Well what is a contract? It's an agreement between consenting people. Why should law trump that?
Because some people are deceived or coerced into agreement, rather than truly agreeing in full knowledge of the consequences of that agreement. Moreover, two people agreeing to do evil to a third does not make it right or legal; laws trump contracts precisely to stop a whole range of truly nasty abuses. (If only there were fewer bad laws, but that's separate from the nature of contracts.)
disabled access seems to be quite expensive for mass transit systems to provide, and yet they pay a LOWER fare.
It's a moral decision by society overall to treat those who are less bodily able better. It's a mark of charity, and a sign that most people aren't just psychopathic jerks. (Moreover, it's cheaper to put in access to mass transit at the time it is being built than later on, and there are bus designs that have particularly low floors and hydraulic lowering mechanisms that make the bus easier to board for the disabled. Everyone else benefits too, except for bus owners who have to put up with a more capital-intensive operation that increases customer satisfaction...)
as far as I know google scholar is still alive and well...
It certainly is — for me, it's the fourth item on the main Google page's More menu, between Finance and Blogs — and it's often tied in with many institutions' own journal licensing setups. For example, when I'm at work, GS will find stuff in my local academic library specially and will provide links directly to the full text of articles on the sites of journals (which are definitely paywalled; I don't intend to debate whether that's right, but it's how it is). By contrast, when I search from home without using the work VPN, I get a greatly curtailed version.
As Stallman's been saying for the last few years, having software freedom is about having control over your computing, and that requires that your computing is done on *your* computer
But it's not practical to put a copy of everything on my computer, or on yours. Some things are too large to host on anything other than a specialist system, others are too complex. Because of that, it's more important to think in terms of having the data on devices you control (though you can duplicate it elsewhere if you want) and on having multiple options for each processing step; where relevant, OSS is a good candidate for building much of the processing and also for the managing of the data itself, but its key power lies in ensuring that there is a choice for people, that they are not beholden.
We need OSS, and Open APIs, and Open Data Formats; they work together.
I think it could be done if it was done unit-by-unit. Start with volume - we already buy 2 liter sodas, replacing the lingering pint, quart and gallon items shouldn't be hard.
Then, distance. Most people use it in terms of speed - miles per hour - to stay within (or mostly within) speed limits. Simply change all the limits to metric, the other uses of the mile will follow.
Temperature will be the hardest, since there's few personal reasons to switch. Save it for last, so you can make the argument that "this is the last one holding us back".
Experience from the partially-metricated UK is that temperature is actually easier — ensure that everyone's thermometers have both scales and persuade weather forecasts to make the celsius scale the "main one" (I don't suggest you use use kelvin though) — and that volumes and distances are harder because they're embedded in a lot more physical devices. Changing the distance measure means changing a lot of roadsigns and cars. Changing volume measurements impacts on many industries.
Currently, the UK is almost entirely celsius-based for temperature, and has most of the changes for volume measurement done with a couple of exceptions, namely milk and alcoholic drinks; I have no idea why milk is taking so long to transition, but with alcohol it's to do with some extremely tough laws on weights and measures in the area. There's absolutely no plan in the UK to change distance measures at the scale of a mile and above (NB: our cars mostly have dual-scale speedometers) but smaller measures are mostly metric now.
You can't trust the software, even open source unless you've personally reviewed it all including the compiler.
You can't even really trust it then, since vulnerabilities could have been introduced when the compiler was being compiled. About the only thing you can trust is stuff you've "toggled in on the front panel" yourself. But then you'll have a system without the benefit of decades of effort spent on bootstrapping stuff to higher levels of sophistication: you'll be secure because nobody will care to do anything at all with you.
Basically, guaranteed trust is a myth.
The amazing thing is that most people are actually trustworthy, even most corporations (within strict limits; they do want your money after all, but they want it not just today but in the future too). Society largely works, even if not perfectly. Perfect trust isn't needed.
So did Steve Jobs. That's why there are no focus groups at Apple. No market research.
