That's a classic one, and something I see all the time at work. People are hostile to / frightened of change, so they make up excuses to avoid it. Each excuse is replaced by another equally far-fetched one as the previous is solved, because they real problem is they just don't want to change.
Don't get me wrong - I think IDSN is cool. It is _not_, however, hot new tech or the way the world is going. I expect IDSN to be around for a _very_ long time, and it's an eminently sensible technology to choose for current and future deployments.
Telstra, unfortunately, talks about IDSN like it's the new hot thing, and prices it to match. Until very recently they were pushing BRI IDSN as an "ideal broadband solution". Sweet gems in their business PRI ISDN pricing include timed local calls with an initial charge that's as much as a normal local call (local calls in Australia are normally flat-free untimed) and expensive timed data calls.
It looks like it'll be cheaper to get eighteen (!!) individual lines and use business-grade ADSL for 'net access rather than get PRI IDSN. Isn't that really rather lame?
It actually looks like it might be worth getting fibre trenched out to our premises as there's a fibre profider with a run nearby, then using VoIP. I'm not enthusiastic about that - IDSN is reliable. Phones need to be reliable. VoIP is unlikely to match the reliability of IDSN. On the other hand, ISDN from Telstra (pretty much the only option) is _most_ unattractively priced.
You make some interesting points, but I can't say I entirely agree. If somebody cannot stop in time to avoid hitting something doing 1/2 the normal speed, they can't stop in time to avoid hitting a broken down car or a stupid child, either. That's called reckless driving, and it tends to land you in a small bare room for a long stay.
As for design issues - agreed. I strongly prefer to use cycleways or cycle lanes, and where that's not possible nice wide roads. Sometimes that's not possible, and it's unavoidable to use a road poorly suited to cycling. It is a driver's responsibility to drive safely, and it is reasonable to expect that they do so, much as it's reasonable to expect cyclists to stay out of the way or off the roads where possible and to be considerate of drivers.
As for riding in the middle of the lane - it's the right thing to do when you know there is not enough room for somebody to pass safely but suspect they might try anyway. The road rules permit it, and while it's generally impolite it's necessary on poorly designed roads. I prefer to avoid such roads instead, but if it's a choice between being wiped out by a moron or forcing the moron to (*gasp*) slow down for a minute, I'll force them to slow down.
The other design issue is that such short-sighted design decisions were made in the first place. Here, even now they're narrowing the roads and adding blocked parking on the sides of the roads, making it a frightening gauntlet to ride between traffic and parked cars with people who may open doors without warning. These roads USED TO BE SAFE. It is beyond me why in a world of high fuel costs, pollution problems, national weight and fitness problems, etc city councils continue to design infrastructure that is actively hostile to cyclists.
You'd be a tad more credible if you felt like putting your name to your post.
As it happens, I agree with you. Cyclists and major roads/highways/freeways do not mix. The difference is that instead of saying "so cyclists should get off the road" I say "I'm paying for this, and can expect decent transport infrastructure too." When roads are upgraded, it's not hard to add a bike path or a cycle lane or two, nor does it take much space. In general, I strongly prefer to stick to cycleways anyway, and in Perth (Western Australia) that's usually a viable option.
On minor roads, however - a cyclist has as much right to use the road as anybody else. It is entirely reasonable to expect not to be wiped out by morons just because they can't be bothered looking where they're going. Paying attention isn't hard. When it comes to the speed issue, at worst people have to slow down for 30s until there's a decent place to pass, and only because that bit of road is too stupidly cramped. Deal with it. Seriously.
As for the insults and generalisations, my thoughts about giant-truck driving redneck hicks are similar - but I'm not making the assumption that you are one like you've made unreasonable assumptions about me (despite the strong temptation to do so).
I must note, also, that I've met more than a few winy idiots myself. Some have been cyclists - and really, painfully bitchy about it. The sort of people who will tell someone who lives 30km from their work in a country that hits 40C in summer that they don't need a car. The temptation to beat them to death is strong. A similar temptation exists for intolerant morons who assume all cyclists are like that and who think they're the only ones whose needs matter.
US$6000/year? A company I know of pays AU$3000/month for our phone service (eighteen lines). That's US$27300/year.
Go VoIP - nah. Our local telco is being _most_ unhelpful with VoIP, and continue to insist that ISDN is the way of the future. Riiiight.
For all the faults of the US telecomms system, at least you have some competition instead of a single private company (gov't owns 51 percent, but like to pretend they don't) that basically owns the system.
"equatorial regions warmed while the arctic grows colder."
No matter what we do, we'll be affecting the climate somehow. Power generation _is_ energy conversion and transfer. If we have to affect the climate, that seems like a darn good direction to push it in.
The equatorial regions will get a bit toasty, but at least it'll stop the icecaps melting. It's hardly ideal, but if you have to pick between a set of bad choices, this one sounds like one of the less awful ones.
