It may be peer-reviewed, but it's from 1994! That's 12 years ago - haven't you got something a bit more recent? (given the large amount of research work that has happened on climate change since then, I think it's reasonable to be skeptical of old articles).
Peer review is fine - but it can only cover the state of the art at the time of the review. Plenty of science that was good for its time has been superseded since...
...but the big challenge is getting mass out of the Earth's gravity well at the moment (for which an ion engine is not useful) so adding extra mass is a problem.
If extra grids (of very low mass) added to the ion engine can significantly improve its impulse, that can be a very effective way to improve the deep space characteristics while still having something that can be launched in one piece from the Earth.
Actually, I'd say the opposite is true (to a point). If you imagine that a manufacturer support agreement will get you out of a hole faster than your own talented staff who know the source of the software you use, you're naive.
My own experience with vendor support is that very often they waste a lot of your time while they persuade you that it's your fault (your hardware, wrong firmware, wrong drivers, security patches you've installed) rather than providing instant access to high-level technical resources.
If you are using an open-source product, then you may be better off maintaining in-house expertise (and a support backstop) but the support contract won't help you in an emergency as much as your own staff will.
The substring functions aren't that inefficient - if you set the storage type to EXTERNAL (ALTER TABLE foo ALTER COLUMN bar SET STORAGE EXTERNAL) then substring operations on large values will only read the minimum number of chunks required (rather than the whole value). This makes streaming a vlaue realistic.
The tradeoff is increased storage space (but for some sorts of data it may not be very compressible anyway) and therefore increased disk time to load it.
No, it all goes to the BBC (minus about 6% which is the cost of collection). And yes, it also pays for radio and online services as well.
The BBC's total income for the year ended 31st March 03 was around 3500 million pounds. BBC Annual Report and Accounts. 2659 million was from the license fee, the remainder from commercial ventures (e.g. selling books & videos; renting facilities; technical consultancy and specialist services).
Ok. For data over a certain size, individual sheets of paper are more error-prone than computer files. As you saw in the Florida election, just having the votes in hand doesn't mean you know what the total is. For nations the size of the US, counting votes can be a monthlong procedure- and that's with a significant chance of error on each one (better form design can reduce it greatly- no butterfly+chad). The inabliity to count & recount quickly is itself a kind of buginess.
Actually, speed of count is in no way related to country size, because you should really be counting on a distributed local level and counting in parallel. Despite using entirely paper-based, hand (not machine!) counting, the UK manages to deliver final election declarations for the majority of the country within 12 hours of close of poll. For very rural areas it takes up to 24 hours.
It seems to me that the US has less need of rapid counting than the UK. Our national administration changes as soon as the result becomes clear - it would matter very little if a US presidential election took 48 or 72 hours to count (can't comment on other US elections). I can't see why it need take any longer than that if there was the willpower there to do it!
It seems that there is pressure to make voting cheap. If you think that election of the president of the USA is of some importance, (and is an infrequent event) maybe it's not so bad to spend some money on it.
It has been argued that this was the correct behaviour - it was a bulletproof car designed to protect him. If it opened the doors every time there was a bug...
I accept the value of goverment bonds in the currency system, but I'm sceptical that financial futures are important to the economy. I would go so far as to assert that economic speculation disconnected from any productive activity is harmful to the productive economy.
Cue lengthy arguments about Tobin taxes (or not - this may not be a turn-on for people...)
We have all seen during the cold war, specifically the Soviet and American space race, that competition gets the best results. I think the case in point is the moon landings of the Apollo missions, NASA's (and the world's) ultimate space faring success. They just need more money. Money is what worked then, and it will work now if they had more. Space is expensive no matter what.
Of course, space exploration is expensive - human spaceflight more so. This leads to a vicious circle where commercial contractors each wanting money spent on their technology cause the final vehicle to be a curious hybrid whose implementation doesn't live up to the promise of the theory. It's a marvellously intricate piece of construction, but is so complex that it is time-consuming and expensive to maintain or upgrade.
And if you're going to spend more money on space, you need to ask what you're trying to achieve - in the cold war it was national prestige and public confidence that was being bought. I think our priorities may have become a bit more pragmatic since - and technology has made unmanned flight useful for many activities.
Routine human spaceflight is an aspirational goal rather than a practical solution - and aspirations are often expensive. (This is not an objection to it).
