Blair is so bad that my father even admitted to voting Tory last year. *shudder*
So bad? The economy the best it has been for over a century. Economic advances like the removal of interest rate changes from government control, political changes like devolution and the reform of the House of Lords, major human rights advances like civil partnerships, equal age of consent, incorporation of European Human Rights into UK law, better rights for new fathers, the setting up of a minimum wage.
Britain is a Parlimentary System, and as such Tony Blair is elected, in truth, only by the members of his particular party.
Wrong. First, he had to be elected as an MP voters in his constituency. These voters have also had the chance to reject him while he was prime minister.
Contrary to a common assumption that Microsoft is anti open source, the reality is not so black and white. Certainly, most customers don't live in that either/or world. They choose a technology - an operating system or an application - based on its ability to solve a particular problem and to serve a certain business need, not based on its development model.
I think he is wrong. My opinion and experience is that many people choose a technology neither for a particular problem or business, or because of its development model. There are quite different reasons, like having a political preference for multi-vendor support for products, or.... in the case of Microsoft, having considerable experience of being let down by a vendor over many years in terms of security issues and long-term API support. I believe that customers frequently choose technologies for much broader and longer-term reasons than individual problems or needs.
.NET has been very successful, and quite a number of businesses are picking up on it.
The only way it has been really successful is in taking over the roles of existing MS development tools that have had support dropped for them, such as VC++ and VB6. Developers who want to stick with Microsoft tools for development have little choice but to migrate to.NET. However, it has largely failed in what it was intended to do - to provide leverage for increased server-side use of Windows by pushing out Java. However, server-side, Java/J2EE dominates and even has a strong presense on Microsoft server platforms..NET was intended to compete with Java. Instead it has successfully competed with Visual Basic 6 and Visual C++, and simply held back Java from the client side, a sitation which seems to be changing.
I said that the shock from deceleration was not the important factor in wether bacteria could survive, and a small rifle bullet doesn't even show that bacteria can survive meteor type decelerations.
And I totally disagree. A small rifle bullet if fired at high velocity at a rock target under conditions intended to simulate impact of a meteor simulates the conditions of impact of a meteor. It does shows that they can survive meteor type decelerations. When a large meteor hits the ground, a considerable amount of material is stressed and heated - there are all kinds of interesting ejecta. However, you can also often find bits of unchanged material from the meteor - indicating neither intense heat or phenomenal deceleration.
I'm arguing because you insisted on saying that mass was irrelevant and that it was a question of force and not heating. I have shown you that it's been a factor of heating and therefore mass/velocity/density/composition that has put a large constraint on the size of meteors that could get to the planet surface.
You were arguing that heating was a result of mass. Well, when a meteor hits the ground, it is. But through the atmosphere (which is what we were talking about, it isn't. It is a result of things like angle of attack and cross-section hitting the atmosphere. Above a certain size (providing the thing doesn't explode due to the pressure or surface stress), the centre will be hardly touched - there is not enough time for heat to be conducted in. You can get damage from abrasion, sure, and rocky meteors are fragile. But rock meteors are a different thing. As I said, interesting chemical analyses can be done on the centres of meteors, which would be impossible if they had been subject to substantial heat.
Which is why I believe that what matters is deceleration, not heat. And projectile experiments with bacteria have shown beyond doubt that they can survive phenomenal decelarations.
You said it was force that mattered, not mass. But now you keep talking about "larger meteors". It's nice to know that you have changed your mind without even realising it. So my original point about the rifle bullet analogy being irrelevant still stands.
Of course mass matters! I mean, a meteor the size of a grain of sand is going not to get through the atomosphere is it? What I meant was that above a certain size, mass is, of course, irrelevant. It has nothing to do with the so-called 'friction' (which is actually adiabatic heating), which was what you were stating.
The point about the rifle bullet is only that it indicated that bacteria can survive tremendous decelerations. You brought up the irrelevant matters of heat and mass. I don't know why. Do you have some philosphical objection to the idea of meteor delivery of microorganisms to planetary surfaces?
I fail to see why you are arguing - everything I have stated is perfectly clear. The fact that the centres of meteors retain their original structures provide conclusively that there is no major heating of their centres - there simply isn't time. And this effect (above, of course, a minimal size) is not affected by mass.