Protected memory and not crashing are hardly the domain of new software. VMS, anybody?
The market just chose crap software, or chose based on marketing and short-term feature list differences rather than real quality.
As for usability features, the ones you've mentioned are no doubt nice... but also almost certainly possible even on a 486 with an ancient X server. Sure, they'd be less flashy, but they'd work.
I'm not the sort who thinks we should all stil be using 486s, but I do think it's pretty silly people find three year old hardware "too slow".
The issue is mostly expanding binaries, too - I can very comfortably run Linux on a PII-300 _if_ I upgrade it to have a decent amount of RAM to hold those monster, leaky binaries. The same doesn't seem to be true of OS/X (non-optional eye candy) and WinXP (general bloat) though.
I think what will eventually push us to new media will be something more durable, more convenient (CDs are a pain), probably with the potential to be made cheaper, and hopefully with lower environmental impact.
Hopefully, anyway. Perhaps some memory card format - they're getting big enough now to be good for general use, though they're definitely not cheap enough yet.
I can't say I've had that much of a problem with the Windows search myself - finding it roughly the same as what I've used on the macs here - but then I'm comparing a G4/400 to a 2GHz Athlon;-) . I'm also talking about NTFS - FAT32 is so slow it's incredible.
As for the 'searching the whole disk' thing, really that's all Windows is doing too. All the information it's looking up is in the disk allocation tables, it pretty much has to be for the filesystem to function. I suspect Windows is just doing a much worse job, but also is dealing with an order of magnitude or two more files than the Mac because of the way Windows systems and programs are structured.
Meh. My preferred search is 'find' and 'locate'. Not exactly standard Windows fare, and not something most folks will use on OS/X;-)
If I recall correctly, MacOS X has a disk indexing service that defaults to 'on'.
I think XP has one but it defaults to 'off'.
I strongly suspect that may be part of the reason for the significant differences you're seeing in search times. It's a little like comparing using 'find / -name blah' on OS/X to 'locate' on Linux and using that to say that your Linux filesystem of choice is faster - it doesn't make much sense.
That said, these 'notification-triggered' indexers like Spotlight sound interesting, and are much nicer than the disk crawlers found on all majors OSes at present. I'll be watching to see where this and similar efforts go with interest.
You make some good points regarding the general damage caused by other generation methods. That is something I was aware of an should have touched on.
I am not any more of a fan of conventional energy generation than I am of nuclear energy. Nor am I entirely comfortable with hydro-electric power, though in some situations I think it makes good sense.
I think our key viewpoint difference is on the classification of risk. I view the risk that the plant operators will penny-pinch on safety as high, and the risk that the regulators and inspectors will be under-resourced, denied access when they need it, and blocked by political force as significant also. I view the chance of the government oversight of the process failing to do the job as very high. I think these risks make an otherwise acceptably safe technology unsafe at this time.
I would still choose nuclear power over coal/oil/gas power. I just don't like it and don't trust it - not the technology (which I think is acceptable), but the way it would be applied. Let me make this clear - I think it's possible from a technical viewpoint to run a safe nuclear programme, I just don't think there's any organisation that can be trusted to do it right.
Personally, I'd like to see us do what we can with solar, wind, and when appropriate hydro power, then look into very open, very well monitored and very well funded nuclear power to provide the (considerable) remainder of power needs. I'd also like to see more incentives for people to reduce their own power consumption in the first place.
My problem with this is that current governments are moving toward secrecy and hiding the operation of critical infrastructure like this from their citizens. I think it's critical that acess to information on plant operation and inspection be increased, not decreased, in the interests of safety and to provide citizens with the ability to monitor and evaluate the performance of the oversight authorities.
I think we need to resolve the secrecy issue and other issues with operating methods before a serious large-scale move to nuclear power can be reasonably considered. Simply building reactors as a technical solution is in my view not acceptable until the social and political issues are addressed.
In the end, I think nuclear power generation technology, viewed alone, is fairly safe. Unfortunately, I think that when viewed in the social and political context of its real-world implementation and operation that no longer holds true.
I find just the opposite to be true. No one wants to believe that there are things about yourself that you cannot change.
Good point. I should've been clearer - the/tester/ usually has this belief, in that they want to believe they can test other people and get some easy, accurate answer. Few will have the same belief when viewing the test as a subject.
Of course, most who run the test will also have sat it, but we all know how good people are at distorting results to suit themselves, conveniently forgetting things, etc.
As for the rest of your comment - thanks for the interesting discussion. You reminded me of how much I've forgotten:-( and how much I'm going to enjoy returning to Uni next year. Your classification of the possible failing points is particularly interesting.
I also agree that the use of a set of tests, so long as you understand what they actually mean and the caveats involved in their interpretation, can definitely be useful in the right context. As you say, hiring is one viable option. My fear is only that many people who are arranging such programs won't have a sufficient knowledge of what the tests actually mean and what their limitations are.
