I thought it was a public record that George W Bush had been in potentially serious legal trouble on several occasions, but had escaped any results against him after special investigators appointed by his father decided there wasn't enough evidence to proceed? (This is from memory, so possibly not technically correct, but that was the general idea.)
Fortunately, as I'm in the UK, it'll be tricky for certain US Treasury Department employees to harrass me for repeating this. Then again, we seem to be creating a new branch of the FBI this week...
Expect whoever's hiring you to perceive any changes to legal documents as threats at first, just as you would modifications to some legalese in your contract by your employer to be some attempt at them fucking you over.
I don't see why it should be perceived as a threat. When my current employers made the job offer, I was sent a nice letter and a summary of the Ts&Cs (salary, holiday entitlement, etc.) and told I'd be expected to sign a full contract when I arrived.
Alarm bells rang, so I replied saying I was definitely interested but would like to see the full contract first (to check for exactly the type of "We Own Everything" clause mentioned here, coincidentally). The following evening, the guy who runs about half the company turned up on my doorstep about 7pm, handed me a copy of their standard contract, and basically said, "Got your letter, no problem, it's not unusual for new starters with experience elsewhere to ask."
Incidentally, my contract does have a clause that explicitly indicates the company having no claim over anything I do off company time and without company equipment. In the OP's position, I would immediately ask to have one inserted in place of the current wording, mentioning my situation regarding the existing other business if necessary. If they won't acknowledge your concern and insert the clause pretty much immediately, well, now you know what kind of employer they are and that you don't want to work for them.:o)
Re:Constraints on type parameters
on
How C# Was Made
·
· Score: 1
If constraints were in C++, I think they should be done on a member function basis. That is, you explicity state which member functions must be implemented.
But that doesn't work either...
You can (and probably should) implement binary operators as free functions, for example, if necessary making them friends of the class in question. When I write if(a==b), C++ just needs to know what the == operator means in this context. It doesn't, and shouldn't, matter exactly how it's defined, or whether it's a member function; again, imposing such a limitation gains you nothing and damages support for generic programming.
I think he understands just fine
on
How C# Was Made
·
· Score: 1
What you say is certainly true up to a point: the highest level code in a particular subsystem, right behind its interface, should indeed define any exceptions it's going to allow out of the subsystem as part of that interface. If there is language-level support for enforcing that definition, it probably makes sense to use it.
However, within the subsystem itself, you often don't want to be translating exceptions at every layer. To do so defeats one of the major advantages of exceptions: you can throw an exception up through several layers at once to a handler, and intervening layers don't have to care unless they're taking some action to handle (or part-handle) the particular exception you threw. In that environment, you could easily wind up with many possible exceptions being thrown from different contexts, but handled in a small number of places. Even with inhertiance and catching groups of related exceptions at once to keep the complexity down, having to give a full specification for everything would soon become unwieldy.
Constraints on type parameters
on
How C# Was Made
·
· Score: 2, Interesting
AIUI, the major reason many of the big names in C++ don't like language-supported constraints on type parameters is that just having "derives from this base class" isn't a sufficiently general condition for good support of generic programming.
C++ enforces any necessary interfaces at compile-time, by checking whether the instantiation of a function template can be interpreted properly with the functions available. Requiring every class that supports == to derive from EqualityComparable is just a cumbersome extra for no benefit in a generic, C++ based world, where there are more ways to define and implement interfaces than just inheritance. If it doesn't support ==, the function template will fail to compile at the point of instantiation anyway. (OK, strictly speaking export messes up the compile-time/link-time distinction slightly, but the basic point is still valid.)
Incidentally, it is possible to use template wizardry in C++ to enforce various plausible constraints on type parameters if you really want to restrict them more than their usage within the function template does anyway, perhaps to guard against a possible misinterpretation of the template code in some cases. It's relatively straightforward to do this, and requires little more code than the formal constraints used by some of the other languages under discussion.
In any true object-oriented language, you should be able to write almost everything as object.method(argument), and this is not the case with Java unfortunately...
I see your point, but of course that "purist" view is also why OO struggles so hard to model some simple concepts, which frequently boil down to examples of multiple dispatch.
If languages are going to support objects, they'll be improved by having a uniform syntax for all of:
method calls
multimethod calls
non-member function calls
accessing members/properties directly.
You can have the really useful thing from OO -- the power of implicit type coercion and polymorphism -- without creating artificial distinctions between actions that are really serving the same purpose. It's just that many languages -- notably including C++, Java and C# -- don't.
