the only stipulation we have is that you provide it back for free?
Thats not quite true; GPLv3 changed the rules a bit, in ways that Nokia probably doesn't like. SIM locking is entirely bogus and a pander to the incumbents, and ultimately a detrimental move to profits and sales as complementary prices like contracts remain higher than they should. But it's also the only way to sell phones today. Still, for a guy who seems to understand the philosophy, you'd think he'd know when to shut the hell up.
Speaking of purchasing power, anyone bought an OpenMoko?
Your logic is circular here at best. People can be uninformed and make choices. They're not "informed" decisions though. I read a fantastic paper about informing students in the Dominican republic about the underinformation students in the 8th grade (last government required grade) had about earnings and education. The worst guessers of the different in wages were those who claimed to get their information from people in their community. Were they making informed decisions?
The problem here is complex. A healthy consumer market would be responsive to changes in pricing plans. A prerequisite for this is for most cell phones to be able to switch providers, so that providers can compete on price and terms. However this is not the case, and I think that "the market rates" are high as a consequence -- there's less incentive to compete on price since the most lucrative customers are locked into 2 year contracts with a phone they can't use anywhere else. So when you say AT&T's plans aren't out of market bounds, that's business as usual and the intended effect.
Flashing the firmware is one step in the chain of demonstrating against the practice of subverting the doctrine of first sale. Do ATT SIM-locked phones come with a big warning sticker that says "Warning, you can't use this phone on other networks"? Boycotts and open hardware are another method, but today only get curious stares from early adopters. Starting your own wireless company might be another approach, if you have that kind of resources and talent.
I've noticed similar problems at my place, and I think it's less about burst packeting and more about fair queuing. Bittorrent opens up tons of connections and VoIP doesn't. It's not that there's no time to send communications on a regular interval, it's that the VoIP app isn't getting them. In my case, I'd been pondering the ins and outs of Tomato's QoS but I mostly just throttled Deluge and called it a day when that did the job.
The thing about UML is that it's supposed to be a step on the path to designing software automatically. Unfortunately, it very slanted towards translation into C++ (and maybe Java). This bleeds into the problems you mention with strictly defined semantics of visual cues developers treat loosely at best. Moreover, OOP is itself slowly dying for all the right reasons, though not as quickly as you might think.
I've seen an interesting theory floating around in multiple places that dataflow languages, visually represented as DAGs, might be a suitable replacement for OO/UML. The trouble is they've all bought into web centric software and AJAX, which runs like ass in the best cases, and looks like ass in the worst. I've been pondering recently how I might formalize the computing technique and generalize it to effectively define a dataflow middle language that could be implemented in whatever language and platform you like, unlike the heavy C++ slant of UML.
I think there's more to it than that on the sales front. The thing is, insurance companies don't need brick and mortar in retail places. Many find large office space somewhere just outside a major city's suburbs to locate a call center, hire adjusters on contract, and advertise a phone number. The internet does reduce staffing costs and such, but they're not quite burying State Farm etc. Of course, the real cost of insurance is actually paying out on claims.
Shipping's one cost, but there's a matter of margins. Margins + shipping is why amazon only recently introduced food to their stores, and why you can't find decent video cards or RAM at Best Buy anymore. They simply cannot compete with lower margins, wider offerings and light shipping charges. I think most computer components are a good Internet candidate: the customer base likely already has a computer connected to the internet, purchases are planned out and not usually time critical, and is an easy customer to advertise to. Of course, everyone knows this, so theres a shit-ton of stores out there, competing on margins.
The most ridiculous part of that NPR story has to be the investment bankers who started picking up terribly administered mortgages once told someone else's trading desk did, and then amplifying that stupidity with "tranches." On the other hand, what NPR didn't cover is how the hell the ratings agencies decided to rate these CDO offerings, and I'd love to understand just what the hell happened.
Crude regular expressions only work for crude regular languages. You mention parsing XHTML, but what evidence do we have that regex don't choke on html just as often as XML parsers? HTML is complicated and often broken enough in distribution that I'm not sure you can define a universal accepted "well formed-ness" checker. And it's not that regex waste CPU -- you can implement them in time linear to the input, it's that they break quite easily if there's any variety in your input. I recall someone complaining that the Wordpress XML parser was regex written to handle their own output, and when someone wanted to migrate from one to the other, their valid XML didn't work because it assumed a certain order of elements.
