Many newer cell phones actually include a GPS antenna / chip. It's actually not that hard or expensive; you can source them for somewhere between 20 and 40 dollars each from what I recall.
As for battery, the problem is not that the GPS is such a large burden on the battery, but that smartphones are battery hungry on the best of days. So you really can't rely on them for a week at a time without something clever like a USB solar panel. But if you have one, then it's probably lighter and cheaper carrying that than carrying separate phone and GPS gadgets.
The point of that clause isn't about papal infalliability so much as laying out expectations. There is no inquisition called upon you for calling out SABDFL on technical or moral grounds. Instead we expect leadership to set a higher example, and hold them accountable to it. The key phrase you failed to quote is "expected to be perfect".
I admit though, that the ways in which you can hold leadership accountable is limited. All members vote on the Community Council, and all developers vote on the Technical Board, but SABDFL has a permanent seat on both. This is one of many ways in which Canonical ultimately owns Ubuntu. They run the servers, they own the trademarks and they dominate the committees. However, Shuttleworth has stated multiple times that he recognizes that the use of veto power is a gamble with volunteers: the vote-with-feet can't be removed after all. Nobody's calling for you to be a slave!
The other benefit of going contractless is that you can cut and cut deep without a huge penalty. Just pick the cheapest plan you want, and put the phone on the credit card if you can't afford it. Or don't buy it. Seems simple to me.
Once someone's trapping the message flow, it's trivial to search for particular triggers. The biggest defence is current generations of routers not sending every message to every machine on the local net, but that's not really much of a defence at all. Encryption stops these trivial attacks.
Of course, wireless via wifi and cellphones breaks this assumption. Seems every bank and credit card is jumping to offer a mobile version of their website. Non-broadcast isn't much of a defense anymore.
Firstly, unlocked phones running on AT&T at EDGE is better that nothing. It's one of the reasons I favor GSM; even though there's only a few carriers/MVNOs, there's at least some alternatives. R-UIM theoretically could do the same thing, but I gather US CDMA carriers don't want it. No matter how you look at it though, there's a slightly broader market for used unlocked phones.
Secondly, the take a look at T-Mobile's Even More+ and Even More plans. One is month-to-month, the other comes with a phone and a 2 year contract. Here's the important part: put an unlocked phone on the month-to-month plan and it's ten dollars cheaper per month. Generally the equation works out to a 200 dollar discount in exchange for that bump in pricing. So basically over two years, you pay 40 dollars beyond the discount. If you take that as a 'finance charge', then if I did my math right, it works out to about an interest rate of 18 percent. If you have a better rate, say on your credit card, grab an unlocked phone. Plus, there's more retail competition for unlocked phones, which can make the locked phone an even worse deal.
Finally, without a contracted plan, you're free to choose what's most cost effective for you; I use a prepaid plan that costs me perhaps 10 dollars a month, and NO data. Wifi gets me far enough currently. Just having the flexibility to change the contract without penalty can help consumers save money.
The difference is that these blocks have gradients, so there's a computational optimization aspect to it. Given a set of similar but gradient, how do you arrange and orient them to best reproduce an image? The gradient aspect leads to a better quality image than a simple grid of pixels.
Actually, you'll love this. In Japan, there have been authorized flashcarts. Convenience stores will have kiosks where you can pay to overwrite your cart with a new game. Apparently Japan has sufficiently more convenience stores than any other nation to make this digitial distribution method work. I guess it's closed in favor of internet downloads now.
Nintendo does sell old games; they used them in the GBA Classics series, and now the Wii "virtual console" has tons of games. The bone of contention, I think, is Nokia advertising their system as a platform to pirate Nintendo games with a weak "don't forget to buy the original cartridge!". Which would be substantially undermined by Nokia's marketing team not even following the rules.
It's hard to remember, but there was a time when Nokia directly competed with Nintendo in the hardware market with the N-Gage. They've since moved to a different model, but the fact remains that they're a competitor. Hell, the n900 is powerful enough that one could probably build decent DS emulators. Partly, Nintendo should be worried that you can get a much better device than their own DSi. Higher quality cameras, more RAM, 3G, bluetooth, IR, etc.
But mostly Nintendo's problem isn't a competing platform though; the problem they have is the absolute lack of DRM built into emulators. No point in putting games in the Ovi Store if Nokia marketing hints that you can get it all for free.