At my last job interview, we were discussing general background in computing for a minute, and I mentioned being karma-capped on Slashdot as a joke (made sense in context). That was followed immediately by a knowing smile from one of my interviewers, and shortly afterwards by a job offer.:-)
I'm always wary of giving credit to someone proclaiming the Next Great Thing, when they've spent the last several years as a leading advocate of the Previous Great Thing, then suddenly start finding all these previously overlooked drawbacks in the older technology when the newer one arrives. Don Box seems awfully close to that position right now.
The case of the two kids who murdered a todller is prove of a crime that would have not be solved with out cctv.
Assuming you're referring to the Jamie Bulger case, how do you know? Sure, the CCTV showed that the kidnappers didn't fit the usual profile and allowed the police to shift their search. OTOH, traditionally police have asked witnesses (of whom there were probably rather a lot in that particular case) to obtain such information, and there's no reason to believe they couldn't have done so here as well.
CCTV may have saved the police some time, but you can't extrapolate from that that they wouldn't have solved the crime anyway. In fact, if you look at overall crime rates before and after CCTV is installed, you find that it helps reduce certain types of crime where it's installed, but actually makes surprisingly little difference to the proportion of crimes solved overall. OTOH, it also results in an increase in some types of crime in neighbouring areas.
I prefer that we use DNA profiling and CCTV to catch the real criminals.
That's fine, taken in isolation. However, as with most civil liberties objections, the key point here is that you are not working in isolation. You have to consider not only the effect on criminals/suspects, but also the effect on everyone else: mistaken identity and voyeurism are both serious concerns, with numerous examples of each identified in relation to the CCTV. For a technology so open to abuse, there's precious little evidence to show it's really helping enough to justify that downside.
Incidentally, the story concludes with, "Despite this, people still seem to prefer the cameras." I don't know who the author talked to, but I doubt it was anyone I know.
And don't blame someone for linking to it, since that is what the WWW is for.
Nice theory, but about 15 years out of date with reality.
Luckily, we aren't talking about a DDoS and therefore that has nothing to do with this conversation.
If knowingly directing vast amounts of traffic at a site from many sources, with the reasonably predictable result being that site's failure, isn't a DDoS attack, what is?
Slashdotting isn't an accident. The eds know damn well that if they post a link in a front page story, that page is going to take zillions of hits. It doesn't take a genius to spot that some servers are better able to handle this than others, and which ones they are likely to be. Deliberately slashdotting a site is no more ethical than distributing a self-contained program to hundreds of computers that then bombards the site with requests from numerous directions at a co-ordinated time. The mechanism is much the same, the result is the same, and it's done just as knowingly.
You are still confusing what you can do with what it's ethical to do, and ignoring the well-established and widely employed legal ideas of reasonable behaviour and fair expectation.
Well, my argument to that is: you never know when you will become popular, so being shortsighted when signing up is not a very good argument for blaming the people coming to your site that you publicly posted.
By your argument, almost no-one would bother to make their work available for free on the web without an infinite resource backing them, and the Internet would collapse. There is a reason terms like "reasonable expectation" appear is so many laws, and actions like co-ordinating a DDoS attack are illegal.
I see unfounded and baseless claims about Microsoft on/. all the time, nobody complains or feels aggrieved about those.
That simply isn't true. A significant number of/. readers are here for useful discussion and information gathering. If someone posts crap on a technical subject, whether it's about Linux, Microsoft, or otherwise, it's a good bet that someone who knows what they're talking about will reply to set the record straight, and be modded up accordingly.
Much easier to pronounce, but not the same thing at all. If that site was Googled to death, then a Googleplexing would pretty much take out the entire Internet, if you think about it...
Now if only we'd all learn to spell googol, we could quite the bad jokes already.;-)
That depends on whether a web site that expected 5,000 hits this month just got 500,000 in one day because some idiot posted its address on Slashdot, and the webmaster is paying $1/MB for everything past the first 200MB this month, I guess. (The answer is around $20,000 that day, in case you're having trouble working it out.)
The key concepts a lot of people seem to be missing here are the intent of a web site and the reasonable expectations of its administrator. I don't believe that sites like Slashdot or Google have the right to knowingly direct huge amounts of traffic to a site that wouldn't expect it, without that site's explicit permission. Doing so is essentially just a deliberate DDoS attack on the site, and should be treated as such. Expecting a mainstream news web site to have serious bandwidth is reasonable. Expecting www.bigisp.com/~littleuser/aboutme.html to have serious bandwidth is not.