For most XML doctypes, I think it's useful to have a generated parser available. If you want your input to be something akin to XML and you want to accept as many files as possible, grammars and parsers are your only valid option.
b) an earlier poster pinned it - double-blind testing is required to prove allergies. With Wi-Fi, this would be easy.
Actually, it would be quite hard, for exactly A reason. It's hard to find or create quiet areas. You'd need a testing center somewhere in the Australian outback, a source of wifi signal and a spectrum analyzer to determine how much exposure the subject is recieving.
What if it's two dollars difference between fresh local fruit and chemically preserved fruit from miles away? It's important to understand how costly such preferences are.
This flipping crap is why I'm not enthusiastic about assisting those in foreclosure. In 2006 Los Vegas, for example, where prices were screaming upwards and the whole place was growing incredibly fast, something like 1 in 4 loans was interest only. That screams classic bubble and I think we need to make it clear that bad investing should not be rewarded.
Investors are not "spooked". Have you ever wondered why oil went as high as it is in the futures market, without a contraction of supply? Sure, China and other worldwide growth is part of it, but I think a substantial part of it is investors heading for commodities after the subprime market did a number on both bonds and stocks. The smart ones aren't risk adverse -- they're busy making the next bubble =(
A lot of upstreams don't want to support anything but the latest release, and apparently get annoyed when people file bugs at sourceforge for stable releases. Normally they should just ignore outdated releases but more seem to be getting angry that Ubuntu is so slow to pick up changes. Fundamentally, I think many advanced users and developers hate stable release cycles and would rather see a Debian unstable process. I'm not sure how well this works for library transitions. Probably pretty bad.
That cuts both ways. If you change the package, you're not reflecting some crazy will of the developer or making them look bad etc. If you don't upgrade it often enough, the developer (and some users) get angry that you're still shipping old code of theirs. If you just ship upstream releases 0day, shit breaks. I just witnessed a package in universe complain that Hardy didn't ship their latest version, even though it was basically a surprise release well after FeatureFreeze and a week or so before the FinalFreeze. Nevermind that their first cut at the release totally broke the program and they had to release a version bump a day later. Many upstreams are terrible at release management, and distributions are valuable because they do that work.
Bottom line is if you want the distro to help you, understand the distro first.
The problem Shuttleworth is addressing is upstream packages that are out of date. They have random or at least highly unpredictable release schedules that don't seem to work well for anyone. By suggesting a specific date for releases, Shuttleworth is attempting to shift a small burden to upstream to relieve a much larger strain in Ubuntu. Once the Freezes are past, it takes significant work to backport a bugfix patch to the feature frozen package. I think Shuttleworth's case would be greatly helped if he could point to a smallish project and how exactly it synchronizes with Ubuntu. I suspect his goal is KDE however, so a more general "this is healthy" essay is all we'll get.
Strategically, if upstreams were to do this on Ubuntu's cycle, Ubuntu would essentially be the Open Source release engineers, and many more upstreams would be invested in Ubuntu's release process. Some people recognize this and ask why they should follow Ubuntu over their own distro of choice, to which Shuttleworth said if any two distros could agree on a date, Canonical would follow suit.
At least for my model, the bios is a matter of pressing F1 at the right time. Not an Award though, which people might be familiar with. Linux has similar BIOS tools available. Out of curiousity, which BIOS settings aren't available from the boot menu?
No engineer I think was ever claiming that engineers ignored money. The OP was crying out that engineers are not valued in a company, and that translates into people wandering off into finance and marketing / advertising rather than put up with the kind of companies allow to happen. It's a far cry from "we should teach more of this business math", and I think you brought in some recent personal experience to the conversation unannounced.
In fact, much of the math they rely on was done by brilliant guys who studied alongside engineers in traditional math. For example, the Black-Scholes option pricing formula is based on the holy grails of engineering, rocket guidance and partial differential equations, and it led to both obscene profits for a while, and a quite large crisis after LTCM's secret got out and profits fell while management was screwed. Similar to the subprime hedge fund collapse that saw stocks oversold. We don't need maths specific to "business" (finance, everyone really means) in high school. Calculating a P:E ratio is easy, and even once you have it, what do you do with it? Lets leave the math track alone, and start instead with a basic microeconomics course in schools. It's far more practical for every business not running a bank to understand how competition works and why eliminating taxes won't do much to solve the high price of gas.
I think you misunderstand. One of the lines flagged did something worthless (add uninitalized memory into the pool or something I cannot recall) and the other did something important (add sources of entropy). The patch that fixes it undoes only one of the comments. The other has been argued to be of little consequence either way. Adding a fixed constant to an entropy pool doesn't make it less random, which is basically what it does since the memory's unlikely to have been touched by that program.
