And in fact, seeing it portrayed is often the genesis of a real world deisgn. Go back to 2002's "Minority Report", where Tom Cruise is controlling a computer with his hands, waving arms around. Five years later, Apple's released the touch-based iPhone, Google's purchased the touch-based AndroidOS and is getting that ready to launch, Microsoft's working with the touch-based Surface. In fact, Apple representatives were claiming "It'll feel like 'Minority Report'" at the launch of the iPhone. (see http://en.wikipedia.org/wiki/Technologies_in_Minority_Report).
I wouldn't claim "Minority Report" as prior art.. if for no other reason than none of these things were good enough. But it's often some work of fiction, a presentation as a futurist conference, or some other thing that gets all sorts of engineers and designers, working independently, thinking in very much the same direction. When you factor in pre-existing UIs like the PalmOS, with it's rows of rounded-edge square icons, it's actually kind of surprising that the touch-screen phone and tablet OSs are as different from one another as they are.
There's a big difference between the whole thing and the very tiny pieces of that thing Apple has seen to patent. Of course it took years of real work to deliver an iPhone or iPad... or a Galaxy S2. In fact, it took Samsung the same level of hardware design work, even having seen the iPad (and the other similar tablets) before they did.
There is no requirement that a patent itself be the result of lots of hard work. And in fact, it's specifically precluded from being tied to commercial success... you have exactly one year from introducing your new thing in any way (previews, research papers, developers conferences, or actual shipments) in the USA to file the patents. All that's required is that your invention be an actual invention, something at the time unique. If it's too similar to other things that already exist, it's simply not that class of invention -- not patentable.
And a patent is another thing -- very, very specific. No one's claiming the sum of the many things that lead to your iPod experience were obvious looking at Fidler's work, the very iPhone-like Japanese phone that Samsung also introduced, etc. All that's at stake here are very specific, very individual patents, such as the shape of the iPad, entirely independent of what it does when you turn it on (that's the main thing being held against Samsung).
( the irony of your CPU name and your harsh view of Jobs was also not lost on me ) The 6502 was designed at MOS Technology by Chuck Peddle, and powered not only Commodore's own computers and the Apple ][, but a good deal of the early personal/home computers. There's no need for a 6502 fan to have any love for Jobs.
Yet, I can't get past this prior art thing... I know it's not an objective question, but... Apple could win this on one question to the jury: Had you ever seen anything like the iPhone when it came out?
That doesn't win it for them. The only basis for their lawsuit is the design patent. If real prior art is exposed here, doesn't matter if it was know to millions or just a handful, that could easily invalidate Apple's design patent. And keep in mind, it's not just Fidler's tablet being presented here... Samsung has a number of different people presenting visually similar devices.
It is necessary for the validity of a patent to be based on something unique and not obvious. If similar enough prior art exists, it doesn't have to be identical -- Apple's stuff compared to that might be obvious tweaks to the earlier design, and thus, not patentable.
That's correct -- Commodore was never sued by Apple for anything AmigaOS related. We did it all differently (not "different") enough. Apple had a few UI patents, but the basis for most of their lawsuits was the notion of a visual copyright. If you didn't look enough like MacOS, they couldn't sue. But Microsoft WAS a special case, due to the code sharing agreement.
In the case of the Mac emulators, it was very small potatoes... it's not as if Commodore were selling a MacOS solution. But it was kind of cool to see an A3000 running MacOS, and watch the disc performance benchmark go past all existing Macs, then off the screen.
IBM did sue Commodore over the Amiga. In those days, IBM was really concerned about being sued themselves, so they tried to get cross licenses with everyone in the PC industry. And of course, given the patents they had on the IBM PC, getting a license from a cloner was a no-brainer.... they charged the same for three patents as 3,000. They threw about 30 patents over the wall to Commodore, most of which either didn't apply (being very IBM PC specific -- they were kind of just guessing how various parts of the Amigas worked) or were obviously bad patents. One example of the latter: they claimed to have patented cut and paste between text buffers, in 1984. They demonstrated our infringement of this by showing MicroEmacs cut and paste between text buffers. I had to point out to the lawyers that the exact same key sequence would have worked on TECO Emacs, back in the 70s.
The thing about IBM, though... after you go through that first stack of 30, they can hit you with another, and another, and another. Early at IBM, patents were done like you might expect -- some engineer thinks they have something cool, they get the lawyers involved, and maybe a patent comes out of it. But in the early 80s, they had what amounted to a Patent Factory. Everything they did went to this huge office in Boca Raton, FL, where the lawyers had become experts at patenting anything that COULD be patented. They worked the system, knowing that the PTO had no software engineers (but were granting software patents), knowing that the PTO didn't do prior art searches outside of the pool of existing patents, etc.
No, actually, it's pretty easy to understand. People can be for personal freedom and against the "freedom" for powerful corporations to compromise individual freedom. As any here ought to understand well, there isn't simply one kind of freedom... the old "free as in beer" vs. "free as in speech" argument certainly applies. An ISP's freedom to "do as thou wilt" with their network will impede my freedom to use that network as I choose.... the ISPs have already guaranteed this to us.
And given that, in most areas, the ISP is already a monopoly (you're very lucky if you have more than one actual choice), I can't even exercise the freedom to choose. That's intentional... some of the big ISPs have unofficial non-complete agreements: we'll limit or stop build-outs in your area as long as you're doing the same for us.
You need to stop using that 1940's encyclopedia as your lone source of political information. The Democrats did support the racist "Dixiecrats" ages ago. But they embraced Civil Rights, starting in the very early 60s with John F. Kennedy and then formally in 1964 with Lyndon B. Johnson passing the Civil Rights Act, followed by the Voting Rights Act in 1965.
That's not to say the Southern Republicans were any better. The Civil Rights act was opposed by 93% of Southern Democrats and 100% of Southern Republicans in the House, 95% of Southern Democrats and 100% of Southern Republicans in the Senate. But this cost the Democrats the racists and the South.... and Johnson knew that going into it -- "I know the risks are great and we might lose the South, but those sorts of states may be lost anyway." That "South" was a reference to the "Solid South", which had voted Democratic since the Civil War (since, of course, The Republicans had been the Party of Lincoln). The racist element had been slowly moving Republican in the South since the 1930s, and this was basically the final big push of the Old South out of the Democratic Party.
This didn't actually have a profound impact on the 1964 re-election of Johnson, largely because Johnson and his opponent, Barry Goldwater, had and kept a gentlemen's agreement to leave racial politics out of the election. Think about that for a moment -- has there been any election in recent memory in which two candidates agreed to leave something alone for the benefit of the country? I can't think of any example.
However, racism was back on the table for the 1968 election, and Nixon milked it for all it was worth, formerly jettisoned the little remaining "Party of Lincoln" policies left in the Republican party, and very effectively leveraged racial fears to a very narrow victory. And don't forget (or do learn) that this was a three-way race... open racist George Wallace, backed by the openly racist former Dixiecrats, carried five states in the deep South. After the election, they pretty migrated to the Republican Party, and within ten years, were in leadership positions.