It's a pragmatic reason: the federal government already requires income tax. You can piggy back on the federal income filings easily, and get their mandatory employer filings and whatnot to combat tax cheats.
In contrast, the sales tax is somewhat hard to prove. In a famous legal case, the Crazy Eddie's electronics franchise got in trouble for slowly stopping their practice of underreporting sales and pocketing the tax they collected from consumers. When you mix that in with categorized sales tax (ie food is free, and booze is extra), and outright sales tax exemptions for charitable organizations and education, every argument you've brought forth applies just as much to sales tax as income tax.
Ideally, you wouldn't invest time in algorithmic layouts. Some big institutions would invest in smart people with advanced degrees and a broad understanding of physics, mathematics and aesthetics to construct a Domain Specific Language for UI layout. Which has been done once in HTML, and adobe is doing it again with Adam and Eve. In a presentation given at Google, Sean Parent argues that this process dramatically reduces the code, and in doing so prevents some bugs and makes debugging others easier.
If nobody has time to implement it, it's because management is too timid to make the case for investing in the business they happen to run.
If you are responsible for the GUI on a large project, then no. Calculus isn't gonna do you a bit of good.
Arguably, math is useful for algorithmic layout engines; even calculus can be used for spring based models. See graphviz for some more impressive layouts and Bezier curve rendering. Geometry, graph theory and calculus are all handy, if you have the gumption to bother applying them.
Or you could just add another ComboBox. People love that shit.
gksudo does not do this. It's literally a GTK frontend to sudo. If you try to do something via gui that fails, gksudo will not suddenly run and ask you for prompts.
There are existing systems that do do this, like update-manager and PolicyKit.
I've believed for a long time that their consumer products division was mainly an engineering training ground. When you look at it, they don't own much; they buy GPS chips from other manufacturers, they license maps and some of their newer stuff runs Linux underneath, and their phone offerings are all outsourced. As you mention in closing, they still have industrial products, mainly avionics, where their competitive advantage is probably in FAA compliance, i.e. paperwork and testing.
Part of the problem is that young students fresh out of high school have no pet datasets. For many, they're buying a new laptop for college and keeping, at most, their music. Chat logs, banking, browsing history; it hasn't occurred to them to keep these things. Hell, I doubt few CS students make backups of their own computers. I know I didn't.
Without a personal dataset of interest to maintain and process, you'll find little demand from students for classes on large dataset computations. Unless they enjoy astronomy, or biology, or whatever, in which case they're likely in a different major. If we want to train CS majors to help in other fields, we need to promote and identify personal data first.
"When was the last time you heard of a hazing in IT?"
I recall hearing many stories of people snapshotting a desktop, making it the background and then removing the icons. And then justify it as teaching employees a lesson about locking workstations when not in use. I last heard this specific form of prank discussed on Slashdot perhaps a month ago.
The trick is to have a massive fucking database of pictures for any given term, and then just throw away anything with a complicated background. Even 2 percent of a thousand pictures gives you plenty alternatives to choose from. And over a dozen colors to choose from.
You can think flickr for the images, Google for the image tag search, and China for completely ignoring photographer copyright;)
You have to distinguish between hedge funds and investment banking, which primarily operate in Wall Street (and London), and Savings & Loans type banks with a physical branch and loan officers. The latter generally didn't lower their lending standards, and therefore didn't lead into the subprime mess. They are however, still on the hook for losses in places where market downturns have exceeded conforming loan downpayment percentages.
However, these S&L places are truly dumb. Partly because they're high profile targets, and partly because of reasons the GP stated. A friend of my brother's related a story to me of how he was offered a CIO position at the bank he worked at as a junior loan officer. He's attending some podunk college and earning a bachelor's in business. Offered pay: $35,000. For the CIO of a bank (he declined, stating mainly that he wasn't interested in the work at any compensation). If this guy has even fucking heard of 2 factor authentication, it's a miracle. And since banking IT is essentially figuring out how to not do anything, it's the most boring kind of job, that nobody wants. Sarbox hasn't helped this.
(If I were more cynical, I'd assume the person in charge of the bank wants poor IT security to aid in covering outright theft.)
They do adjust the price. There's entire sites dedicated to price drops. In essence, the games are priced at 60 dollars because someone is dumb enough to pay it. They might blame piracy, but only to keep their cash cows ignorant of their discounting practices. Game companies have almost every incentive to charge everyone the highest price they'll pay. I say almost, because it's discriminatory to charge Bob Megabucks $60 and Jimmy Cheapskate $30 for the same product, and generally illegal.