"The heat generated by compression of air in front of the meteor as it travels through the atmosphere is immense, and most meteors completely burn up or explode before they can reach the ground." http://en.wikipedia.org/wiki/Tunguska_event - scroll down to Meteor Airburst
We aren't talking about most meteors. We are talking about meteors large enough to deliver microorganisms. Also, contrary to your assertion, it is larger meteors that can do this, as they can get through the atmosphere without exploding. You need meteors large enough so that they can resist the pressure change of impacting the atmosphere.
Remember, large meteors get through the entire atmosphere in a matter of only a few seconds. If you think that this is enough time for heat to penetrate to the centre of a reasonably sized meteor, you don't know physics.
Proof that it is not is given by the fact that scientists study the interiors of meteors that have hit the ground in order to investigate their original state. Remember the Martian meteorite with apparent forms of life that was discovered in the 90s? Well, that had delicate iron and carbon compounds inside.
Oh no wait, wikipedia is wrong according to you? Sorry my mistake.
Ah - this from the guy who assumed air got hot due to 'friction'.
Actually, as I am sure you well know, Wikipedia is not always definitive.
I think you've lost track of the point i have been trying to make. That is, the kinetic energy mostly is transferred into heating the object, and therefore the mass is directly related to that. YOu go off on a tangent explaining something related but not actually a relative to anything I've been trying to get across.
You were the one mentioning irrelevant terms like 'friction'.
Actually, little of the energy goes into heating the rock, as there is insufficient time (this is why the air gets hot, as there is no time for the heat to leak away, including into the rock).
I really don't think you understand basic physics. Tackle the problem like this: K.E.=.5m(v^2) Most of the K.E. has to be transferred into heating, unless you have some other idea of where the kinetic energy is going? As you said compression of the air causes heating, I mean friction of air against air. With very fast moving objects hitting into an atmosphere, the air doesn't have time to move out of the way.
I'll ignore the insult and explain. Compression heating is not friction. It is called 'adiabatic heating' - technically the work done on the air by compression is transferred to internal energy. That is where the kinetic energy goes. The heat does not have time to escape, so the air heats up. In the popular media this is often called 'friction', but it really isn't.
The thing that annoys me is this use of the word 'community'. This implies that FOSS people are one coherent group, or disabled people are one coherent groups. We are all individuals.
I disagree with the grandparent's view that you should accept the view of experts, since I consider this another form of fundamentalism.
Utter tosh. What is the point of being human if you don't rely on the practical experience of others? Do you intend to personally test all foods and medicines you eat or take? Do you intend to design your own car or plane? If not, they you are relying in the views and experience of experts. To arbitrarily then pick a certain field of knowledge and refuse to listen to experts (I did not say that you should accept their views unquestioningly) is hypocritical.
I tend to not take the word of experts on anything. Maybe it's just a genetic trait I have or something.
but you do, in practise. Do you drive a car? Or travel on a plane? If so, you are taking the word of experts that these things work.
Nutritional recommendations seem to change daily, for example. Experts don't always know.
True, by my point is that experts know better than most. To ignore experts on everything because of the occasional mistakes seems to be (as we British say) 'throwing the baby out with the bathwater'.
I think mass is more important than force. The force isn't in question, the friction and therefore mass is. A heavier object causes more friction as it takes more energy to slow down.
No, this is poor physics. The friction is nothing whatsoever to do with the mass. It is to do with the roughness of the surface. Also, smaller meteors don't get hot because of friction - they get hot because they compress the air in front of them, and compressed things heat up. Larger meteors get hot because the energy of motion get changed to heat when they hit the ground.
Bacteria are more resitant to force than they are to high temperatures.
True.
And also, meteors are asteroids that have simply started raining down onto the Earth. Generally small asteroids the size of rifle bullets won't make it through the atmosphere due to the friction.
The only point of mentioning rifle bullets was because that was the delivery mechanism for bacteria onto rock in an experiment to test how much deceleration they could survive. No-one is suggesting that bacteria can only travel between planets on pieces of rock the size of rifle bullets. Providing the meteor is the apropriate size, it will get slowed down by the atmophere and can deliver bacteria to a planet. The 'rifle bullet' test showed that even meteors that hit a planet's surface with quite a bang can still deliver bacteria.
And of course if the asteroid is too big then there's an airburst or a impact into the ground. Both of which could easily kill cells due to the extreme heat and pressure.
The point of the 'rifle bullet' experiment was to show that ground impacts aren't necessarily fatal to microorganisms.
You fail to understand the magnitude of these impacts. Hint: rifle-bullets do not impact at 24 miles/second, they leave the muzzle with something between 0.2 and 1 miles/second (old pistol to kinetic-kill armour-piercing), the impact-speed is even lower due to air-resistance. (depends on distance though)
The rifle bullets in the experiments weren't fired from a rifle - they were fired at much higher speed.