Replacing the shuttle with a new design from scratch might well be better, but there are plenty of vested interests in the shuttle programme to contend with - and a case of not wanting to write off existing investment in the shuttle.
The politics involved may also be "let my state keep its shuttle jobs". Of course, this happens on all sorts of military-style projects (another consequence of the Cold War space race was to tie US and Russian space programs rather too closely to the military.) NASA seems to become a political football between various vested interests.
The remaining theories focus on an apparent problem in movement of a mechanism within the instrument that rapidly vibrates a gamma-ray source back and forth.
"The Mossbauer spectrometer on Spirit is working, and even if we don't come up with a way to improve its performance, we'll be able to get scientific information out of the data it sends us from Mars," Squyres said. "But it's a very flexible instrument, with lots of parameters we can change. We have high hopes that over the coming months we'll be able to understand exactly what's happened to it and make adjustments that will improve its performance. And if the Mossbauer spectrometer on Opportunity behaves on Mars the way it did today, we'll get beautiful data from that instrument."
This implies that they don't necessarily need to fix the software, just recalibrate/ change the configuration parameters. So something is not up to spec, but that doesn't really matter because they can tweak other parameters to compensate.
And I wasn't joking - I was involved in a university "improvised STM" project once, so I can just about remember how to do it. The piezo element was nearly the most expensive bit of the thing for that one as well.
I suspect that it would be a problem in some cases - their example (clustering of metal atoms) is possibly less sensitive to excess electrons passing through than some others might be. I'm not sure that it's necessarily quantum interference either.
The other thing worth mentioning is that depending on the process you're observing, the 1 frame/30 seconds offered by competitor technologies may not really be a problem. Some reactions are slow! (About 10-15 years ago, the oxidisation of Si(111) 7x7 was one of the questions that people were interested in, and you could slow down the process by letting the oxygen into your vacuum chamber more slowly.)
Also, bear in mind that electron transmission is only one thing you might want to measure - atomic force microscopes and scanning tunnelling microscopes measure surface properties (e.g. STMs measure the local density of electon states). Now if they can just get the scanning faster....
This is still interesting stuff. Some day when I get bored, I'll build myself an STM...
For singlemode fibre, TIR is not relevant. It's all transmission modes. When the core gets narrow enough only one mode can exist.
The issue with bending is that you change the geometry and encourage coupling between the core mode and the cladding mode (the cladding being 125 micron it carries plenty of modes.
I see four ways to solve this: 1) replace the fibers with narrower fibers,
That doesn't necessarily help in a transmission system - it alters a whole lot of the characteristics. Bear in mind also that singlemode fibre (9 micron core) is much harder to terminate properly than 62.5 multimode.
2) replace the cladding with cladding that can take the heat dissipation,
Interesting materials problem. We currently use silica glass because it is flexible, strong and has good optical properties. , I think
3) use a lower transmission power, 4) have someone go out and assess each place where the fiber bends, and make it bend at a shallower angle if necessary.
Option 3 is pretty much impossible, since you need higher power to get a higher data rate (this is, after all, why the powers keep increasing). I think option 4 is pretty much the best shot.
Higher data rates (for same pulse width) have higher duty cycles, so the average power (which is what matters thermally) increases. You're right: fixing the paths to avoid bend loss is the best solution, really. You could always use multiple fibres as well (admittedly although the silica uis cheap, the bits to hang on the ends are more expensive)
As always, Slashdot is the home of the gross oversimplification.
Most nations can't make laws as they see fit - not even the US - because they have committed by treaty to behave in a particular way[*]. States in the US are likewise constrained by the existence of a federal constitution and legislative system.
In that context, the European states have decided to go down a particular route because they see a benefit to it (e.g. harmonisation of standards makes interstate commerce easier). They have only given up sovereignty on specific issues. (I won't go into the whole story of Qualified Majority Voting here!)
[*] Of course, you can break the treaty, but people may choose to impose sanctions on you for doing so... just as these countries will probably be fined if they don't comply.
Yes, it is obviously slower. But that's a tradeoff that many people are willing to make. And I didn't say that libxml2 was the fastest parser - I'm not qualified to say that, as I've not benchmarked it properly.