The fact that it is technically possible to do this to anyone with a web site is not the point. It is technically possible for me to run you over with my car when you step into the road without looking, but if I do so when I could have stopped, I will rightly be charged with (insert your local motoring offence here).
The EU guys will not understand this as they have always lived in a society where you had to get permission to do anything. You had no real rights only permits. The founders of the USA so opposed such a concept that they ran it out of our land on a rail. The return of it is the return to what brought on the Dark Ages. (Something the EU guys might know a about)
Why the snide comments about the EU? And what on earth makes you think that we don't understand the concepts of rights and freedoms?
I find it deeply ironic that the United States, a country that prides itself on its Constitution and the rights of its citizens, is also the place where: you have giant corporations controlling your government and laughing at your legal system; you have the right to free speech, as long as you can afford the lawyers to defend it; the constitutional safeguards over copyright are being trampled by the aforementioned big corps; you hold hundreds of people indefinitely and without charge, based on an accusation and a technicality of international law that no-one else recognises; and you have a President of dubious mandate, taking your country to war supported by dubious intelligence, resulting in the deaths of thousands of innocents, the destruction of a whole country's infrastructure, the deaths of numerous American servicemen and women, and did I mention some rather lucrative rebuilding contracts for major corps with whom your senior leadership has intimate ties?
The UK government has become increasingly abusive of its authority, particularly since Tony Blair's lot came to power, with Jack Straw and then David Blunkett as Home Secretary. However, we can't even approach your level of legal impotence and government abuse, and you're busy trying to inflict it on the rest of the world! And at least at our general election next year, we'll have candidates to vote for who don't all say the same thing, which is as bad as who we've got at the moment anyway...
You make a lot of good points in your post, and I agree with much of what you say, but with all due respect, you seem to have a serious lack of perspective on the world outside. For all our knowledge of the Dark Ages, as far as rights and responsibilities go, I'd still far prefer to be living in the EU than the US right now.
I'm a great fan of OpenOffice. Nevertheless, I'm also a realist, and I have no hesitation in describing Excel as a vastly superior application to OpenOffice Calc. The latter is OK for basic spreadsheeting, and I do use it often for little things like budgeting, but for serious stuff I'd choose Excel every time, at least for now.
Joel's main complaint appears to be about cross version compatibility (b) and all the cruft you need to install to get.net executables working (uh....Net itself:-)) (a). I understand this in the short term, but over the long term I think it's going to be less of an issue.
I think Joel's point, which half the flamers on this thread don't seem to get, was that if you're writing software today, and you're trying to get customers to download it tomorrow, writing for.Net is probably not a smart move.
I realise that.Net theoretically has a decent versioning system. However, as someone who's been responsible for investigating new versions of Visual Studio at two different companies now, I've seen plenty of problems in the real world. VS.Net 2002 and VS.Net 2003 really don't play nicely together for a start, and if Microsoft themselves can't get it right with an application whose entire focus is on.Net, what hope does Joe Developer have? I'd be very wary of risking my whole application by trusting a technology so unproven to get the version control right.
And as of today, I think Joel's point about download sizes is also entirely reasonable. I bet the critics here are all happily sitting at home, probably in the US, probably with free local phone calls and always-on broadband. I bet most of them also don't realise that most of the world doesn't have those facilities today, and nor are many of them likely to any time within the next few years. I still have the same 56k dial-up modem I bought when I left uni five years ago, because broadband still has too many practical problems here, and I'm a professional developer and part-time geek who spends several hours on-line some days. I only got a flat-rate package about 3 years ago (before that I was paying per-minute) and even now I'm capped in the number of hours I can use each month, so big downloads are not high on my list of Good Things.
IOWs, as you say, short term Joel's points are very relevant. Later on -- probably 5-7 years away at a minimum -- if 95+% of the Windows user base has.Net installed and/or many more people have broadband, then it might be less so. But 5-7 years is an eternity in software development. For short- to medium-term developments, Joel's comments were right on the money. Using C++ with a good set of libraries is still far ahead of.Net if the size of your executable matters.
News flash - new technology requires new, usually larger, downloads. BFD.
