to render anything meaningful, and the program had an 80% chance of crashing before it finished anyway. And when it didn't crash, it was Florida before UPSes became affordable, so 2-second power outages were almost guaranteed to nuke any multi-day rendering job before it finished even if the program DIDN'T crash. Such was life on the bleeding razor's edge of computer graphics ~20 years ago. Sigh.
> And introverts don't necessarily love the bustle of the city.
It depends... I'd argue that to introverts, "bustle" is just background noise that's easy to ignore, and maybe vaguely comforting in a passive-social-life kind of way. In contrast, the invasive social bullying you'll encounter in a small town "where everyone knows everybody" is positively soul-crushing. Being left alone when you're one in 4 million is pretty easy. Being left alone when you're one of 137 people living within 20 miles is a bit more challenging, especially if one or more of the 137 takes an invasive & annoying interest in you, or decides you don't neatly conform to their world view.
In a big city, you can buy a townhouse. It might share walls with 2 neighbors, but those walls are probably concrete and (if built within the past 50 years or so) devoid of openings & wall penetrations. If your back yard isn't surrounded by an 8-foot concrete wall, it probably COULD be done legally if you wanted one. Contrast that with typical suburbia, where you might have 12 windows and 10 feet of airspace separating you from your neighbors, and you probably aren't allowed to build anything more substantial than a 6' wall with 8' hedge before the HOA will come after you for creating an "eyesore" or blocking your neighbor's "view" (of your yard). In a city, your neighbors are closer, but you're also allowed to do more effective things to buffer yourself from them.
Narrow townhomes often have three, four, or more stories. Some people hate it, but for others, there's just something totally cool about a zig-zag floor plan where the front rooms are offset by 1/2 floor from the rear rooms & the stairs do double-duty as both stairways and interior hallways. Bonus points if you have a basement room that opens up into a walk-out excavated light court ( http://www.busyboo.com/2013/02/20/home-extension-uk-garden/ ), and official "jackpot" status if your garage has a pit-type hydraulic car stacker like this one (which allows you to independently park two cars in a single-car garage with high ceiling and pit) : http://www.youtube.com/watch?v=b3aJjCI0JSE :-D
> since NTT DoCoMo is where much of mobile innovation starts
Seriously?!? Japan is probably the only country on Earth whose mobile phone network is more locked down & arbitrarily gimped by carriers than America's. NTT even makes fsck'ing *Verizon* look like the shining light of open, interoperable Android freedom.
Wait... you're saying the HI-LINK HLK-RM04 is a complete SoC, and not just a wi-fi/serial bridge module?!? Now I'm intrigued & might need to order one ;-)
You're right about not needing a GPU to open the garage door and control lights, but remember... when you're talking about ASICs, there are MASSIVE economies of scale. It's cheaper to make 5 million chips with GPU 2 million buyers don't care about than it is to make 3 million with GPU and 2 million without. Eventually, there's a point where it becomes cheaper, but for something complex like a GPU, there's also a huge area where it's still cheaper to make them all with it, test them, bin the ones with defective (or just untested) GPUs, and sell THOSE as the "GPU-less" version.
Interesting Google sidetrip: I don't remember the exact chip, but I know a few grad students (from Michigan, I think... maybe Arkansas, Kansas, and/or Oklahoma) have already built prototypes of low-cost short-range (~10-25 mile) weather radar arrays using basically three off-the-shelf chips (Broadcom agile baseband processor, TI DSP, and ARM-based SoC) with a 250-watt class-D amp tuned for ~5GHz to build an exciter with software-defined receiver that needs little more than a rotating high-gain antenna and power supply to be hardware-complete. Obviously there's a huge amount of software development remaining just to get it to the point where you can make the output look like current weather radar data & view it with off-the-shelf apps, but when you're taking about building something that could profitably retail for $10,000, and would currently be classified as "unbelievably cheap" with a $400,000 price tag, well... that's definitely dotcom-level "disruptive innovation" ;-)
> i don't know who would have evacuated south
Demographics. Remember, it was 1992... "two cities, sharing the same patch of ground, but otherwise existing in parallel, largely oblivious to each other's existence". Statistically, if you were a gringo from east-central Dade or South Beach, your friends and family members were overwhelmingly likely to live south of Miller Drive. North of 836, the area west of 27th avenue was like a foreign country, and the area east of 27th avenue was a riot-torn warzone. Yeah, that's an exaggeration... but not much. Miami was a VERY different place 21 years ago.