The problem with focus groups is that they tell you what you should already know, not where you need to go in the future. What you want is the focus groups from X years in the future (where X is your product development time, which varies by market) but you don't get that, so you instead need to get a good team of designers to lead, and a good team of engineers to keep the designers grounded in what is possible at all (merely very difficult is OK, physically or legally impossible or horribly expensive... less so), and a good marketing team to persuade the public to buy what you have produced. Apple was really very good at that (though the jury's out on whether they'll manage to keep that balance in the future). Microsoft has not had that sort of flair in recent years, though they've done better than I expected in some areas (not mobile devices though).
In all, I think if Babbage's design had succeeded, it may have made the computer revolution happen 30 or 40 years earlier, in which case, I would have been born in the the mostly ignorant generation of kids comprising the social-networking and internet revolution, and not in the more down-to-earth generation of 8-bit gaming, Q-BASIC and assembler-programming, personal computer revolution folk.
It's not clear to me that communications technology would have advanced any faster even with computers being more advanced. A lot of the key parts are really optical and follow on from quantum mechanics and semiconductors. There's not much point in trying to invent WLAN when the network backbone can only support 8kbps connections, and mobile phones can't be mechanical (well, not without some seriously exotic techniques that are we are currently only looking at the edges of).
Again, IANAL, but "bear" arms presumably means, you know, to actually carry them. (That is, in fact, the definition of the transitive.) Although the SCOTUS has yet to decide on this issue, it's pretty clear cut to me that we ought to be able to carry guns basically anywhere per the constitution.
Yes, but you definitely don't have the right to go onto private property without invitation. If the owner of that property says "You may only enter my property if you are unarmed" then you have to leave you weapon behind in order to enter legally. It's your choice whether you do that, or to stay off their property and armed (or to break the law by trespassing, in which case you'd better be ready to deal with the consequences).
It also depends on what you're doing. Badly written code in one language is slower than well-written code in another (no surprise!) and benchmarks are hard to write so to give a fair picture. Hell, they're hard within the scope of a single language, and across languages you have the problem that most people who suggest benchmarks tend to suggest ones that are on things that their language does well and which others have a problem with (or think to be totally irrelevant).
Then you'd also need to change the length of terms for congressmen. 2 years isn't really long enough to learn how to do the job effectively; you'd end up with the whole place being a mix of noobs and stitch-ups between a pair of people who alternate holding a seat (ugh, that wouldn't be better...) A four-year term would work better, and a limited number of reelections could work too; if it's good enough for the President, it's good enough for the Representative House.
A map that you carry with you at all times - or only in your car? And what is it a map of? Say you live in CA, but you're going to NV on a trip - do you buy an NV map ahead of time?
Those of us who memorize maps for fun find your whole attitude amusing. A quick glance at a map before we go (on paper or on a computer) and we're good for the whole trip, including rerouting around problems and/or to make interesting side-trips. Eidetic memory has some benefits, though perhaps not as many as you might imagine...
In the third world country where I live in, vendor-driven and free events usually have their marketing guys do all the talking. This means biased information. On the other hand, paid events are the ones that have real knowledgeable guys in it.
That's very often the case in the rest of the world too. The best events tend to be the ones that attract people who are professionally in the field, and yet where the sponsors are restricted/absent. That means they're usually held during the working week in substantial hotels in cities and so tend to cost a lot (hotels charging what they do, and not much sponsor money to bear the cost). Yes, this is hard but its just how it works; getting the people that make it really worth it just tends to push costs up, and many conferences are run on a razor edge in terms of costs, with even a small error in terms of income estimation — attendance really — leading to ruin or high profitability. (I'm on the committee for a small conference, but thankfully don't have to deal with the financial side.)
I'd go as far as saying that a pencil is more dangerous than every single one of [the list of dangerous toys].
Agreed, and a pen is more dangerous still. You can't sign a declaration of war with a pencil; nobody would take you seriously, not even your own troops!
No something more effective would be to permanently and irreversibly revoke their copyright for that particular work if found guilty of filing a false DMCA take down notice. Combine that with a duty to also publish an article in all major newspapers about said revoked copyright.
Amend that to "knowingly filing a false DMCA take down" for that sort of penalty, though I suppose it should actually be punishable as false witness, i.e. perjury.
