I'm amazed so few people have mentioned B5. OK, it was a series with some extended episodes rather than a movie, but...
This was pretty much the first time space-based sci-fi went completely CG for the effects, with no models at all. They also did it very realistically: not only does the station rotate to give artificial gravity, it launches its fighters using that rotation, and the fighters themselves perform manoeuvres that are realistic in zero-g, rather than the remarkably atmospheric effects you see in most sci-fi TV/movies. When a ship gets hit by a beam, it doesn't explode into a huge ball of fire, it separates out into thousands of little pieces, which fly apart in all directions. Hell, according to JMS there was significant interaction with NASA -- in both directions -- about spacecraft design...
I think using CG allowed the various space effects to look a lot more realistic than anything that had gone before. Star Wars, Star Trek, all the movies people have mentioned above that used models... They all had their place, but none of them had either the realism or the sheer scale of the big space effects in B5.
It really helps if you have a nice flexable definition of "speeding" To most people that just means driving at a speed greater than the speed limit but they could just as easily be defining it as "driving too fast for conditions".
I've seen figures similar to what the grandparent post mentioned, which were for "excessive speed" as a "major contributory factor" in the accidents concerned, i.e., one without which the accident probably would not have occurred at all.
IOWs, they already thought of your point and they agree with you.:-)
The UK motoring laws may be silly and out of date in some cases, but the driving test isn't one of them. You are not required to take professional tuition before going for a test; pretty much any reasonably experienced and qualified driver can accompany a learner on a public road. Your figures are exaggerated anyway.
Generally speaking, the driving tests here are fair and reasonable. At the end of the test, you receive a detailed report indicating for various areas whether you had minor faults or a major or dangerous fault. There are clear criteria defining how many faults translate into a failure, and test examiners have to justify why they fail every person they don't pass -- which is quite a significant proportion of the tests they take. There is scope for appeal and for having an independent witness riding with you in cases of suspected foul play (not that these are often used other than for an instructor to sit in with a student who repeatedly fails to help identify the persistent problem).
Sure, there's some variability with examiners, but almost all are fair and reasonable. IMHO, the UK driving test is a pretty good base line for deciding whether or not someone is safe to let out on the roads unaccompanied.
Re the silly test failure you mentioned, you shouldn't believe everything you hear about other countries without getting your own information first-hand. What you describe would almost certainly not qualify as a failure in itself, so there's pretty much guaranteed more to it than the story told. If I believed everything I was told about the US, I'd think it was full of capitalist, litigious, war-mongering, overly religious pricks with no regard for anyone outside their own country, who consume nothing but burgers, donuts and sodas, say "Have a nice day" all the time without ever really meaning it, shoot anyone who comes into range and have a penchant for Britney Spears in a schoolgirl uniform. Fortunately, I know better. See my point?
Few car-pedestrian collisions occur without the car making any attempt to avoid the pedestrian. The car is usually travelling much slower when it hits the pedestrian than its normal speed. Remind me again what the logical or statistical validity of your relative survivability argument was?
I would favour a blanket 20mph on urban and surburban streets, except dual carriageways and arterial roads.
I live in a city where the local authorities are a little overzealous in imposing 20mph limits, and very evasive when questioned about their actions. (Plenty of local authorities across the UK currently seem to be overruling the recommendations of traffic experts and putting in 20mph limits anyway, by the way.)
On one particular road that comes to mind, not a single driver keeps to the recently imposed 20mph limit. Not one. Never. At least not in my experience. Why? Because it's a silly and offensive limit. You have excellent visibility, the road is long and straight, the traffic is light, the only significant hazards are a school and a small number of junctions. People now drive faster along that road than when it had a 30 limit, presumably because they figure if there's no point sticking to the limit you might as well drive at what you perceive to be a safe speed. Incidentally, AFAICT the road also has a perfect safety record going back years to before half the buildings were even there.
And your blanket absolute speed limit is a good idea why, again?
Oh, and just give up with the 80-on-a-motorway argument, would you please? Go learn about the magic figure 85% as it relates to speeds and accidents, look at the safety record of motorways in the UK (and perhaps in Germany, where they have no limit at all but have an exemplary safety record) and come back when you have an argument that's more than "I think it should be this, because."