The anger came from the fact that lots of people were told that they were going to die unless they evacuated, went to stay with friends in areas that were NOT in evacuation zones, got hit directly and went through complete HELL, then came home to neighborhoods that were barely affected... and found out that government officials had known for at least a day before landfall that Andrew's eyewall wasn't going anywhere NEAR South Beach or downtown unless some truly freakish path change occurred that would have been inconceivable under every known model.
I have friends who worked for Dade County. The topic of Miami's lopsided demographics came up over and over again in private during the 48 hours before Andrew's landfall. County officials knew damn well where everyone was going to go if they evacuated, and intentionally decided it was more important to be politically correct & pretend people were going to evacuate to public shelters in neighborhoods they normally wouldn't even DRIVE THROUGH than to tell people point blank, "Look , we know your friends and family members live down south... don't go there, because they're more likely to get hit directly than YOU are. If you don't want to evacuate to Liberty City or Hialeah, go to Orlando."
The RPi ALSO has a CPU & lots of ram, so it takes the place of BOTH the serial-ethernet bridge AND 8-bit MCU. The boards you mentioned are interesting alternatives to the w5100 (though I couldn't find anything about their inherent security, or lack thereof, vs w5100), but I'm pretty sure that a RPi is probably still the most cost-effective way to get internet-connectivity that includes SSL regardless of whether or not you care about HDMI.
In the grand scheme of things, HDMI adds very little to the RPi's cost. HDMI is just a collection of fast serial ports with inline op-amps to balance the signals, and the entire implementation besides the port itself is inside the SoC anyway. A Pi board without HDMI might cost a whopping 50c less, but if you actually needed HDMI, you'd end up spending half as much as the entire Pi for a HDMI port on a 100-mil breakout board by the time you factor in the purchase price and shipping.
Miami has two seasons: Summer and February.
In case you haven't noticed, Miami is normally about 10 degrees hotter than Daytona... especially in the evening. North of Orlando, the temperature goes down a bit after sundown. In Miami, the thermometer just kind of laughs at us and stays where it is. Or worse, in the winter, we get wacky nighttime heat inversions where it'll be in the upper 70s as the sun is going down, then start going up again around 9-11pm until it's in the mid-80s by midnight. Your air conditioner runs a lot for half the day. Our air conditioners don't get as much of a break. Especially when you're trying to cool the interior down to 20 degrees or more below ambient outside temperature.
In Central Florida, a heat pump makes sense. In South Florida, you'll get more usefulness out of a dual-speed compressor that can run longer at reduced speed to wring more humidity out of the air, but kick into high gear in the mid-afternoon and keep the house cool in scenarios where a properly-sized air conditioner would actually be fighting a losing battle against the sun & your indoor temperature would be creeping up by a degree every hour or two EVEN WITH the air conditioner running constantly. Putting the whole "heat thing" into perspective, I've literally gone for entire "winters" without ever turning on the heat. Over the past year, I think we've had MAYBE 4 or 5 days when it was genuinely cool AND DRY enough outside to open the windows... at night. MAYBE two days in February or March when it was still cool and non-humid enough during the day. Even when it's in the upper 60s, we STILL have insane humidity that ruins it.
I still remember joyfully watching a huge cold front making its way south a few months ago (January?), I think it even snowed in Jacksonville, then... DAMMIT! The cold front NEVER MADE IT SOUTH OF WEST PALM BEACH. It's like it just ran into our wall of heat & humidity & fizzled out after dumping 3 feet of snow on the entire eastern seaboard and midwest.
