The academic estimates I've seen for current code are in the range of one bug for every 20-40 lines of code. Mind you, I think that's very vulnerable to definition: if code doesn't do what the requirement says, theoretically that's a bug, but if there is no requirement separate from the code (the normal FOS case) you've reduced the number of bugs without increasing code quality. I think a more important point is that we used to be able to associate a bug with a line of code, or a region of the code. Now a bug is more likely to be something like "3G modem type XYZ indicates an invalid SIM after 20s of poor signal strength", and is probably not connected to a code error per se. If we allow that sort of definition, then based on the asymptotic decrease I see in finding bugs in a particular supplier's mature code over a couple of years hammering, I think about one bug per 300 lines is about right.
BTW, I'm not sure what the 300M lines refers to. The total corpus in current use worldwide is way more than that. I've written about 100k lines myself for production code, and I'm sure there are more than 3000 programmers out there! Did you mean the source for Windows, perhaps?
My favourite bit of code was about 22 years ago. It was supposed to control some scientific equipment that we had bought from Denmark, and I had insisted on getting the source because of previous problems with the supplier. In summary:
- 6000 lines of Pascal
- 200 global variables
- 3 local variables
- 1 comment - the single word "midlertidig"
Oh, and one bug. Code really was less buggy back then.
I'm not familiar with this theory that there was a successful revolution in the North American colonies in 1776. Can you send me an example where the theory accurately predicts known experimental results? Also, what experiments would you posit to prove or disprove the theory? I eagerly await your reply.
Seriously, scientific method works well, but has a limited domain of application that excludes it from most areas that people are actually interested in. Mathematical method gives even more robust results, but is even more limited in application and does not use prediction/experiment. Historical method, in its domain, works about as well as scientific method (as an ex-scientist, I'm a bit cynical about the frequency with which scientific results are fiddled), but for obvious reasons does not rely on prediction and experiment. For pretty well everything else, you cannot use even slightly formal methods. Got a theory about the current political outlook of the leaders of North Korea? Good luck trying to build a robust experimental test to verify it.
It used to be common to launch spotter seaplanes off short (30 foot) catapults, running sideways across a battleship or even mounted on top of a gun turret, then using a crane to get them back on board. There's a picture of a Supermarine Walrus being launched halfway down this page. That particular type could land in pretty rough seas: I've seen film of a landing in 4-5' chop. See here for an open water landing. The father of a friend of mine got pulled out of the drink by rescue Walrus (which I have to be careful not to call by its common name of "Shagbat" in his presence) after he bailed out of a Spitfire. You might also be interested in an account of the first trans-Atlantic flight, which involved forced landings on the sea (this was about two weeks before Alcock and Brown, who had the first non-stop flight).
By the way, there was also one British submarine seaplane carrier, the M2.
Not a daft question, but it's not a separate standard. SIP is the closest thing we have to a VoIP standard. IMS is the specification of how to fit SIP into a GSM network. There's a lot that SIP doesn't specify - e.g. authentication and authorisation (GSM networks need to use SIMs and the corresponding HLR/HSS back end databases), accounting and billing, gateways to circuit-switched voice etc.
I know it sound odd, but there is a small market for just that in the UK. I work for a mobile phone company, and I have two phone numbers. One is a normal mobile phone number, and you pay to ring it as usual. The other is a "landline" number - you pay at landline rates to ring it, and my company picks up the difference. There's also some fairly sophisticated PABX functionality on the "landline" number - hunt groups, black/white listing, out of hours handoffs etc. Personally I never use it, but some customers do find it useful.
You know why they reckon Harley riders ride in threes? One to work on the bike, one to pass the tools, and one to flag down passing riders for more hammers.
If it's a "da-da-daah da-da-daah da-da-daah" interference pattern, that's a location update. Why yours is doing it so often, I don't know. Other than an odd manufacturer, my main guess would be that you are in poor coverage. If the phone drops out of coverage, it will call in when it re-associates. Sending or receiving an SMS would cause it, as would using GPRS (do you poll for email?).
