The problem is VW Smalltalk, not OOP. Its very hard to disentangle apps written in Smalltalk from their development environment for shipment, and the process is encumbered by PPD's clumsy licensing schemes.
I still think VW Smalltalk is the best development environment ever, and still manages to be faster then Java in spite of the obstacles, but thats another issue.
We do have a lot of public cameras. They're in the city center of most major towns and cities now. I hate to put words in the mouth of the majority, but I don't think people here see this as a civil liberties issue. The cameras are actually quite popular because they make city centers safer, especially in places where there were a lot of fights at night.
Personally I actually don't see the strong link most American libertarians make between privacy (especially in public places) and liberty.
Yes the UK is as directly involved in Echelon as the USA. This is a Bad Thing. Our secret intelligence services are just as much a rule unto themselves as anyone else's.
We also have a voluntary (but doubtless eventually essential) photo driver's license in the pipe. Once again I'm not sure this is a problem.
Britain is a hard country to rate when it comes to things like this. I'm reluctant to hold it up as an exemplar of civil libertarianism, because we lack any written gaurantee of anything at all. OTOH we seem to have maintained a generally better record than almost any other large or medium sized power when it comes to domestic affairs. A puzzle. Possibly the fact that executive power technically lies with the legislature, and is therefore subject to public debate, has had something to do with it. If so, the fact this has been eroded over the last twenty (or two hundred) years is a Very Bad Thing.
1. OO has three features, inheritance, encapsulation and polymorphism. You can (and should) encapsulate data types in any language. The other two are clumsy and/or unsafe to try to do in any non-OO strongly typed language.
2. Java doesn't do non-blocking I/O. If you want to do something while waiting on a socket, you start another thread. If you are trying to do otherwise you have a paradigm/language mismatch. Traditional Unix style socket-centric programming is impossible in Java. Thats not necessarily a bad thing, and you can get the functionality in other (more OO) ways.
1. Generics are nice. Templates aren't proper generics. You cannot write arbitrary code that does not specify the values of the type parameters using templates without templatising the class or function it lies in. That means you cannot do any type checking untill instantiation time. That sucks. The implementation can either be done using a text preprocessor or at link time. No compiler I've seen can report errors properly in code with templates. Either way often generates too much code by recompiling for every use. Check out Ada, Eiffel, GJ (modified Java) or (sort of) Standard ML for generic programming done right. Has the ANSI standard fixed the portability problems yet ?
2. Deep implementation hierarchys suck, especially with multiple inheritance. Deep subtyping hierarchies are useful. C++ makes it impossible to enforce a distinction.
3. The libraries have improved. To be blunt, at least Java *has* libraries. STL is a start, but not much of one.
4. Garbage collection works. If you don't get any benefit from it you aren't releasing your references. A bad idea for other reasons, regardless of GC. If you find it slow, thats another matter.
How I see Sun killing the Java innovation is in its insistence on making it fit only in a few environments, in a few ways. Who needs a big wad of "Enterprise" features? Only a handful of those APIs are really needed. Bundling them into an expensive package might get Sun some money, but it doesn't make Java more useful to very many people (even to Enterprise developers).
Sun are trying to fill every niche they can find in the hope that the platform sticks in as many of them as possible. To do this they have a range of platforms from the various attempts at embedded implementations to the new "enterprise edition". Obviously each new API adds a bit less value to the platform than the last, and we've already reached a point with 1.2 where it is nearly impossible to create a cleanroom JDK implementation. You can still have a mostly compliant platform without all the new features, but you then lose the trademarkability.
What is not clear to me is that there is any alternative approach that would work. They want the platform to be portable and they want it to be useful. That means bundling as much stuff, especially potentially OS dependant stuff, into the libraries as possible. Everything that is not in the standard libraries is something else applications have to lug around with them, or cannot access at all if it is OS dependant. While I do not wholly like the way Sun handles the issue of platform integrity, I don't see that they have much choice.
In a similar vein, 100% Pure Java is useless to me except as a portability tool... and I happen to know how to write portable C/C++ code, as well as portable Java code. Sun wants a market in portable Java more than it wants a market in a hybrid of portable Java/C code. I think most of us are coming at it the other way around; look at how, say, Linux works (hybrid of portable C/assembler).