And the Republicans have been the Party of Racism and Old White Rich Guys ever since.
That is correct... T-Mo currently does voice and 2G at 1900MHz (they don't have an 850MHz slot), and 1700/2100MHz for 3G/HSPA/etc.
That may be changing, though. They're supposedly planning to phase out 2G entirely, move 3G/HSPA to the 1900MHz band (and... 2100MHz? There's not enough bandwidth at 1900MHz for a full 3G data connection), and then start up LTE at 1700MHz. This follows their deal with AT&T... after the merger failed to pass regulators, AT&T was required to pay T-Mobile a huge pile of cash. Much of that's being paid in AWS spectrum formerly owned by AT&T. Presumably, T-Mobile's getting enough to launch a viable LTE service at 1700MHz... I have not seen any fine detail on this yet.
Last I checked, the only phone in the US that has full 3G support for both AT&T and T-Mobile is the Google Galaxy Nexus. Not surprisingly, this is also the only phone really being sold as unlocked by default ($350 on the Google Play store). If you get a GSM phone through AT&T, it's pretty certain to now support T-Mobile's AWS bands (1700MHz and 2100MHz).
AT&T has or at least had coverage problems in some areas. Before AT&T was bought by Cingular, and then re-named back to AT&T, there actually was the AT&T Wireless company. They used a radio protocol dubbed "TDMA"... that's just Time-Division Multiple Access, same scheme used by GSM 2G. But not compatible. The real name of their protocol was DAMPS: Digital Advanced Mobile Phone Service. AMPS was the US analog phone system; DAMPS was supposed to the upgrade.
Anyway, AT&T used DAMPS, and when Cingular took over, they transitioned to GSM, finally shutting down the DAMPS network in 2008 or so. Former DAMPS towers became GSM/HSPA towers. But there was one problem: DAMPS had a greater useful range than GSM. So a AT&T area with perfect DAMPS coverage would have holes in it, once converted to GSM. The only solution: add more towers. Which works, but the problem is, that's a worst-case need for a new tower... to boost the smallest possible hole in coverage. So some of these holes still exist.
Now hold on a minute...as an actual resident of the boondocks, let me 'splain something. A city with 540K people isn't anything close to the boondocks. That's a thriving metropolis. Sure, there are even more thriving metropolia.
I live in a town that's about 40 square miles, with under 3,500 people (http://en.wikipedia.org/wiki/Upper_Pittsgrove_Township,_New_Jersey). That's in New Jersey, the most densely populated state in the USA. So this is really just an edge-case boondocks. Think about the middle of the country. And yet, we do have towers in the area, supported by all four major carriers. But only Verizon and AT&T are actually acceptable at my house (Verizon sometimes actually makes it to the cellar). No 4G yet, but no big shock -- and we'll never see it from T-Mobile (even when they do actually offer real 4G), probably not from Sprint.
T-Mobile is certainly unacceptable in a real boondocks. So is Sprint. The main problem is distance and foliage. T-Mobile and Sprint only have high frequency bands, other than Sprint's old Nextel (now being converted to LTE) bands. The Nextel stuff is in the 800MHz band, which should do pretty well when compared to AT&T and Verizon on 700MHz for LTE and 850MHz (as an option) for voice and data (though AT&T's also stuck using 1900MHz for 3G, since there's not enough bandwidth at 850MHz, HSPA needs 5MHz each way, while EvDO does two-way data using only 2.5MHz, same as voice). Sprint's Wi-Max is at 2500MHz, which makes it a few feet into a forest. T-Mo's got 3G on the AWS bands, 1700MHz and 2100MHz -- they have plans to move 3G to 1900MHz and put LTE at 1700MHz, but that's still not going to work well in the boondocks. And that's just foliage (sure, less of an issue in Kansas, I suppose)... the free-space loss at higher frequencies also mandates more towers to cover the same area.
Well, they did actually improve Unity in 12.04. I had the same impressions. Well, close... I'm not a huge fan of Unity, but at least it's not as forced as in 11.04. I had 11.04 on a test machine for a year or so, and I did come close to breaking it out of frustration. No Windows -- what's that all about? Not sure if there was an easy fix to the whole full-screen disease in 11.04, but it's definitely the case that 12.04 doesn't come that way out of the box. The dock... well, all docks suck, one way or another. This one's not substantially worse than others I've used, once you get used to it.
And at least with Linux, you do have a choice. Even Ubuntu -- you can run Gnome or KDE or whatever you want. As an expert, not a problem. But also one of the things keeping Linux as an experts-only OS, mostly. It would take some big company doing a push like Google did with Android to see Linux on the desktop take off as even a minor threat to Windows. I've felt this way for years, only now, with Windows getting so dramatically worse with Windows 8, it's a real shame that's not happening. There is a Window (sic) for it.
I thought Win2K was just dandy... I used it well into the XP years.
I guess the main complaint was, over time, Microsoft never kept it up-to-date on the key consumer technologies like DirectX, because, well, it wasn't being sold as a consumer OS. So you could use it for work all day, professional CAD and video and audio applications ran on it fine, but not the latest video games. I don't know of any other major complaint. I even downgraded a laptop I bought in XP days to Win2K, due to some evil issue or another they didn't address until a service pack.
I did get XP on my Win2K machine, ultimately, and I recall it being forced, for some hardware or another that was never going to be supported on Win2K.
He's got that weird flavor of US Libertarians who, like Ryan, idolize Any Rand, but couldn't tell you the difference between John Stuart Mill and a Wind Mill. But I think Romney had them, or at least those who don't toss away their vote on the actual US Libertarian Party.
Of course, like most fringers, Ryan manages to claim to be both an Objectivist (or something like it) and a way far right Catholic. I mean, he's on the fringe of that group... denies climate change, he's sponsored both anti-gay and Fetus Personhood amendments (yeah, to the Constitution) as a Congresscritter. His stand on personal freedoms would chill the blood of any real Libertarian.
The one thing I did appreciate about Rand... she understood there's no place for religion and other superstitions in a modern society. No good ever came of them.
Maybe Romney will try to paint himself as more of a moderate now?
That was the popular theory up until the point at which he chose Paul Ryan. Mittens could easily have swung to the middle -- most of his actual far right talk has, after all, just been talk. That's not the way he functioned as governor. He would have had little trouble picking up fence sitters and those just tired of Obama.
With Ryan on the ticket, though, he's made something of a promise to the lunatic fringe of the Tea Party. And of course, ensure even more money from the Koch Brothers -- Ryan's their boy. No good can come of this, but it sure makes things more interesting. And I'd have bet good money that nothing Romney ever has or did in his life would really qualify as interesting. Well, maybe the way he transports his dog....