No, that's not the case. A GSM phone will only call in every few hours; when it is switched on or off; when it needs to call out or send an SMS; when it is asked to call in; or when or when it moves between areas covered by different MSC/VLRs. An MSC/VLR covers a large area of a country with thousands of base stations. The bit about "asked to call in" is interesting. The network knows that the handset is in the area covered by an MSC/VLR, but not where, so it broadcasts a request for contact over the base stations in the area. The handset responds, localising itself to a base station. The point is to minimise signally costs and battery power consumption.
Yes, you can use the information about the last localisation in legal investigations, because the network keeps track of where and when you were last seen. It's also possible to send a silent SMS to get the phone to localise. However there is no continuous tracking of handsets by default.
It will have faded, but less than other colour films. IIRC, the estimated colour lifetime of Kodakchrome was about 60 years, vs 30 years for E6 process film and 200 years for conventional B&W negatives.
And checking the density is part of what they do to assay it. Look, check it for yourself: if you want to sell a gold bar, it either has to have been in controlled storage (i.e. you can't have kept it yourself) or it needs to be re-assayed, which will cost you money. This isn't an argument about physics, I'm just telling you the way the trade works.
I looked at getting some gold a couple of years ago as a precaution when the banks started having problems. Two issues: firstly, it's difficult to sell unless it has never left the vault of a registered gold dealer. Because it's easily adulterated, it has to be re-assayed unless there is a chain of trust. Coins are a bit easier to sell, but sell at a heavy discount. Secondly, the price is wildly unstable, so that it's only suitable for speculation, not as a precautionary investment. IMHO the best precautionary investment is some dollars, yen, euros and pounds held as banknotes. Obviously you will lose some value to inflation, but that's low at the moment and if you view that as your "insurance" charge, it's not that expensive.
I work in data product development for a major international mobile phone company. No, we absolutely do *not* want to land you with an unexpectedly large bill. We've never had a foul-up of that order: our worst was a man roaming in Germany who got a £600 bill when our product showed him that he should have got a bill for about a third of that amount. We waived that and spent a lot of time working out what could have gone wrong. There were several changes in the product following that: in particular we took the responsibility for reporting charges out of the PC client and provided a web-based system in the home network, because we thought he had probably inadvertently used the SIM with a different client or on a different PC. I know no-one is likely to believe this, but we do have serious conversations about what is best for the user.
There are a few approaches for mitigating the problems of bill shock: for instance someone mentioned CAMEL up-thread. Unfortunately we can't always make solutions water-tight due to limitations in either the home or the visited network, and for some cases there is a trade-off where we can't get real-time information and being over-cautious might prevent access by a business traveller who really does understand that they will be charged heavily. I used to travel extensively when data was much more expensive, and yes I have racked up much more than $100 in a session, but this was a rational business decision at the time. Not everyone is a private user, and we don't always have the information to tell what sort of customer we are dealing with.
In this specific case, AT&T could certainly have done better, but the real problem is that the cruise ship should not have had its system switched on. I'm pretty sure this is illegal, as they wouldn't have a spectrum licence. This would imply that AT&T would have grounds to void the interconnect charge, so its not surprising that they eventually voided the charge to the user. As to the huge figures involved: that's probably because the visited network was charging for satellite costs, which can be very expensive in some cases. The assumption is that you know that you are on a satellite network and modify your behaviour accordingly: I'd be interested to know if there was anything on the ship which warned of how expensive the traffic was when you were at sea.
Capital punishment solves nothing, and just feeds the basest desire of humans for revenge.
I am reluctant to argue for capital punishment, but I take issue with your word just. I believe that our form of law is a covenant between the state and the people. The people consent to be governed, and to give up certain rights of self-protection and retaliation which they believe themselves to have, in exchange for the protection of an equitable system of law. If the law ceases to satisfy its side of the covenant, the people will believe that the right to revolt, and the rights of self protection and retaliation revert to them.
Yes, stringing them up would be a simplistic, brutal reversion to the dark ages. But maybe that is the point? These people undermined the law itself, and I believe that is as serious as treason. While I would not sentence them to death myself, I wouldn't shed a tear if their heads ended up on spikes outside the county court.
Hmm, I'm getting depressingly right-wing in my senescence.
Ah, well I'm more into bikes, so you can't wind me up that way. Now I could mention the reason that afficionados of a certain American motorcycle travel in groups of at least three is so that one can work on the bike, one can pass the hammers, and a third can flag down passers-by to borrow more hammers. But that would be off-topic, so I won't.