Yep, it sucks. My recommended work commute by Denver's Regional Transportation District takes three transfers, 2.5 hours, and is followed by "walk the remaining 3 miles" (yes, really). I can drive the same route in 40 minutes most days, so I do.
I don't blame you for driving that. (Have you considered having work and home more closely co-located? Mind you, I'm a fine one to talk there...) But that doesn't mean that public transit should suck so much, or that it is so ill-suited for so many of your fellow drivers. Of course, the real problem in the US is that most cities are laid out for cars at lots of scales (suburban sprawl, spread out facilities/stores/etc, large car lots in front; so many aspects that are wrong) and changing is going to be expensive and difficult.
In the mean time, still don't drive while intoxicated or when thoroughly distracted. How you do that is your business, but moving cars are dangerous (to yourself, to others) and you should pay good attention and take care when in charge of one.
You won't build any community among your customers if they feel locked out of key pieces of the product. If anything, they will be resentful. Jon Oosterhout tried it with TCL (Scriptics). Ransom Love tried it with OpenLinux (Caldera). Both failed.
Scriptics didn't fail. They got bought by Interwoven (at a point when Scriptics weren't in financial trouble either; it wasn't a firesale, it was just a good offer on the basis of some of the Tcl-powered products being sold at the time) and then sold on the Tcl parts to ActiveState a short while later (because Interwoven didn't want to write or support language-related tools) who are still going fine. Moreover, Tcl itself was transferred to being fully community-managed before the Interwoven takeover; nobody was locked out of anything.
I don't know anything about OpenLinux, but I do know that you've got at least some of your facts wrong.
you can't simply "alter your code" to transition to Cocoa. Cocoa can only be used with Objective-C. if your code it written in C/C++ then you need to trash it all.
Not true. You can call Obj-C APIs fairly easily from C provided you don't mind the complexity of assembling the message and dispatching it. You'd just be doing what Obj-C does for you behind the scenes. And if you can do it from C, you bet you can do it from C++.
It seems some java coders think: Why just print a string when you could instead instantiate a new string-writer class implementing an abstract string writer factory with a text-writer interface? And instead of hardcoding a constant value, use an xml configuration file, even if you will never actually change the value.
That's actually bad practice, or at least insufficiently good practice. They should be using an XML file to control a library that instantiates and connects the classes together while pulling configuration strings from a separate file somewhere else that's been set up to handle localization (without, you know, actually supplying multiple localizations).
Such monstrosities, where you're in a maze of object-oriented complexity that tries its hardest to conceal exactly what is happening and where, I term them Ravioli Code. Maybe they're discreet parcels rather than the tangled mass of spaghetti code, but they're still pasta in sauce and they're still a mess.
And considering that massive cost, it might make sense to invest massive effort trying to find bugs before software is released. At least for popular and/or mission-critical software...
You'd think so, but evidence says otherwise. After all, many OSes are still released with bugs in!!!
(Runtime patching seems to have become a bit easier over the years, which reduces the pressure to get it right first time. This is both good and bad I suppose; releasing sooner is a good thing...)
What would it be like if we could run native code inside the browser?
The massive swamp of security vulnerabilities that was ActiveX?
I'd agree, except that Google actually appear to have made real progress on the problem of sandboxing native code. I'm not 100% sure that they've solved it — alas, I've not had the time to study it in enough depth — but it at least looks quite likely. Key things that are restricted include access to the parts of the OS that read and write files, with much of the sandboxing focusing on ensuring that nothing can jump to a bad instruction. Without unrestricted access to the OS, a process really can't do that much that's harmful (except chew CPU cycles until something shoots it in the head).
Very clever. Conceptually portable to other platforms too, but I don't know how much work that would be.
Large, hairy, complex solutions should eventually result in a new programming language that makes those sorts of things much smaller, far smoother, and simpler.
Or switching to an existing programming language that is better able to express the problem than the ones you've been used to using. Just because you don't know it doesn't mean it doesn't exist.
Slashdot Mirror
Well what is a contract? It's an agreement between consenting people. Why should law trump that?