For the morbidly-curious, here's a book that might give you somewhat of an idea of what USED to be involved with interfacing a microcontroller with a network over Ethernet pre-Wiznet w5100, and give the benefit of context to understand why that module (and its descendants) have been so wildly popular among embedded developers working with 8-bit microcontrollers.
http://www.amazon.com/Networking-Internetworking-Microcontrollers-Fred-Eady/dp/0750676981/ref=wl_it_dp_o_pC_S_nC?ie=UTF8&colid=75OKCKDXZ6YI&coliid=I2PABIRD1YO96X
The Microchip ENC28J60 falls somewhere between the older chips written about in that book and a "plug & play" module like the W5100. With the older chips, you were lucky to hack together your own personal networking protocol that (barely) managed to coexist on the same wire as NETBIOS, TCP/IP, and IPX/SPX. The ENC28J60 does for networking kind of what the ATI Rage Theater chipset did for MPEG-2 video compression... it accelerates and automates some of the grunt work of interacting with signals on the cable so you can pay attention to bigger details, like your actual protocol. I've never personally used it, but from what I've read, ENC28J60 TCP/IP is "do-able, but with a few cautions and limits". By comparison, the W5100 is pure black magic... to your embedded app, it turns the Internet and/or your local LAN into a big virtual serial cable.
When the w5100 came out ~5-6 years ago, embedded developers were LITERALLY dancing in the streets, because it was dirt cheap and "just worked". Security wasn't even a CONSIDERATION until 2-3 years later, when the consequences of exposing the serial ports of devices with no security besides physical access to the port started to really sink in... and the devices themselves had almost no serial-port security, because pre-Wiznet, an ethernet-serial adapter cost somewhere between $250 and $400... at RESELLER prices. Pre-w5100, serial ports just plain didn't get exposed to the internet, because the adapters to do it were too expensive to even contemplate.
> And why is that?
Because they'll cut your AC precisely on the hottest days of the year when you need it the most. But wait, it gets worse. If you cut the compressor, but allow the fan to run, you're effectively running dry air over a pool of water in the evaporator pan. If your compressor can only run for 20 minutes per hour, it's 100+ degrees outside, and your thermostat is set to 74, the interior temperature is probably going to go up more in 40 minutes than it can be cooled in the remaining 20, which means your blower is going to run nonstop for hour after hour -- raising the humidity in your house, allowing the temperature to creep up, AND using almost half as much power running the blower alone for an hour that you WOULD have used if they'd allowed the compressor + blower to run normally for 10-15 minutes two or three times per hour.
Years ago, I rented an apartment that had a FPL on-demand box installed. In return for a piddling discount of something like $5/month off an electric bill that normally exceeded $180-220, they left me completely MISERABLE several dozen times per year (2-3 times per month during the month or two we call "winter", AT LEAST 3 or 4 times PER WEEK during the remainder of the year).
Even worse, I had to fight with them for MONTHS to get it removed, and finally had to get the Florida PSC involved to force them to remove it. They apparently figured that since I was renting, if they just stonewalled me long enough, eventually I'd move out with the box still in place.
Any "savings" were completely neutralized by the fact that I had to cool the house down to 70 before 11am and keep it there all afternoon as a defensive measure against having it disabled when I needed it the most. The irony is that my bills went DOWN by ~$20-30/month (even after the discount was removed), because I no longer had to supercool the house daily before noon to subvert them.
Oh, there will be PLENTY of such designs coming over the next year or two.
The RPi is fundamentally game-changing now that they've finally gotten the shipping charges down to non-anal-rape levels, because it means you can finally get an ARM with enough ram to do SSL via software for less than the cost of an Atmel Atmega 2560 + W5100.
In the past, it was the need for external RAM that killed everyone who tried to homebrew ARM boards with it. RAM looks deceptively easy on a schematic, but it's a BIATCH to actually hand-solder a .5mm TQFP MCU to a board with 30MHz+ bus and make it work reliably. A $30 ARM9 with a few megs of ram, glue chips, and voltage regulator would be a hell of a deal, even if the board literally did nothing else besides break out the remaining pins on the MCU to 100-mil headers.