No, the X-type is the modern small Jaguar, and mine used a Ford diesel engine. Reliable, economical, but a bit tinny and un-interesting. The worst thing about it was that the headlight bulbs only lasted about 25k miles and they were a bit awkward to change.
I run a Mini Cooper Diesel. At 104g/km, it's the second lowest CO2 on the UK market, and it does a consistent 71 mpg (US), 89mpg (Imperial) cruising at 60 on the motorway averaged over the 12 miles in to work. But the reason I got it rather than the Prius (which I hired for a week) was that the Mini is just more fun. The top speed isn't that remarkable, perhaps 120mph, but it goes through corners like a rat up a drainpipe. There's more room in the front than in my old X-Type Jag, and while it's not huge in the back, passengers seem ok for an hour or two.
Why are we still required to pay fifteen dollars for a CD?
You aren't. If you don't want to pay fifteen dollars, no-one's forcing you to buy the CD. If you want it, the only financial definition of the value of the CD is what people are prepared to pay for it.
One thing I really don't like about/. is the way libertarians turn in to communists when it comes to free-loading music (and I apologise if I am over-generalising in your particular case, comrade). Guys, just because the ??AA are litigious bottom-feeding parasites doesn't give you the right to set a value on someone else's property before you buy (or take) it.
Let me put it another way: we hear plenty on this site about getting music cheaply or free, but no-one is making the same point about commercial software, even though it can be eye-wateringly expensive. I'd love a copy of Photoshop, which I can't afford, but I bet that if I boasted of having a pirate copy here, the reaction would be far more negative than if I said that I'd got a large pirate MP3 collection. Instead the general reaction is to support open source software, and in some cases to contribute to it. Why? Is it because most of us are in some area of the IT industry and have a natural inclination to value other programmer's work rather than rip it off, whereas few of us work in the commercial music industry?
640kb wasn't an MS/DOS limit, even given this. It was specific to PC/DOS and the IBM PC. Early MS/DOS machines had different limits: for instance I remember that on the RM Nimbus as being 768kb due to different use of high memory.
Conversely, there used to be a letter yogh in Scots, which for similar reasons became written with a "z" character. This is why some names and place names are pronounced oddly - for instance Menzies is pronounced "Mingis"
No, the original objective for the Harrier family was to have a supersonic fighter, but the money ran out and they settled for what was feasible given the budget. In any case, only some Harriers were for ground support: carrier-based air defense is also a major role.
Bring back the English Electric Lightning - the original supercruiser. Mainly 'cos it was basically two huge engines with a couple of control surfaces tacked on the sides and the pilot dangling his feet in the inlet tract.
Slashdot Mirror
The academic estimates I've seen for current code are in the range of one bug for every 20-40 lines of code. Mind you, I think that's very vulnerable to definition: if code doesn't do what the requirement says, theoretically that's a bug, but if there is no requirement separate from the code (the normal FOS case) you've reduced the number of bugs without increasing code quality. I think a more important point is that we used to be able to associate a bug with a line of code, or a region of the code. Now a bug is more likely to be something like "3G modem type XYZ indicates an invalid SIM after 20s of poor signal strength", and is probably not connected to a code error per se. If we allow that sort of definition, then based on the asymptotic decrease I see in finding bugs in a particular supplier's mature code over a couple of years hammering, I think about one bug per 300 lines is about right.
BTW, I'm not sure what the 300M lines refers to. The total corpus in current use worldwide is way more than that. I've written about 100k lines myself for production code, and I'm sure there are more than 3000 programmers out there! Did you mean the source for Windows, perhaps?
- 6000 lines of Pascal
- 200 global variables
- 3 local variables
- 1 comment - the single word "midlertidig"
Oh, and one bug. Code really was less buggy back then.
Now get off my lawn, you kids.
I'm not familiar with this theory that there was a successful revolution in the North American colonies in 1776. Can you send me an example where the theory accurately predicts known experimental results? Also, what experiments would you posit to prove or disprove the theory? I eagerly await your reply.