24 miles/second is roughly 50 times that speed, so the impact-energy would be roughly 2500 times higher.
No, it won't, as there is atmospheric braking.
So, it's sorta like saying that humans are hardy creatures, they can easily survive a fall of 1 foot, so falling 2500 feet and landing safely should be no problem.
No, it isn't, as the experiment was specifically designed to test meteor impact decelerations.
Once you're talking of microbes on the inside of rock, then impact velocities would be much less important. The rock would absorb much of the impact
Actually, microbes are so tough that there is little need to absorb impact stresses. Some experiments have involved bacteria put inside a rifle bullet and fired at rock (to see if they could survive the decelerations of a meteor impact). The bacteria survived and could reproduce.
This is why there is little need, as this article suggests, to have the rocks containing bacteria travelling slowly.
Incidentally, I'm a fundamentalist, and I lean toward a literal understanding of Genesis and a 6000-year earth (although I'm not adamant about it and easily accept that I might be misunderstanding things)
I hope you don't mind me asking, because I am fascinated by the fundamentalist mindset - but why don't you simply accept that you are indeed misunderstanding things?
I think that one of the harshest lessons anyone who likes think about more than trivia has to learn is humility - that one simply has to take the word of experts, no matter how much one may personally want other things to be true.
Tough bugs, sure, but traveling through space also means withstanding the full bore radation of Mr. Sun, with no atmosphere to protect you. I'm not sure I want to meet one of these in a dark alley.
You probably already have. There are bacteria that can survive and even grow exposed to levels of radioactivity found in some parts of nuclear reactors. It looks like some of these bacteria also live in the human stomach.
The thing is, harsh environments and to things like drying out can cause DNA damage, and the same incredible repair mechanisms that help some species to survive those allow them to survive intense radiation.
Incidentally, bacteria surviving to reach Titan is not that interesting - far more exciting is the possibility of them reaching another moon of Saturn - Enceladus, which probably has liquid water.
Please provide an example of data that can't be cleanly set into the relational model.
Irregular data, such as a polymorphic list of objects of different types. Especially data that is frequently changing (say, new subclasses of objects are added).
Huge data stores in archaic, convoluted formats are difficult to manage when they're in transition, yes, but I really don't buy the argument that XML is a universal solution to the problem. Even if you use XML, you still have to have both ends of the transaction agree on the meaning of the XML, and when you're talking about, for example, two different companies merging, that entails sitting down and merging the two formats via a schema, which isn't really that much different from deciding on your own data format since you still have to define all the atoms.
It is completely different from having to define your own data format, for several reasons. First, you know that there are a huge number of tools out there that can parse and process and transform your data. Secondly, you know that you can easily use information and merge without having to sit down and agree on anything. You can say 'here is my bit with my namespace' and 'here is your bit with your namespace' and just process the bits you want to without conflict. One of the main design considerations of XML was specifically to enable this sort of merging!
If one company has data that needs to be converted into the markup of another, then there are standard ways to do describe how this should be done (XSLT).
To re-invent all this from scratch with your own data format would be a ridiculous waste of time.
But, again, that's a reference to data description not data storage. GIF formats provide a description, but filesystems provide the storage for the GIF files. XML formats provide a description of the data, but you must still choose some other mechanism to manage that data, whether it be a DBMS, filesystem, or something else.
No, you are missing a fundamental point. XML is designed from the start to be human readable, and users are encouraged to use meaningful markup. The idea is that the data should be, to as large an extent as possible, self describing. It makes sense without having software. XML can make sense on paper.... and sorry, but bringing up filesystems seems wildly irrelevant to me - after all, filesystems don't store data - magnetic domains on the surface of a disk do....
I don't deny that XML can be a useful tool at times, I just don't accept that it's nearly the breakthrough solution some people portray it as
Having dealt with incompatible formats and formats which have changed (breaking existing uses) for decades, I have no doubt that it is indeed a breakthrough solution. I have seen different organisations XML data combined together seamlessly, and work with each other's software.
I suppose once there are more experts in XML who have honed their XML-ninja skills to the point that they can recite the various pieces of the standards backwards
You don't need such expertise. XML itself is a trivially simple format, and even DTD and schemas aren't that hard.
and it certainly has no place in applications or data management as a controlling factor where people are trying to push it now.
I think that it certainly has - it has barely started to be used to the extent where its real benefits are felt throughout the IT industry.