The spy-in-the-car will either fall at the first hurdle or people will just accept it.
The Association of Chief Police Officers' official guideline for prosecution in speeding cases is that the vehicle was travelling at more than 10%+2mph over the speed limit. Why? Because if you tried to prosecute everyone who broke the absolute speed limit, the court system would grind to a halt.
Things like spy-in-the-car technology or speed limiters will fail for much the same reason, even if they're ever implemented successfully in the first place. Given that they can't even set up cameras in London that reliably detect people in the new congestion charging zones, there's not much danger of Big Brother sitting on my back seat any time soon.
You're so right about speed limits; they should be set by independent people who know what they're doing and study the facts, not on the whims of government departments who stand to make money from a beneficial adjustment.
For the benefit of those outside the UK, the "school run" where parents load their little darlings into a 4x4 and then set in solid traffic for 45 minutes, rather than simply walking them the half-mile to school as we always used to, is one of the single worst things about the UK road system today. Providing an effective school bus service to alleviate this would do more than almost any other measure that's been proposed to lift the burden of congestion (as magically happens during school holidays, even though clearly most of the same people are using most of the same roads as much as any other time).
Hence, I'm in favour of virtually ANY regulation of motorists.
/me files poster under "fool with good intentions". If you want to improve things, don't advocate what you think "ought to work", based on your own presumptions or bias. Do the research into what does work, much of which is readily available when it comes to transport policy, and advocate change to those things that do.
Over-regulation of motorists, and the consequent resources required to enforce those regulations, are precisely the reason there aren't enough police officers available to pull over the dangerous drivers any more; they're too busy fining people for doing 90 on a clear motorway.
The vast majority of drivers, left to their own devices, will drive reasonably safely and without doing really stupid things. Modest improvements in driver education, particularly in the few months following passing a test once a driver has some "real world" experience, have been shown to make things much safer still. Throwing up so-called safety cameras and speed bumps everywhere isn't the answer, it's just a set of irritating distractions spun with a healthy dose of very poor statistical analysis.
None of the above is supposition on my part. It's all well known, and you can read about it in numerous sites on the web if you care to Google into the subject. If you want to help, try getting your local reps to look at the facts, instead of pursuing their own well-intentioned but ultimately more dangerous agendas.
Nobody has a god-given right to carry themselves about in two tons of steel - especially in a small and crowded country like the UK with an adequate public transport infrastructure.
You mean the infrastructure featuring train companies who advise their commuters to take the rest of this week off, so they can get on with necessary maintenance work without the travellers taking 2 hours to finish a journey that usually takes a quarter of that? Yeah, we have a great public transport infrastructure.
As to "driving at 35 when the limit is 30"... in any case, either of those speeds is too fast on urban UK roads.
Like hell it is. There are plenty of urban UK roads where it is perfectly reasonable to do those speeds. Of course, part of the problem is that you're thinking as though one speed limit could possibly be appropriate for all conditions. There are roads where doing 60 is safe for 90+% of the day, but during an hour when kids are leaving a nearby school doing half that is reckless. Absolute speed limits are a daft idea, end of story.
There's also a massive problem with illegaly-owned, uninsured cars and people driving them without the proper license.
Who, coincidentally, also make up most of the excessive speeders, runners of red lights, tailgaters, etc.
The UK has (very) restrictive gun laws for precisely this reason, and they enjoy huge majority support.
That's debatable. Amongst the informed population, i.e., those who know about guns and choose to research the effects of having an armed population based on what happens in other countries in the world (not just the US, please), it's a very different picture.
"If you make guns illegal, only criminals will have guns," chant the pro-gun lobby. It isn't as simple as that of course, and personally I wouldn't want to see complete legalisation of firearms out of the blue in this country for a variety of reasons, but there's more than a touch of truth in that chant.
If you have a dodgy CGI script on an IIS server, may god have mercy on your soul. With Apache your chances are much better.