News flash: users on dial-up (that's most of the world, BTW) don't like big downloads, and will mostly stick to older technology in preference to spending hours downloading something new and unproven. Big downloads loses you large chunks of user base. If you're doing it commercially, that means large chunks of cash, which is a BFD.
As a company, the email and the equipment is owned by the company, and the fact that the employee feels they have the right to use them for personal business is rediculous.
But is it?
Of course, as an employer, you're entitled to expect your employees to do their jobs to the best of their ability, in exchange for whatever compensation you agreed. That is not in question.
However, you also have to recognise that you're employing real people with real lives. Some things basically have to be done during office hours, and if you're employing someone during those hours every day, it's only common sense that they can, e.g., make a five-minute personal phone call to a bank or mail order firm now and then.
By the same token, I think it's reasonable to expect, unless explicitly stated otherwise, that an e-mail account may be used for personal reasons, provided that use is not an excessive drain on either the employee's time or the company's resources/chequebook.
The question of monitoring communications is a tricky one. My argument, which I believe is enshrined in law in some countries fairly directly, would be that an employee should be entitled to reasonable privacy. If they're not abusing the system, they shouldn't be subject to monitoring or having their mail read by others, end of story.
Obviously, some times an employer will genuinely have reason to believe that an employee is doing something inappropriate, and must have some recourse in that event. However, IMHO that should be done with the support of a court, just as any wire tap or other invasion of privacy would be (well, aside from things like the Patriot XXX Act, and so on:-/) and not just because an employer randomly decides to take advantage of their access. If it's serious enough that it needs an invasion of privacy, it's surely also serious enough that someone's future employment and reputation is in question, and that's serious enough to do things properly, through legal channels.
Incidentally, I agree entirely with one or two of the other posters here: a smart employer won't rely on things like intercepting communications anyway, because it just breeds discontent amongst the staff, and that will damage productivity at best. It's also the surest way to ensure your good staff are the first to leave you for a more humane employer.
We were talking about a company. About doing things on company time. [...] And I put disclaimers in both a login script and the company handbook about email being the property of the company, any admin that doesnt probably shouldnt be an admin.
Ah, so that's your bias here. You're not by any chance also based in the US, are you? This heartfelt attitude of "employees are slaves to their employers" seems to be peculiar to (management in) that part of the world.
Fortunately, just because you wrote something doesn't necessarily make it legally binding, no matter how officially you state it. Even more fortunately, for some of us at least, most countries in the west explicitly recognise that employees are human beings and not robots, and provide certain safeguards that employers cannot overrule. One of these is often the right to privacy in communications; this varies from country to country, but in most places intercepting phone calls or e-mail traffic without good cause would be against the rules, and an employer with your attitude would get his ass handed to him in court.
I believe by UK law that is not the case. software programmed by you that has no relevance to your job whatsoever is still owned by you.
Really? If, as the grandparent post said, the software was written on work time, I'd assume (but it is an assumption) that it belonged to my employer.
My employment contract does state explicitly that the company has no claim over anything I write using neither work resources nor work time, which seems a reasonable compromise on the "who owns it" question. I wasn't aware of anything in UK law that overrode contractual agreements in this case, but if anyone would like to provide relevant citations...
Slashdot Mirror
Binary? BINARY? When I was young, we programmed with just 0s, and we were grateful.
I thought it was a public record that George W Bush had been in potentially serious legal trouble on several occasions, but had escaped any results against him after special investigators appointed by his father decided there wasn't enough evidence to proceed? (This is from memory, so possibly not technically correct, but that was the general idea.)
Fortunately, as I'm in the UK, it'll be tricky for certain US Treasury Department employees to harrass me for repeating this. Then again, we seem to be creating a new branch of the FBI this week...
I don't see why it should be perceived as a threat. When my current employers made the job offer, I was sent a nice letter and a summary of the Ts&Cs (salary, holiday entitlement, etc.) and told I'd be expected to sign a full contract when I arrived.
Alarm bells rang, so I replied saying I was definitely interested but would like to see the full contract first (to check for exactly the type of "We Own Everything" clause mentioned here, coincidentally). The following evening, the guy who runs about half the company turned up on my doorstep about 7pm, handed me a copy of their standard contract, and basically said, "Got your letter, no problem, it's not unusual for new starters with experience elsewhere to ask."