> For criminy's sake. TLS is *there*. It's *free*. Why the hell aren't these guys using it??
Quite a few embedded home automation devices are built around 8-bit MCUs like the Atmel AVR family. You'd be massively challenged to get even a minimal subset of TCP/IP working with a chip like the Microchip ENC28J60 ethernet controller and an Atmel Atmega 128. SSL/TLS? ROFLMAO. It's not happening. You could probably kludge something with more chips and sram, but by that point, you'd be better off throwing in the towel and buying a RPi board.
Pre-RPi, ARM boards with additional RAM were pretty expensive (at least $80-150), so a $10 AVR plus $15 Wiznet board represented a huge cost savings. Now that you can get a RPi for $30, it's kind of stupid to keep building controllers with 8-bit MCUs and ethernet-serial bridge boards... but a year ago, the RPi basically didn't exist, and even 6 months ago, it was pretty expensive once you factored in rape-level shipping charges to the US. Genuinely cheap ARM chips with external RAM are game-changing for anything that involves communication over the internet.
In almost every case, the big, impressive waves arrive LONG before the storm itself, and the weather deteriorates & becomes unpleasant long before it actually becomes *dangerous* to be within sight of the waves. A hurricane isn't an event, it's a process. A *tornado* is an event... the storm quickly rolls in, the funnel drops from the sky (while the condensation cloud forms and makes it visible from the ground up), and it's *there*. Hurricanes? Hours, if not DAYS, of getting bitchslapped by rainbands that generally get worse and worse, until the eyewall FINALLY makes landfall and passes nearby (or directly above). Then, you get hit with a tornado-like windfield, have a few minutes of calm, another hit by the windfield (in the opposite direction), followed by a few more hours/days of the process in reverse.
Now... the authorities might certainly *prefer* that everyone just leave instead of making their jobs more complicated by having to manage tourists and traffic as a hurricane is approaching, but it's not like the beach suddenly becomes a lethal place to be just because a hurricane is approaching and will be there in a few hours.
The local road network matters, too. In Southwest Florida, yeah... there's a ~25 mile stretch of beach between Bonita Springs and Fort Myers with a 2-lane bridge at each end, another 10-15 miles of traffic just to get to I-75, and literally no other route to the mainland in between. Getting everyone "off the islands" is a major exercise in logistics. Contrast that with South Beach, where there's a 6-lane freeway and multiple other roads connecting it to the mainland, and a quarter of the population drives ashore every weekday over the span of 2-3 hours just getting to work. If anything, at the "Miami" end, the problem isn't evacuating Miami Beach to the mainland... it's the fact that people who live in western Dade County -- many of whom vividly remember being lied to by the government(*) -- trying to evacuate to Orlando (or worse, get frustrated when northbound I-75, Turnpike, and I-95 turn into parking lots, and head WEST across I-75 towards Naples, not realizing that they're going to encounter even WORSE northbound traffic on 75 long before they even make it across the county line).
(*)When Andrew hit, the county authorities were ADAMANT that it was going to directly hit South Beach, because they wanted to scare people who lived there into evacuating. The National Hurricane Center wasn't happy about it, but grudgingly went along with it because the county's rationale sounded reasonable. The problem was, it was a total lie from the first press release, and they knew it. As a result, people who lived in places like northern Coral Gables, Coconut Grove, Brickell, and downtown Miami fled SOUTH to the homes of friends and family members... and ended up evacuating INTO the area that was the hardest-hit. NOBODY who remembers Andrew trusts the authorities anymore. It's 80% of the reason why I learned how to run GFS myself and got into stormchasing in the first place... so I could independently run my own sims and do my own fact-checking instead of being forced to take the authorities' word at face value and hope they weren't lying again.
Connecting HA gear to the internet in a way that's both secure and works (especially with the manufacturer's own Android/iOS/ActiveX software) is actually pretty hard to do with real-world equipment, mainly because the overwhelming majority of stuff that's affordable (and shocking amounts of stuff that's supposedly top of the line) gets implemented with little more than symbolic security that's the equivalent of a TSA lock.