First of all, it's misleading to say that RDMBSs work in terms of "sets". They work in terms of relations, and a relation can be as restricted as a single attribute and tuple pairing, or as complex as the entire set of data.
Fair enough.
Secondly, RDBMSs have no sense of "order" becuase they don't need it.
Well, RDBMSes don't need it, because they store data that doesn't need it. Not all data fits neatly into relational form.
Third, XML imposes no constraints.
Not of itself, but it has the ability to.
Relational algebra is an actual, tangible thing which, with a limited number of inputs, can be used to restrict data to ensure consistency and atomicity.
True.
XML is just whatever data you want to toss into a text file.
This can apply equally to data stored in relational form. You can have a table that consists of nothing more than, say, a timestamp and big text field. To get the benefits you need to define how the data is divided into meaningful columns to get the benefit. Same with XML - unless you define the meaning and relationships of tags you don't get the benefit.
You can build a DTD or Schema, sure, but how is that any different, really, from just building a complicated application specific to that data to manage the data for you and then tossing it into a text file without the XML?
There is all the difference in the world. There are clearly defined and widely used standards for DTDs etc. One of the primary design considerations of XML was to avoid loss of data because applications go out of date. This is why things like DTDs are standardised. Everyone can open your XML file and be able to be sure it is valid according to the rules you have expressed in your DTD. They can even extend the format with their own namespace and be sure that they haven't broken existing uses.
This is a huge step forward from old formats that did indeed require specific software (and specific versions of that software) to intepret a file format, and my experience is that surprisingly few developers realise it.
But data formats aren't storage, they're representation.
But then relational database files aren't storage - they are representation. If you want to proceed down that path, storage is nothing but magnetic domains on disk.
There are many inherent enforcement in relational systems. There also happen to be many psuedo-relational system masquerading as something they're not.
Very true.
That's not true, you can order your datasets in any way you want when you retrieve them as long as you put them in in a meaningful way to begin with (or, if your order is totally meaningless, keyed it to an autonumber/identity attribute), they're just not stored on the physical medium in an "ordered" fashion of your choosing, which is hardly relevant because a saved XML file probably isn't either.
By 'stored' I mean what you get back by opening the storage medium and retrieving it. If you open a database, you don't get back things in a specific order unless you ask for it. If you open an XML file you do get things back in a specific order.
* need to order chronologically? Store and inspect a creation date. Need to order alphabetically? Trivial. Need to order by byte size? No problem. There's no way that a reasonably built RDBMS can't order data, except randomly, which you can always accomplish with the presence of an identity attribute.
Here is your own argument back at you! That is not storage - that is retrieval. Relational theory has no idea of intrinsic order in storage.
So, XML can impose more constraints than relational databases, as relational stores have no idea about order, and work only in terms of sets.
Yes, but it's not actually a storage system, so it still has to be stored in one, so it's still going to be subject to the problems you mention.
No, it is not addressing the issues of storage systems going out of date, it is addressing the issues of data formats going out of date.
But you have to define that on a case-by-case basis, it's not an inherent enforcement like one finds in relational systems.
There are few inherent enforcements in relational systems - things constraints and foreign keys are optional. Also, you can't impose in a relational system the range of constraints you can impose with XML, such as ordering of data.
So this certainly doesn't seem like it is coming out of nowhere, as you might be concerned -- there have been a lot of other big simulations done recently in the 100,000 atom and up range.
Looks like I am way out of date! When I was dealing with these things, it was on a scale more of hundreds of atoms rather than hundreds of thousands.
I can begin to see where you are coming from. However, I still don't quite agree. Storage is about more than just live databases. It can be about repositories and archives. This can be an area where XML is very important, as it has been specifically designed as a format that is resistant to becoming unreadable over time, perhaps due to the software that writes a format becoming unsupported. Doing long-term archives as, say, Oracle export dumps just isn't good enough!
On the matter of XML being error prone - that need not be the case. XML allows validation of structure and replication through the use of DTDs and Schemas. If you want to avoid such problems, you can prevent them. Also, if additional information is added that can be enforced as being through new validated namespaces.
XML is not just any arbitrary hierarchical format.
Slashdot Mirror
Blair is so bad that my father even admitted to voting Tory last year. *shudder*
So bad? The economy the best it has been for over a century. Economic advances like the removal of interest rate changes from government control, political changes like devolution and the reform of the House of Lords, major human rights advances like civil partnerships, equal age of consent, incorporation of European Human Rights into UK law, better rights for new fathers, the setting up of a minimum wage.