Seriously, scientific method works well, but has a limited domain of application that excludes it from most areas that people are actually interested in. Mathematical method gives even more robust results, but is even more limited in application and does not use prediction/experiment. Historical method, in its domain, works about as well as scientific method (as an ex-scientist, I'm a bit cynical about the frequency with which scientific results are fiddled), but for obvious reasons does not rely on prediction and experiment. For pretty well everything else, you cannot use even slightly formal methods. Got a theory about the current political outlook of the leaders of North Korea? Good luck trying to build a robust experimental test to verify it.
By the way, there was also one British submarine seaplane carrier, the M2.
Not a daft question, but it's not a separate standard. SIP is the closest thing we have to a VoIP standard. IMS is the specification of how to fit SIP into a GSM network. There's a lot that SIP doesn't specify - e.g. authentication and authorisation (GSM networks need to use SIMs and the corresponding HLR/HSS back end databases), accounting and billing, gateways to circuit-switched voice etc.
No use unless the people you are communicating with do the same. Even then, it doesn't help with traffic analysis.
I know it sound odd, but there is a small market for just that in the UK. I work for a mobile phone company, and I have two phone numbers. One is a normal mobile phone number, and you pay to ring it as usual. The other is a "landline" number - you pay at landline rates to ring it, and my company picks up the difference. There's also some fairly sophisticated PABX functionality on the "landline" number - hunt groups, black/white listing, out of hours handoffs etc. Personally I never use it, but some customers do find it useful.
You know why they reckon Harley riders ride in threes? One to work on the bike, one to pass the tools, and one to flag down passing riders for more hammers.
(with best wishes from a Triumph rider)
If it's a "da-da-daah da-da-daah da-da-daah" interference pattern, that's a location update. Why yours is doing it so often, I don't know. Other than an odd manufacturer, my main guess would be that you are in poor coverage. If the phone drops out of coverage, it will call in when it re-associates. Sending or receiving an SMS would cause it, as would using GPRS (do you poll for email?).
No, that's not the case. A GSM phone will only call in every few hours; when it is switched on or off; when it needs to call out or send an SMS; when it is asked to call in; or when or when it moves between areas covered by different MSC/VLRs. An MSC/VLR covers a large area of a country with thousands of base stations. The bit about "asked to call in" is interesting. The network knows that the handset is in the area covered by an MSC/VLR, but not where, so it broadcasts a request for contact over the base stations in the area. The handset responds, localising itself to a base station. The point is to minimise signally costs and battery power consumption.
Yes, you can use the information about the last localisation in legal investigations, because the network keeps track of where and when you were last seen. It's also possible to send a silent SMS to get the phone to localise. However there is no continuous tracking of handsets by default.
S'funny: Windows for Satellites won't let me enter those coords, and Clippy keeps asking if I'm trying to enter a support call.
It will have faded, but less than other colour films. IIRC, the estimated colour lifetime of Kodakchrome was about 60 years, vs 30 years for E6 process film and 200 years for conventional B&W negatives.
And checking the density is part of what they do to assay it. Look, check it for yourself: if you want to sell a gold bar, it either has to have been in controlled storage (i.e. you can't have kept it yourself) or it needs to be re-assayed, which will cost you money. This isn't an argument about physics, I'm just telling you the way the trade works.
I looked at getting some gold a couple of years ago as a precaution when the banks started having problems. Two issues: firstly, it's difficult to sell unless it has never left the vault of a registered gold dealer. Because it's easily adulterated, it has to be re-assayed unless there is a chain of trust. Coins are a bit easier to sell, but sell at a heavy discount. Secondly, the price is wildly unstable, so that it's only suitable for speculation, not as a precautionary investment. IMHO the best precautionary investment is some dollars, yen, euros and pounds held as banknotes. Obviously you will lose some value to inflation, but that's low at the moment and if you view that as your "insurance" charge, it's not that expensive.
There are a few approaches for mitigating the problems of bill shock: for instance someone mentioned CAMEL up-thread. Unfortunately we can't always make solutions water-tight due to limitations in either the home or the visited network, and for some cases there is a trade-off where we can't get real-time information and being over-cautious might prevent access by a business traveller who really does understand that they will be charged heavily. I used to travel extensively when data was much more expensive, and yes I have racked up much more than $100 in a session, but this was a rational business decision at the time. Not everyone is a private user, and we don't always have the information to tell what sort of customer we are dealing with.