Because some people are deceived or coerced into agreement, rather than truly agreeing in full knowledge of the consequences of that agreement. Moreover, two people agreeing to do evil to a third does not make it right or legal; laws trump contracts precisely to stop a whole range of truly nasty abuses. (If only there were fewer bad laws, but that's separate from the nature of contracts.)
disabled access seems to be quite expensive for mass transit systems to provide, and yet they pay a LOWER fare.
It's a moral decision by society overall to treat those who are less bodily able better. It's a mark of charity, and a sign that most people aren't just psychopathic jerks. (Moreover, it's cheaper to put in access to mass transit at the time it is being built than later on, and there are bus designs that have particularly low floors and hydraulic lowering mechanisms that make the bus easier to board for the disabled. Everyone else benefits too, except for bus owners who have to put up with a more capital-intensive operation that increases customer satisfaction...)
as far as I know google scholar is still alive and well...
It certainly is — for me, it's the fourth item on the main Google page's More menu, between Finance and Blogs — and it's often tied in with many institutions' own journal licensing setups. For example, when I'm at work, GS will find stuff in my local academic library specially and will provide links directly to the full text of articles on the sites of journals (which are definitely paywalled; I don't intend to debate whether that's right, but it's how it is). By contrast, when I search from home without using the work VPN, I get a greatly curtailed version.
As Stallman's been saying for the last few years, having software freedom is about having control over your computing, and that requires that your computing is done on *your* computer
But it's not practical to put a copy of everything on my computer, or on yours. Some things are too large to host on anything other than a specialist system, others are too complex. Because of that, it's more important to think in terms of having the data on devices you control (though you can duplicate it elsewhere if you want) and on having multiple options for each processing step; where relevant, OSS is a good candidate for building much of the processing and also for the managing of the data itself, but its key power lies in ensuring that there is a choice for people, that they are not beholden.
We need OSS, and Open APIs, and Open Data Formats; they work together.
I think it could be done if it was done unit-by-unit. Start with volume - we already buy 2 liter sodas, replacing the lingering pint, quart and gallon items shouldn't be hard.
Then, distance. Most people use it in terms of speed - miles per hour - to stay within (or mostly within) speed limits. Simply change all the limits to metric, the other uses of the mile will follow.
Temperature will be the hardest, since there's few personal reasons to switch. Save it for last, so you can make the argument that "this is the last one holding us back".
Experience from the partially-metricated UK is that temperature is actually easier — ensure that everyone's thermometers have both scales and persuade weather forecasts to make the celsius scale the "main one" (I don't suggest you use use kelvin though) — and that volumes and distances are harder because they're embedded in a lot more physical devices. Changing the distance measure means changing a lot of roadsigns and cars. Changing volume measurements impacts on many industries.
Currently, the UK is almost entirely celsius-based for temperature, and has most of the changes for volume measurement done with a couple of exceptions, namely milk and alcoholic drinks; I have no idea why milk is taking so long to transition, but with alcohol it's to do with some extremely tough laws on weights and measures in the area. There's absolutely no plan in the UK to change distance measures at the scale of a mile and above (NB: our cars mostly have dual-scale speedometers) but smaller measures are mostly metric now.
I won't go into how many certificate authority breeches in the last year.
Yeah, they're all just pants.
He said he wants an IT department. Not an IT-Security department.
So it's different justifications.
You can't trust the software, even open source unless you've personally reviewed it all including the compiler.
You can't even really trust it then, since vulnerabilities could have been introduced when the compiler was being compiled. About the only thing you can trust is stuff you've "toggled in on the front panel" yourself. But then you'll have a system without the benefit of decades of effort spent on bootstrapping stuff to higher levels of sophistication: you'll be secure because nobody will care to do anything at all with you.
Basically, guaranteed trust is a myth.
The amazing thing is that most people are actually trustworthy, even most corporations (within strict limits; they do want your money after all, but they want it not just today but in the future too). Society largely works, even if not perfectly. Perfect trust isn't needed.
Firstly, he thinks that consumers are stupid
So did Steve Jobs. That's why there are no focus groups at Apple. No market research.