You probably wouldn't want Windows on that machine, anyway... not that you have the choice. Only OEMs are getting Windows for ARM. It's as locked down as an iPad.
Thing is, you're completely correct on the software pricing thing. Microsoft isn't giving this stuff away, and they're going to want typical MS licensing fees. In fact, at least to date, they're only offering Windows on ARM bundled with Office. That's going to be a significant chunk of change for anyone building a tablet and pricing it at a point that'll actually sell (eg, much less than an iPad, unless they have "luxury branding" that can compete with Apple). So MS is already setting up a 20% or so price/profit advantage between their tablets and any other Windows tablets. Makes them potentially more competitive with Apple.. only, the "Microsoft" brand absolutely doesn't have that luxury branding in the CE market -- no one's going to pay Apple prices for a Microsoft product. Maybe it works better in some businesses?
I agree... keep Ballmer where he is. He's doing so much good for everyone who's NOT Microsoft (well... and Nokia, though really, that level of stupidity came primarily from Elop, not Ballmer), it would be risky to remove him.
And they do seem very much to be doing for the desktop, with Windows 8, what they did somehow with Windows Mobile. The one real problem, though: where's the alternative? WinMo sucked and didn't evolve, sure. But it also took the iPhone and Android to really clobber Microsoft.
With MacOS both itself evolving in bad ways, and being an Apple product, that can't be a large scale alternative.
And sure, there's desktop Linux. But there had been handtop Linux as well, lots of them, in fact, and they didn't really go anywhere, either. Linux, at least as presented in Android, won the handset simply because there was just one binary applications socket (eg, no need for source distribution) and, of course, it was all very consumer friendly. That's very not the case with Linux right now. And every time it looks like something might happen in that direction, bad things happen. Like Ubuntu going all weird last year. Without a 900lbs gorilla like Google pushing the buttons, I don't think desktop Linux will ever be a serious threat to Windows.
Enemy of my enemy is my friend? Nook is a Barnes and Noble tablet, of course, not an Amazon product.
But Amazon isn't directly in competition with Wal-Mart on many things. Wal-Mart doesn't really try on books, for example; they have high prices and small selection.
I have a tablet with a Pixel Qi screen -- not as sweet as eInk, but perfectly readable at noon on a beach. This kind of thing will eventually go mainstream, as the daylight-readable technologies get more comparable to regular LCDs when used indoors (at present, the Pixel Qi with backlight on is a bit unsaturated, more like a magazine than a computer monitor).
A 3G chipset costs about $25 in volume. A bit more for 4G. But keep in mind, the chip companies make over a billion of these things every year. They get crazy volume on them. The prices have been dropping fast. One reason the newer chips have multi-procotol and 4G -- as a chip company (Qualcomm, for example), you have to offer some coolness that's not such a commodity, to get any sort of premium for your particular phone chips.
The carriers don't pay anywhere near retail for the phones they supply -- they have a negotiated percentage off the MSRP. That's why the MSRPs are so high -- this keeps customers from buying their own devices. It's completely artificial. But the OEMs have little choice; if you don't sell via a carrier, you may not sell much at all. Google does it, with the GSM/HSPA+ version of the Galaxy Nexus -- lower price ($349), but they're not retailing it via AT&T or T-Mobile, either.
I predict they cancel all iPods. They'll have a new thing that looks like it ought to be called the iPod Touch, but built on the iPhone 5 technology, including 4" screen. This will, instead, be dubbed the iPad Nano, and because of the name and the simple fact Apple's all about the marketing, they'll up the base MSRP to around $250. The new iPad, of course, remains at an entry level $499 price, and sure, the new-old stock iPad 2 is still around, but that's not necessarily going to remain so. The 7" iPad Mini will go out at around $350-375.
Apple isn't going to price these low. For one thing, they don't need to -- they're Apple. They're the only real Luxury band left in consumer electronics, given the bad fortunes of Sony in recent times. Think back to the iPad introduction -- long time Apple pundits saw that as the end of Apple-as-they-knew-it, because the iPad cost half the price of the lower-end MacBooks.
On the other hand, how much should it cost? An iPad or similar tablet starts with the same screen you'll find in a netbook, more or less. You take out RAM, you take out storage, you take out ports, cut the battery size in half, toss out an Intel Atom and put in a cheaper ARM SOC, toss the keyboard, put it in a cheaper case. The only additional expense is a touchscreen (though some netbooks have these), about $25 cost if you're not buying in Apple quantities.
Another datapoint: Apple's selling a boatload of iPhones and iPads, not so many Macs. And yet, they did a gross margin of 45% last quarter. So they're not even close to dropping markets on "i" devices, relative to their perhaps more obviously overpriced PCs.
As for the iPhone, like every smartphone or even dumb phone, the MSRP is total fiction. I mean, how could Nokia sell a dumb candybar phone for $150? Well... they don't. The MSRP is a reaction to the fact that Apple's customer for the iPhone isn't you or me, it's Verizon or AT&T. Like every other phone sold, Apple's dependent on these carriers to re-sell the phones, and get them into customer hands. They'd have a hard-to-impossible time doing their on their own, without the phones being featured in the carriers' stores. The carriers, meanwhile, want you to sign in blood for two years to get your cheap, subsidized phone. And they want you to not want to buy one without a contract, and to think you're getting a great deal on the hardware. So they buy phones from Apple, Motorola, HTC, Samsung, etc. based on a percentage of the MSRP. A fairly small percentage. If Apple priced the iPhone at the same 40-50% margin they price the iPod Touch at, they'd basically be paying AT&T and Verizon to take the iPhone. So they make a profit, and that results in a crazy high MSRP.
Here's a good way to think about it, since Apple makes this easy. At any given technology node so far, the iPod Touch is simply an iPhone with some stuff left off. What's left off? Depends on the model, certainly, but for one, the cellular modem. That's about $25 in cost if you're not Apple -- I'm sure Apple pays less. On the current version, the camera isn't as good, so that's another $3 or so. No microphone, that's under $0.50. The iPhone has a larger battery, another $5. So figure less than $40 difference in cost, so that's $60-$80 MSRP tops. If they could sell an iPod Touch for $230, an iPhone of the same generation and flash capacity ought to run $320 or less. But it doesn't, for the reasons stated.
Is there another example? Sure is... the Galaxy Nexus. That's a smartphone at the same basic technology node as the iPhone 4S (the 4S has a faster GPU and better camera, the Nexus a faster CPU and better screen). Google, making far fewer of these than Apple, sells them direct on the Google Play store for $399. That's an unlocked GSM/HSPA model, also just like the iPhone, but unlike the the iPhone or pretty much any other phone, it fully supports both AT&T and T-Mobile (2G on 850MHz or 1900MHz, 3G on 850MHz, 1700MHz, 1900MHz, and 2100MHz is various possible combinations...and maybe even the European frequencies, 900MHz and 1800MHz). The Verizon version of this original sold, on contract, for $299, and MSRPed at around $600. But Google's not trying to sell to AT&T or T-Mobile, so they can do this.