Well, most Java developers seem to want "fire and forget" binary portability. They want their users on randon-platform-X to not need a compiler, and not need to ask for support. Its not quite there yet, but its the potential real value in Java-style portability. It matters especially in big companies for enterprise-wide applications (timecards apps... yawn). Java can also do the kind of portability you mention, but in that context its just a nice YAPL.
To put it a bit differently: Sun's had its chance. We appreciate the good start, and the protection from Miccrosoft. Now can we please start developing Java as an Open (tm) systems technology? It's what we developers always wanted. And it's about time.
There are issues with trying to do this. Firstly, Sun are paranoid about platform integrity. Secondly, the class libraries are big and in many cases poorly defined - there is a real risk of them running out of control if they were opened to changes from every man and his dog. Lastly, Sun's "community source license" is not only a bogus attempt to get some free kudos - they really do listen, and they really do accept suggestions. Obviously they listen harder if you pay them some money, but thats life.
Ummmm... unless the clock chip actually malfunctions when the date rolls over (which would be more than usually crappy design) just including a clock chip in a product that does not use the date is not going to cause it to fail.
Oh, and the GPS roll-over is a red herring. GPS readouts depend only on relative time, not absolute, so while you might get one incorrect location reading or a crash, it won't cause a long term failure except in cases of truly dreadful design. The only potential problem is if you receiver tries to give an obsulute time readout based on the time signal from the satellite. In that case it'll just be wrong, and the designer deserves to die.
I think that with code, as with the Mona Lisa, you cannot separate the thing you create from the process that goes into creating it. Bad processes lead to bad creations, and good ones to good creations.
Both the code, and the result of compiling it and running it are results of the process, so they're both parts of the art.
I agree that most toilets under plastic bubbles are not art, but neither are most of the painting and drawings people make in art classes, or the writing they do in creative writing classes.
As the guy who started this thread said, art is in the process of creation. If the process is imitative, or just cheaply exploitative (like those toilets under plastic bubbles, I suspect) then its not art, but if you really put your soul into something - then its art, even if it also a toilet under a plastic bubble. Thats sometimes applies to code.
Obviously thats not a positive definition. Since you're so confident you know what art is, I'll leave it up to you to come up with one.
IIRC there is a part of the European copyright code that says essentially the same thing as this Australian law: you can decompile or otherwise reverse engineer software in order to develop a compatible product.
I don't know if there is any case law about this yet.
The web browser is a bad example in general, because its pretty much the worst case for ICA. The whole web-page pane of the web browser can only be represented in ICA as a bitmap (or rather, the changes to it get transmaitted as bitmap operations). X at least gets the oportunity to manipulate some part of the page as windows.
This may extend to other applications too, but I wouldn't bet on it. YMMV of course.
1) Bullshit. Anyone can join XFree86, who wield substantial influence.
I don't know what XFree requires of its developers, but the first point is that requiring anything at all slows down development.
However, the XFree86 guys have another obstacle in their way which is that they have to comply with the X standard. To join the consortium that defines this standard, you do have to pay money. Before calling bullshit in future, I suggest you try reading what you are replying to.
1) Bullshit. Anyone can join XFree86, who wield substantial influence.
I don't know what XFree requires of its developers, but the first point is that requiring anything at all slows down development.
However, the XFree86 guys have another obstacle in their way which is that they have to comply with the X standard. To join the consortium that defines this standard, you do have to pay money. Before calling bullshit in future, I suggest you try reading what you are replying to.
Interesting article. I'm sure none of Linux's current 'kindly uncles' would carry out the kind of 'embrace and extend' operation being anticipated, since they are mostly firms that are either obliged to act in their customers interests because they are very small (Corel), or firms that have comitted relationships with a customer base they really don't want to annoy (IBM). Most radical changes made within the community are in users' better interests, and that forms a basis on which companies and the developers of a project could handle them. What is more, most companies involved are trying to leaverage Linux's momentum to restore or add to their own. That means they have no interest in encouraging or supporting fragmentation.