The problem with focus groups is that they tell you what you should already know, not where you need to go in the future. What you want is the focus groups from X years in the future (where X is your product development time, which varies by market) but you don't get that, so you instead need to get a good team of designers to lead, and a good team of engineers to keep the designers grounded in what is possible at all (merely very difficult is OK, physically or legally impossible or horribly expensive... less so), and a good marketing team to persuade the public to buy what you have produced. Apple was really very good at that (though the jury's out on whether they'll manage to keep that balance in the future). Microsoft has not had that sort of flair in recent years, though they've done better than I expected in some areas (not mobile devices though).
In all, I think if Babbage's design had succeeded, it may have made the computer revolution happen 30 or 40 years earlier, in which case, I would have been born in the the mostly ignorant generation of kids comprising the social-networking and internet revolution, and not in the more down-to-earth generation of 8-bit gaming, Q-BASIC and assembler-programming, personal computer revolution folk.
It's not clear to me that communications technology would have advanced any faster even with computers being more advanced. A lot of the key parts are really optical and follow on from quantum mechanics and semiconductors. There's not much point in trying to invent WLAN when the network backbone can only support 8kbps connections, and mobile phones can't be mechanical (well, not without some seriously exotic techniques that are we are currently only looking at the edges of).
Again, IANAL, but "bear" arms presumably means, you know, to actually carry them. (That is, in fact, the definition of the transitive.) Although the SCOTUS has yet to decide on this issue, it's pretty clear cut to me that we ought to be able to carry guns basically anywhere per the constitution.
Yes, but you definitely don't have the right to go onto private property without invitation. If the owner of that property says "You may only enter my property if you are unarmed" then you have to leave you weapon behind in order to enter legally. It's your choice whether you do that, or to stay off their property and armed (or to break the law by trespassing, in which case you'd better be ready to deal with the consequences).
[Python] is much faster than either Ruby or Tcl.
It also depends on what you're doing. Badly written code in one language is slower than well-written code in another (no surprise!) and benchmarks are hard to write so to give a fair picture. Hell, they're hard within the scope of a single language, and across languages you have the problem that most people who suggest benchmarks tend to suggest ones that are on things that their language does well and which others have a problem with (or think to be totally irrelevant).
CPython's pretty quick at some things though.
I'd like to see "no consecutive terms".
Then you'd also need to change the length of terms for congressmen. 2 years isn't really long enough to learn how to do the job effectively; you'd end up with the whole place being a mix of noobs and stitch-ups between a pair of people who alternate holding a seat (ugh, that wouldn't be better...) A four-year term would work better, and a limited number of reelections could work too; if it's good enough for the President, it's good enough for the Representative House.
A map that you carry with you at all times - or only in your car? And what is it a map of? Say you live in CA, but you're going to NV on a trip - do you buy an NV map ahead of time?
Those of us who memorize maps for fun find your whole attitude amusing. A quick glance at a map before we go (on paper or on a computer) and we're good for the whole trip, including rerouting around problems and/or to make interesting side-trips. Eidetic memory has some benefits, though perhaps not as many as you might imagine...
In the third world country where I live in, vendor-driven and free events usually have their marketing guys do all the talking. This means biased information. On the other hand, paid events are the ones that have real knowledgeable guys in it.
That's very often the case in the rest of the world too. The best events tend to be the ones that attract people who are professionally in the field, and yet where the sponsors are restricted/absent. That means they're usually held during the working week in substantial hotels in cities and so tend to cost a lot (hotels charging what they do, and not much sponsor money to bear the cost). Yes, this is hard but its just how it works; getting the people that make it really worth it just tends to push costs up, and many conferences are run on a razor edge in terms of costs, with even a small error in terms of income estimation — attendance really — leading to ruin or high profitability. (I'm on the committee for a small conference, but thankfully don't have to deal with the financial side.)
I'd go as far as saying that a pencil is more dangerous than every single one of [the list of dangerous toys].
Agreed, and a pen is more dangerous still. You can't sign a declaration of war with a pencil; nobody would take you seriously, not even your own troops!
No something more effective would be to permanently and irreversibly revoke their copyright for that particular work if found guilty of filing a false DMCA take down notice. Combine that with a duty to also publish an article in all major newspapers about said revoked copyright.