All of these advances were opposed by the Tories.
A truly awful time to be living in Britain? Not!
Britain is a Parlimentary System, and as such Tony Blair is elected, in truth, only by the members of his particular party.
Wrong. First, he had to be elected as an MP voters in his constituency. These voters have also had the chance to reject him while he was prime minister.
Indeed. TB seems to think he knows better, and when the houses rightly reject his bills, he wants to have some method for forcing them through.
Whether or not he knows better, at least he was elected.
From the article:
.... in the case of Microsoft, having considerable experience of being let down by a vendor over many years in terms of security issues and long-term API support. I believe that customers frequently choose technologies for much broader and longer-term reasons than individual problems or needs.
Contrary to a common assumption that Microsoft is anti open source, the reality is not so black and white. Certainly, most customers don't live in that either/or world. They choose a technology - an operating system or an application - based on its ability to solve a particular problem and to serve a certain business need, not based on its development model.
I think he is wrong. My opinion and experience is that many people choose a technology neither for a particular problem or business, or because of its development model. There are quite different reasons, like having a political preference for multi-vendor support for products, or
.NET has been very successful, and quite a number of businesses are picking up on it.
.NET. However, it has largely failed in what it was intended to do - to provide leverage for increased server-side use of Windows by pushing out Java. However, server-side, Java/J2EE dominates and even has a strong presense on Microsoft server platforms. .NET was intended to compete with Java. Instead it has successfully competed with Visual Basic 6 and Visual C++, and simply held back Java from the client side, a sitation which seems to be changing.
The only way it has been really successful is in taking over the roles of existing MS development tools that have had support dropped for them, such as VC++ and VB6. Developers who want to stick with Microsoft tools for development have little choice but to migrate to
I said that the shock from deceleration was not the important factor in wether bacteria could survive, and a small rifle bullet doesn't even show that bacteria can survive meteor type decelerations.
And I totally disagree. A small rifle bullet if fired at high velocity at a rock target under conditions intended to simulate impact of a meteor simulates the conditions of impact of a meteor. It does shows that they can survive meteor type decelerations. When a large meteor hits the ground, a considerable amount of material is stressed and heated - there are all kinds of interesting ejecta. However, you can also often find bits of unchanged material from the meteor - indicating neither intense heat or phenomenal deceleration.
I'm arguing because you insisted on saying that mass was irrelevant and that it was a question of force and not heating. I have shown you that it's been a factor of heating and therefore mass/velocity/density/composition that has put a large constraint on the size of meteors that could get to the planet surface.
You were arguing that heating was a result of mass. Well, when a meteor hits the ground, it is. But through the atmosphere (which is what we were talking about, it isn't. It is a result of things like angle of attack and cross-section hitting the atmosphere. Above a certain size (providing the thing doesn't explode due to the pressure or surface stress), the centre will be hardly touched - there is not enough time for heat to be conducted in. You can get damage from abrasion, sure, and rocky meteors are fragile. But rock meteors are a different thing. As I said, interesting chemical analyses can be done on the centres of meteors, which would be impossible if they had been subject to substantial heat.
Which is why I believe that what matters is deceleration, not heat. And projectile experiments with bacteria have shown beyond doubt that they can survive phenomenal decelarations.
You said it was force that mattered, not mass. But now you keep talking about "larger meteors". It's nice to know that you have changed your mind without even realising it. So my original point about the rifle bullet analogy being irrelevant still stands.
Of course mass matters! I mean, a meteor the size of a grain of sand is going not to get through the atomosphere is it? What I meant was that above a certain size, mass is, of course, irrelevant. It has nothing to do with the so-called 'friction' (which is actually adiabatic heating), which was what you were stating.
The point about the rifle bullet is only that it indicated that bacteria can survive tremendous decelerations. You brought up the irrelevant matters of heat and mass. I don't know why. Do you have some philosphical objection to the idea of meteor delivery of microorganisms to planetary surfaces?
I fail to see why you are arguing - everything I have stated is perfectly clear. The fact that the centres of meteors retain their original structures provide conclusively that there is no major heating of their centres - there simply isn't time. And this effect (above, of course, a minimal size) is not affected by mass.
"The heat generated by compression of air in front of the meteor as it travels through the atmosphere is immense, and most meteors completely burn up or explode before they can reach the ground." http://en.wikipedia.org/wiki/Tunguska_event - scroll down to Meteor Airburst
We aren't talking about most meteors. We are talking about meteors large enough to deliver microorganisms. Also, contrary to your assertion, it is larger meteors that can do this, as they can get through the atmosphere without exploding. You need meteors large enough so that they can resist the pressure change of impacting the atmosphere.