Viewting things as a potential job candidate I find the idea of personalty tests somewhat disconcerting, as they can not measure what really matters - my job performance and attitude.
As a potential employer, however, I can see the attraction and the fact that they can be a useful preselection tool with proper application.
Indeed. Most decent intellgence tests (well, decent as far as intellgence tests go, anyway) do produce a set of scores for different areas. Most also provide a single "big number" because that is, in the end, what people want. Recent events have shown just how interested people are in thinking about subtle issues with no single clear answer.
The desire for a single number reflects the rather interesting fallacy that people have some single attribute, "intelligence". There really isn't any such single, clearly defined thing - there are many different sorts of intelligence, and they no more distill down to one thing in a person than they do in a test.
In a sense, I guess I do keep my mail in a database - the Cyrus IMAPd mail store. There it's indexed (body and header indexes) and sorted. Sure, its a heirachal database otherwise known as "the file system" with additional files for the mail-specific header and body indexes, but it works very well.
Personally, I don't think a full RDBMS is necessarily the right answer, but I can see the appeal in some form of client-side database-like functionality, yes.
(I've also replied to another reply to your post with some more information, btw.)
There's at least one IMAP server that does use a database - Exchange. I don't know about you, but I don't find the results thrilling at least from an admin's perspective.
There has been repeated discussion on the Cyrus IMAPd list about mail-in-database storage, but the general consensus is that it would gain nothing over how Cyrus currently does things. Cyrus keeps its own header index and an optional body text index (searching a non-locally-cached 30,000 message mailbox in < 1 second, over dial-up = nice) and does rather well with that approach.
In other words, the facilities are already there. You could use Cyrus for an IMAP server for the kind of client you describe. All you'd need is to write a client that used the IMAP SEARCH command to get folder contents.
That's my #1 reason - the IMAP support works. I'm the administrator of a Cyrus IMAPd server for work, and I am _all_ too familar with the pain Outlook causes IMAP server admins and Outlook IMAP users.
Grepping the Cyrus sources for 'outlook' is less hilarious than I would've expected, but then Cyrus isn't big on comments. *sigh*. Eudora takes the cake anyway - and no surprise, that client is pure evil.
I actually use Evolution myself (but it won't do you any good - no Windows port), but Thunderbird would be my next choice, mostly for stability and because it's a tolerant, well behaved client that I can expect to just work without worrying about it.
I've now had a very quick look at some of the tests offered online.
My god.
At least as interpreted by the online services (rarely known for good test design and conduct - *lol*) it's so full of social context as to be downright scary.
"A uniform"
What the fuck. How many different meanings can "a uniform" have to someone? To have authority, to have authority force something upon you, to have power, to have someone in power use it on you, to be restricted or restrained in choice, a role or task, to be performing a role or task, menial labour, low paying employment, etc etc etc.
It is possible that the test takes that into account, or that the question is there as a dummy / control, etc. I'm going to hope so.
This sort of issue is also very common in IQ tests. Most are packed full of cultural and social context, as well as requiring significant language skills. They test language, education, and cultural knowledge much more than intelligence. If you understand this, they're still useful, but most people don't and treat them as intelligence tests.
There is no such thing as an accurate and unchangeable psychological test. Thankfully, people aren't that simple.
First, a negative may very obviously later turn into a positive. People go crazy, they aren't generally born crazy (though they sometimes are with some conditions). A lot can happen to you after you're eight, and it's far from impossible that your core ideas, thinking patterns and values could be affected in adverse ways.
Second, a positive may sometimes turn in to a negative in later life. I don't accept that if you have psychological problems they can not be addressed. Sometimes (usually) this will require professional treatment, but there are going to be cases where it does not. There are also going to be cases where you simply "get better".
Third, almost no test of something not directly observable is perfectly accurate. Psychological tests are renowned for being difficult to design and carry out accurately. They're also renowned for people's tendency to want to belive them despite known inaccuracies and problems. IQ tests (which actually measure one thing: your IQ, a number that describes how good you are at an IQ test, not your intelligence) are a great example of this problem in action.
Fourth, many (even most) psychological tests are easily gamed if you understand what the test is for and how it works. Doing so is also a lot of fun:-) . Don't put that past an eight year old - especially the sort they want to identify and "catch".
Finally, I think it is ethically unacceptable to write somebody off based on a test that can not be 100% accurate, on results that may change in future, and for a condition that may be correctable anyway.
It looks to me like the Hartman Value Profile has entered the realm of pop psychology. It's something poeople want to have, so they'll believe it. Pop psychology is dangerous, though, because people tend to belive in these things in very black and white terms, with little room for considerations of the issues and inaccuracies involved.