The thing the article raises that is interesting is the question of what would happen is an actively hostile company, one whose operating systems are competing with Linux, were to produce an incompatible version and try to get people to use that. That could be quite destructive, but I think it could be dealt with. The GPL forces anyone deriving their software from a version of Linux to publish their source-code. That published source code would form a basis for the Linux community to work to restore compatibility.
Thats from Mao, only I think it was actually "freedom comes from the barrel of a gun", which is something I suspect an Objectivist might have some trouble with if the gun belonged to the state, but would agree with wholeheartedly if it was his own.
But surely all the index pages on machines listed in DNS on port 80 + all the pages they (recursively) link to is more than 20% of the web's content ? Almost all sites are both on port 80 and in DNS.
Someone else (who is still down at 0, because he posted it anonymously) came up with a much better answer, which is that the hardware and bandwidth required to index 100% of static content is extremely large, and anyway most content is not static. Its this last point, I think, which is most important - by definition nothing you read daily is static content.
I may be being a bit slow here, but what is the problem which prevents coverage of the entire wbe by search engines ?
Surely if you just hit port 80 of every machine registered in DNS, and search recursively from the pages retrieved by that, you'll get a greater number of pages than the 10-20 percent most search engines have ?
Or is it the case that the problem is in the indexing of the data, and searching it quickly enough, rather than retrieving it ?
I agree completely that this whole Libertarian "its not obviosuly coercion so we have to leave it alone" line is a highly suspect piece of rhetoric which really begs the quetion: "what is coercion if it isn't making people do things they don't like ?".
However, I think you are wrong about the IMF. Their problem is that the governments they are lending the money to are corrupt, and the western agencies who are actually meant to do the development work are on a massive gravy train. The economics they push on people is pretty much irrelevant when the governments they are pushing it on are too corrupt and incompetent to implement it.
Re:I'm looking forward to the day they ditch X
on
Some KDE news
·
· Score: 1
I cannot agree that X is efficient or flexible. I agree we should be very cautious about replacing it, but is definitely showing its age. It'll be interesting to see if the Berlin guys ever come out with anything useful - they do have some good ideas.
I think you should check again. Look up communism in the dictionary, and compare it with modern China. Lets see:
communism \Com"mu*nism\, n. [F. communisme, fr. commun common.] A scheme of equalizing the social conditions of life; specifically, a scheme which contemplates the abolition of inequalities in the possession of property, as by distributing all wealth equally to all, or by holding all wealth in common for the equal use and advantage of all. Note: At different times, and in different countries, various schemes pertaining to socialism in government and the conditions of domestic life, as well as in the distribution of wealth, have been called communism. Source: Webster's Revised Unabridged Dictionary
Equalising conditions of life ? no. Eliminating inequality of property ? no. Holding all wealth in common ? no (they don't even have universal state ownership any more). Not that anyone has ever come close to acheiving these goals, but the Chinese aren't even pretending to try any more.
NSI had a time-limited monopoly, much like you get with a patent, which the government granted more-or-less by accident. They campaigned like crazy to keep it, but understandably it got taken away.
The action the banks, etc, are taking is precautionary. They don't, by and large, know the full extent of the problem in their software, and to maintain the confidence of their customers they have be seen to examine all their own code and contact all their suppliers and get them to promise theirs is OK too.
Taking all these measures, and employing thousands of consultants to check the code, does not mean they have a problem. It means they want to play safe (and be seen to play safe) and they have money to burn.
When the British or American government talks about compliance, they mean a state where an organisation knows for sure it does not have any problems. This is extremely hard to do. It means the organisation has to go through the following:
Locate all their computer systems
Locate the suppliers for those from outside sources.
Extract promises of compliance and/or software and hardware updates from all the suppliers.