Incidentally, my contract does have a clause that explicitly indicates the company having no claim over anything I do off company time and without company equipment. In the OP's position, I would immediately ask to have one inserted in place of the current wording, mentioning my situation regarding the existing other business if necessary. If they won't acknowledge your concern and insert the clause pretty much immediately, well, now you know what kind of employer they are and that you don't want to work for them. :o)
But that doesn't work either...
You can (and probably should) implement binary operators as free functions, for example, if necessary making them friends of the class in question. When I write if(a==b), C++ just needs to know what the == operator means in this context. It doesn't, and shouldn't, matter exactly how it's defined, or whether it's a member function; again, imposing such a limitation gains you nothing and damages support for generic programming.
What you say is certainly true up to a point: the highest level code in a particular subsystem, right behind its interface, should indeed define any exceptions it's going to allow out of the subsystem as part of that interface. If there is language-level support for enforcing that definition, it probably makes sense to use it.
However, within the subsystem itself, you often don't want to be translating exceptions at every layer. To do so defeats one of the major advantages of exceptions: you can throw an exception up through several layers at once to a handler, and intervening layers don't have to care unless they're taking some action to handle (or part-handle) the particular exception you threw. In that environment, you could easily wind up with many possible exceptions being thrown from different contexts, but handled in a small number of places. Even with inhertiance and catching groups of related exceptions at once to keep the complexity down, having to give a full specification for everything would soon become unwieldy.
AIUI, the major reason many of the big names in C++ don't like language-supported constraints on type parameters is that just having "derives from this base class" isn't a sufficiently general condition for good support of generic programming.
C++ enforces any necessary interfaces at compile-time, by checking whether the instantiation of a function template can be interpreted properly with the functions available. Requiring every class that supports == to derive from EqualityComparable is just a cumbersome extra for no benefit in a generic, C++ based world, where there are more ways to define and implement interfaces than just inheritance. If it doesn't support ==, the function template will fail to compile at the point of instantiation anyway. (OK, strictly speaking export messes up the compile-time/link-time distinction slightly, but the basic point is still valid.)
Incidentally, it is possible to use template wizardry in C++ to enforce various plausible constraints on type parameters if you really want to restrict them more than their usage within the function template does anyway, perhaps to guard against a possible misinterpretation of the template code in some cases. It's relatively straightforward to do this, and requires little more code than the formal constraints used by some of the other languages under discussion.
I see your point, but of course that "purist" view is also why OO struggles so hard to model some simple concepts, which frequently boil down to examples of multiple dispatch.
If languages are going to support objects, they'll be improved by having a uniform syntax for all of:
You can have the really useful thing from OO -- the power of implicit type coercion and polymorphism -- without creating artificial distinctions between actions that are really serving the same purpose. It's just that many languages -- notably including C++, Java and C# -- don't.
At my last job interview, we were discussing general background in computing for a minute, and I mentioned being karma-capped on Slashdot as a joke (made sense in context). That was followed immediately by a knowing smile from one of my interviewers, and shortly afterwards by a job offer. :-)
I'm always wary of giving credit to someone proclaiming the Next Great Thing, when they've spent the last several years as a leading advocate of the Previous Great Thing, then suddenly start finding all these previously overlooked drawbacks in the older technology when the newer one arrives. Don Box seems awfully close to that position right now.
Assuming you're referring to the Jamie Bulger case, how do you know? Sure, the CCTV showed that the kidnappers didn't fit the usual profile and allowed the police to shift their search. OTOH, traditionally police have asked witnesses (of whom there were probably rather a lot in that particular case) to obtain such information, and there's no reason to believe they couldn't have done so here as well.
CCTV may have saved the police some time, but you can't extrapolate from that that they wouldn't have solved the crime anyway. In fact, if you look at overall crime rates before and after CCTV is installed, you find that it helps reduce certain types of crime where it's installed, but actually makes surprisingly little difference to the proportion of crimes solved overall. OTOH, it also results in an increase in some types of crime in neighbouring areas.
That's fine, taken in isolation. However, as with most civil liberties objections, the key point here is that you are not working in isolation. You have to consider not only the effect on criminals/suspects, but also the effect on everyone else: mistaken identity and voyeurism are both serious concerns, with numerous examples of each identified in relation to the CCTV. For a technology so open to abuse, there's precious little evidence to show it's really helping enough to justify that downside.
Incidentally, the story concludes with, "Despite this, people still seem to prefer the cameras." I don't know who the author talked to, but I doubt it was anyone I know.
Nice theory, but about 15 years out of date with reality.