OK, your turn: post an example or STFU. How are your chances better with Apache, exactly?
But this has nothing to do with the underlying operating system. Apache can run on windows, too.
Indeed, but in the case of my example, it was a Linux box set up by geeks who were generally competent. They just missed something, and as a result once a cracker had some access, they got root. So much for Linux being inherently more secure, it's user-specific privilege model limiting the damage, yada yada.
That is irrelevant. What the grandparent is saying is that an average Unix power user could potentially block a virus on his own, wheras with Windows this would be next to impossible.
Again, post an example or STFU. I consider myself, amongst other things, a Windows power user. I have never, to my knowledge, had a virus on a system I run, nor been infected with the various worms, etc. Go figure.
And as for Windows not going anything... You do know what Windows Update is, right? Microsoft released a patch that would have stopped MSBlast dead a month before it came out. Smart users installed it, and we don't have MSBlast. Go figure some more.
Unix is simply designed and developed much more with security and securability in mind.
And yet typical Linux installations from the major distros don't do much to, for example, stop you getting root once you've got some system access. The last machine I know was rooted was a shared Linux box, running Apache, that had a dodgy CGI script on it. And yes, it's run by people who have a clue what they're doing, and have since taken steps to fix that problem. But it was still rooted.
Can the average WinDOS power user come up with a script to disable Win32 IPC off the cuff?
The average Windows power user doesn't suffer from MSBlast etc. any more than the average Linux power user, because the system will be patched, running a personal firewall, etc.
Windows does nothing to encourage tools or practices that can easily be deployed as roadblocks to malware.
Oops, it seems I've been trolled. Sorry guys, my mistake.
Tools like Apache and -- perhaps more relevant -- Linux aren't invulnerable, y'know. My club's whole web site has been messed up in recent weeks because someone rooted the host server. How? Another club hosted on it apparently put up a dodgy CGI and the system wasn't set up properly to lock out anyone who got through it.
If you really think Linux and co don't get large numbers of security problems themselves, you need to subscribe to more of the big security mailing lists.
I didn't have ANY trouble with SoBig.. or Blaster.. why, because I patched my system and secured it..
Likewise for the patching.
Unfortunately, I also help to run a society with 2,000 people on its mailing list. Guess how many of them have got the virus at some point, and how many other people are now getting the virus with a spoofed From: header containing the mailing list address? I know nobody got it from that address, because there is exactly one person in the world who can authorise posts to that list, and my system is clean. Doesn't stop the irritating automatic replies from "clever" ISPs though.:-(
Unfortunately, they aren't nearly so cheap here. We use them at work sometimes, and to get one worth having for backup purposes, we seem to be looking at closer to ten times the figure you gave.
As for practical uses, not everyone in the world works with Word documents and CPP files, though as it happens, I do. However, we also have overnight builds running on a dozen platforms, with a full set of automated test for each, resulting in hundreds of megs of output each night. It's very useful to have those executables and test results available months later when you're trying to track down when some regression started, but at the moment, hard drives or CD-Rs seem to be the only media up to the job at a cost-effective rate and an acceptable ease of use.
It's a shame "pen drives" and the like are still so expensive. A portable, robust hard disk that basically just sits in your pocket and plugs into a USB port when needed has a lot going for it...
And of course there's also the CD media. If you bought the 10 cent bulk discs and expected them to last, shame on you. I record at slow speed to the old dark-blue verbatims whenever I can, and after 7 years I havn't lost data yet.
I guess just like audio and video cassettes, buying cheap is a false economy, and you want a quality brand for anything you're not going to throw away shortly afterwards. I was thinking about this just yesterday, while considering which of the vast array of CD-RW media before me to buy.
The question is, how do you tell what the quality brands are? Does buying in bulk always imply cheap and nasty, and buying a well-presented set of 10 CD-Rs with jewel cases always imply high quality? I doubt it.
I, too, have a set of the nice blue Verbatim disks, which have given me no problems thus far. I've also used the Master "Audio" 80min brand to record (legally!) music onto CDs for use at dancing events, and have had no problems with these. Both cost more than most brands on the shelf at the time I bought them.