Of course, none of this in any way argues that the test doesn't work - I know too little about it for that, and the chances are it's as effective as any other psychological test - somewhat, if the results are interpreted and explained by someone who understands the test and all the issues involved. My argument is merely that it'd be very stupid to rely on it for anything important.
If some company wants to use it to select sales people, that seems reasonable enough. They're no doubt willing to simply accept the risks of being gamed, and the risks of inaccuracies and misinterpretations. Of course, it's also very likely they just don't understand how psychological testing works and have been "sold" it as some sort of magic selection tool. *sigh*. Business is full of that sort of thing - just read any management book from your local bookshop to get an idea of how full.
I will research this particular test further, as I'm now interested in determining what the full situation is. Unfortunately, Google appears to be flooded with information from firms offering services based upon these ideas, and I don't have access to any online journal services from home.
In closing, simply let me caution you against believing any psychological test to be accurate as a stand-alone evaluation. Many are useful as guides, and many can when properly conducted produce information that can be interpreted in informative ways, but it's not likely that a given test will really be able to tell you some aspect of someone's personality "out of the box". People aren't that simple, and neither is psychology.
(NOTE: I am not a trained psychologist, though I have completed some formal and casual study in the area. It's a fascinating area, and well worth reading about - but make sure you get proper study materials, not the reams of pop psych crap.)
One other issue is that while there's little question a nuclear reactor _can_ be operated very safely, there are entirely reasonable doubts over whether a government can effectively ensure that a large collection of nuclear reactors _are_ operated safely.
Sloppy operation is a risk, as is insufficient inspection and monitoring. Another potential issue is underfunding (especially with privatised operation) of safety measures. Even if abundant government funds were thrown at the task, there's the risk they'd be abused or redirected to other things and still leave safety underfunded and undermonitored.
The Japanese government was unsuccessful. If they can't do it, there's no chance in hell we (Australia) can with our current government. I wouldn't rate the US's chances as even worth considering, especially as right now a reactor program would probably be contracted out to private industry (buddies, no doubt), probably with terms that paid well but didn't do as much for proper regulation, inspection, and safety requirements as required.
My big problem with nuclear power is that it only takes one big fuck up to do major amounts of essentially permanent damage. If you're running several thousand reactors for extended periods of time, almost no risk is only questionably good enough.
If I thought I could trust an organisation to get it right, I'd be all for it anyway - I do think it is possible to make it low risk enough. I don't think there is any such organisaion, and people being what they are (as you noted yourself) I'm not sure there ever will be.
What's really depressing is that this is _exactly_ the same as Australia, except Australia was even more pronounced.
Howard won by a large margin. To a large extent, this was because... what was his name again?... Mark Latham was so wet and lacking in personality, spine, and credibility. Even I only voted for him (well, my preferences went to his party before Howard's party) because I felt it was critically important to get Howard out.
Latham ignored the economic issue, and Howard gutted him on it. Latham was focusing on real issues, though rather poorly, as Howard cut him to pieces with scaremongering. It works. Howard made it simple, black and white, and scary - so he won.
Don't forget pushing their absurd environmental and IP agenda on the rest of the world. There's also attempting to suppress dissent through media and political power, and attempting to stilfe free speech both in the US and abroad.
Regarding free speech - a good example is that recent Bin Laden video. The US attempted to use its political muscle to prevent it from being shown by Al Jazeera. WTF?!?
From the outside - specifically Australia (we have our own shame named Howard, of course) - this is not just a shame, it's terrifying. The most common word phrase I've heard in the last hour has been "oh shit, bush won."
It's bad enough for the rest of the world, but Howard will take this as validation of his own beliefs and policies and tie us even more firmly to the runaway train that is US foreign policy.
I'd be scared for the stupidity involved if I wasn't so miserable about the whole thing. Howard winning over here was quite bad enough:-(
Re:EC over IP I have been doing this for years.
on
Replacing TCP?
·
· Score: 1
Thanks for your reply.
There's one thing I still don't understand, though. You had a 60kbps video data stream being delivered over a 64kbps link with 60kbps actual throughput - fine. If you're sending 600kps, you are as you say going to lose 90% of your packets.
That said, I don't see how you can achieve a 60kbps data stream from 60kbps final throughput from your protocol, as you can't guarantee that the/right/ packets will arrive, and you have no room for redundancy. Are you sure the video codec wasn't just correcting for missed packets?
I also don't understand your comments on packet loss. Do you mean that the link was a theoretial 64kpbs link that had 90% packet loss? Or was it _effective_ 64kbps after packet loss was taken into account? If the former, I still don't see how your scheme could work - I just can't see how you can turn 6kbps of data into 60.
Slashdot Mirror
That's a classic one, and something I see all the time at work. People are hostile to / frightened of change, so they make up excuses to avoid it. Each excuse is replaced by another equally far-fetched one as the previous is solved, because they real problem is they just don't want to change.