If knowingly directing vast amounts of traffic at a site from many sources, with the reasonably predictable result being that site's failure, isn't a DDoS attack, what is?
Slashdotting isn't an accident. The eds know damn well that if they post a link in a front page story, that page is going to take zillions of hits. It doesn't take a genius to spot that some servers are better able to handle this than others, and which ones they are likely to be. Deliberately slashdotting a site is no more ethical than distributing a self-contained program to hundreds of computers that then bombards the site with requests from numerous directions at a co-ordinated time. The mechanism is much the same, the result is the same, and it's done just as knowingly.
You are still confusing what you can do with what it's ethical to do, and ignoring the well-established and widely employed legal ideas of reasonable behaviour and fair expectation.
By your argument, almost no-one would bother to make their work available for free on the web without an infinite resource backing them, and the Internet would collapse. There is a reason terms like "reasonable expectation" appear is so many laws, and actions like co-ordinating a DDoS attack are illegal.
No, I don't think littleuser can afford one, at whatever price per gig over his monthly quota his ISP charges.
That simply isn't true. A significant number of /. readers are here for useful discussion and information gathering. If someone posts crap on a technical subject, whether it's about Linux, Microsoft, or otherwise, it's a good bet that someone who knows what they're talking about will reply to set the record straight, and be modded up accordingly.
Much easier to pronounce, but not the same thing at all. If that site was Googled to death, then a Googleplexing would pretty much take out the entire Internet, if you think about it...
Now if only we'd all learn to spell googol, we could quite the bad jokes already. ;-)
That depends on whether a web site that expected 5,000 hits this month just got 500,000 in one day because some idiot posted its address on Slashdot, and the webmaster is paying $1/MB for everything past the first 200MB this month, I guess. (The answer is around $20,000 that day, in case you're having trouble working it out.)
The key concepts a lot of people seem to be missing here are the intent of a web site and the reasonable expectations of its administrator. I don't believe that sites like Slashdot or Google have the right to knowingly direct huge amounts of traffic to a site that wouldn't expect it, without that site's explicit permission. Doing so is essentially just a deliberate DDoS attack on the site, and should be treated as such. Expecting a mainstream news web site to have serious bandwidth is reasonable. Expecting www.bigisp.com/~littleuser/aboutme.html to have serious bandwidth is not.
The fact that it is technically possible to do this to anyone with a web site is not the point. It is technically possible for me to run you over with my car when you step into the road without looking, but if I do so when I could have stopped, I will rightly be charged with (insert your local motoring offence here).
Why the snide comments about the EU? And what on earth makes you think that we don't understand the concepts of rights and freedoms?
I find it deeply ironic that the United States, a country that prides itself on its Constitution and the rights of its citizens, is also the place where: you have giant corporations controlling your government and laughing at your legal system; you have the right to free speech, as long as you can afford the lawyers to defend it; the constitutional safeguards over copyright are being trampled by the aforementioned big corps; you hold hundreds of people indefinitely and without charge, based on an accusation and a technicality of international law that no-one else recognises; and you have a President of dubious mandate, taking your country to war supported by dubious intelligence, resulting in the deaths of thousands of innocents, the destruction of a whole country's infrastructure, the deaths of numerous American servicemen and women, and did I mention some rather lucrative rebuilding contracts for major corps with whom your senior leadership has intimate ties?
The UK government has become increasingly abusive of its authority, particularly since Tony Blair's lot came to power, with Jack Straw and then David Blunkett as Home Secretary. However, we can't even approach your level of legal impotence and government abuse, and you're busy trying to inflict it on the rest of the world! And at least at our general election next year, we'll have candidates to vote for who don't all say the same thing, which is as bad as who we've got at the moment anyway...
You make a lot of good points in your post, and I agree with much of what you say, but with all due respect, you seem to have a serious lack of perspective on the world outside. For all our knowledge of the Dark Ages, as far as rights and responsibilities go, I'd still far prefer to be living in the EU than the US right now.
I'm a great fan of OpenOffice. Nevertheless, I'm also a realist, and I have no hesitation in describing Excel as a vastly superior application to OpenOffice Calc. The latter is OK for basic spreadsheeting, and I do use it often for little things like budgeting, but for serious stuff I'd choose Excel every time, at least for now.
That's "Junior Sanitation Engineer". A little respect, please.
I think Joel's point, which half the flamers on this thread don't seem to get, was that if you're writing software today, and you're trying to get customers to download it tomorrow, writing for .Net is probably not a smart move.