One of the most common ways embedded hardware gets connected to the Internet is via Wiznet w5100 modules (and variants). Basically, the w5100 is a bridge between UDP and a serial port. Data arrives on some port, and gets blindly sent to the serial port. Data comes in through the serial port, and gets sprayed out via UDP. There's some minimal logic that implements a half-assed 8-character "password" that -- at best -- might be equivalent to a 64-bit random number IF you hacked their setup utility to accept arbitrary byte values instead of just letters and digits (effectively turning it into more like a 40-bit random value). Keep in mind that there's no rate-limiting or lockout, so the only limiting factor at which someone can try to bruteforce you is your internet connection.
A few months ago, I estimated that an attacker who knows you have something specific behind a Wiznet interface that responds to a known command with a known response would take about 1-3 months to bruteforce if they kept the rate low enough to not noticeably affect your internet access or attract undue attention, and less than a weekend if they just all-out hammered you as fast as they could, trying only 8-character alphanumeric values and starting with those that begin with digits & plaintext English words.
That itself isn't necessarily the problem per se... there's nothing that says you can't encrypt the data being sent via UDP and in response, and implement stronger authentication and authorization checks on your own... except nothing actually DOES.
99.9% of the time, you have a circuit with almost no real network-level security that was developed with the assumption that someone with physical access to the serial port has already demonstrated some level of authorization, connected to a serial-ethernet bridge whose "security" is almost a complete sham, with predictable results: disaster.
Short answer: if you want to connect consumer gear over the internet, buy a Raspberry Pi, and use it as a middleware gateway device that accepts incoming connections via https, enforces its own strong authentication, passes no raw commands directly between the internet and embedded device, or at least requires that any raw value be signed with a pinned certificate. Then connect THAT to the embedded device through the Wiznet serial-ethernet adapter. Never, ever, EVER expose a serial port directly to the internet through a serial-ethernet adapter... I can almost guarantee that any such adapter that ISN'T built around a RPi and costs less than $200 is inherently insecure and a hack attack waiting to happen.
If you absolutely MUST expose some consumer-grade device with insecure ethernet-serial interface over the internet, at least hide it behind a router running OpenWRT/Tomato/dd-WRT and use something like port knocking and IP range-blocking to temporarily unblock access to your mobile device's current IP address for short periods of time when you intentionally enable it (keeping in mind that with many wireless providers, switching between HSPA/EVDO and LTE will radically change your IP address, and your address might change from request to request ANYWAY.
>That's the same reason smartphones don't have replaceable batteries much anymore:
> by the time the battery wears out, 99% of users are going to be looking for a new phone anyways
Within ~4 hours of purchase? That's about how long a stock battery *might* last for me... if I don't use it much.
User-replaceable batteries mean you can toss the wimpy 1800mAH toy into a drawer & snap in a nice, big, beefy 4,000-8,000mAH extended battery that *might* last until at least midnight. Or swap the stock battery for a spare every few hours as they die.
And unfortunately, eBook reading is STILL a completely *dreadful* experience for anything besides cover-to-cover novels and magazines.
E-ink takes an eternity to flip pages, tablets are eye-searing, low-contrast, low-res, or some combination... and both totally suck for random-access reference-type "flip back and forth between a few sections of the book" reading. I'm as guilty as anyone of buying eBooks when I need some book RIGHT NOW, but damn, I wish somebody could make ebooks not suck so miserably.
> Let's say patents were strong and eternal. This could easily lead to land grabs - there's a rush to discover new areas
> that people aren't even thinking about right now and bring them to market so that they can be patented. The value of a
> patent portfolio is high, and so companies may innovate to create such portfolios. Cross-licensing would be the order
> of the day in that environment, but it would hardly mean innovation would be dead.