That said, I always burn at well under the maximum quoted recording speed of my drive to reduce error. There ought to be a disclosure law on CD recorders/rewriters that using the much-advertised higher speeds will reduce the effective lifetime of the media.
Yesterday, I bought a set of 10 imation "1x-4x compatible" 650MB CD-RWs. Again, they were one of the more expensive brands on display, they're from a known brand and they look to be decent kit, but how do I tell? Certainly the label on the back of the pack -- basically a disclaimer in case you lose data because the disks are bad -- isn't very reassuring.
So, given the pool of knowledge we have here on Slashdot, perhaps this is a good place to ask: which CD media brands do you find are reliable, and which have caused bad experiences?
CDRs, if they are high quality and are properly stored, can last many years.
Um... Wasn't the whole point of the story that this isn't necessarily the case?
As for hard drives, yes, some of the newer brands that push the boundaries do fail with amazing regularity. However, if you stick to tried and tested kit, that doesn't push all the boundaries and puts reliability ahead of that extra 10GB per platter, then problems are still relatively rare.
General concensus is that you don't need to (and shouldn't) learn C first if C++ is your ultimate goal. By all means learn C if you want to use C, but if you want to use C++ then just go straight there.
Nobody ever said anything about not learning these things at all, not learning the entire language. That'd be stupid.
In fairness, Accelerated C++ has a lot going for it and does indeed break new ground (after an amazingly long time with an obvious hole in the market, I thought -- shoulda written it myself!) in how C++ is taught. OTOH, I find a lot of the comments about it somewhat annoying. It isn't perfect, far from it.
Basically, it's too trendy. Sure, it covers lots of STL things, but in practice a lot of people won't be using STL. (Except for fairly recent projects, it's far more likely that some other class library's containers will be the status quo on a project, and even some quite recent compilers don't support templates and/or don't have a good STL implementation.) The coverage of exceptions -- an equally if not more important topic than templates -- is almost non-existant. That was a great disappointment, in a book trying hard to look at C++ for its unique features and not the ones it borrows from C.
I think the book is much better for those who've programmed in other languages and then come to C++ than for beginners. As anyone who follows the C++ newsgroups can tell you, beginners who meet AC++ without other programming exposure are frequently overpowered by its fast-paced approach, and give up in frustration. Those with a "frame of reference" seem to get much more out of it.
Finally, it's written by the in-crowd. Many of the reviews were also written by the in-crowd. Draw your own conclusions about the objectivity of those reviews. This is my big problem with the ACCU book reviews, which are otherwise excellent IMHO. It's also a problem with the book's reputation on-line; so many "helpful" people on the newsgroups were bowing down at the creation of K&M that they recommended it with barely a second thought, and obviously without reading it first in many cases.
In summary, I think Accelerated C++ has a lot going for it, but it's way over-rated (mostly through no fault of its own) and it's not a book that stands alone (because it's in a short-but-sweet series, not a comprehensive one).
So let me get this straight. Gilette always had the best of intentions, and did not encourage the use of the system for photographing customers. Tesco said what the hell, and did it anyway. Now they've canned the trial after protests in Cambridge.
But wait! Now Tesco are going to try the same trick with DVDs at their Sandhurst store. Did they really listen at all?:-(
Oh, and since it's obligatory to bitch, I told them this last weekend.:-)
2003-08-17 01:48:32 Gilette/Tesco RFID trial ends in failure? (articles,slashback) (rejected)
I think perhaps we actually agree about most of the discussion here, we just looked at it from different perspectives initially.:-)
The one point where I do disagree with you comes from this, though:
The code and improvements based on the code will be kept accessible for the community. Why should we allow a company to proffit from the effort of others without giving anything back?
I think this is where I don't agree with your basic premise. If I take public domain code, work on it, and release the results, then what I am offering is the extra work. Everyone already has the public domain code, to do with as they wish, and nothing I do can change that.
The only question is what my extra work is worth. If it's valuable, people will pay for it, and it may be profitable for me. If it's not, anyone is free to ignore it and use the original, public domain code instead.