Remember, large meteors get through the entire atmosphere in a matter of only a few seconds. If you think that this is enough time for heat to penetrate to the centre of a reasonably sized meteor, you don't know physics.
Proof that it is not is given by the fact that scientists study the interiors of meteors that have hit the ground in order to investigate their original state. Remember the Martian meteorite with apparent forms of life that was discovered in the 90s? Well, that had delicate iron and carbon compounds inside.
Oh no wait, wikipedia is wrong according to you? Sorry my mistake.
Ah - this from the guy who assumed air got hot due to 'friction'.
Actually, as I am sure you well know, Wikipedia is not always definitive.
I think you've lost track of the point i have been trying to make. That is, the kinetic energy mostly is transferred into heating the object, and therefore the mass is directly related to that. YOu go off on a tangent explaining something related but not actually a relative to anything I've been trying to get across.
You were the one mentioning irrelevant terms like 'friction'.
Actually, little of the energy goes into heating the rock, as there is insufficient time (this is why the air gets hot, as there is no time for the heat to leak away, including into the rock).
I really don't think you understand basic physics. Tackle the problem like this: K.E.=.5m(v^2)
Most of the K.E. has to be transferred into heating, unless you have some other idea of where the kinetic energy is going? As you said compression of the air causes heating, I mean friction of air against air. With very fast moving objects hitting into an atmosphere, the air doesn't have time to move out of the way.
I'll ignore the insult and explain. Compression heating is not friction. It is called 'adiabatic heating' - technically the work done on the air by compression is transferred to internal energy. That is where the kinetic energy goes. The heat does not have time to escape, so the air heats up. In the popular media this is often called 'friction', but it really isn't.
The thing that annoys me is this use of the word 'community'. This implies that FOSS people are one coherent group, or disabled people are one coherent groups. We are all individuals.
I disagree with the grandparent's view that you should accept the view of experts, since I consider this another form of fundamentalism.
Utter tosh. What is the point of being human if you don't rely on the practical experience of others? Do you intend to personally test all foods and medicines you eat or take? Do you intend to design your own car or plane? If not, they you are relying in the views and experience of experts. To arbitrarily then pick a certain field of knowledge and refuse to listen to experts (I did not say that you should accept their views unquestioningly) is hypocritical.
I tend to not take the word of experts on anything. Maybe it's just a genetic trait I have or something.
but you do, in practise. Do you drive a car? Or travel on a plane? If so, you are taking the word of experts that these things work.
Nutritional recommendations seem to change daily, for example. Experts don't always know.
True, by my point is that experts know better than most. To ignore experts on everything because of the occasional mistakes seems to be (as we British say) 'throwing the baby out with the bathwater'.
I think mass is more important than force. The force isn't in question, the friction and therefore mass is. A heavier object causes more friction as it takes more energy to slow down.
No, this is poor physics. The friction is nothing whatsoever to do with the mass. It is to do with the roughness of the surface. Also, smaller meteors don't get hot because of friction - they get hot because they compress the air in front of them, and compressed things heat up. Larger meteors get hot because the energy of motion get changed to heat when they hit the ground.
Bacteria are more resitant to force than they are to high temperatures.
True.
And also, meteors are asteroids that have simply started raining down onto the Earth. Generally small asteroids the size of rifle bullets won't make it through the atmosphere due to the friction.
The only point of mentioning rifle bullets was because that was the delivery mechanism for bacteria onto rock in an experiment to test how much deceleration they could survive. No-one is suggesting that bacteria can only travel between planets on pieces of rock the size of rifle bullets. Providing the meteor is the apropriate size, it will get slowed down by the atmophere and can deliver bacteria to a planet. The 'rifle bullet' test showed that even meteors that hit a planet's surface with quite a bang can still deliver bacteria.
And of course if the asteroid is too big then there's an airburst or a impact into the ground. Both of which could easily kill cells due to the extreme heat and pressure.
The point of the 'rifle bullet' experiment was to show that ground impacts aren't necessarily fatal to microorganisms.
The rifle analogy isn't really appropriate.
There was no rifle analogy - the rifle bullet was simply the mechanism for delivery of bacter
Firstly, it's of several orders of magnitude lighter.
Which is irrelevant - what matters is force, not mass.
It also doesn't travel at tens of km per second like an asteroid does.