If your carrier is still running GSM-only cells, they're well behind the usual curve, at least for the USA. The CDMA2000 carriers (Sprint and Verizon) pretty much had their networks all 3G nearly a decade ago. And it's actually more important on GSM... 2G uses hard-carrier handoffs, which is where all those dropped calls come from. UMTS/HSPA uses the same soft handoffs used by CDMA2000, so, less in the way of dropped calls.
If they're talking about HSPA+ upgrades finished by Spring 2013, this must be T-Mobile, and much of that's going to be HSPA->HSPA+ upgrades. Pretty much everyone else is spending their upgrade money on LTE. Supposedly (and given where I live, I'll be able to tell you), Verizon will have every cell upgraded to LTE sometime in 2013. AT&T may be a bit behind, Sprint's further still, but they're making progress. LTE is particularly important to a carrier like Sprint, since like T-Mobile, they've been at a frequency disadvantage against Verizon and AT&T. Sprint's CDMA2000/EvDO is only 1900MHz (like T-Mo's GSM 2G), and their WiMax (in partnership with Clear and Comcast) is at 2500MHz, even worse for long distances. Their LTE is going out on the old 800MHz Nextel band, so they're getting much the same range and penetration (foliage, building) advantage as the two big guys.
T-Mobile is at a big disadvantage, since they only have 3G on 1700/2100MHz (AWS band) as well as their 2G at 1900MHz. They got some money from the collapse of the AT&T merger, which AT&T is paying back partially in AWS spectrum. So sometimes next year, T-Mobile will be starting to build LTE out in the AWS band and push UMTS/HSPA (and HSPA+, if there's actually room for it) into the more crowded 1900MHz (PCS) band.... that's their incentive for dumping 2G entirely, more than anything. So their getting a little worse before they get any better. And they're just going to have to live with the frequency disadvantage.
In fact, they're pretty even. Apple's gross profits last quarter were about 45%. The various telcos have been complaining, lately, about their profits not always hitting the 45% margin. Coming from the PC industry, where a good profit margin is more like 10-15% if you're lucky (and often much less), it's pretty obvious why neither telcos nor Apple are seen as offering much value for the money.
At least, when you buy an Apple product, you pretty much know what you're getting, and that next year's will be a little better than this year's. The telcos keep inventing ways to charge you more for less, like most of them nixing unlimited data plans, and now this year, charging more for a 1GB of shared data (with no carryover) than they did for unlimited data back in the day (or for those of us grandfathered in).
They're also the folks -- the telecoms, not Apple or Samsung, responsible for the crazy inflated pricing on smartphones sold without a contract. A high enough smartphone will probably retail at $300-$400, tops, if it were priced the way most any other consumer electronics item is priced. But the telcos force these $550-$750 MSRPs, so that the average Joe will see the $200-$300 price they pay on-contract as a good deal.
All smartphones these days run the cellular baseband on a separate, small ARM SOC, not the main application processor. That code, like any other code, is subject to the occasional bug. It's also quite possible the same Qualcomm (or whoever) code that's running on everyone else's phone with the same wireless chip. The application level OS can be smart about detecting a baseband crash and rebooting the baseband chip. But it's not always obvious. So yeah, you're going to have occasional reboots for "no signal" on practically any smartphone. Unless you're lucky enough to get one with a bug-free-enough baseband.
This thing never happened on my O.G. Droid. It's happened two or three times on my Galaxy Nexus. No biggie. In the dumb phone days, your phone would have probably hit a watchdog timeout of some kind and just rebooted -- you might never have even noticed it.
I double majored in EE and Mathematics.. I had something like six really math heavy courses, starting with Calc2, and that's not counting some of the applied courses in EE. I didn't use much of that for 15 years, but I did eventually. It's an important background, because you just never know.
If you're only interested in learning a specific kind of programming, that's for a technical school, not a University. But the thing is, what was hot in the early 80s when I was in school isn't so hot anymore. Teaching a programming language, that's a technical school thing at best -- it should really be a couple of days, tops, for an experienced software engineer (I had to learn a few of them on my last SW project).
A University degree should give you the mental tools to adapt to any new job requirement. Math is an important part of that. The computer languages I learned in college (Pascal, LISP, SNOBOL4, APL, InterCal:-), etc).. not terribly in demand today. The math... calculus and other numerical methods are a bit easier to do, thanks to computer tools, but they're just as applicable today.
Games sure. But also embedded control systems. Robotics. Cryptographic systems. Signal processing: digital video, digital audio, software radio, etc. I think the need for "real math" is on the rise.
When I got out of school, I got into computer systems design (things like the Commodore Amiga 2000)... not a great deal of complex math in there. But I actually did double major, EE and Mathematics, so I had that toolbox of stupid math tricks available, once things got a little more demanding than just calculating worse-case timing and power budgets.
Which they did. One thing I did, which many folks think is cool (and it is, particularly if you don't care about getting paid that well)... I got into startups. When you're at a startup, you find you wear just about any hat that fits, and you also get pretty good at tailoring those that don't. Having a proper background in math will definitely help you at this. And it's not unusual, particularly if you're a hardware guy at a small startup, that you're expected to solve EVERY possible hardware problem... or at least know it well enough that you can solve things with the occasional help of a consultant.
So since Commodore, I've done control systems and sensor systems (robotics), digital radio systems (R/C cars, more robotics, general purpose mesh routers). I wrote an NTSC decoder for a software radio.... all sorts of different math in there (FFTs, filters, etc)... and knowing math well enough is critical to building things like this well... you can find dozens of books on the analog TV signals, but doing that all in software, you really need to understand things. I've also been working in cryptography, more SDR, and other kinds of signal processing... even wrote a few audio plug-ins for my own use, which may one day get released somewhere.
Of course, with great tools like MathCAD, you may not find yourself actually "doing" the math they way you might have back in school. No pencil and paper, you have tons of primitives that do complex things for you. But you need to understand how it all works to able to use such tools.. and without that knowledge, you'll be unable to adapt. The other guy who wants that cool new job won't have that limitation.
And in fact, the very best thing I learned out of my more-expensive-than-Dad-had-budgeted college (and those extra ten years of monthly payments) was the ability to adapt. The specifics we learned in the early 1980s (Pascal and LISP programming, TTL logic, etc) all but vanished. The big ideas don't. An education shouldn't worry too much about today's specifics... you have no idea today what kind of engineering you'll be doing in 5, 10, 20 years. Even if you don't change jobs, the nature of the work can change out from under you. I went from HW design last summer to a software project (and the need to learn three new languages), then back to hardware this summer -- with some pretty deep mathematics in the mix now.
Slashdot Mirror
Yup.