All software development is, to an extent, built on what has gone before, the previous state of the art. The game is all about ways to offer improved value. Amongst other things, you can
introduce a more advanced product
offer customisation services to clients wanting to specialise the basic product
offer supporting products to interact with a basic product the client already has.
It's these extras that are valuable, if everyone has what you started with anyway. IMHO, if the extras aren't paid for by the public, there is no reason why they should be publicly available under the same conditions as the original work that was publicly subsidised. Rather, the person or people adding the value should be the ones who benefit from it if they wish to do so.
Anyone who doesn't think the extras are worth the asking price is free not to use them; they lose nothing else as a result. It's really little different to writing a product from scratch, and the moral issue is the same to me: if someone does fair work, they should be entitled to fair benefits from it. If the work is publicly supported, it should be given back to the public when done. Simple as that, really.:-)
Slashdot Mirror
<ahem> Web services, anyone?
Oh, puleeeeze. RORing is so, like, 1990s, man. Everyone knows about it. Today's discriminating L337 H8x0r uses ROLing instead, for added security.
I'm amazed so few people have mentioned B5. OK, it was a series with some extended episodes rather than a movie, but...
This was pretty much the first time space-based sci-fi went completely CG for the effects, with no models at all. They also did it very realistically: not only does the station rotate to give artificial gravity, it launches its fighters using that rotation, and the fighters themselves perform manoeuvres that are realistic in zero-g, rather than the remarkably atmospheric effects you see in most sci-fi TV/movies. When a ship gets hit by a beam, it doesn't explode into a huge ball of fire, it separates out into thousands of little pieces, which fly apart in all directions. Hell, according to JMS there was significant interaction with NASA -- in both directions -- about spacecraft design...
I think using CG allowed the various space effects to look a lot more realistic than anything that had gone before. Star Wars, Star Trek, all the movies people have mentioned above that used models... They all had their place, but none of them had either the realism or the sheer scale of the big space effects in B5.
I've seen figures similar to what the grandparent post mentioned, which were for "excessive speed" as a "major contributory factor" in the accidents concerned, i.e., one without which the accident probably would not have occurred at all.
IOWs, they already thought of your point and they agree with you. :-)
The UK motoring laws may be silly and out of date in some cases, but the driving test isn't one of them. You are not required to take professional tuition before going for a test; pretty much any reasonably experienced and qualified driver can accompany a learner on a public road. Your figures are exaggerated anyway.
Generally speaking, the driving tests here are fair and reasonable. At the end of the test, you receive a detailed report indicating for various areas whether you had minor faults or a major or dangerous fault. There are clear criteria defining how many faults translate into a failure, and test examiners have to justify why they fail every person they don't pass -- which is quite a significant proportion of the tests they take. There is scope for appeal and for having an independent witness riding with you in cases of suspected foul play (not that these are often used other than for an instructor to sit in with a student who repeatedly fails to help identify the persistent problem).
Sure, there's some variability with examiners, but almost all are fair and reasonable. IMHO, the UK driving test is a pretty good base line for deciding whether or not someone is safe to let out on the roads unaccompanied.
Re the silly test failure you mentioned, you shouldn't believe everything you hear about other countries without getting your own information first-hand. What you describe would almost certainly not qualify as a failure in itself, so there's pretty much guaranteed more to it than the story told. If I believed everything I was told about the US, I'd think it was full of capitalist, litigious, war-mongering, overly religious pricks with no regard for anyone outside their own country, who consume nothing but burgers, donuts and sodas, say "Have a nice day" all the time without ever really meaning it, shoot anyone who comes into range and have a penchant for Britney Spears in a schoolgirl uniform. Fortunately, I know better. See my point?
Few car-pedestrian collisions occur without the car making any attempt to avoid the pedestrian. The car is usually travelling much slower when it hits the pedestrian than its normal speed. Remind me again what the logical or statistical validity of your relative survivability argument was?
I live in a city where the local authorities are a little overzealous in imposing 20mph limits, and very evasive when questioned about their actions. (Plenty of local authorities across the UK currently seem to be overruling the recommendations of traffic experts and putting in 20mph limits anyway, by the way.)