*sigh*
Don't get me wrong - I think IDSN is cool. It is _not_, however, hot new tech or the way the world is going. I expect IDSN to be around for a _very_ long time, and it's an eminently sensible technology to choose for current and future deployments.
Telstra, unfortunately, talks about IDSN like it's the new hot thing, and prices it to match. Until very recently they were pushing BRI IDSN as an "ideal broadband solution". Sweet gems in their business PRI ISDN pricing include timed local calls with an initial charge that's as much as a normal local call (local calls in Australia are normally flat-free untimed) and expensive timed data calls.
It looks like it'll be cheaper to get eighteen (!!) individual lines and use business-grade ADSL for 'net access rather than get PRI IDSN. Isn't that really rather lame?
It actually looks like it might be worth getting fibre trenched out to our premises as there's a fibre profider with a run nearby, then using VoIP. I'm not enthusiastic about that - IDSN is reliable. Phones need to be reliable. VoIP is unlikely to match the reliability of IDSN. On the other hand, ISDN from Telstra (pretty much the only option) is _most_ unattractively priced.
You make some interesting points, but I can't say I entirely agree. If somebody cannot stop in time to avoid hitting something doing 1/2 the normal speed, they can't stop in time to avoid hitting a broken down car or a stupid child, either. That's called reckless driving, and it tends to land you in a small bare room for a long stay.
As for design issues - agreed. I strongly prefer to use cycleways or cycle lanes, and where that's not possible nice wide roads. Sometimes that's not possible, and it's unavoidable to use a road poorly suited to cycling. It is a driver's responsibility to drive safely, and it is reasonable to expect that they do so, much as it's reasonable to expect cyclists to stay out of the way or off the roads where possible and to be considerate of drivers.
As for riding in the middle of the lane - it's the right thing to do when you know there is not enough room for somebody to pass safely but suspect they might try anyway. The road rules permit it, and while it's generally impolite it's necessary on poorly designed roads. I prefer to avoid such roads instead, but if it's a choice between being wiped out by a moron or forcing the moron to (*gasp*) slow down for a minute, I'll force them to slow down.
The other design issue is that such short-sighted design decisions were made in the first place. Here, even now they're narrowing the roads and adding blocked parking on the sides of the roads, making it a frightening gauntlet to ride between traffic and parked cars with people who may open doors without warning. These roads USED TO BE SAFE. It is beyond me why in a world of high fuel costs, pollution problems, national weight and fitness problems, etc city councils continue to design infrastructure that is actively hostile to cyclists.
You'd be a tad more credible if you felt like putting your name to your post.
As it happens, I agree with you. Cyclists and major roads/highways/freeways do not mix. The difference is that instead of saying "so cyclists should get off the road" I say "I'm paying for this, and can expect decent transport infrastructure too." When roads are upgraded, it's not hard to add a bike path or a cycle lane or two, nor does it take much space. In general, I strongly prefer to stick to cycleways anyway, and in Perth (Western Australia) that's usually a viable option.
On minor roads, however - a cyclist has as much right to use the road as anybody else. It is entirely reasonable to expect not to be wiped out by morons just because they can't be bothered looking where they're going. Paying attention isn't hard. When it comes to the speed issue, at worst people have to slow down for 30s until there's a decent place to pass, and only because that bit of road is too stupidly cramped. Deal with it. Seriously.
As for the insults and generalisations, my thoughts about giant-truck driving redneck hicks are similar - but I'm not making the assumption that you are one like you've made unreasonable assumptions about me (despite the strong temptation to do so).
I must note, also, that I've met more than a few winy idiots myself. Some have been cyclists - and really, painfully bitchy about it. The sort of people who will tell someone who lives 30km from their work in a country that hits 40C in summer that they don't need a car. The temptation to beat them to death is strong. A similar temptation exists for intolerant morons who assume all cyclists are like that and who think they're the only ones whose needs matter.
US$6000/year? A company I know of pays AU$3000/month for our phone service (eighteen lines). That's US$27300/year.
Go VoIP - nah. Our local telco is being _most_ unhelpful with VoIP, and continue to insist that ISDN is the way of the future. Riiiight.
For all the faults of the US telecomms system, at least you have some competition instead of a single private company (gov't owns 51 percent, but like to pretend they don't) that basically owns the system.
Oh, crap. I'm a cyclist. Or should I say soon-to-be-ex-cyclist if this hits Australia.
The sort of people this is already a problem with are _exactly_ the sort of people who would watch sitcom update snippets. *uggh*.
"equatorial regions warmed while the arctic grows colder."
No matter what we do, we'll be affecting the climate somehow. Power generation _is_ energy conversion and transfer. If we have to affect the climate, that seems like a darn good direction to push it in.