So instead, they use discounting to create a legal price discrimination. Games cost 60 dollars at release, and the people willing to pay $60 complete the transaction, they reduce the price and capture the next bucket of consumers. The cost accounting "oh well, retailers get a cut, and so does " is a sham, a cover for this truth of the industry. Think about it: if retailing took a $12 cut per game, and production cost $10, Platinum Hits games shouldn't exist in the market place at $20. And yet, even though they have the same retail margins, the same marginal costs of manufacture, Gears of War, Mass Effect and Forza 2 are all sitting there at $19.99. I'm glad at least that the "high game budgets cause high prices" canard has been dispensed with. Fixed costs like game programming and design determine whether the project is done at all, not the price point that maximizes revenue.
The only reason game companies don't adjust the price starting higher is the fear of consumer backlash: if you price a game higher than the norm at launch, you risk fewer sales before discounting starts, and possibly fewer sales a month later when the price drops. Retailer purchasers get antsy about a high price point unless your game is destined to succeed. Additionally, many games today exhibit "network effects"; high sales of Halo 3 cements demand for Halo 3 via network multiplayer. It might be worth skipping a higher MSRP to capture bigger multiplayer audiences.
RAID has two points: eliminating single point of failure, and multichannel IO. The RAID design you choose for a given system depends on your desired balance of performance and reliability. To claim the "whole point of RAID is to provide a first-line defense against catastrophic data loss" ignores half of the point and the decisions that should be made when designing systems.
Sure, AVR and other 8bit stuff is great, as long as you don't actually have a UI. But the word "embedded" comes with no hard limit on processing power. Eventually these half-assed while true systems will have to give way to better power management anyways, at which point you might as go along for the ride with separate address spaces.
Nevertheless, I never used the word 'embedded'. Instead I used the word 'embedding,' as in putting into firmware. There's a EEPROM, and it gets written with a Linux image. Close enough to embedded for me, but not really germane to my point.
Slashdot Mirror
N900 is substantially smaller than the N810.
Many newer cell phones actually include a GPS antenna / chip. It's actually not that hard or expensive; you can source them for somewhere between 20 and 40 dollars each from what I recall.
As for battery, the problem is not that the GPS is such a large burden on the battery, but that smartphones are battery hungry on the best of days. So you really can't rely on them for a week at a time without something clever like a USB solar panel. But if you have one, then it's probably lighter and cheaper carrying that than carrying separate phone and GPS gadgets.
The point of that clause isn't about papal infalliability so much as laying out expectations. There is no inquisition called upon you for calling out SABDFL on technical or moral grounds. Instead we expect leadership to set a higher example, and hold them accountable to it. The key phrase you failed to quote is "expected to be perfect".
I admit though, that the ways in which you can hold leadership accountable is limited. All members vote on the Community Council, and all developers vote on the Technical Board, but SABDFL has a permanent seat on both. This is one of many ways in which Canonical ultimately owns Ubuntu. They run the servers, they own the trademarks and they dominate the committees. However, Shuttleworth has stated multiple times that he recognizes that the use of veto power is a gamble with volunteers: the vote-with-feet can't be removed after all. Nobody's calling for you to be a slave!
The other benefit of going contractless is that you can cut and cut deep without a huge penalty. Just pick the cheapest plan you want, and put the phone on the credit card if you can't afford it. Or don't buy it. Seems simple to me.
Once someone's trapping the message flow, it's trivial to search for particular triggers. The biggest defence is current generations of routers not sending every message to every machine on the local net, but that's not really much of a defence at all. Encryption stops these trivial attacks.
Of course, wireless via wifi and cellphones breaks this assumption. Seems every bank and credit card is jumping to offer a mobile version of their website. Non-broadcast isn't much of a defense anymore.
Firstly, unlocked phones running on AT&T at EDGE is better that nothing. It's one of the reasons I favor GSM; even though there's only a few carriers/MVNOs, there's at least some alternatives. R-UIM theoretically could do the same thing, but I gather US CDMA carriers don't want it. No matter how you look at it though, there's a slightly broader market for used unlocked phones.