And in fact, seeing it portrayed is often the genesis of a real world deisgn. Go back to 2002's "Minority Report", where Tom Cruise is controlling a computer with his hands, waving arms around. Five years later, Apple's released the touch-based iPhone, Google's purchased the touch-based AndroidOS and is getting that ready to launch, Microsoft's working with the touch-based Surface. In fact, Apple representatives were claiming "It'll feel like 'Minority Report'" at the launch of the iPhone. (see http://en.wikipedia.org/wiki/Technologies_in_Minority_Report).
I wouldn't claim "Minority Report" as prior art.. if for no other reason than none of these things were good enough. But it's often some work of fiction, a presentation as a futurist conference, or some other thing that gets all sorts of engineers and designers, working independently, thinking in very much the same direction. When you factor in pre-existing UIs like the PalmOS, with it's rows of rounded-edge square icons, it's actually kind of surprising that the touch-screen phone and tablet OSs are as different from one another as they are.
There's a big difference between the whole thing and the very tiny pieces of that thing Apple has seen to patent. Of course it took years of real work to deliver an iPhone or iPad... or a Galaxy S2. In fact, it took Samsung the same level of hardware design work, even having seen the iPad (and the other similar tablets) before they did.
There is no requirement that a patent itself be the result of lots of hard work. And in fact, it's specifically precluded from being tied to commercial success... you have exactly one year from introducing your new thing in any way (previews, research papers, developers conferences, or actual shipments) in the USA to file the patents. All that's required is that your invention be an actual invention, something at the time unique. If it's too similar to other things that already exist, it's simply not that class of invention -- not patentable.
And a patent is another thing -- very, very specific. No one's claiming the sum of the many things that lead to your iPod experience were obvious looking at Fidler's work, the very iPhone-like Japanese phone that Samsung also introduced, etc. All that's at stake here are very specific, very individual patents, such as the shape of the iPad, entirely independent of what it does when you turn it on (that's the main thing being held against Samsung).
( the irony of your CPU name and your harsh view of Jobs was also not lost on me )
The 6502 was designed at MOS Technology by Chuck Peddle, and powered not only Commodore's own computers and the Apple ][, but a good deal of the early personal/home computers. There's no need for a 6502 fan to have any love for Jobs.
Yet, I can't get past this prior art thing... I know it's not an objective question, but...
Apple could win this on one question to the jury: Had you ever seen anything like the iPhone when it came out?
That doesn't win it for them. The only basis for their lawsuit is the design patent. If real prior art is exposed here, doesn't matter if it was know to millions or just a handful, that could easily invalidate Apple's design patent. And keep in mind, it's not just Fidler's tablet being presented here... Samsung has a number of different people presenting visually similar devices.
It is necessary for the validity of a patent to be based on something unique and not obvious. If similar enough prior art exists, it doesn't have to be identical -- Apple's stuff compared to that might be obvious tweaks to the earlier design, and thus, not patentable.
That's correct -- Commodore was never sued by Apple for anything AmigaOS related. We did it all differently (not "different") enough. Apple had a few UI patents, but the basis for most of their lawsuits was the notion of a visual copyright. If you didn't look enough like MacOS, they couldn't sue. But Microsoft WAS a special case, due to the code sharing agreement.
In the case of the Mac emulators, it was very small potatoes... it's not as if Commodore were selling a MacOS solution. But it was kind of cool to see an A3000 running MacOS, and watch the disc performance benchmark go past all existing Macs, then off the screen.
IBM did sue Commodore over the Amiga. In those days, IBM was really concerned about being sued themselves, so they tried to get cross licenses with everyone in the PC industry. And of course, given the patents they had on the IBM PC, getting a license from a cloner was a no-brainer.... they charged the same for three patents as 3,000. They threw about 30 patents over the wall to Commodore, most of which either didn't apply (being very IBM PC specific -- they were kind of just guessing how various parts of the Amigas worked) or were obviously bad patents. One example of the latter: they claimed to have patented cut and paste between text buffers, in 1984. They demonstrated our infringement of this by showing MicroEmacs cut and paste between text buffers. I had to point out to the lawyers that the exact same key sequence would have worked on TECO Emacs, back in the 70s.
The thing about IBM, though... after you go through that first stack of 30, they can hit you with another, and another, and another. Early at IBM, patents were done like you might expect -- some engineer thinks they have something cool, they get the lawyers involved, and maybe a patent comes out of it. But in the early 80s, they had what amounted to a Patent Factory. Everything they did went to this huge office in Boca Raton, FL, where the lawyers had become experts at patenting anything that COULD be patented. They worked the system, knowing that the PTO had no software engineers (but were granting software patents), knowing that the PTO didn't do prior art searches outside of the pool of existing patents, etc.
No, actually, it's pretty easy to understand. People can be for personal freedom and against the "freedom" for powerful corporations to compromise individual freedom. As any here ought to understand well, there isn't simply one kind of freedom... the old "free as in beer" vs. "free as in speech" argument certainly applies. An ISP's freedom to "do as thou wilt" with their network will impede my freedom to use that network as I choose.... the ISPs have already guaranteed this to us.
And given that, in most areas, the ISP is already a monopoly (you're very lucky if you have more than one actual choice), I can't even exercise the freedom to choose. That's intentional... some of the big ISPs have unofficial non-complete agreements: we'll limit or stop build-outs in your area as long as you're doing the same for us.
You need to stop using that 1940's encyclopedia as your lone source of political information. The Democrats did support the racist "Dixiecrats" ages ago. But they embraced Civil Rights, starting in the very early 60s with John F. Kennedy and then formally in 1964 with Lyndon B. Johnson passing the Civil Rights Act, followed by the Voting Rights Act in 1965.
That's not to say the Southern Republicans were any better. The Civil Rights act was opposed by 93% of Southern Democrats and 100% of Southern Republicans in the House, 95% of Southern Democrats and 100% of Southern Republicans in the Senate. But this cost the Democrats the racists and the South.... and Johnson knew that going into it -- "I know the risks are great and we might lose the South, but those sorts of states may be lost anyway." That "South" was a reference to the "Solid South", which had voted Democratic since the Civil War (since, of course, The Republicans had been the Party of Lincoln). The racist element had been slowly moving Republican in the South since the 1930s, and this was basically the final big push of the Old South out of the Democratic Party.
This didn't actually have a profound impact on the 1964 re-election of Johnson, largely because Johnson and his opponent, Barry Goldwater, had and kept a gentlemen's agreement to leave racial politics out of the election. Think about that for a moment -- has there been any election in recent memory in which two candidates agreed to leave something alone for the benefit of the country? I can't think of any example.
However, racism was back on the table for the 1968 election, and Nixon milked it for all it was worth, formerly jettisoned the little remaining "Party of Lincoln" policies left in the Republican party, and very effectively leveraged racial fears to a very narrow victory. And don't forget (or do learn) that this was a three-way race... open racist George Wallace, backed by the openly racist former Dixiecrats, carried five states in the deep South. After the election, they pretty migrated to the Republican Party, and within ten years, were in leadership positions.