On one particular road that comes to mind, not a single driver keeps to the recently imposed 20mph limit. Not one. Never. At least not in my experience. Why? Because it's a silly and offensive limit. You have excellent visibility, the road is long and straight, the traffic is light, the only significant hazards are a school and a small number of junctions. People now drive faster along that road than when it had a 30 limit, presumably because they figure if there's no point sticking to the limit you might as well drive at what you perceive to be a safe speed. Incidentally, AFAICT the road also has a perfect safety record going back years to before half the buildings were even there.
And your blanket absolute speed limit is a good idea why, again?
Oh, and just give up with the 80-on-a-motorway argument, would you please? Go learn about the magic figure 85% as it relates to speeds and accidents, look at the safety record of motorways in the UK (and perhaps in Germany, where they have no limit at all but have an exemplary safety record) and come back when you have an argument that's more than "I think it should be this, because."
The Association of Chief Police Officers' official guideline for prosecution in speeding cases is that the vehicle was travelling at more than 10%+2mph over the speed limit. Why? Because if you tried to prosecute everyone who broke the absolute speed limit, the court system would grind to a halt.
Things like spy-in-the-car technology or speed limiters will fail for much the same reason, even if they're ever implemented successfully in the first place. Given that they can't even set up cameras in London that reliably detect people in the new congestion charging zones, there's not much danger of Big Brother sitting on my back seat any time soon.
You're so right about speed limits; they should be set by independent people who know what they're doing and study the facts, not on the whims of government departments who stand to make money from a beneficial adjustment.
For the benefit of those outside the UK, the "school run" where parents load their little darlings into a 4x4 and then set in solid traffic for 45 minutes, rather than simply walking them the half-mile to school as we always used to, is one of the single worst things about the UK road system today. Providing an effective school bus service to alleviate this would do more than almost any other measure that's been proposed to lift the burden of congestion (as magically happens during school holidays, even though clearly most of the same people are using most of the same roads as much as any other time).
That's a good link; the information there seems pretty complete and it cites the major Acts that are relevant.
NB: The summary provided is somewhat misleading, so if you really want to know, do follow that link.
/me files poster under "fool with good intentions". If you want to improve things, don't advocate what you think "ought to work", based on your own presumptions or bias. Do the research into what does work, much of which is readily available when it comes to transport policy, and advocate change to those things that do.
Over-regulation of motorists, and the consequent resources required to enforce those regulations, are precisely the reason there aren't enough police officers available to pull over the dangerous drivers any more; they're too busy fining people for doing 90 on a clear motorway.
The vast majority of drivers, left to their own devices, will drive reasonably safely and without doing really stupid things. Modest improvements in driver education, particularly in the few months following passing a test once a driver has some "real world" experience, have been shown to make things much safer still. Throwing up so-called safety cameras and speed bumps everywhere isn't the answer, it's just a set of irritating distractions spun with a healthy dose of very poor statistical analysis.
None of the above is supposition on my part. It's all well known, and you can read about it in numerous sites on the web if you care to Google into the subject. If you want to help, try getting your local reps to look at the facts, instead of pursuing their own well-intentioned but ultimately more dangerous agendas.
You mean the infrastructure featuring train companies who advise their commuters to take the rest of this week off, so they can get on with necessary maintenance work without the travellers taking 2 hours to finish a journey that usually takes a quarter of that? Yeah, we have a great public transport infrastructure.
Like hell it is. There are plenty of urban UK roads where it is perfectly reasonable to do those speeds. Of course, part of the problem is that you're thinking as though one speed limit could possibly be appropriate for all conditions. There are roads where doing 60 is safe for 90+% of the day, but during an hour when kids are leaving a nearby school doing half that is reckless. Absolute speed limits are a daft idea, end of story.
Who, coincidentally, also make up most of the excessive speeders, runners of red lights, tailgaters, etc.
That's debatable. Amongst the informed population, i.e., those who know about guns and choose to research the effects of having an armed population based on what happens in other countries in the world (not just the US, please), it's a very different picture.