In this specific case, AT&T could certainly have done better, but the real problem is that the cruise ship should not have had its system switched on. I'm pretty sure this is illegal, as they wouldn't have a spectrum licence. This would imply that AT&T would have grounds to void the interconnect charge, so its not surprising that they eventually voided the charge to the user. As to the huge figures involved: that's probably because the visited network was charging for satellite costs, which can be very expensive in some cases. The assumption is that you know that you are on a satellite network and modify your behaviour accordingly: I'd be interested to know if there was anything on the ship which warned of how expensive the traffic was when you were at sea.
Capital punishment solves nothing, and just feeds the basest desire of humans for revenge.
I am reluctant to argue for capital punishment, but I take issue with your word just. I believe that our form of law is a covenant between the state and the people. The people consent to be governed, and to give up certain rights of self-protection and retaliation which they believe themselves to have, in exchange for the protection of an equitable system of law. If the law ceases to satisfy its side of the covenant, the people will believe that the right to revolt, and the rights of self protection and retaliation revert to them.
Yes, stringing them up would be a simplistic, brutal reversion to the dark ages. But maybe that is the point? These people undermined the law itself, and I believe that is as serious as treason. While I would not sentence them to death myself, I wouldn't shed a tear if their heads ended up on spikes outside the county court.
Hmm, I'm getting depressingly right-wing in my senescence.
Ah, well I'm more into bikes, so you can't wind me up that way. Now I could mention the reason that afficionados of a certain American motorcycle travel in groups of at least three is so that one can work on the bike, one can pass the hammers, and a third can flag down passers-by to borrow more hammers. But that would be off-topic, so I won't.
No, the X-type is the modern small Jaguar, and mine used a Ford diesel engine. Reliable, economical, but a bit tinny and un-interesting. The worst thing about it was that the headlight bulbs only lasted about 25k miles and they were a bit awkward to change.
I run a Mini Cooper Diesel. At 104g/km, it's the second lowest CO2 on the UK market, and it does a consistent 71 mpg (US), 89mpg (Imperial) cruising at 60 on the motorway averaged over the 12 miles in to work. But the reason I got it rather than the Prius (which I hired for a week) was that the Mini is just more fun. The top speed isn't that remarkable, perhaps 120mph, but it goes through corners like a rat up a drainpipe. There's more room in the front than in my old X-Type Jag, and while it's not huge in the back, passengers seem ok for an hour or two.
Turbines were used extensively for high performance marine transport. Here is where it started, in 1894.
You aren't. If you don't want to pay fifteen dollars, no-one's forcing you to buy the CD. If you want it, the only financial definition of the value of the CD is what people are prepared to pay for it.
One thing I really don't like about /. is the way libertarians turn in to communists when it comes to free-loading music (and I apologise if I am over-generalising in your particular case, comrade). Guys, just because the ??AA are litigious bottom-feeding parasites doesn't give you the right to set a value on someone else's property before you buy (or take) it.
Let me put it another way: we hear plenty on this site about getting music cheaply or free, but no-one is making the same point about commercial software, even though it can be eye-wateringly expensive. I'd love a copy of Photoshop, which I can't afford, but I bet that if I boasted of having a pirate copy here, the reaction would be far more negative than if I said that I'd got a large pirate MP3 collection. Instead the general reaction is to support open source software, and in some cases to contribute to it. Why? Is it because most of us are in some area of the IT industry and have a natural inclination to value other programmer's work rather than rip it off, whereas few of us work in the commercial music industry?
--
I do believe in intangible property
640kb wasn't an MS/DOS limit, even given this. It was specific to PC/DOS and the IBM PC. Early MS/DOS machines had different limits: for instance I remember that on the RM Nimbus as being 768kb due to different use of high memory.
Conversely, there used to be a letter yogh in Scots, which for similar reasons became written with a "z" character. This is why some names and place names are pronounced oddly - for instance Menzies is pronounced "Mingis"
No, the original objective for the Harrier family was to have a supersonic fighter, but the money ran out and they settled for what was feasible given the budget. In any case, only some Harriers were for ground support: carrier-based air defense is also a major role.
Bring back the English Electric Lightning - the original supercruiser. Mainly 'cos it was basically two huge engines with a couple of control surfaces tacked on the sides and the pilot dangling his feet in the inlet tract.