Secondly, the take a look at T-Mobile's Even More+ and Even More plans. One is month-to-month, the other comes with a phone and a 2 year contract. Here's the important part: put an unlocked phone on the month-to-month plan and it's ten dollars cheaper per month. Generally the equation works out to a 200 dollar discount in exchange for that bump in pricing. So basically over two years, you pay 40 dollars beyond the discount. If you take that as a 'finance charge', then if I did my math right, it works out to about an interest rate of 18 percent. If you have a better rate, say on your credit card, grab an unlocked phone. Plus, there's more retail competition for unlocked phones, which can make the locked phone an even worse deal.
Finally, without a contracted plan, you're free to choose what's most cost effective for you; I use a prepaid plan that costs me perhaps 10 dollars a month, and NO data. Wifi gets me far enough currently. Just having the flexibility to change the contract without penalty can help consumers save money.
Nobless Oblige. I mean can your iPhone order missile strikes?
And you don't think their clients suffer from Christmas peak loads? Even if you don't sell things, it's a time of year with a lot of vacation time.
The difference is that these blocks have gradients, so there's a computational optimization aspect to it. Given a set of similar but gradient, how do you arrange and orient them to best reproduce an image? The gradient aspect leads to a better quality image than a simple grid of pixels.
Actually, you'll love this. In Japan, there have been authorized flashcarts. Convenience stores will have kiosks where you can pay to overwrite your cart with a new game. Apparently Japan has sufficiently more convenience stores than any other nation to make this digitial distribution method work. I guess it's closed in favor of internet downloads now.
Nintendo does sell old games; they used them in the GBA Classics series, and now the Wii "virtual console" has tons of games. The bone of contention, I think, is Nokia advertising their system as a platform to pirate Nintendo games with a weak "don't forget to buy the original cartridge!". Which would be substantially undermined by Nokia's marketing team not even following the rules.
It's hard to remember, but there was a time when Nokia directly competed with Nintendo in the hardware market with the N-Gage. They've since moved to a different model, but the fact remains that they're a competitor. Hell, the n900 is powerful enough that one could probably build decent DS emulators. Partly, Nintendo should be worried that you can get a much better device than their own DSi. Higher quality cameras, more RAM, 3G, bluetooth, IR, etc.
But mostly Nintendo's problem isn't a competing platform though; the problem they have is the absolute lack of DRM built into emulators. No point in putting games in the Ovi Store if Nokia marketing hints that you can get it all for free.
It's a pragmatic reason: the federal government already requires income tax. You can piggy back on the federal income filings easily, and get their mandatory employer filings and whatnot to combat tax cheats.
In contrast, the sales tax is somewhat hard to prove. In a famous legal case, the Crazy Eddie's electronics franchise got in trouble for slowly stopping their practice of underreporting sales and pocketing the tax they collected from consumers. When you mix that in with categorized sales tax (ie food is free, and booze is extra), and outright sales tax exemptions for charitable organizations and education, every argument you've brought forth applies just as much to sales tax as income tax.
Ideally, you wouldn't invest time in algorithmic layouts. Some big institutions would invest in smart people with advanced degrees and a broad understanding of physics, mathematics and aesthetics to construct a Domain Specific Language for UI layout. Which has been done once in HTML, and adobe is doing it again with Adam and Eve. In a presentation given at Google, Sean Parent argues that this process dramatically reduces the code, and in doing so prevents some bugs and makes debugging others easier.
If nobody has time to implement it, it's because management is too timid to make the case for investing in the business they happen to run.
Arguably, math is useful for algorithmic layout engines; even calculus can be used for spring based models. See graphviz for some more impressive layouts and Bezier curve rendering. Geometry, graph theory and calculus are all handy, if you have the gumption to bother applying them.
Or you could just add another ComboBox. People love that shit.
gksudo does not do this. It's literally a GTK frontend to sudo. If you try to do something via gui that fails, gksudo will not suddenly run and ask you for prompts.
There are existing systems that do do this, like update-manager and PolicyKit.
Keepass 2.0 is a .NET application. I don't think you'll see an android version for that reason.
I've believed for a long time that their consumer products division was mainly an engineering training ground. When you look at it, they don't own much; they buy GPS chips from other manufacturers, they license maps and some of their newer stuff runs Linux underneath, and their phone offerings are all outsourced. As you mention in closing, they still have industrial products, mainly avionics, where their competitive advantage is probably in FAA compliance, i.e. paperwork and testing.
Part of the problem is that young students fresh out of high school have no pet datasets. For many, they're buying a new laptop for college and keeping, at most, their music. Chat logs, banking, browsing history; it hasn't occurred to them to keep these things. Hell, I doubt few CS students make backups of their own computers. I know I didn't.