And the Republicans have been the Party of Racism and Old White Rich Guys ever since.
That is correct ... T-Mo currently does voice and 2G at 1900MHz (they don't have an 850MHz slot), and 1700/2100MHz for 3G/HSPA/etc.
That may be changing, though. They're supposedly planning to phase out 2G entirely, move 3G/HSPA to the 1900MHz band (and ... 2100MHz? There's not enough bandwidth at 1900MHz for a full 3G data connection), and then start up LTE at 1700MHz. This follows their deal with AT&T... after the merger failed to pass regulators, AT&T was required to pay T-Mobile a huge pile of cash. Much of that's being paid in AWS spectrum formerly owned by AT&T. Presumably, T-Mobile's getting enough to launch a viable LTE service at 1700MHz... I have not seen any fine detail on this yet.
Last I checked, the only phone in the US that has full 3G support for both AT&T and T-Mobile is the Google Galaxy Nexus. Not surprisingly, this is also the only phone really being sold as unlocked by default ($350 on the Google Play store). If you get a GSM phone through AT&T, it's pretty certain to now support T-Mobile's AWS bands (1700MHz and 2100MHz).
AT&T has or at least had coverage problems in some areas. Before AT&T was bought by Cingular, and then re-named back to AT&T, there actually was the AT&T Wireless company. They used a radio protocol dubbed "TDMA"... that's just Time-Division Multiple Access, same scheme used by GSM 2G. But not compatible. The real name of their protocol was DAMPS: Digital Advanced Mobile Phone Service. AMPS was the US analog phone system; DAMPS was supposed to the upgrade.
Anyway, AT&T used DAMPS, and when Cingular took over, they transitioned to GSM, finally shutting down the DAMPS network in 2008 or so. Former DAMPS towers became GSM/HSPA towers. But there was one problem: DAMPS had a greater useful range than GSM. So a AT&T area with perfect DAMPS coverage would have holes in it, once converted to GSM. The only solution: add more towers. Which works, but the problem is, that's a worst-case need for a new tower ... to boost the smallest possible hole in coverage. So some of these holes still exist.
Now hold on a minute...as an actual resident of the boondocks, let me 'splain something. A city with 540K people isn't anything close to the boondocks. That's a thriving metropolis. Sure, there are even more thriving metropolia.
I live in a town that's about 40 square miles, with under 3,500 people (http://en.wikipedia.org/wiki/Upper_Pittsgrove_Township,_New_Jersey). That's in New Jersey, the most densely populated state in the USA. So this is really just an edge-case boondocks. Think about the middle of the country. And yet, we do have towers in the area, supported by all four major carriers. But only Verizon and AT&T are actually acceptable at my house (Verizon sometimes actually makes it to the cellar). No 4G yet, but no big shock -- and we'll never see it from T-Mobile (even when they do actually offer real 4G), probably not from Sprint.
T-Mobile is certainly unacceptable in a real boondocks. So is Sprint. The main problem is distance and foliage. T-Mobile and Sprint only have high frequency bands, other than Sprint's old Nextel (now being converted to LTE) bands. The Nextel stuff is in the 800MHz band, which should do pretty well when compared to AT&T and Verizon on 700MHz for LTE and 850MHz (as an option) for voice and data (though AT&T's also stuck using 1900MHz for 3G, since there's not enough bandwidth at 850MHz, HSPA needs 5MHz each way, while EvDO does two-way data using only 2.5MHz, same as voice). Sprint's Wi-Max is at 2500MHz, which makes it a few feet into a forest. T-Mo's got 3G on the AWS bands, 1700MHz and 2100MHz -- they have plans to move 3G to 1900MHz and put LTE at 1700MHz, but that's still not going to work well in the boondocks. And that's just foliage (sure, less of an issue in Kansas, I suppose)... the free-space loss at higher frequencies also mandates more towers to cover the same area.
Well, they did actually improve Unity in 12.04. I had the same impressions. Well, close... I'm not a huge fan of Unity, but at least it's not as forced as in 11.04. I had 11.04 on a test machine for a year or so, and I did come close to breaking it out of frustration. No Windows -- what's that all about? Not sure if there was an easy fix to the whole full-screen disease in 11.04, but it's definitely the case that 12.04 doesn't come that way out of the box. The dock... well, all docks suck, one way or another. This one's not substantially worse than others I've used, once you get used to it.
And at least with Linux, you do have a choice. Even Ubuntu -- you can run Gnome or KDE or whatever you want. As an expert, not a problem. But also one of the things keeping Linux as an experts-only OS, mostly. It would take some big company doing a push like Google did with Android to see Linux on the desktop take off as even a minor threat to Windows. I've felt this way for years, only now, with Windows getting so dramatically worse with Windows 8, it's a real shame that's not happening. There is a Window (sic) for it.
I thought Win2K was just dandy... I used it well into the XP years.
I guess the main complaint was, over time, Microsoft never kept it up-to-date on the key consumer technologies like DirectX, because, well, it wasn't being sold as a consumer OS. So you could use it for work all day, professional CAD and video and audio applications ran on it fine, but not the latest video games. I don't know of any other major complaint. I even downgraded a laptop I bought in XP days to Win2K, due to some evil issue or another they didn't address until a service pack.
I did get XP on my Win2K machine, ultimately, and I recall it being forced, for some hardware or another that was never going to be supported on Win2K.
He's got that weird flavor of US Libertarians who, like Ryan, idolize Any Rand, but couldn't tell you the difference between John Stuart Mill and a Wind Mill. But I think Romney had them, or at least those who don't toss away their vote on the actual US Libertarian Party.
Of course, like most fringers, Ryan manages to claim to be both an Objectivist (or something like it) and a way far right Catholic. I mean, he's on the fringe of that group... denies climate change, he's sponsored both anti-gay and Fetus Personhood amendments (yeah, to the Constitution) as a Congresscritter. His stand on personal freedoms would chill the blood of any real Libertarian.
The one thing I did appreciate about Rand... she understood there's no place for religion and other superstitions in a modern society. No good ever came of them.
Maybe Romney will try to paint himself as more of a moderate now?
That was the popular theory up until the point at which he chose Paul Ryan. Mittens could easily have swung to the middle -- most of his actual far right talk has, after all, just been talk. That's not the way he functioned as governor. He would have had little trouble picking up fence sitters and those just tired of Obama.
With Ryan on the ticket, though, he's made something of a promise to the lunatic fringe of the Tea Party. And of course, ensure even more money from the Koch Brothers -- Ryan's their boy. No good can come of this, but it sure makes things more interesting. And I'd have bet good money that nothing Romney ever has or did in his life would really qualify as interesting. Well, maybe the way he transports his dog....