I realise that .Net theoretically has a decent versioning system. However, as someone who's been responsible for investigating new versions of Visual Studio at two different companies now, I've seen plenty of problems in the real world. VS.Net 2002 and VS.Net 2003 really don't play nicely together for a start, and if Microsoft themselves can't get it right with an application whose entire focus is on .Net, what hope does Joe Developer have? I'd be very wary of risking my whole application by trusting a technology so unproven to get the version control right.
And as of today, I think Joel's point about download sizes is also entirely reasonable. I bet the critics here are all happily sitting at home, probably in the US, probably with free local phone calls and always-on broadband. I bet most of them also don't realise that most of the world doesn't have those facilities today, and nor are many of them likely to any time within the next few years. I still have the same 56k dial-up modem I bought when I left uni five years ago, because broadband still has too many practical problems here, and I'm a professional developer and part-time geek who spends several hours on-line some days. I only got a flat-rate package about 3 years ago (before that I was paying per-minute) and even now I'm capped in the number of hours I can use each month, so big downloads are not high on my list of Good Things.
IOWs, as you say, short term Joel's points are very relevant. Later on -- probably 5-7 years away at a minimum -- if 95+% of the Windows user base has .Net installed and/or many more people have broadband, then it might be less so. But 5-7 years is an eternity in software development. For short- to medium-term developments, Joel's comments were right on the money. Using C++ with a good set of libraries is still far ahead of .Net if the size of your executable matters.
News flash: users on dial-up (that's most of the world, BTW) don't like big downloads, and will mostly stick to older technology in preference to spending hours downloading something new and unproven. Big downloads loses you large chunks of user base. If you're doing it commercially, that means large chunks of cash, which is a BFD.
But is it?
Of course, as an employer, you're entitled to expect your employees to do their jobs to the best of their ability, in exchange for whatever compensation you agreed. That is not in question.
However, you also have to recognise that you're employing real people with real lives. Some things basically have to be done during office hours, and if you're employing someone during those hours every day, it's only common sense that they can, e.g., make a five-minute personal phone call to a bank or mail order firm now and then.
By the same token, I think it's reasonable to expect, unless explicitly stated otherwise, that an e-mail account may be used for personal reasons, provided that use is not an excessive drain on either the employee's time or the company's resources/chequebook.
The question of monitoring communications is a tricky one. My argument, which I believe is enshrined in law in some countries fairly directly, would be that an employee should be entitled to reasonable privacy. If they're not abusing the system, they shouldn't be subject to monitoring or having their mail read by others, end of story.
Obviously, some times an employer will genuinely have reason to believe that an employee is doing something inappropriate, and must have some recourse in that event. However, IMHO that should be done with the support of a court, just as any wire tap or other invasion of privacy would be (well, aside from things like the Patriot XXX Act, and so on :-/) and not just because an employer randomly decides to take advantage of their access. If it's serious enough that it needs an invasion of privacy, it's surely also serious enough that someone's future employment and reputation is in question, and that's serious enough to do things properly, through legal channels.
Incidentally, I agree entirely with one or two of the other posters here: a smart employer won't rely on things like intercepting communications anyway, because it just breeds discontent amongst the staff, and that will damage productivity at best. It's also the surest way to ensure your good staff are the first to leave you for a more humane employer.
Ah, so that's your bias here. You're not by any chance also based in the US, are you? This heartfelt attitude of "employees are slaves to their employers" seems to be peculiar to (management in) that part of the world.
Fortunately, just because you wrote something doesn't necessarily make it legally binding, no matter how officially you state it. Even more fortunately, for some of us at least, most countries in the west explicitly recognise that employees are human beings and not robots, and provide certain safeguards that employers cannot overrule. One of these is often the right to privacy in communications; this varies from country to country, but in most places intercepting phone calls or e-mail traffic without good cause would be against the rules, and an employer with your attitude would get his ass handed to him in court.
Really? If, as the grandparent post said, the software was written on work time, I'd assume (but it is an assumption) that it belonged to my employer.
My employment contract does state explicitly that the company has no claim over anything I write using neither work resources nor work time, which seems a reasonable compromise on the "who owns it" question. I wasn't aware of anything in UK law that overrode contractual agreements in this case, but if anyone would like to provide relevant citations...
Think of it as a service payment for all those great pr0n sites you've introduced him to.