Bear in mind that many uses of XML are for data interchange, where speed is less important than compatibility. XML gives you more potential to add extra data into the format and still use a mix of old and new tools. Binary formats generally require that all programs using the format be upgraded if the format changes.
The "structure of the file". Obviously, binary files have no evident structure to humans without special tools - so if you have a mixture of content (e.g. floating point, variable length text) you have to rely on remembering the order of fields and be able to translate their values.
Honestly, I'm not saying there's not a place for binary formats (there is and I'd be happy to use them) but for many of the applications XML is used for, particularly data interchange, human readability is a big bonus. The XML I've been involved with shoudl be pretty readable to anyone with a knowledge of the domain.
Manipulating XML may be cheaper than you think. libxml2 is very fast (IME) - I've used it with PostgreSQL for doing XPath queries on database columns and it is fast enough to make an XPath search (which involves building a DOM, parsing the XPath query and then executing it, for each row) across 1200 rows sufficiently fast to be useful. (It was a fraction of a second IIRC - obviosuly dependent on the nature of our XML docs).
Yeah, I was surprised too.
I disagree about the human readable/writable bit. It is easily human readable/writable if it's properly structured (if it's complex because the information is complex, that's an inevitability. Make the data model simpler, if that's a problem to you). In terms of efficiency - sure, binary formats are more efficient, but they are much harder to debug when they go wrong.
I agree that XML documents are not necessarily self-documenting. That isn't surprising. XML is about syntax, not semantics. You can use XSD to provide basic (integer vs char) semantics, but anything more complicated comes back to human understanding and agreed specification. If you understand the objects in your schema, XML can provide a good presentation of those objects.
I accept that the development cost is still high - but the reuse of a common platform (for basic comms, power, navigation (if applicable) infrastructure) helps. I notice that ESA has built Mars Express on a standard "bus" that they hope can be reused for other interplanetary missions (e.g. to Venus).
The consequence of increasing knowledge of space technology is that parts of it can maybe be a little more routine (though still scary/exciting when you think that your code - possibly bugs and all - is in space!)
... is that it is allowing students (albeit mainly postgraduates) to actually try out new satellite concepts on a space platform.
This could have very positive effects for the overall cost of satellite technology - Mistakes are currently sooooo expensive! (Hopefully you can launch a little cluster of these fairly cheaply by piggybacking on other payloads).
See, for example: this ESA article:
Solar-electric propulsion is ESA's new spacecraft engine. It does not burn fuel as chemical rockets do; instead the technique converts sunlight into electricity via solar panels and uses it to electrically charge heavy gas atoms, which accelerate from the spacecraft at high velocity. This drives the spacecraft forwards. In a chemical rocket, burning the fuel creates gas that is expelled relatively slowly compared to electric thrusters. However, in an ion engine, the gas is ejected at large velocities, which makes it generally much more efficient, so less fuel is required.
Because propulasion works by conservation of momentum, if you can fire the ions out the back fast enough you don't need too many of them. The problem with normal jet propulsion is that the jets aren't very high velocity.
They reported a comment made by a "credible source" in the intelligence services suggesting that there had been political pressure to "sex up" the dossier. They clearly attributed this to a single source - they (the BBC) did not accuse the government of lying. In fact, it was the other way around - Alistair Campbell accused BBC reporters of telling lies.
As for the proof of it - that would be difficult, given we're talking intelligence services. They might be right, they might be wrong about the detail, but surely it is a matter of genuine public interest if the government were manipulating intelligence for political end, and the intelligence services were unhappy about it.
Except that the camera does lie, not just through manipulating frames, but especially through choosing carefully what you show, and from what angle. Stage magicians are very good at this, and the audience for a show has got a much wider field of view than you get on a TV. Beware.
Perhaps you could use this instant data as evidence in court, with you taking an oath - but a reputable broadcaster shouldn't touch it with a bargepole? (i.e. all the direputable ones will be wall-to-wall "grim reality" TV).
Of course, it was a stupid example to use to promote 3G (and thus carefully chosen by the article's author to mock Dixon). Editing is everything...
Quite so. Our wish to "do more" proceeds along with technology - why do governments need expensive computer systems to administer income tax? Why do businesses use computers now when armies of clerks and typists served them well a hundred years ago.
Well, these things happened because technology gave governments and companies a way to realise an ambition - and that ambition centred on "If only we knew more about..." and an accumulation of details which had been ignored before because it was too difficult to collect and manage them. The problem comes when you keep collecting data and can no longer cope with it.