Replace anything externally supplied that cannot be gauranteed to be compliant
Carry out a risk assessment on all their internal systems
For those systems deemed to pose too high a risk, carry out an in-depth bug hunt and propose fixes
Carry out all the fixes
Perform vast quantities of testing to integrate the new systems
The extent to which all this is actually necessary is basically a complete unknown. Noone has a clue how widespread the bug is, or which systems it is most likely to reside in. The point is that non-compliant organisations, for whatever reasons, may very well not have had a problem anyway of if they do it may not be severe. This means that while non-completion of Y2K projects poses some risk of inconvenience or worse, it does not mean that civilisation will grind to a halt.
The consequences of the bug, remember, depend on the system it is in. Even in safety critical systems (a tiny minority of cases) the consequences of date overflow may be as mild as an incorrect clock display. Of course it goes the other way too. One power-station under (simulated) test started to overheat because the air conditioning system had a bug.
I would guess that the majority of systems either have no problem or only cosmetic problems. Nonetheless I will be spending my new year as far from civilisation as I can get, but mostly to avoid rampaging nuts, not exploding computers.
What have WinModems got to do with anything ? A box like this will either come with its modem on the motherboard, in which case the hardware manufacturer would have to be insane to tie it to a particular OS from a particular manufacturer, or it will plug in over USB, in some as-yet-unestablished manner. WinModems are PCI or ISA cards, usually.
What is experimental about the USB and PnP support in 2.2 ? PnP seems just fine (I don't mess with USB just now).
Firstly, my objection to spam is that it is a waste of my time and computing resources. My objection to unsolicited proseletysing is similar - it is a waste of my time and a waste of the mental effort required to try to argue coherently with people who have been trained in advertising strategies that would make the Saatchis jealous. The fact is that I was raised a Christain and deliberately stopped attending church, because I feel the whole basis for the organised religion is mistaken - you cannot argue that I do not know the 'truth' - I know it and have decided against it, so proseletysing me is a waste of everyone's time.
You say these Christians are acting out of charity. I don't believe you. The churches here (and I'm not an Amercian - YMMV) that go in for this style of 'missionary work' tend to be the ones that teach that personal salvation depends on spreading the word to others. Many of these 'missionaries' seem to me to be acting out of fear, and a great deal of 'Chrisitian' thought seems to be motivated similarly. The Jehovah's Witnesses seem to organise their whole church as a kind of giant religious pyramid scheme.
Slashdot Mirror
The problem is VW Smalltalk, not OOP. Its very hard to disentangle apps written in Smalltalk from their development environment for shipment, and the process is encumbered by PPD's clumsy licensing schemes.
I still think VW Smalltalk is the best development environment ever, and still manages to be faster then Java in spite of the obstacles, but thats another issue.
We do have a lot of public cameras. They're in the city center of most major towns and cities now. I hate to put words in the mouth of the majority, but I don't think people here see this as a civil liberties issue. The cameras are actually quite popular because they make city centers safer, especially in places where there were a lot of fights at night.
Personally I actually don't see the strong link most American libertarians make between privacy (especially in public places) and liberty.
Yes the UK is as directly involved in Echelon as the USA. This is a Bad Thing. Our secret intelligence services are just as much a rule unto themselves as anyone else's.
We also have a voluntary (but doubtless eventually essential) photo driver's license in the pipe. Once again I'm not sure this is a problem.
Britain is a hard country to rate when it comes to things like this. I'm reluctant to hold it up as an exemplar of civil libertarianism, because we lack any written gaurantee of anything at all. OTOH we seem to have maintained a generally better record than almost any other large or medium sized power when it comes to domestic affairs. A puzzle. Possibly the fact that executive power technically lies with the legislature, and is therefore subject to public debate, has had something to do with it. If so, the fact this has been eroded over the last twenty (or two hundred) years is a Very Bad Thing.
Linus is infallible
Linus has occasionally said mildly bad things about Java.
ergo Java is a bad idea.
I think thats the core mentality here.
For what it matters, I agree with you. Java and Linux have more to gain than lose from one another, and its sad that some people can't see that.
1. OO has three features, inheritance, encapsulation and polymorphism. You can (and should) encapsulate data types in any language. The other two are clumsy and/or unsafe to try to do in any non-OO strongly typed language.