You probably wouldn't want Windows on that machine, anyway... not that you have the choice. Only OEMs are getting Windows for ARM. It's as locked down as an iPad.
Thing is, you're completely correct on the software pricing thing. Microsoft isn't giving this stuff away, and they're going to want typical MS licensing fees. In fact, at least to date, they're only offering Windows on ARM bundled with Office. That's going to be a significant chunk of change for anyone building a tablet and pricing it at a point that'll actually sell (eg, much less than an iPad, unless they have "luxury branding" that can compete with Apple). So MS is already setting up a 20% or so price/profit advantage between their tablets and any other Windows tablets. Makes them potentially more competitive with Apple.. only, the "Microsoft" brand absolutely doesn't have that luxury branding in the CE market -- no one's going to pay Apple prices for a Microsoft product. Maybe it works better in some businesses?
I agree... keep Ballmer where he is. He's doing so much good for everyone who's NOT Microsoft (well... and Nokia, though really, that level of stupidity came primarily from Elop, not Ballmer), it would be risky to remove him.
And they do seem very much to be doing for the desktop, with Windows 8, what they did somehow with Windows Mobile. The one real problem, though: where's the alternative? WinMo sucked and didn't evolve, sure. But it also took the iPhone and Android to really clobber Microsoft.
With MacOS both itself evolving in bad ways, and being an Apple product, that can't be a large scale alternative.
And sure, there's desktop Linux. But there had been handtop Linux as well, lots of them, in fact, and they didn't really go anywhere, either. Linux, at least as presented in Android, won the handset simply because there was just one binary applications socket (eg, no need for source distribution) and, of course, it was all very consumer friendly. That's very not the case with Linux right now. And every time it looks like something might happen in that direction, bad things happen. Like Ubuntu going all weird last year. Without a 900lbs gorilla like Google pushing the buttons, I don't think desktop Linux will ever be a serious threat to Windows.
Then why do I see Nooks at WalMart?
Enemy of my enemy is my friend? Nook is a Barnes and Noble tablet, of course, not an Amazon product.
But Amazon isn't directly in competition with Wal-Mart on many things. Wal-Mart doesn't really try on books, for example; they have high prices and small selection.
I have a tablet with a Pixel Qi screen -- not as sweet as eInk, but perfectly readable at noon on a beach. This kind of thing will eventually go mainstream, as the daylight-readable technologies get more comparable to regular LCDs when used indoors (at present, the Pixel Qi with backlight on is a bit unsaturated, more like a magazine than a computer monitor).
A 3G chipset costs about $25 in volume. A bit more for 4G. But keep in mind, the chip companies make over a billion of these things every year. They get crazy volume on them. The prices have been dropping fast. One reason the newer chips have multi-procotol and 4G -- as a chip company (Qualcomm, for example), you have to offer some coolness that's not such a commodity, to get any sort of premium for your particular phone chips.
The carriers don't pay anywhere near retail for the phones they supply -- they have a negotiated percentage off the MSRP. That's why the MSRPs are so high -- this keeps customers from buying their own devices. It's completely artificial. But the OEMs have little choice; if you don't sell via a carrier, you may not sell much at all. Google does it, with the GSM/HSPA+ version of the Galaxy Nexus -- lower price ($349), but they're not retailing it via AT&T or T-Mobile, either.
I predict they cancel all iPods. They'll have a new thing that looks like it ought to be called the iPod Touch, but built on the iPhone 5 technology, including 4" screen. This will, instead, be dubbed the iPad Nano, and because of the name and the simple fact Apple's all about the marketing, they'll up the base MSRP to around $250. The new iPad, of course, remains at an entry level $499 price, and sure, the new-old stock iPad 2 is still around, but that's not necessarily going to remain so. The 7" iPad Mini will go out at around $350-375.
Apple isn't going to price these low. For one thing, they don't need to -- they're Apple. They're the only real Luxury band left in consumer electronics, given the bad fortunes of Sony in recent times. Think back to the iPad introduction -- long time Apple pundits saw that as the end of Apple-as-they-knew-it, because the iPad cost half the price of the lower-end MacBooks.
On the other hand, how much should it cost? An iPad or similar tablet starts with the same screen you'll find in a netbook, more or less. You take out RAM, you take out storage, you take out ports, cut the battery size in half, toss out an Intel Atom and put in a cheaper ARM SOC, toss the keyboard, put it in a cheaper case. The only additional expense is a touchscreen (though some netbooks have these), about $25 cost if you're not buying in Apple quantities.
Another datapoint: Apple's selling a boatload of iPhones and iPads, not so many Macs. And yet, they did a gross margin of 45% last quarter. So they're not even close to dropping markets on "i" devices, relative to their perhaps more obviously overpriced PCs.
As for the iPhone, like every smartphone or even dumb phone, the MSRP is total fiction. I mean, how could Nokia sell a dumb candybar phone for $150? Well... they don't. The MSRP is a reaction to the fact that Apple's customer for the iPhone isn't you or me, it's Verizon or AT&T. Like every other phone sold, Apple's dependent on these carriers to re-sell the phones, and get them into customer hands. They'd have a hard-to-impossible time doing their on their own, without the phones being featured in the carriers' stores. The carriers, meanwhile, want you to sign in blood for two years to get your cheap, subsidized phone. And they want you to not want to buy one without a contract, and to think you're getting a great deal on the hardware. So they buy phones from Apple, Motorola, HTC, Samsung, etc. based on a percentage of the MSRP. A fairly small percentage. If Apple priced the iPhone at the same 40-50% margin they price the iPod Touch at, they'd basically be paying AT&T and Verizon to take the iPhone. So they make a profit, and that results in a crazy high MSRP.
Here's a good way to think about it, since Apple makes this easy. At any given technology node so far, the iPod Touch is simply an iPhone with some stuff left off. What's left off? Depends on the model, certainly, but for one, the cellular modem. That's about $25 in cost if you're not Apple -- I'm sure Apple pays less. On the current version, the camera isn't as good, so that's another $3 or so. No microphone, that's under $0.50. The iPhone has a larger battery, another $5. So figure less than $40 difference in cost, so that's $60-$80 MSRP tops. If they could sell an iPod Touch for $230, an iPhone of the same generation and flash capacity ought to run $320 or less. But it doesn't, for the reasons stated.
Is there another example? Sure is... the Galaxy Nexus. That's a smartphone at the same basic technology node as the iPhone 4S (the 4S has a faster GPU and better camera, the Nexus a faster CPU and better screen). Google, making far fewer of these than Apple, sells them direct on the Google Play store for $399. That's an unlocked GSM/HSPA model, also just like the iPhone, but unlike the the iPhone or pretty much any other phone, it fully supports both AT&T and T-Mobile (2G on 850MHz or 1900MHz, 3G on 850MHz, 1700MHz, 1900MHz, and 2100MHz is various possible combinations.. .and maybe even the European frequencies, 900MHz and 1800MHz). The Verizon version of this original sold, on contract, for $299, and MSRPed at around $600. But Google's not trying to sell to AT&T or T-Mobile, so they can do this.