So, back to gadgets and our free time. We think - if only I could escape this tedious drudge (which might only take 5 minutes a week) then things would be better. You then spend 3 hours reading the manual trying to understand how to work it. The marketing of high-tech gadgets frequently offers high aspirations (as someone said above) and of course they work for some people. The trick is that the sellers want you to feel like one of those people, and it's only once you've bought the gadget and played with it that you find it doesn't make you the most fulfilled person on Earth. Unsurprisingly, the gadget isn't an instant solution for unhappiness.
We're sold a life of success, satisfaction and instant gratification - these things don't come cheap and the cost is often paid by people working insanely long hours on short-term contracts who reckon that if they could have things delivered to *them* right now, they'd be happier....
Slashdot Mirror
Peer review is fine - but it can only cover the state of the art at the time of the review. Plenty of science that was good for its time has been superseded since...
...but the big challenge is getting mass out of the Earth's gravity well at the moment (for which an ion engine is not useful) so adding extra mass is a problem. If extra grids (of very low mass) added to the ion engine can significantly improve its impulse, that can be a very effective way to improve the deep space characteristics while still having something that can be launched in one piece from the Earth.
Actually, I'd say the opposite is true (to a point). If you imagine that a manufacturer support agreement will get you out of a hole faster than your own talented staff who know the source of the software you use, you're naive. My own experience with vendor support is that very often they waste a lot of your time while they persuade you that it's your fault (your hardware, wrong firmware, wrong drivers, security patches you've installed) rather than providing instant access to high-level technical resources. If you are using an open-source product, then you may be better off maintaining in-house expertise (and a support backstop) but the support contract won't help you in an emergency as much as your own staff will.
The substring functions aren't that inefficient - if you set the storage type to EXTERNAL (ALTER TABLE foo ALTER COLUMN bar SET STORAGE EXTERNAL) then substring operations on large values will only read the minimum number of chunks required (rather than the whole value). This makes streaming a vlaue realistic. The tradeoff is increased storage space (but for some sorts of data it may not be very compressible anyway) and therefore increased disk time to load it.
No, silver is.
Gold is not quite as good, but doesn't tarnish, which is why it gets used so much.
I believe the order goes: Silver, Copper, Gold, Sodium (!)). Of these Gold is best for exposed contacts.
The BBC's total income for the year ended 31st March 03 was around 3500 million pounds. BBC Annual Report and Accounts. 2659 million was from the license fee, the remainder from commercial ventures (e.g. selling books & videos; renting facilities; technical consultancy and specialist services).
Actually, speed of count is in no way related to country size, because you should really be counting on a distributed local level and counting in parallel. Despite using entirely paper-based, hand (not machine!) counting, the UK manages to deliver final election declarations for the majority of the country within 12 hours of close of poll. For very rural areas it takes up to 24 hours.
It seems to me that the US has less need of rapid counting than the UK. Our national administration changes as soon as the result becomes clear - it would matter very little if a US presidential election took 48 or 72 hours to count (can't comment on other US elections). I can't see why it need take any longer than that if there was the willpower there to do it!
It seems that there is pressure to make voting cheap. If you think that election of the president of the USA is of some importance, (and is an infrequent event) maybe it's not so bad to spend some money on it.
It has been argued that this was the correct behaviour - it was a bulletproof car designed to protect him. If it opened the doors every time there was a bug...
Cue lengthy arguments about Tobin taxes (or not - this may not be a turn-on for people...)
They're forever telling me I've been pre-approved (subject to status) but I've never been sent an unsolicited card...
We have all seen during the cold war, specifically the Soviet and American space race, that competition gets the best results. I think the case in point is the moon landings of the Apollo missions, NASA's (and the world's) ultimate space faring success. They just need more money. Money is what worked then, and it will work now if they had more. Space is expensive no matter what.
Of course, space exploration is expensive - human spaceflight more so. This leads to a vicious circle where commercial contractors each wanting money spent on their technology cause the final vehicle to be a curious hybrid whose implementation doesn't live up to the promise of the theory. It's a marvellously intricate piece of construction, but is so complex that it is time-consuming and expensive to maintain or upgrade.