At any rate, this was definitely a case of an overzealous perfectionist removing memory debugger warnings, and an upstream that does a borderline practice without explanation in code or the FAQ.
It really doesn't happen as often as you'd think. Apparently the difference between a troll tag and a flamebait tag is that trolls have bad grammar and poor spelling.
Emphasis mine:
Because the point of a business - at any level - is making money. It's not building and designing and thinking and believing or anything This is where I disagree. If you don't believe in ethical business, you're part of the problem. Every problem I mentioned was caused by placing the pursuit of money ahead of customer's or even the company's interests. I never said that unnecessary projects should go on, or even unimportant ones. Engineers have a responsibility to protect the public from bad designs. It's their job and ethical code. If it means the company makes less money for the need for testing, too bad. If the difference between a profitable project and a losing one is understaffing or less testing or unmeetable deadlines, it means it's not a project your company should be doing.
We codify these ethics in law by making engineers responsible for failures of oversight, and hold companies liable for selling dangerous products against the recommendations of their engineers. It is of course also good long run business to make safe designs for customers, but a shit ton of management is not in it for the long run. It's probably easier to ship a faulty design, put it on the resume as a success and find a new job before the whole thing falls apart. The right way is to go back to the customer like a debtor to a bank and say "hey, we can't meet this deadline" or "we can't meet this requirement at this price" and work something out.
No, the whole point is to create a copyright license that allows collaboration between developers, without being able to pull an evil trick and renege on your half of the agreement. The difference between a fork and a branch is easy to spot: hostilities directly before the fork. I think their website is quite clear on where they stand.
Toshiba BIOS settings are weird, but in their defense they have been doing this for longer than anyone else. They did were doing ACPI stuff before it was formalized. We see this over and over again -- someone comes up with an okay idea on their own. Then an industry comittee or Microsoft or someone else defines a standard. Now you have two interfaces and neither party have incentives to resolve the situation. It happened with ACPI and with wireless drivers, probably more. But when you say userland, do you mean "not in kernel" or "not from BIOS itself?" My laptop has a goofy configuration screen that handles most everything I'm aware of. I wouldn't be surprised to hear there's more out there that the boot menu doesn't touch, as the whole thing scares me a bit and I stay away from messing with it when possible.
In Linux, big changes like this mostly only happen when there's universal support. Open Source makes it possible to see how the old stuff worked and incorporate the changes to a new design.
Slashdot Mirror
Thats not quite true; GPLv3 changed the rules a bit, in ways that Nokia probably doesn't like. SIM locking is entirely bogus and a pander to the incumbents, and ultimately a detrimental move to profits and sales as complementary prices like contracts remain higher than they should. But it's also the only way to sell phones today. Still, for a guy who seems to understand the philosophy, you'd think he'd know when to shut the hell up.
Speaking of purchasing power, anyone bought an OpenMoko?
Your logic is circular here at best. People can be uninformed and make choices. They're not "informed" decisions though. I read a fantastic paper about informing students in the Dominican republic about the underinformation students in the 8th grade (last government required grade) had about earnings and education. The worst guessers of the different in wages were those who claimed to get their information from people in their community. Were they making informed decisions?
The problem here is complex. A healthy consumer market would be responsive to changes in pricing plans. A prerequisite for this is for most cell phones to be able to switch providers, so that providers can compete on price and terms. However this is not the case, and I think that "the market rates" are high as a consequence -- there's less incentive to compete on price since the most lucrative customers are locked into 2 year contracts with a phone they can't use anywhere else. So when you say AT&T's plans aren't out of market bounds, that's business as usual and the intended effect.
Flashing the firmware is one step in the chain of demonstrating against the practice of subverting the doctrine of first sale. Do ATT SIM-locked phones come with a big warning sticker that says "Warning, you can't use this phone on other networks"? Boycotts and open hardware are another method, but today only get curious stares from early adopters. Starting your own wireless company might be another approach, if you have that kind of resources and talent.
Believe it or not people worry about whether they'll be employable with tatoos.
I've noticed similar problems at my place, and I think it's less about burst packeting and more about fair queuing. Bittorrent opens up tons of connections and VoIP doesn't. It's not that there's no time to send communications on a regular interval, it's that the VoIP app isn't getting them. In my case, I'd been pondering the ins and outs of Tomato's QoS but I mostly just throttled Deluge and called it a day when that did the job.