I was talking about meteors, not asteroids. Smaller meteors can get substantially slowed by the atmosphere.
You fail to understand the magnitude of these impacts. Hint: rifle-bullets do not impact at 24 miles/second, they leave the muzzle with something between 0.2 and 1 miles/second (old pistol to kinetic-kill armour-piercing), the impact-speed is even lower due to air-resistance. (depends on distance though)
The rifle bullets in the experiments weren't fired from a rifle - they were fired at much higher speed.
24 miles/second is roughly 50 times that speed, so the impact-energy would be roughly 2500 times higher.
No, it won't, as there is atmospheric braking.
So, it's sorta like saying that humans are hardy creatures, they can easily survive a fall of 1 foot, so falling 2500 feet and landing safely should be no problem.
No, it isn't, as the experiment was specifically designed to test meteor impact decelerations.
Once you're talking of microbes on the inside of rock, then impact velocities would be much less important. The rock would absorb much of the impact
Actually, microbes are so tough that there is little need to absorb impact stresses. Some experiments have involved bacteria put inside a rifle bullet and fired at rock (to see if they could survive the decelerations of a meteor impact). The bacteria survived and could reproduce.
This is why there is little need, as this article suggests, to have the rocks containing bacteria travelling slowly.
Incidentally, I'm a fundamentalist, and I lean toward a literal understanding of Genesis and a 6000-year earth (although I'm not adamant about it and easily accept that I might be misunderstanding things)
I hope you don't mind me asking, because I am fascinated by the fundamentalist mindset - but why don't you simply accept that you are indeed misunderstanding things?
I think that one of the harshest lessons anyone who likes think about more than trivia has to learn is humility - that one simply has to take the word of experts, no matter how much one may personally want other things to be true.
Tough bugs, sure, but traveling through space also means withstanding the full bore radation of Mr. Sun, with no atmosphere to protect you. I'm not sure I want to meet one of these in a dark alley.
You probably already have. There are bacteria that can survive and even grow exposed to levels of radioactivity found in some parts of nuclear reactors. It looks like some of these bacteria also live in the human stomach.
The thing is, harsh environments and to things like drying out can cause DNA damage, and the same incredible repair mechanisms that help some species to survive those allow them to survive intense radiation.
Incidentally, bacteria surviving to reach Titan is not that interesting - far more exciting is the possibility of them reaching another moon of Saturn - Enceladus, which probably has liquid water.
Please provide an example of data that can't be cleanly set into the relational model.
Irregular data, such as a polymorphic list of objects of different types. Especially data that is frequently changing (say, new subclasses of objects are added).
Huge data stores in archaic, convoluted formats are difficult to manage when they're in transition, yes, but I really don't buy the argument that XML is a universal solution to the problem. Even if you use XML, you still have to have both ends of the transaction agree on the meaning of the XML, and when you're talking about, for example, two different companies merging, that entails sitting down and merging the two formats via a schema, which isn't really that much different from deciding on your own data format since you still have to define all the atoms.
It is completely different from having to define your own data format, for several reasons. First, you know that there are a huge number of tools out there that can parse and process and transform your data. Secondly, you know that you can easily use information and merge without having to sit down and agree on anything. You can say 'here is my bit with my namespace' and 'here is your bit with your namespace' and just process the bits you want to without conflict. One of the main design considerations of XML was specifically to enable this sort of merging!
If one company has data that needs to be converted into the markup of another, then there are standard ways to do describe how this should be done (XSLT).
To re-invent all this from scratch with your own data format would be a ridiculous waste of time.
But, again, that's a reference to data description not data storage. GIF formats provide a description, but filesystems provide the storage for the GIF files. XML formats provide a description of the data, but you must still choose some other mechanism to manage that data, whether it be a DBMS, filesystem, or something else.
No, you are missing a fundamental point. XML is designed from the start to be human readable, and users are encouraged to use meaningful markup. The idea is that the data should be, to as large an extent as possible, self describing. It makes sense without having software. XML can make sense on paper.... and sorry, but bringing up filesystems seems wildly irrelevant to me - after all, filesystems don't store data - magnetic domains on the surface of a disk do....
I don't deny that XML can be a useful tool at times, I just don't accept that it's nearly the breakthrough solution some people portray it as
Having dealt with incompatible formats and formats which have changed (breaking existing uses) for decades, I have no doubt that it is indeed a breakthrough solution. I have seen different organisations XML data combined together seamlessly, and work with each other's software.