Amend that to "knowingly filing a false DMCA take down" for that sort of penalty, though I suppose it should actually be punishable as false witness, i.e. perjury.
Yep, it sucks. My recommended work commute by Denver's Regional Transportation District takes three transfers, 2.5 hours, and is followed by "walk the remaining 3 miles" (yes, really). I can drive the same route in 40 minutes most days, so I do.
I don't blame you for driving that. (Have you considered having work and home more closely co-located? Mind you, I'm a fine one to talk there...) But that doesn't mean that public transit should suck so much, or that it is so ill-suited for so many of your fellow drivers. Of course, the real problem in the US is that most cities are laid out for cars at lots of scales (suburban sprawl, spread out facilities/stores/etc, large car lots in front; so many aspects that are wrong) and changing is going to be expensive and difficult.
In the mean time, still don't drive while intoxicated or when thoroughly distracted. How you do that is your business, but moving cars are dangerous (to yourself, to others) and you should pay good attention and take care when in charge of one.
You won't build any community among your customers if they feel locked out of key pieces of the product. If anything, they will be resentful. Jon Oosterhout tried it with TCL (Scriptics). Ransom Love tried it with OpenLinux (Caldera). Both failed.
Scriptics didn't fail. They got bought by Interwoven (at a point when Scriptics weren't in financial trouble either; it wasn't a firesale, it was just a good offer on the basis of some of the Tcl-powered products being sold at the time) and then sold on the Tcl parts to ActiveState a short while later (because Interwoven didn't want to write or support language-related tools) who are still going fine. Moreover, Tcl itself was transferred to being fully community-managed before the Interwoven takeover; nobody was locked out of anything.
I don't know anything about OpenLinux, but I do know that you've got at least some of your facts wrong.
you can't simply "alter your code" to transition to Cocoa. Cocoa can only be used with Objective-C. if your code it written in C/C++ then you need to trash it all.
Not true. You can call Obj-C APIs fairly easily from C provided you don't mind the complexity of assembling the message and dispatching it. You'd just be doing what Obj-C does for you behind the scenes. And if you can do it from C, you bet you can do it from C++.
It seems some java coders think: Why just print a string when you could instead instantiate a new string-writer class implementing an abstract string writer factory with a text-writer interface? And instead of hardcoding a constant value, use an xml configuration file, even if you will never actually change the value.
That's actually bad practice, or at least insufficiently good practice. They should be using an XML file to control a library that instantiates and connects the classes together while pulling configuration strings from a separate file somewhere else that's been set up to handle localization (without, you know, actually supplying multiple localizations).
Such monstrosities, where you're in a maze of object-oriented complexity that tries its hardest to conceal exactly what is happening and where, I term them Ravioli Code. Maybe they're discreet parcels rather than the tangled mass of spaghetti code, but they're still pasta in sauce and they're still a mess.
And considering that massive cost, it might make sense to invest massive effort trying to find bugs before software is released. At least for popular and/or mission-critical software...
You'd think so, but evidence says otherwise. After all, many OSes are still released with bugs in!!!
(Runtime patching seems to have become a bit easier over the years, which reduces the pressure to get it right first time. This is both good and bad I suppose; releasing sooner is a good thing...)
The latest for Java is JPA
Which is damn close to Hibernate, but standardized so there can be multiple implementations of it.
What would it be like if we could run native code inside the browser?
The massive swamp of security vulnerabilities that was ActiveX?
I'd agree, except that Google actually appear to have made real progress on the problem of sandboxing native code. I'm not 100% sure that they've solved it — alas, I've not had the time to study it in enough depth — but it at least looks quite likely. Key things that are restricted include access to the parts of the OS that read and write files, with much of the sandboxing focusing on ensuring that nothing can jump to a bad instruction. Without unrestricted access to the OS, a process really can't do that much that's harmful (except chew CPU cycles until something shoots it in the head).
Very clever. Conceptually portable to other platforms too, but I don't know how much work that would be.
Large, hairy, complex solutions should eventually result in a new programming language that makes those sorts of things much smaller, far smoother, and simpler.
Or switching to an existing programming language that is better able to express the problem than the ones you've been used to using. Just because you don't know it doesn't mean it doesn't exist.