No, it means we'd be in eternal stagnant IP feudalism, where only companies the size of Sony, Samsung, Apple, and Microsoft were allowed to innovate anything meaningful commercially. Everyone else would be -- at best -- IP sharecroppers forced to kneel before them and grant their patents for a pittance in return for the right to make commercial use of them at all, because just about any conceivable implementation would otherwise infringe upon two dozen other patents.
Think about it. When's the last time a huge corporation held by risk-averse institutional shareholders concerned with maximizing the next quarter's profits truly innovated *anything* that could be described as "disruptive"? The infant Microsoft would have been thrown into the well and drowned by IBM, DEC, Burroughs, Wang, and all the other companies that used to dominate the American computer industry back in the 1970s/80s. Woz would have gotten served a C&D 3 days after showing off his new prototype to the computer club. Commodore would have never existed, because its whole economic foundation was a semiconductor firm who built what was practically a 6800 clone. Fairchild *and* Shockley would have been eternal vassals of Bell Labs.
Big companies owned by institutional investors are almost genetically incapable of disruptive innovation, because it involves risk, and risk means the next quarter's profits might be disrupted. When alliances of similarly risk-averse companies have de-facto veto power over everyone else by virtue of strong patent protection, disruptive innovation stagnates.
In a way, the former Soviet Union is a perfect indirect example. In the Soviet Union, risk-averse bureaucrats had de-facto (if not de-jure) veto power over everything. The same way a company like Sony can unleash its lawyers to stop disruptive media technologies in their tracks in the holy name of IP law, the Soviet government could unleash its army of law enforcement officers and bureaucrats to prevent a Soviet citizen with a disruptive invention from ever doing anything with it. There was actually quite a bit of underground innovation in the Soviet Union... it's just that it was all basically masturbation, because their efforts were stymied at every level (by buraucratic indifference if they were lucky, by aggressive political backlash if they were perceived as presenting a risk to someone in a position of power, no matter how petty or meaningless).
Going back further to the dark ages, how much real innovation occurred when everyone was forced (by law, if not practical necessity) into being vassals under feudalism that subordinated everyone and everything to landed nobility and the Church? Yes, we had people like Kepler, Newton, and Galileo... and their contemporary influence was almost nonexistent. Their discoveries were talked about privately, behind closed doors, and most of their effort was spent trying to avoid getting crushed by those who cared mainly about preserving the status quo. They're revered today, centuries after their deaths, but in their day, they were basically viewed as godless heretics who deserved to burn for their sins, or at least spend a few decades repenting them in a dark prison cell somewhere.
People forget that the US itself was founded as a reaction to the old world order. Pre-Berne, American IP law was almost at a 90-degree angle to European IP law. European IP law viewed IP as a moral right, like being granted land by a sovereign king acting through the grace of God. American IP law, in stark contrast, was scandalously utilitarian. Up until t
> I wish there were some type of test suite that phone makers could offer to emulate their products,
There is... kind of... the pile of old Android phones most Android developers have accumulated by now ;-)
My pile: HTC HeroC, Samsung Epic4G, Motorola Photon, Galaxy S3 (current phone). Of the phones in the pile, only the Photon sits (battery removed) unloved and hated in a drawer, awaiting the day someone cracks that poor gimped phone's Motorola-permalocked 2.3.5 bootloader, breaks its chains & shackles, and lets it play Martin Luther King's "Free at Last!" exclamation while booting into Cyanogen.
> So if Google announced a major new version tomorrow how long would it be before I
> could exclusively target that version?
Depends upon whether your new app needs immediate sales, and whether you're talking about an expensive app that needs best-of-breed hardware to run acceptably well, anyway. If your app will suck on anything less than a GNex/S3/Note2/Nexus4/OneX, you won't lose many real sales by requiring 4.1, and probably won't lose many more by aiming for 4.2.
Just don't make the mistake of setting the hardware/version bar high, then releasing an app that sucks. Aim high, knock everybody's socks off, and you'll do OK. Aim low & avoid sucking, and you'll probably make some cash anyway. Aim high & suck, and you'll get buried under bad reviews & refunds.
^^^ Argh. replace 'corneal' with 'retinal'. It's late.