The equatorial regions will get a bit toasty, but at least it'll stop the icecaps melting. It's hardly ideal, but if you have to pick between a set of bad choices, this one sounds like one of the less awful ones.
Protected memory and not crashing are hardly the domain of new software. VMS, anybody?
... but also almost certainly possible even on a 486 with an ancient X server. Sure, they'd be less flashy, but they'd work.
The market just chose crap software, or chose based on marketing and short-term feature list differences rather than real quality.
As for usability features, the ones you've mentioned are no doubt nice
I'm not the sort who thinks we should all stil be using 486s, but I do think it's pretty silly people find three year old hardware "too slow".
The issue is mostly expanding binaries, too - I can very comfortably run Linux on a PII-300 _if_ I upgrade it to have a decent amount of RAM to hold those monster, leaky binaries. The same doesn't seem to be true of OS/X (non-optional eye candy) and WinXP (general bloat) though.
I think what will eventually push us to new media will be something more durable, more convenient (CDs are a pain), probably with the potential to be made cheaper, and hopefully with lower environmental impact.
Hopefully, anyway. Perhaps some memory card format - they're getting big enough now to be good for general use, though they're definitely not cheap enough yet.
... will no doubt make theirs out of plastic.
Fair enough.
;-) . I'm also talking about NTFS - FAT32 is so slow it's incredible.
;-)
I can't say I've had that much of a problem with the Windows search myself - finding it roughly the same as what I've used on the macs here - but then I'm comparing a G4/400 to a 2GHz Athlon
As for the 'searching the whole disk' thing, really that's all Windows is doing too. All the information it's looking up is in the disk allocation tables, it pretty much has to be for the filesystem to function. I suspect Windows is just doing a much worse job, but also is dealing with an order of magnitude or two more files than the Mac because of the way Windows systems and programs are structured.
Meh. My preferred search is 'find' and 'locate'. Not exactly standard Windows fare, and not something most folks will use on OS/X
If I recall correctly, MacOS X has a disk indexing service that defaults to 'on'.
I think XP has one but it defaults to 'off'.
I strongly suspect that may be part of the reason for the significant differences you're seeing in search times. It's a little like comparing using 'find / -name blah' on OS/X to 'locate' on Linux and using that to say that your Linux filesystem of choice is faster - it doesn't make much sense.
That said, these 'notification-triggered' indexers like Spotlight sound interesting, and are much nicer than the disk crawlers found on all majors OSes at present. I'll be watching to see where this and similar efforts go with interest.
You make some good points regarding the general damage caused by other generation methods. That is something I was aware of an should have touched on.
I am not any more of a fan of conventional energy generation than I am of nuclear energy. Nor am I entirely comfortable with hydro-electric power, though in some situations I think it makes good sense.
I think our key viewpoint difference is on the classification of risk. I view the risk that the plant operators will penny-pinch on safety as high, and the risk that the regulators and inspectors will be under-resourced, denied access when they need it, and blocked by political force as significant also. I view the chance of the government oversight of the process failing to do the job as very high. I think these risks make an otherwise acceptably safe technology unsafe at this time.
I would still choose nuclear power over coal/oil/gas power. I just don't like it and don't trust it - not the technology (which I think is acceptable), but the way it would be applied. Let me make this clear - I think it's possible from a technical viewpoint to run a safe nuclear programme, I just don't think there's any organisation that can be trusted to do it right.
Personally, I'd like to see us do what we can with solar, wind, and when appropriate hydro power, then look into very open, very well monitored and very well funded nuclear power to provide the (considerable) remainder of power needs. I'd also like to see more incentives for people to reduce their own power consumption in the first place.
My problem with this is that current governments are moving toward secrecy and hiding the operation of critical infrastructure like this from their citizens. I think it's critical that acess to information on plant operation and inspection be increased, not decreased, in the interests of safety and to provide citizens with the ability to monitor and evaluate the performance of the oversight authorities.
I think we need to resolve the secrecy issue and other issues with operating methods before a serious large-scale move to nuclear power can be reasonably considered. Simply building reactors as a technical solution is in my view not acceptable until the social and political issues are addressed.
In the end, I think nuclear power generation technology, viewed alone, is fairly safe. Unfortunately, I think that when viewed in the social and political context of its real-world implementation and operation that no longer holds true.
I find just the opposite to be true. No one wants to believe that there are things about yourself that you cannot change.
/tester/ usually has this belief, in that they want to believe they can test other people and get some easy, accurate answer. Few will have the same belief when viewing the test as a subject.
:-( and how much I'm going to enjoy returning to Uni next year. Your classification of the possible failing points is particularly interesting.
Good point. I should've been clearer - the
Of course, most who run the test will also have sat it, but we all know how good people are at distorting results to suit themselves, conveniently forgetting things, etc.