"If you make guns illegal, only criminals will have guns," chant the pro-gun lobby. It isn't as simple as that of course, and personally I wouldn't want to see complete legalisation of firearms out of the blue in this country for a variety of reasons, but there's more than a touch of truth in that chant.
OK, your turn: post an example or STFU. How are your chances better with Apache, exactly?
Indeed, but in the case of my example, it was a Linux box set up by geeks who were generally competent. They just missed something, and as a result once a cracker had some access, they got root. So much for Linux being inherently more secure, it's user-specific privilege model limiting the damage, yada yada.
Again, post an example or STFU. I consider myself, amongst other things, a Windows power user. I have never, to my knowledge, had a virus on a system I run, nor been infected with the various worms, etc. Go figure.
And as for Windows not going anything... You do know what Windows Update is, right? Microsoft released a patch that would have stopped MSBlast dead a month before it came out. Smart users installed it, and we don't have MSBlast. Go figure some more.
And yet typical Linux installations from the major distros don't do much to, for example, stop you getting root once you've got some system access. The last machine I know was rooted was a shared Linux box, running Apache, that had a dodgy CGI script on it. And yes, it's run by people who have a clue what they're doing, and have since taken steps to fix that problem. But it was still rooted.
The average Windows power user doesn't suffer from MSBlast etc. any more than the average Linux power user, because the system will be patched, running a personal firewall, etc.
Oops, it seems I've been trolled. Sorry guys, my mistake.
s/communist/common sense/
:-)
Tools like Apache and -- perhaps more relevant -- Linux aren't invulnerable, y'know. My club's whole web site has been messed up in recent weeks because someone rooted the host server. How? Another club hosted on it apparently put up a dodgy CGI and the system wasn't set up properly to lock out anyone who got through it.
If you really think Linux and co don't get large numbers of security problems themselves, you need to subscribe to more of the big security mailing lists.
Likewise for the patching.
Unfortunately, I also help to run a society with 2,000 people on its mailing list. Guess how many of them have got the virus at some point, and how many other people are now getting the virus with a spoofed From: header containing the mailing list address? I know nobody got it from that address, because there is exactly one person in the world who can authorise posts to that list, and my system is clean. Doesn't stop the irritating automatic replies from "clever" ISPs though. :-(
Unfortunately, they aren't nearly so cheap here. We use them at work sometimes, and to get one worth having for backup purposes, we seem to be looking at closer to ten times the figure you gave.
As for practical uses, not everyone in the world works with Word documents and CPP files, though as it happens, I do. However, we also have overnight builds running on a dozen platforms, with a full set of automated test for each, resulting in hundreds of megs of output each night. It's very useful to have those executables and test results available months later when you're trying to track down when some regression started, but at the moment, hard drives or CD-Rs seem to be the only media up to the job at a cost-effective rate and an acceptable ease of use.
It's a shame "pen drives" and the like are still so expensive. A portable, robust hard disk that basically just sits in your pocket and plugs into a USB port when needed has a lot going for it...
I guess just like audio and video cassettes, buying cheap is a false economy, and you want a quality brand for anything you're not going to throw away shortly afterwards. I was thinking about this just yesterday, while considering which of the vast array of CD-RW media before me to buy.
The question is, how do you tell what the quality brands are? Does buying in bulk always imply cheap and nasty, and buying a well-presented set of 10 CD-Rs with jewel cases always imply high quality? I doubt it.
I, too, have a set of the nice blue Verbatim disks, which have given me no problems thus far. I've also used the Master "Audio" 80min brand to record (legally!) music onto CDs for use at dancing events, and have had no problems with these. Both cost more than most brands on the shelf at the time I bought them.
That said, I always burn at well under the maximum quoted recording speed of my drive to reduce error. There ought to be a disclosure law on CD recorders/rewriters that using the much-advertised higher speeds will reduce the effective lifetime of the media.
Yesterday, I bought a set of 10 imation "1x-4x compatible" 650MB CD-RWs. Again, they were one of the more expensive brands on display, they're from a known brand and they look to be decent kit, but how do I tell? Certainly the label on the back of the pack -- basically a disclaimer in case you lose data because the disks are bad -- isn't very reassuring.