Without a personal dataset of interest to maintain and process, you'll find little demand from students for classes on large dataset computations. Unless they enjoy astronomy, or biology, or whatever, in which case they're likely in a different major. If we want to train CS majors to help in other fields, we need to promote and identify personal data first.
"When was the last time you heard of a hazing in IT?"
I recall hearing many stories of people snapshotting a desktop, making it the background and then removing the icons. And then justify it as teaching employees a lesson about locking workstations when not in use. I last heard this specific form of prank discussed on Slashdot perhaps a month ago.
The trick is to have a massive fucking database of pictures for any given term, and then just throw away anything with a complicated background. Even 2 percent of a thousand pictures gives you plenty alternatives to choose from. And over a dozen colors to choose from.
You can think flickr for the images, Google for the image tag search, and China for completely ignoring photographer copyright ;)
You have to distinguish between hedge funds and investment banking, which primarily operate in Wall Street (and London), and Savings & Loans type banks with a physical branch and loan officers. The latter generally didn't lower their lending standards, and therefore didn't lead into the subprime mess. They are however, still on the hook for losses in places where market downturns have exceeded conforming loan downpayment percentages.
However, these S&L places are truly dumb. Partly because they're high profile targets, and partly because of reasons the GP stated. A friend of my brother's related a story to me of how he was offered a CIO position at the bank he worked at as a junior loan officer. He's attending some podunk college and earning a bachelor's in business. Offered pay: $35,000. For the CIO of a bank (he declined, stating mainly that he wasn't interested in the work at any compensation). If this guy has even fucking heard of 2 factor authentication, it's a miracle. And since banking IT is essentially figuring out how to not do anything, it's the most boring kind of job, that nobody wants. Sarbox hasn't helped this.
(If I were more cynical, I'd assume the person in charge of the bank wants poor IT security to aid in covering outright theft.)
They do adjust the price. There's entire sites dedicated to price drops. In essence, the games are priced at 60 dollars because someone is dumb enough to pay it. They might blame piracy, but only to keep their cash cows ignorant of their discounting practices. Game companies have almost every incentive to charge everyone the highest price they'll pay. I say almost, because it's discriminatory to charge Bob Megabucks $60 and Jimmy Cheapskate $30 for the same product, and generally illegal.
So instead, they use discounting to create a legal price discrimination. Games cost 60 dollars at release, and the people willing to pay $60 complete the transaction, they reduce the price and capture the next bucket of consumers. The cost accounting "oh well, retailers get a cut, and so does " is a sham, a cover for this truth of the industry. Think about it: if retailing took a $12 cut per game, and production cost $10, Platinum Hits games shouldn't exist in the market place at $20. And yet, even though they have the same retail margins, the same marginal costs of manufacture, Gears of War, Mass Effect and Forza 2 are all sitting there at $19.99. I'm glad at least that the "high game budgets cause high prices" canard has been dispensed with. Fixed costs like game programming and design determine whether the project is done at all, not the price point that maximizes revenue.
The only reason game companies don't adjust the price starting higher is the fear of consumer backlash: if you price a game higher than the norm at launch, you risk fewer sales before discounting starts, and possibly fewer sales a month later when the price drops. Retailer purchasers get antsy about a high price point unless your game is destined to succeed. Additionally, many games today exhibit "network effects"; high sales of Halo 3 cements demand for Halo 3 via network multiplayer. It might be worth skipping a higher MSRP to capture bigger multiplayer audiences.
It's entirely possible to have performance goals and whatnot; busybox manages to have modular growth AND memory constraints, after all.
RAID has two points: eliminating single point of failure, and multichannel IO. The RAID design you choose for a given system depends on your desired balance of performance and reliability. To claim the "whole point of RAID is to provide a first-line defense against catastrophic data loss" ignores half of the point and the decisions that should be made when designing systems.
Sure, AVR and other 8bit stuff is great, as long as you don't actually have a UI. But the word "embedded" comes with no hard limit on processing power. Eventually these half-assed while true systems will have to give way to better power management anyways, at which point you might as go along for the ride with separate address spaces.
Nevertheless, I never used the word 'embedded'. Instead I used the word 'embedding,' as in putting into firmware. There's a EEPROM, and it gets written with a Linux image. Close enough to embedded for me, but not really germane to my point.