As for the rest of your comment - thanks for the interesting discussion. You reminded me of how much I've forgotten
I also agree that the use of a set of tests, so long as you understand what they actually mean and the caveats involved in their interpretation, can definitely be useful in the right context. As you say, hiring is one viable option. My fear is only that many people who are arranging such programs won't have a sufficient knowledge of what the tests actually mean and what their limitations are.
Viewting things as a potential job candidate I find the idea of personalty tests somewhat disconcerting, as they can not measure what really matters - my job performance and attitude.
As a potential employer, however, I can see the attraction and the fact that they can be a useful preselection tool with proper application.
Indeed. Most decent intellgence tests (well, decent as far as intellgence tests go, anyway) do produce a set of scores for different areas. Most also provide a single "big number" because that is, in the end, what people want. Recent events have shown just how interested people are in thinking about subtle issues with no single clear answer.
The desire for a single number reflects the rather interesting fallacy that people have some single attribute, "intelligence". There really isn't any such single, clearly defined thing - there are many different sorts of intelligence, and they no more distill down to one thing in a person than they do in a test.
I'm not convinced.
In a sense, I guess I do keep my mail in a database - the Cyrus IMAPd mail store. There it's indexed (body and header indexes) and sorted. Sure, its a heirachal database otherwise known as "the file system" with additional files for the mail-specific header and body indexes, but it works very well.
Personally, I don't think a full RDBMS is necessarily the right answer, but I can see the appeal in some form of client-side database-like functionality, yes.
(I've also replied to another reply to your post with some more information, btw.)
There's at least one IMAP server that does use a database - Exchange. I don't know about you, but I don't find the results thrilling at least from an admin's perspective.
There has been repeated discussion on the Cyrus IMAPd list about mail-in-database storage, but the general consensus is that it would gain nothing over how Cyrus currently does things. Cyrus keeps its own header index and an optional body text index (searching a non-locally-cached 30,000 message mailbox in < 1 second, over dial-up = nice) and does rather well with that approach.
In other words, the facilities are already there. You could use Cyrus for an IMAP server for the kind of client you describe. All you'd need is to write a client that used the IMAP SEARCH command to get folder contents.
That's my #1 reason - the IMAP support works. I'm the administrator of a Cyrus IMAPd server for work, and I am _all_ too familar with the pain Outlook causes IMAP server admins and Outlook IMAP users.
Grepping the Cyrus sources for 'outlook' is less hilarious than I would've expected, but then Cyrus isn't big on comments. *sigh*. Eudora takes the cake anyway - and no surprise, that client is pure evil.
I actually use Evolution myself (but it won't do you any good - no Windows port), but Thunderbird would be my next choice, mostly for stability and because it's a tolerant, well behaved client that I can expect to just work without worrying about it.
I've now had a very quick look at some of the tests offered online.
.
My god.
At least as interpreted by the online services (rarely known for good test design and conduct - *lol*) it's so full of social context as to be downright scary.
"A uniform"
What the fuck. How many different meanings can "a uniform" have to someone? To have authority, to have authority force something upon you, to have power, to have someone in power use it on you, to be restricted or restrained in choice, a role or task, to be performing a role or task, menial labour, low paying employment, etc etc etc
It is possible that the test takes that into account, or that the question is there as a dummy / control, etc. I'm going to hope so.
This sort of issue is also very common in IQ tests. Most are packed full of cultural and social context, as well as requiring significant language skills. They test language, education, and cultural knowledge much more than intelligence. If you understand this, they're still useful, but most people don't and treat them as intelligence tests.
There is no such thing as an accurate and unchangeable psychological test. Thankfully, people aren't that simple.
:-) . Don't put that past an eight year old - especially the sort they want to identify and "catch".
First, a negative may very obviously later turn into a positive. People go crazy, they aren't generally born crazy (though they sometimes are with some conditions). A lot can happen to you after you're eight, and it's far from impossible that your core ideas, thinking patterns and values could be affected in adverse ways.
Second, a positive may sometimes turn in to a negative in later life. I don't accept that if you have psychological problems they can not be addressed. Sometimes (usually) this will require professional treatment, but there are going to be cases where it does not. There are also going to be cases where you simply "get better".
Third, almost no test of something not directly observable is perfectly accurate. Psychological tests are renowned for being difficult to design and carry out accurately. They're also renowned for people's tendency to want to belive them despite known inaccuracies and problems. IQ tests (which actually measure one thing: your IQ, a number that describes how good you are at an IQ test, not your intelligence) are a great example of this problem in action.
Fourth, many (even most) psychological tests are easily gamed if you understand what the test is for and how it works. Doing so is also a lot of fun
Finally, I think it is ethically unacceptable to write somebody off based on a test that can not be 100% accurate, on results that may change in future, and for a condition that may be correctable anyway.