And if you're going to spend more money on space, you need to ask what you're trying to achieve - in the cold war it was national prestige and public confidence that was being bought. I think our priorities may have become a bit more pragmatic since - and technology has made unmanned flight useful for many activities.
Routine human spaceflight is an aspirational goal rather than a practical solution - and aspirations are often expensive. (This is not an objection to it).
Replacing the shuttle with a new design from scratch might well be better, but there are plenty of vested interests in the shuttle programme to contend with - and a case of not wanting to write off existing investment in the shuttle.
The politics involved may also be "let my state keep its shuttle jobs". Of course, this happens on all sorts of military-style projects (another consequence of the Cold War space race was to tie US and Russian space programs rather too closely to the military.) NASA seems to become a political football between various vested interests.
The remaining theories focus on an apparent problem in movement of a mechanism within the instrument that rapidly vibrates a gamma-ray source back and forth.
"The Mossbauer spectrometer on Spirit is working, and even if we don't come up with a way to improve its performance, we'll be able to get scientific information out of the data it sends us from Mars," Squyres said. "But it's a very flexible instrument, with lots of parameters we can change. We have high hopes that over the coming months we'll be able to understand exactly what's happened to it and make adjustments that will improve its performance. And if the Mossbauer spectrometer on Opportunity behaves on Mars the way it did today, we'll get beautiful data from that instrument."
This implies that they don't necessarily need to fix the software, just recalibrate/ change the configuration parameters. So something is not up to spec, but that doesn't really matter because they can tweak other parameters to compensate.
scp new_values.conf science@spirit.nasa.gov:/etc/mossbauer.conf
thinks Now, how do we kill -HUP it :)
And I wasn't joking - I was involved in a university "improvised STM" project once, so I can just about remember how to do it. The piezo element was nearly the most expensive bit of the thing for that one as well.
The other thing worth mentioning is that depending on the process you're observing, the 1 frame/30 seconds offered by competitor technologies may not really be a problem. Some reactions are slow! (About 10-15 years ago, the oxidisation of Si(111) 7x7 was one of the questions that people were interested in, and you could slow down the process by letting the oxygen into your vacuum chamber more slowly.)
Also, bear in mind that electron transmission is only one thing you might want to measure - atomic force microscopes and scanning tunnelling microscopes measure surface properties (e.g. STMs measure the local density of electon states). Now if they can just get the scanning faster....
This is still interesting stuff. Some day when I get bored, I'll build myself an STM...
For singlemode fibre, TIR is not relevant. It's all transmission modes. When the core gets narrow enough only one mode can exist.
The issue with bending is that you change the geometry and encourage coupling between the core mode and the cladding mode (the cladding being 125 micron it carries plenty of modes.
I see four ways to solve this: 1) replace the fibers with narrower fibers,
That doesn't necessarily help in a transmission system - it alters a whole lot of the characteristics. Bear in mind also that singlemode fibre (9 micron core) is much harder to terminate properly than 62.5 multimode.
2) replace the cladding with cladding that can take the heat dissipation,
Interesting materials problem. We currently use silica glass because it is flexible, strong and has good optical properties. , I think
3) use a lower transmission power, 4) have someone go out and assess each place where the fiber bends, and make it bend at a shallower angle if necessary.
Option 3 is pretty much impossible, since you need higher power to get a higher data rate (this is, after all, why the powers keep increasing). I think option 4 is pretty much the best shot.
Higher data rates (for same pulse width) have higher duty cycles, so the average power (which is what matters thermally) increases. You're right: fixing the paths to avoid bend loss is the best solution, really. You could always use multiple fibres as well (admittedly although the silica uis cheap, the bits to hang on the ends are more expensive)
Most nations can't make laws as they see fit - not even the US - because they have committed by treaty to behave in a particular way[*]. States in the US are likewise constrained by the existence of a federal constitution and legislative system.
In that context, the European states have decided to go down a particular route because they see a benefit to it (e.g. harmonisation of standards makes interstate commerce easier). They have only given up sovereignty on specific issues. (I won't go into the whole story of Qualified Majority Voting here!) [*] Of course, you can break the treaty, but people may choose to impose sanctions on you for doing so... just as these countries will probably be fined if they don't comply.
Yes, it is obviously slower. But that's a tradeoff that many people are willing to make. And I didn't say that libxml2 was the fastest parser - I'm not qualified to say that, as I've not benchmarked it properly.