I suppose once there are more experts in XML who have honed their XML-ninja skills to the point that they can recite the various pieces of the standards backwards
You don't need such expertise. XML itself is a trivially simple format, and even DTD and schemas aren't that hard.
and it certainly has no place in applications or data management as a controlling factor where people are trying to push it now.
I think that it certainly has - it has barely started to be used to the extent where its real benefits are felt throughout the IT industry.
First of all, it's misleading to say that RDMBSs work in terms of "sets". They work in terms of relations, and a relation can be as restricted as a single attribute and tuple pairing, or as complex as the entire set of data.
Fair enough.
Secondly, RDBMSs have no sense of "order" becuase they don't need it.
Well, RDBMSes don't need it, because they store data that doesn't need it. Not all data fits neatly into relational form.
Third, XML imposes no constraints.
Not of itself, but it has the ability to.
Relational algebra is an actual, tangible thing which, with a limited number of inputs, can be used to restrict data to ensure consistency and atomicity.
True.
XML is just whatever data you want to toss into a text file.
This can apply equally to data stored in relational form. You can have a table that consists of nothing more than, say, a timestamp and big text field. To get the benefits you need to define how the data is divided into meaningful columns to get the benefit. Same with XML - unless you define the meaning and relationships of tags you don't get the benefit.
You can build a DTD or Schema, sure, but how is that any different, really, from just building a complicated application specific to that data to manage the data for you and then tossing it into a text file without the XML?
There is all the difference in the world. There are clearly defined and widely used standards for DTDs etc. One of the primary design considerations of XML was to avoid loss of data because applications go out of date. This is why things like DTDs are standardised. Everyone can open your XML file and be able to be sure it is valid according to the rules you have expressed in your DTD. They can even extend the format with their own namespace and be sure that they haven't broken existing uses.
This is a huge step forward from old formats that did indeed require specific software (and specific versions of that software) to intepret a file format, and my experience is that surprisingly few developers realise it.
But data formats aren't storage, they're representation.
But then relational database files aren't storage - they are representation. If you want to proceed down that path, storage is nothing but magnetic domains on disk.
There are many inherent enforcement in relational systems. There also happen to be many psuedo-relational system masquerading as something they're not.
Very true.
That's not true, you can order your datasets in any way you want when you retrieve them as long as you put them in in a meaningful way to begin with (or, if your order is totally meaningless, keyed it to an autonumber/identity attribute), they're just not stored on the physical medium in an "ordered" fashion of your choosing, which is hardly relevant because a saved XML file probably isn't either.
By 'stored' I mean what you get back by opening the storage medium and retrieving it. If you open a database, you don't get back things in a specific order unless you ask for it. If you open an XML file you do get things back in a specific order.
* need to order chronologically? Store and inspect a creation date. Need to order alphabetically? Trivial. Need to order by byte size? No problem. There's no way that a reasonably built RDBMS can't order data, except randomly, which you can always accomplish with the presence of an identity attribute.
Here is your own argument back at you! That is not storage - that is retrieval. Relational theory has no idea of intrinsic order in storage.
So, XML can impose more constraints than relational databases, as relational stores have no idea about order, and work only in terms of sets.
Yes, but it's not actually a storage system, so it still has to be stored in one, so it's still going to be subject to the problems you mention.
No, it is not addressing the issues of storage systems going out of date, it is addressing the issues of data formats going out of date.
But you have to define that on a case-by-case basis, it's not an inherent enforcement like one finds in relational systems.
There are few inherent enforcements in relational systems - things constraints and foreign keys are optional. Also, you can't impose in a relational system the range of constraints you can impose with XML, such as ordering of data.
So this certainly doesn't seem like it is coming out of nowhere, as you might be concerned -- there have been a lot of other big simulations done recently in the 100,000 atom and up range.
Looks like I am way out of date! When I was dealing with these things, it was on a scale more of hundreds of atoms rather than hundreds of thousands.
I can begin to see where you are coming from. However, I still don't quite agree. Storage is about more than just live databases. It can be about repositories and archives. This can be an area where XML is very important, as it has been specifically designed as a format that is resistant to becoming unreadable over time, perhaps due to the software that writes a format becoming unsupported. Doing long-term archives as, say, Oracle export dumps just isn't good enough!
On the matter of XML being error prone - that need not be the case. XML allows validation of structure and replication through the use of DTDs and Schemas. If you want to avoid such problems, you can prevent them. Also, if additional information is added that can be enforced as being through new validated namespaces.
XML is not just any arbitrary hierarchical format.