2. Java doesn't do non-blocking I/O. If you want to do something while waiting on a socket, you start another thread. If you are trying to do otherwise you have a paradigm/language mismatch. Traditional Unix style socket-centric programming is impossible in Java. Thats not necessarily a bad thing, and you can get the functionality in other (more OO) ways.
1. Generics are nice. Templates aren't proper generics. You cannot write arbitrary code that does not specify the values of the type parameters using templates without templatising the class or function it lies in. That means you cannot do any type checking untill instantiation time. That sucks. The implementation can either be done using a text preprocessor or at link time. No compiler I've seen can report errors properly in code with templates. Either way often generates too much code by recompiling for every use. Check out Ada, Eiffel, GJ (modified Java) or (sort of) Standard ML for generic programming done right. Has the ANSI standard fixed the portability problems yet ?
2. Deep implementation hierarchys suck, especially with multiple inheritance. Deep subtyping hierarchies are useful. C++ makes it impossible to enforce a distinction.
3. The libraries have improved. To be blunt, at least Java *has* libraries. STL is a start, but not much of one.
4. Garbage collection works. If you don't get any benefit from it you aren't releasing your references. A bad idea for other reasons, regardless of GC. If you find it slow, thats another matter.
How I see Sun killing the Java innovation is in its insistence on making it fit only in a few environments, in a few ways. Who needs a big wad of "Enterprise" features? Only a handful of those APIs are really needed. Bundling them into an expensive package might get Sun some money, but it doesn't make Java more useful to very many people (even to Enterprise developers).
Sun are trying to fill every niche they can find in the hope that the platform sticks in as many of them as possible. To do this they have a range of platforms from the various attempts at embedded implementations to the new "enterprise edition". Obviously each new API adds a bit less value to the platform than the last, and we've already reached a point with 1.2 where it is nearly impossible to create a cleanroom JDK implementation. You can still have a mostly compliant platform without all the new features, but you then lose the trademarkability.
What is not clear to me is that there is any alternative approach that would work. They want the platform to be portable and they want it to be useful. That means bundling as much stuff, especially potentially OS dependant stuff, into the libraries as possible. Everything that is not in the standard libraries is something else applications have to lug around with them, or cannot access at all if it is OS dependant. While I do not wholly like the way Sun handles the issue of platform integrity, I don't see that they have much choice.
In a similar vein, 100% Pure Java is useless to me except as a portability tool ... and I happen to know how to write portable C/C++ code, as well as portable Java code. Sun wants a market in portable Java more than it wants a market in a hybrid of portable Java/C code. I think most of us are coming at it the other way around; look at how, say, Linux works (hybrid of portable C/assembler).
Well, most Java developers seem to want "fire and forget" binary portability. They want their users on randon-platform-X to not need a compiler, and not need to ask for support. Its not quite there yet, but its the potential real value in Java-style portability. It matters especially in big companies for enterprise-wide applications (timecards apps ... yawn). Java can also do the kind of portability you mention, but in that context its just a nice YAPL.
To put it a bit differently: Sun's had its chance. We appreciate the good start, and the protection from Miccrosoft. Now can we please start developing Java as an Open (tm) systems technology? It's what we developers always wanted. And it's about time.
There are issues with trying to do this. Firstly, Sun are paranoid about platform integrity. Secondly, the class libraries are big and in many cases poorly defined - there is a real risk of them running out of control if they were opened to changes from every man and his dog. Lastly, Sun's "community source license" is not only a bogus attempt to get some free kudos - they really do listen, and they really do accept suggestions. Obviously they listen harder if you pay them some money, but thats life.
Ummmm ... unless the clock chip actually malfunctions when the date rolls over (which would be more than usually crappy design) just including a clock chip in a product that does not use the date is not going to cause it to fail.
Oh, and the GPS roll-over is a red herring. GPS readouts depend only on relative time, not absolute, so while you might get one incorrect location reading or a crash, it won't cause a long term failure except in cases of truly dreadful design. The only potential problem is if you receiver tries to give an obsulute time readout based on the time signal from the satellite. In that case it'll just be wrong, and the designer deserves to die.
I think that with code, as with the Mona Lisa, you cannot separate the thing you create from the process that goes into creating it. Bad processes lead to bad creations, and good ones to good creations.