Bear in mind that many uses of XML are for data interchange, where speed is less important than compatibility. XML gives you more potential to add extra data into the format and still use a mix of old and new tools. Binary formats generally require that all programs using the format be upgraded if the format changes.
The "structure of the file". Obviously, binary files have no evident structure to humans without special tools - so if you have a mixture of content (e.g. floating point, variable length text) you have to rely on remembering the order of fields and be able to translate their values.
Honestly, I'm not saying there's not a place for binary formats (there is and I'd be happy to use them) but for many of the applications XML is used for, particularly data interchange, human readability is a big bonus. The XML I've been involved with shoudl be pretty readable to anyone with a knowledge of the domain.
Yeah, I was surprised too.
I disagree about the human readable/writable bit. It is easily human readable/writable if it's properly structured (if it's complex because the information is complex, that's an inevitability. Make the data model simpler, if that's a problem to you). In terms of efficiency - sure, binary formats are more efficient, but they are much harder to debug when they go wrong.
I agree that XML documents are not necessarily self-documenting. That isn't surprising. XML is about syntax, not semantics. You can use XSD to provide basic (integer vs char) semantics, but anything more complicated comes back to human understanding and agreed specification. If you understand the objects in your schema, XML can provide a good presentation of those objects.
My favourite FAQ from the UK site is:
Are we going to die?
Yes. We are all going to die some time. It is, however, very unlikely that the collision of a Near Earth Object will be responsible.
The consequence of increasing knowledge of space technology is that parts of it can maybe be a little more routine (though still scary/exciting when you think that your code - possibly bugs and all - is in space!)
... is that it is allowing students (albeit mainly postgraduates) to actually try out new satellite concepts on a space platform.
This could have very positive effects for the overall cost of satellite technology - Mistakes are currently sooooo expensive! (Hopefully you can launch a little cluster of these fairly cheaply by piggybacking on other payloads).
I wonder if they'll sell me one :)
Because propulasion works by conservation of momentum, if you can fire the ions out the back fast enough you don't need too many of them. The problem with normal jet propulsion is that the jets aren't very high velocity.
They reported a comment made by a "credible source" in the intelligence services suggesting that there had been political pressure to "sex up" the dossier. They clearly attributed this to a single source - they (the BBC) did not accuse the government of lying. In fact, it was the other way around - Alistair Campbell accused BBC reporters of telling lies.
As for the proof of it - that would be difficult, given we're talking intelligence services. They might be right, they might be wrong about the detail, but surely it is a matter of genuine public interest if the government were manipulating intelligence for political end, and the intelligence services were unhappy about it.
Except that the camera does lie, not just through manipulating frames, but especially through choosing carefully what you show, and from what angle. Stage magicians are very good at this, and the audience for a show has got a much wider field of view than you get on a TV. Beware.
Perhaps you could use this instant data as evidence in court, with you taking an oath - but a reputable broadcaster shouldn't touch it with a bargepole? (i.e. all the direputable ones will be wall-to-wall "grim reality" TV).
Of course, it was a stupid example to use to promote 3G (and thus carefully chosen by the article's author to mock Dixon). Editing is everything...
Quite so. Our wish to "do more" proceeds along with technology - why do governments need expensive computer systems to administer income tax? Why do businesses use computers now when armies of clerks and typists served them well a hundred years ago.
Well, these things happened because technology gave governments and companies a way to realise an ambition - and that ambition centred on "If only we knew more about..." and an accumulation of details which had been ignored before because it was too difficult to collect and manage them. The problem comes when you keep collecting data and can no longer cope with it.
So, back to gadgets and our free time. We think - if only I could escape this tedious drudge (which might only take 5 minutes a week) then things would be better. You then spend 3 hours reading the manual trying to understand how to work it. The marketing of high-tech gadgets frequently offers high aspirations (as someone said above) and of course they work for some people. The trick is that the sellers want you to feel like one of those people, and it's only once you've bought the gadget and played with it that you find it doesn't make you the most fulfilled person on Earth. Unsurprisingly, the gadget isn't an instant solution for unhappiness.
We're sold a life of success, satisfaction and instant gratification - these things don't come cheap and the cost is often paid by people working insanely long hours on short-term contracts who reckon that if they could have things delivered to *them* right now, they'd be happier....