If your carrier is still running GSM-only cells, they're well behind the usual curve, at least for the USA. The CDMA2000 carriers (Sprint and Verizon) pretty much had their networks all 3G nearly a decade ago. And it's actually more important on GSM... 2G uses hard-carrier handoffs, which is where all those dropped calls come from. UMTS/HSPA uses the same soft handoffs used by CDMA2000, so, less in the way of dropped calls.
If they're talking about HSPA+ upgrades finished by Spring 2013, this must be T-Mobile, and much of that's going to be HSPA->HSPA+ upgrades. Pretty much everyone else is spending their upgrade money on LTE. Supposedly (and given where I live, I'll be able to tell you), Verizon will have every cell upgraded to LTE sometime in 2013. AT&T may be a bit behind, Sprint's further still, but they're making progress. LTE is particularly important to a carrier like Sprint, since like T-Mobile, they've been at a frequency disadvantage against Verizon and AT&T. Sprint's CDMA2000/EvDO is only 1900MHz (like T-Mo's GSM 2G), and their WiMax (in partnership with Clear and Comcast) is at 2500MHz, even worse for long distances. Their LTE is going out on the old 800MHz Nextel band, so they're getting much the same range and penetration (foliage, building) advantage as the two big guys.
T-Mobile is at a big disadvantage, since they only have 3G on 1700/2100MHz (AWS band) as well as their 2G at 1900MHz. They got some money from the collapse of the AT&T merger, which AT&T is paying back partially in AWS spectrum. So sometimes next year, T-Mobile will be starting to build LTE out in the AWS band and push UMTS/HSPA (and HSPA+, if there's actually room for it) into the more crowded 1900MHz (PCS) band.... that's their incentive for dumping 2G entirely, more than anything. So their getting a little worse before they get any better. And they're just going to have to live with the frequency disadvantage.
In fact, they're pretty even. Apple's gross profits last quarter were about 45%. The various telcos have been complaining, lately, about their profits not always hitting the 45% margin. Coming from the PC industry, where a good profit margin is more like 10-15% if you're lucky (and often much less), it's pretty obvious why neither telcos nor Apple are seen as offering much value for the money.
At least, when you buy an Apple product, you pretty much know what you're getting, and that next year's will be a little better than this year's. The telcos keep inventing ways to charge you more for less, like most of them nixing unlimited data plans, and now this year, charging more for a 1GB of shared data (with no carryover) than they did for unlimited data back in the day (or for those of us grandfathered in).
They're also the folks -- the telecoms, not Apple or Samsung, responsible for the crazy inflated pricing on smartphones sold without a contract. A high enough smartphone will probably retail at $300-$400, tops, if it were priced the way most any other consumer electronics item is priced. But the telcos force these $550-$750 MSRPs, so that the average Joe will see the $200-$300 price they pay on-contract as a good deal.
All smartphones these days run the cellular baseband on a separate, small ARM SOC, not the main application processor. That code, like any other code, is subject to the occasional bug. It's also quite possible the same Qualcomm (or whoever) code that's running on everyone else's phone with the same wireless chip. The application level OS can be smart about detecting a baseband crash and rebooting the baseband chip. But it's not always obvious. So yeah, you're going to have occasional reboots for "no signal" on practically any smartphone. Unless you're lucky enough to get one with a bug-free-enough baseband.
This thing never happened on my O.G. Droid. It's happened two or three times on my Galaxy Nexus. No biggie. In the dumb phone days, your phone would have probably hit a watchdog timeout of some kind and just rebooted -- you might never have even noticed it.
I double majored in EE and Mathematics.. I had something like six really math heavy courses, starting with Calc2, and that's not counting some of the applied courses in EE. I didn't use much of that for 15 years, but I did eventually. It's an important background, because you just never know.
If you're only interested in learning a specific kind of programming, that's for a technical school, not a University. But the thing is, what was hot in the early 80s when I was in school isn't so hot anymore. Teaching a programming language, that's a technical school thing at best -- it should really be a couple of days, tops, for an experienced software engineer (I had to learn a few of them on my last SW project).
A University degree should give you the mental tools to adapt to any new job requirement. Math is an important part of that. The computer languages I learned in college (Pascal, LISP, SNOBOL4, APL, InterCal :-), etc) .. not terribly in demand today. The math... calculus and other numerical methods are a bit easier to do, thanks to computer tools, but they're just as applicable today.
Games sure. But also embedded control systems. Robotics. Cryptographic systems. Signal processing: digital video, digital audio, software radio, etc. I think the need for "real math" is on the rise.
When I got out of school, I got into computer systems design (things like the Commodore Amiga 2000)... not a great deal of complex math in there. But I actually did double major, EE and Mathematics, so I had that toolbox of stupid math tricks available, once things got a little more demanding than just calculating worse-case timing and power budgets.
Which they did. One thing I did, which many folks think is cool (and it is, particularly if you don't care about getting paid that well)... I got into startups. When you're at a startup, you find you wear just about any hat that fits, and you also get pretty good at tailoring those that don't. Having a proper background in math will definitely help you at this. And it's not unusual, particularly if you're a hardware guy at a small startup, that you're expected to solve EVERY possible hardware problem... or at least know it well enough that you can solve things with the occasional help of a consultant.
So since Commodore, I've done control systems and sensor systems (robotics), digital radio systems (R/C cars, more robotics, general purpose mesh routers). I wrote an NTSC decoder for a software radio.... all sorts of different math in there (FFTs, filters, etc)... and knowing math well enough is critical to building things like this well... you can find dozens of books on the analog TV signals, but doing that all in software, you really need to understand things. I've also been working in cryptography, more SDR, and other kinds of signal processing... even wrote a few audio plug-ins for my own use, which may one day get released somewhere.
Of course, with great tools like MathCAD, you may not find yourself actually "doing" the math they way you might have back in school. No pencil and paper, you have tons of primitives that do complex things for you. But you need to understand how it all works to able to use such tools.. and without that knowledge, you'll be unable to adapt. The other guy who wants that cool new job won't have that limitation.
And in fact, the very best thing I learned out of my more-expensive-than-Dad-had-budgeted college (and those extra ten years of monthly payments) was the ability to adapt. The specifics we learned in the early 1980s (Pascal and LISP programming, TTL logic, etc) all but vanished. The big ideas don't. An education shouldn't worry too much about today's specifics... you have no idea today what kind of engineering you'll be doing in 5, 10, 20 years. Even if you don't change jobs, the nature of the work can change out from under you. I went from HW design last summer to a software project (and the need to learn three new languages), then back to hardware this summer -- with some pretty deep mathematics in the mix now.