The thing about UML is that it's supposed to be a step on the path to designing software automatically. Unfortunately, it very slanted towards translation into C++ (and maybe Java). This bleeds into the problems you mention with strictly defined semantics of visual cues developers treat loosely at best. Moreover, OOP is itself slowly dying for all the right reasons, though not as quickly as you might think.
I've seen an interesting theory floating around in multiple places that dataflow languages, visually represented as DAGs, might be a suitable replacement for OO/UML. The trouble is they've all bought into web centric software and AJAX, which runs like ass in the best cases, and looks like ass in the worst. I've been pondering recently how I might formalize the computing technique and generalize it to effectively define a dataflow middle language that could be implemented in whatever language and platform you like, unlike the heavy C++ slant of UML.
I think there's more to it than that on the sales front. The thing is, insurance companies don't need brick and mortar in retail places. Many find large office space somewhere just outside a major city's suburbs to locate a call center, hire adjusters on contract, and advertise a phone number. The internet does reduce staffing costs and such, but they're not quite burying State Farm etc. Of course, the real cost of insurance is actually paying out on claims.
Shipping's one cost, but there's a matter of margins. Margins + shipping is why amazon only recently introduced food to their stores, and why you can't find decent video cards or RAM at Best Buy anymore. They simply cannot compete with lower margins, wider offerings and light shipping charges. I think most computer components are a good Internet candidate: the customer base likely already has a computer connected to the internet, purchases are planned out and not usually time critical, and is an easy customer to advertise to. Of course, everyone knows this, so theres a shit-ton of stores out there, competing on margins.
The most ridiculous part of that NPR story has to be the investment bankers who started picking up terribly administered mortgages once told someone else's trading desk did, and then amplifying that stupidity with "tranches." On the other hand, what NPR didn't cover is how the hell the ratings agencies decided to rate these CDO offerings, and I'd love to understand just what the hell happened.
Crude regular expressions only work for crude regular languages. You mention parsing XHTML, but what evidence do we have that regex don't choke on html just as often as XML parsers? HTML is complicated and often broken enough in distribution that I'm not sure you can define a universal accepted "well formed-ness" checker. And it's not that regex waste CPU -- you can implement them in time linear to the input, it's that they break quite easily if there's any variety in your input. I recall someone complaining that the Wordpress XML parser was regex written to handle their own output, and when someone wanted to migrate from one to the other, their valid XML didn't work because it assumed a certain order of elements.
For most XML doctypes, I think it's useful to have a generated parser available. If you want your input to be something akin to XML and you want to accept as many files as possible, grammars and parsers are your only valid option.
Actually, it would be quite hard, for exactly A reason. It's hard to find or create quiet areas. You'd need a testing center somewhere in the Australian outback, a source of wifi signal and a spectrum analyzer to determine how much exposure the subject is recieving.
It's a bit more insidious than you propose: medicine is now taking credit for smaller tumors you might have destroyed on your own!
What if it's two dollars difference between fresh local fruit and chemically preserved fruit from miles away? It's important to understand how costly such preferences are.
This flipping crap is why I'm not enthusiastic about assisting those in foreclosure. In 2006 Los Vegas, for example, where prices were screaming upwards and the whole place was growing incredibly fast, something like 1 in 4 loans was interest only. That screams classic bubble and I think we need to make it clear that bad investing should not be rewarded.
Investors are not "spooked". Have you ever wondered why oil went as high as it is in the futures market, without a contraction of supply? Sure, China and other worldwide growth is part of it, but I think a substantial part of it is investors heading for commodities after the subprime market did a number on both bonds and stocks. The smart ones aren't risk adverse -- they're busy making the next bubble =(
Sadly, the AJAX interface is kinda non-free. Apparently he doesn't want other free firmwares stealing his work.
A lot of upstreams don't want to support anything but the latest release, and apparently get annoyed when people file bugs at sourceforge for stable releases. Normally they should just ignore outdated releases but more seem to be getting angry that Ubuntu is so slow to pick up changes. Fundamentally, I think many advanced users and developers hate stable release cycles and would rather see a Debian unstable process. I'm not sure how well this works for library transitions. Probably pretty bad.
That cuts both ways. If you change the package, you're not reflecting some crazy will of the developer or making them look bad etc. If you don't upgrade it often enough, the developer (and some users) get angry that you're still shipping old code of theirs. If you just ship upstream releases 0day, shit breaks. I just witnessed a package in universe complain that Hardy didn't ship their latest version, even though it was basically a surprise release well after FeatureFreeze and a week or so before the FinalFreeze. Nevermind that their first cut at the release totally broke the program and they had to release a version bump a day later. Many upstreams are terrible at release management, and distributions are valuable because they do that work.