So, given the pool of knowledge we have here on Slashdot, perhaps this is a good place to ask: which CD media brands do you find are reliable, and which have caused bad experiences?
Um... Wasn't the whole point of the story that this isn't necessarily the case?
As for hard drives, yes, some of the newer brands that push the boundaries do fail with amazing regularity. However, if you stick to tried and tested kit, that doesn't push all the boundaries and puts reliability ahead of that extra 10GB per platter, then problems are still relatively rare.
General concensus is that you don't need to (and shouldn't) learn C first if C++ is your ultimate goal. By all means learn C if you want to use C, but if you want to use C++ then just go straight there.
For somewhat more detailed arguments, have a read of Bjarne Stroustrup's FAQ...
It wasn't iteration, it was tail recursion. :-)
In fairness, Accelerated C++ has a lot going for it and does indeed break new ground (after an amazingly long time with an obvious hole in the market, I thought -- shoulda written it myself!) in how C++ is taught. OTOH, I find a lot of the comments about it somewhat annoying. It isn't perfect, far from it.
Basically, it's too trendy. Sure, it covers lots of STL things, but in practice a lot of people won't be using STL. (Except for fairly recent projects, it's far more likely that some other class library's containers will be the status quo on a project, and even some quite recent compilers don't support templates and/or don't have a good STL implementation.) The coverage of exceptions -- an equally if not more important topic than templates -- is almost non-existant. That was a great disappointment, in a book trying hard to look at C++ for its unique features and not the ones it borrows from C.
I think the book is much better for those who've programmed in other languages and then come to C++ than for beginners. As anyone who follows the C++ newsgroups can tell you, beginners who meet AC++ without other programming exposure are frequently overpowered by its fast-paced approach, and give up in frustration. Those with a "frame of reference" seem to get much more out of it.
Finally, it's written by the in-crowd. Many of the reviews were also written by the in-crowd. Draw your own conclusions about the objectivity of those reviews. This is my big problem with the ACCU book reviews, which are otherwise excellent IMHO. It's also a problem with the book's reputation on-line; so many "helpful" people on the newsgroups were bowing down at the creation of K&M that they recommended it with barely a second thought, and obviously without reading it first in many cases.
In summary, I think Accelerated C++ has a lot going for it, but it's way over-rated (mostly through no fault of its own) and it's not a book that stands alone (because it's in a short-but-sweet series, not a comprehensive one).
So let me get this straight. Gilette always had the best of intentions, and did not encourage the use of the system for photographing customers. Tesco said what the hell, and did it anyway. Now they've canned the trial after protests in Cambridge.
But wait! Now Tesco are going to try the same trick with DVDs at their Sandhurst store. Did they really listen at all? :-(
Oh, and since it's obligatory to bitch, I told them this last weekend. :-)
I think perhaps we actually agree about most of the discussion here, we just looked at it from different perspectives initially. :-)
The one point where I do disagree with you comes from this, though:
I think this is where I don't agree with your basic premise. If I take public domain code, work on it, and release the results, then what I am offering is the extra work. Everyone already has the public domain code, to do with as they wish, and nothing I do can change that.
The only question is what my extra work is worth. If it's valuable, people will pay for it, and it may be profitable for me. If it's not, anyone is free to ignore it and use the original, public domain code instead.
All software development is, to an extent, built on what has gone before, the previous state of the art. The game is all about ways to offer improved value. Amongst other things, you can
It's these extras that are valuable, if everyone has what you started with anyway. IMHO, if the extras aren't paid for by the public, there is no reason why they should be publicly available under the same conditions as the original work that was publicly subsidised. Rather, the person or people adding the value should be the ones who benefit from it if they wish to do so.
Anyone who doesn't think the extras are worth the asking price is free not to use them; they lose nothing else as a result. It's really little different to writing a product from scratch, and the moral issue is the same to me: if someone does fair work, they should be entitled to fair benefits from it. If the work is publicly supported, it should be given back to the public when done. Simple as that, really. :-)