Both the code, and the result of compiling it and running it are results of the process, so they're both parts of the art.
I agree that most toilets under plastic bubbles are not art, but neither are most of the painting and drawings people make in art classes, or the writing they do in creative writing classes.
As the guy who started this thread said, art is in the process of creation. If the process is imitative, or just cheaply exploitative (like those toilets under plastic bubbles, I suspect) then its not art, but if you really put your soul into something - then its art, even if it also a toilet under a plastic bubble. Thats sometimes applies to code.
Obviously thats not a positive definition. Since you're so confident you know what art is, I'll leave it up to you to come up with one.
IIRC there is a part of the European copyright code that says essentially the same thing as this Australian law: you can decompile or otherwise reverse engineer software in order to develop a compatible product.
I don't know if there is any case law about this yet.
The web browser is a bad example in general, because its pretty much the worst case for ICA. The whole web-page pane of the web browser can only be represented in ICA as a bitmap (or rather, the changes to it get transmaitted as bitmap operations). X at least gets the oportunity to manipulate some part of the page as windows.
This may extend to other applications too, but I wouldn't bet on it. YMMV of course.
1) Bullshit. Anyone can join XFree86, who wield substantial influence.
I don't know what XFree requires of its developers, but the first point is that requiring anything at all slows down development.
However, the XFree86 guys have another obstacle in their way which is that they have to comply with the X standard. To join the consortium that defines this standard, you do have to pay money. Before calling bullshit in future, I suggest you try reading what you are replying to.
1) Bullshit. Anyone can join XFree86, who wield substantial influence.
I don't know what XFree requires of its developers, but the first point is that requiring anything at all slows down development.
However, the XFree86 guys have another obstacle in their way which is that they have to comply with the X standard. To join the consortium that defines this standard, you do have to pay money. Before calling bullshit in future, I suggest you try reading what you are replying to.
Interesting article. I'm sure none of Linux's current 'kindly uncles' would carry out the kind of 'embrace and extend' operation being anticipated, since they are mostly firms that are either obliged to act in their customers interests because they are very small (Corel), or firms that have comitted relationships with a customer base they really don't want to annoy (IBM). Most radical changes made within the community are in users' better interests, and that forms a basis on which companies and the developers of a project could handle them. What is more, most companies involved are trying to leaverage Linux's momentum to restore or add to their own. That means they have no interest in encouraging or supporting fragmentation.
The thing the article raises that is interesting is the question of what would happen is an actively hostile company, one whose operating systems are competing with Linux, were to produce an incompatible version and try to get people to use that. That could be quite destructive, but I think it could be dealt with. The GPL forces anyone deriving their software from a version of Linux to publish their source-code. That published source code would form a basis for the Linux community to work to restore compatibility.
Thats from Mao, only I think it was actually "freedom comes from the barrel of a gun", which is something I suspect an Objectivist might have some trouble with if the gun belonged to the state, but would agree with wholeheartedly if it was his own.
But surely all the index pages on machines listed in DNS on port 80 + all the pages they (recursively) link to is more than 20% of the web's content ? Almost all sites are both on port 80 and in DNS.
Someone else (who is still down at 0, because he posted it anonymously) came up with a much better answer, which is that the hardware and bandwidth required to index 100% of static content is extremely large, and anyway most content is not static. Its this last point, I think, which is most important - by definition nothing you read daily is static content.
I may be being a bit slow here, but what is the problem which prevents coverage of the entire wbe by search engines ?
Surely if you just hit port 80 of every machine registered in DNS, and search recursively from the pages retrieved by that, you'll get a greater number of pages than the 10-20 percent most search engines have ?
Or is it the case that the problem is in the indexing of the data, and searching it quickly enough, rather than retrieving it ?
I agree completely that this whole Libertarian "its not obviosuly coercion so we have to leave it alone" line is a highly suspect piece of rhetoric which really begs the quetion: "what is coercion if it isn't making people do things they don't like ?".