Bottom line is if you want the distro to help you, understand the distro first.
The problem Shuttleworth is addressing is upstream packages that are out of date. They have random or at least highly unpredictable release schedules that don't seem to work well for anyone. By suggesting a specific date for releases, Shuttleworth is attempting to shift a small burden to upstream to relieve a much larger strain in Ubuntu. Once the Freezes are past, it takes significant work to backport a bugfix patch to the feature frozen package. I think Shuttleworth's case would be greatly helped if he could point to a smallish project and how exactly it synchronizes with Ubuntu. I suspect his goal is KDE however, so a more general "this is healthy" essay is all we'll get.
Strategically, if upstreams were to do this on Ubuntu's cycle, Ubuntu would essentially be the Open Source release engineers, and many more upstreams would be invested in Ubuntu's release process. Some people recognize this and ask why they should follow Ubuntu over their own distro of choice, to which Shuttleworth said if any two distros could agree on a date, Canonical would follow suit.
At least for my model, the bios is a matter of pressing F1 at the right time. Not an Award though, which people might be familiar with. Linux has similar BIOS tools available. Out of curiousity, which BIOS settings aren't available from the boot menu?
No engineer I think was ever claiming that engineers ignored money. The OP was crying out that engineers are not valued in a company, and that translates into people wandering off into finance and marketing / advertising rather than put up with the kind of companies allow to happen. It's a far cry from "we should teach more of this business math", and I think you brought in some recent personal experience to the conversation unannounced.
In fact, much of the math they rely on was done by brilliant guys who studied alongside engineers in traditional math. For example, the Black-Scholes option pricing formula is based on the holy grails of engineering, rocket guidance and partial differential equations, and it led to both obscene profits for a while, and a quite large crisis after LTCM's secret got out and profits fell while management was screwed. Similar to the subprime hedge fund collapse that saw stocks oversold. We don't need maths specific to "business" (finance, everyone really means) in high school. Calculating a P:E ratio is easy, and even once you have it, what do you do with it? Lets leave the math track alone, and start instead with a basic microeconomics course in schools. It's far more practical for every business not running a bank to understand how competition works and why eliminating taxes won't do much to solve the high price of gas.
I think you misunderstand. One of the lines flagged did something worthless (add uninitalized memory into the pool or something I cannot recall) and the other did something important (add sources of entropy). The patch that fixes it undoes only one of the comments. The other has been argued to be of little consequence either way. Adding a fixed constant to an entropy pool doesn't make it less random, which is basically what it does since the memory's unlikely to have been touched by that program.
At any rate, this was definitely a case of an overzealous perfectionist removing memory debugger warnings, and an upstream that does a borderline practice without explanation in code or the FAQ.
It really doesn't happen as often as you'd think. Apparently the difference between a troll tag and a flamebait tag is that trolls have bad grammar and poor spelling.
We codify these ethics in law by making engineers responsible for failures of oversight, and hold companies liable for selling dangerous products against the recommendations of their engineers. It is of course also good long run business to make safe designs for customers, but a shit ton of management is not in it for the long run. It's probably easier to ship a faulty design, put it on the resume as a success and find a new job before the whole thing falls apart. The right way is to go back to the customer like a debtor to a bank and say "hey, we can't meet this deadline" or "we can't meet this requirement at this price" and work something out.
No, the whole point is to create a copyright license that allows collaboration between developers, without being able to pull an evil trick and renege on your half of the agreement. The difference between a fork and a branch is easy to spot: hostilities directly before the fork. I think their website is quite clear on where they stand.
Toshiba BIOS settings are weird, but in their defense they have been doing this for longer than anyone else. They did were doing ACPI stuff before it was formalized. We see this over and over again -- someone comes up with an okay idea on their own. Then an industry comittee or Microsoft or someone else defines a standard. Now you have two interfaces and neither party have incentives to resolve the situation. It happened with ACPI and with wireless drivers, probably more. But when you say userland, do you mean "not in kernel" or "not from BIOS itself?" My laptop has a goofy configuration screen that handles most everything I'm aware of. I wouldn't be surprised to hear there's more out there that the boot menu doesn't touch, as the whole thing scares me a bit and I stay away from messing with it when possible.
In Linux, big changes like this mostly only happen when there's universal support. Open Source makes it possible to see how the old stuff worked and incorporate the changes to a new design.