It looks to me like the Hartman Value Profile has entered the realm of pop psychology. It's something poeople want to have, so they'll believe it. Pop psychology is dangerous, though, because people tend to belive in these things in very black and white terms, with little room for considerations of the issues and inaccuracies involved.
Of course, none of this in any way argues that the test doesn't work - I know too little about it for that, and the chances are it's as effective as any other psychological test - somewhat, if the results are interpreted and explained by someone who understands the test and all the issues involved. My argument is merely that it'd be very stupid to rely on it for anything important.
If some company wants to use it to select sales people, that seems reasonable enough. They're no doubt willing to simply accept the risks of being gamed, and the risks of inaccuracies and misinterpretations. Of course, it's also very likely they just don't understand how psychological testing works and have been "sold" it as some sort of magic selection tool. *sigh*. Business is full of that sort of thing - just read any management book from your local bookshop to get an idea of how full.
I will research this particular test further, as I'm now interested in determining what the full situation is. Unfortunately, Google appears to be flooded with information from firms offering services based upon these ideas, and I don't have access to any online journal services from home.
In closing, simply let me caution you against believing any psychological test to be accurate as a stand-alone evaluation. Many are useful as guides, and many can when properly conducted produce information that can be interpreted in informative ways, but it's not likely that a given test will really be able to tell you some aspect of someone's personality "out of the box". People aren't that simple, and neither is psychology.
(NOTE: I am not a trained psychologist, though I have completed some formal and casual study in the area. It's a fascinating area, and well worth reading about - but make sure you get proper study materials, not the reams of pop psych crap.)
"Then store it away somewhere."
;-)
It's the "somewhere" that's the problem
One other issue is that while there's little question a nuclear reactor _can_ be operated very safely, there are entirely reasonable doubts over whether a government can effectively ensure that a large collection of nuclear reactors _are_ operated safely.
Sloppy operation is a risk, as is insufficient inspection and monitoring. Another potential issue is underfunding (especially with privatised operation) of safety measures. Even if abundant government funds were thrown at the task, there's the risk they'd be abused or redirected to other things and still leave safety underfunded and undermonitored.
The Japanese government was unsuccessful. If they can't do it, there's no chance in hell we (Australia) can with our current government. I wouldn't rate the US's chances as even worth considering, especially as right now a reactor program would probably be contracted out to private industry (buddies, no doubt), probably with terms that paid well but didn't do as much for proper regulation, inspection, and safety requirements as required.
My big problem with nuclear power is that it only takes one big fuck up to do major amounts of essentially permanent damage. If you're running several thousand reactors for extended periods of time, almost no risk is only questionably good enough.
If I thought I could trust an organisation to get it right, I'd be all for it anyway - I do think it is possible to make it low risk enough. I don't think there is any such organisaion, and people being what they are (as you noted yourself) I'm not sure there ever will be.
What's really depressing is that this is _exactly_ the same as Australia, except Australia was even more pronounced.
... what was his name again? ... Mark Latham was so wet and lacking in personality, spine, and credibility. Even I only voted for him (well, my preferences went to his party before Howard's party) because I felt it was critically important to get Howard out.
Howard won by a large margin. To a large extent, this was because
Latham ignored the economic issue, and Howard gutted him on it. Latham was focusing on real issues, though rather poorly, as Howard cut him to pieces with scaremongering. It works. Howard made it simple, black and white, and scary - so he won.
I fucking hate people.
Don't forget pushing their absurd environmental and IP agenda on the rest of the world. There's also attempting to suppress dissent through media and political power, and attempting to stilfe free speech both in the US and abroad.
Regarding free speech - a good example is that recent Bin Laden video. The US attempted to use its political muscle to prevent it from being shown by Al Jazeera. WTF?!?
From the outside - specifically Australia (we have our own shame named Howard, of course) - this is not just a shame, it's terrifying. The most common word phrase I've heard in the last hour has been "oh shit, bush won."
:-(
It's bad enough for the rest of the world, but Howard will take this as validation of his own beliefs and policies and tie us even more firmly to the runaway train that is US foreign policy.
I'd be scared for the stupidity involved if I wasn't so miserable about the whole thing. Howard winning over here was quite bad enough
Thanks for your reply.
/right/ packets will arrive, and you have no room for redundancy. Are you sure the video codec wasn't just correcting for missed packets?
There's one thing I still don't understand, though. You had a 60kbps video data stream being delivered over a 64kbps link with 60kbps actual throughput - fine. If you're sending 600kps, you are as you say going to lose 90% of your packets.
That said, I don't see how you can achieve a 60kbps data stream from 60kbps final throughput from your protocol, as you can't guarantee that the
I also don't understand your comments on packet loss. Do you mean that the link was a theoretial 64kpbs link that had 90% packet loss? Or was it _effective_ 64kbps after packet loss was taken into account? If the former, I still don't see how your scheme could work - I just can't see how you can turn 6kbps of data into 60.