However, I think you are wrong about the IMF. Their problem is that the governments they are lending the money to are corrupt, and the western agencies who are actually meant to do the development work are on a massive gravy train. The economics they push on people is pretty much irrelevant when the governments they are pushing it on are too corrupt and incompetent to implement it.
I cannot agree that X is efficient or flexible. I agree we should be very cautious about replacing it, but is definitely showing its age. It'll be interesting to see if the Berlin guys ever come out with anything useful - they do have some good ideas.
I think you should check again. Look up communism in the dictionary, and compare it with modern China. Lets see:
communism \Com"mu*nism\, n. [F. communisme, fr. commun common.] A scheme of equalizing the social conditions of life; specifically, a scheme which contemplates the abolition of inequalities in the possession of property, as by distributing all wealth equally to all, or by holding all wealth in common for the equal use and advantage of all. Note: At different times, and in different countries, various schemes pertaining to socialism in government and the conditions of domestic life, as well as in the distribution of wealth, have been called communism. Source: Webster's Revised Unabridged DictionaryEqualising conditions of life ? no. Eliminating inequality of property ? no. Holding all wealth in common ? no (they don't even have universal state ownership any more). Not that anyone has ever come close to acheiving these goals, but the Chinese aren't even pretending to try any more.
NSI had a time-limited monopoly, much like you get with a patent, which the government granted more-or-less by accident. They campaigned like crazy to keep it, but understandably it got taken away.
So there isn't really any evidence that its the thing in the photo, and not the Unix box, a that is answering the pings and serving the web pages.
The action the banks, etc, are taking is precautionary. They don't, by and large, know the full extent of the problem in their software, and to maintain the confidence of their customers they have be seen to examine all their own code and contact all their suppliers and get them to promise theirs is OK too.
Taking all these measures, and employing thousands of consultants to check the code, does not mean they have a problem. It means they want to play safe (and be seen to play safe) and they have money to burn.
When the British or American government talks about compliance, they mean a state where an organisation knows for sure it does not have any problems. This is extremely hard to do. It means the organisation has to go through the following:
The extent to which all this is actually necessary is basically a complete unknown. Noone has a clue how widespread the bug is, or which systems it is most likely to reside in. The point is that non-compliant organisations, for whatever reasons, may very well not have had a problem anyway of if they do it may not be severe. This means that while non-completion of Y2K projects poses some risk of inconvenience or worse, it does not mean that civilisation will grind to a halt.
The consequences of the bug, remember, depend on the system it is in. Even in safety critical systems (a tiny minority of cases) the consequences of date overflow may be as mild as an incorrect clock display. Of course it goes the other way too. One power-station under (simulated) test started to overheat because the air conditioning system had a bug.
I would guess that the majority of systems either have no problem or only cosmetic problems. Nonetheless I will be spending my new year as far from civilisation as I can get, but mostly to avoid rampaging nuts, not exploding computers.
What have WinModems got to do with anything ? A box like this will either come with its modem on the motherboard, in which case the hardware manufacturer would have to be insane to tie it to a particular OS from a particular manufacturer, or it will plug in over USB, in some as-yet-unestablished manner. WinModems are PCI or ISA cards, usually.
What is experimental about the USB and PnP support in 2.2 ? PnP seems just fine (I don't mess with USB just now).
Firstly, my objection to spam is that it is a waste of my time and computing resources. My objection to unsolicited proseletysing is similar - it is a waste of my time and a waste of the mental effort required to try to argue coherently with people who have been trained in advertising strategies that would make the Saatchis jealous. The fact is that I was raised a Christain and deliberately stopped attending church, because I feel the whole basis for the organised religion is mistaken - you cannot argue that I do not know the 'truth' - I know it and have decided against it, so proseletysing me is a waste of everyone's time.
You say these Christians are acting out of charity. I don't believe you. The churches here (and I'm not an Amercian - YMMV) that go in for this style of 'missionary work' tend to be the ones that teach that personal salvation depends on spreading the word to others. Many of these 'missionaries' seem to me to be acting out of fear, and a great deal of 'Chrisitian' thought seems to be motivated similarly. The Jehovah's Witnesses seem to organise their whole church as a kind of giant religious pyramid scheme.