Thanks for a really interesting discussion. It's not often such well written technical material makes it's way onto/.. As a Java developer though I have read this thread and thanked my lucky stars that I will never ever have to deal with or know anything about these problems.
Why are they bothering? This is going to be greeted with all the enthusiasm of someone breaking wind in a swimming pool. It's great that Google have realized that it has people who want to use it's services that run linux but unless they are going to do the porting job properly I don't think they should do it at all.
Perhaps that's a little harsh but I don't want some clunky Windows app with a ton of Wine libs following it around cluttering up my system. Personally I find digiCam to be as good if not better than Picasa so I think I'll stick with that - certainly on Linux I feel that is the application they are competing against.
I certainly don't have a problem with the idea that we don't know everything. Perhaps my problem with dark matter is that it is reported in the press almost as if it is almost fact and yet in reality we haven't got any direct evidence. All we really have to go on is the fact that if dark matter didn't exist things, such as galaxies, wouldn't look like they do. If I had gone to my supervisor with an argument like that when I was doing chemistry he would probably have laughed himself stupid right before he sacked me. As for it not interacting with radiation thinking about it even if it did it could still be very hard to see. After all the universe is very big - you could easily hide something in it.
In the spirit of good science hence forth I am going believe that the FSM holds galaxies together with his noodly appendages. The reason the speed with which the universe is expanding is increasing is easly solved by saving that the universe is created by the FSM using lazy initialization. We gain the ability to see further faster so the FSM has to push on the edge of the universe harder hence making the universe expand faster. Simple really.
We had something similar happen to us. The bank were only calling to find out if we wanted some stupid over priced service or something but wouldn't tell us what it was unil they could verify who we were. We insisted on calling them back in order to verify they were in actual fact the bank and the guy on the other end started to get the hump. After arguing with him for about 10 minutes and explaining basic security ideas very slowly he gave up and told us his number. It wasn't a regular bank number so we called the fraud department who got back to us a couple of hours later to tell us it was indeed a bank employee. Apparently he wasn't meant to be asking security questions and had been given a serious talking to by the fraud office. I wondered if he was an employee that had gone bad. Just goes to show that you even have to be suspicious of the bank sometimes.
I'm sure that there are lots of people that are far more clued up on this than I am that can find holes in what I am about to say but I always felt like dark matter was a bit of a fudge because we don't understand what is happening.
My problem with dark matter is that it's almost as difficult to believe in as God. The only real proof we have is that the universe doesn't appear to move correctly without it. If that's as good as we can do then we might as well say God (or the FSM) is holding the universe together. To my mind it is a big leap from "the universe isn't moving as we expect" to "90% of the universe is made of something we can't see". Surely if the universe was full of this stuff we would be able to detect it because it would block radiation from distant galaxies - or is dark matter conveniently transparent?
You joke but here it is JOS. I think this is actually a midly interesting project - a whole OS (well most anyway) written in a very high level OO language.
There has never been a machine as pretty at the TI-99 in it's brushed metal case. The keyboard was fantastic and, of course, it had a cartridge loading system so no waiting for tapes to load. I wish I had learnt more about it now.
After that I moved onto a BBC B and then a 386 (that was quite a leap).
I understand basically how this system will work but what I would like to know though is how they will spot machines that have cloned keys? Do the trusted devices phone home or something? If not then they would have to actually get hold of at least two actual devices that they suspect are cloned and extract the keys to compare them.
Seems to me it would take a long time for them to find cloned machines and therefore cloning might be worthwhile if it can be done cheaply enough (electron microscpoe not withstanding). Assume for a moment that an organised gang such as the mafia are doing the cloning. If they only produced 100 clones for every machine they bought the chances that the enforcement agencies would find two cloned machines from the same batch is pretty small. Revocation is only a really good threat if the criminal makes thousands of clones because in one fell swoop they could end the criminals business. If the criminal is a bit clever a revocation would only hit a small percentage of their business.
Without the phoning home part of the loop I struggle to see how the revocation side of this system will work very well.
To get even further from the original topic... the game that didn't work on 2000 but did work on 98 and XP probably used some horrbile hack that was introducted in 98 (or before). 2000 was M$es switch to the NT technology and it didn't 100% support the old 98 style technology. They improved backwards compatability in XP and probably fixed what support bug was stopping the game working in 2000. I wouldn't be supprised if the game stopped working again in Vista.
Perhaps I'm being a nut job conspiracy theorist type here but I could easily see this being one of the things that starts a major, perhaps world, war between primarily the middle east and the west. There has been a great deal of tension between these two groups for a long time which isn't helped by the middle east having something the west wants badly (oil).
I think the general feeling of the west is that the middle east is somewhere that we unfortunately have to deal with because they have oil. No sane person is really willing to go to war over oil yet because the middle east is playing ball with us. It isn't a happy relationship but one that both parties are happy enough to just live with.
A few gangs of thugs smashing up the local town centre and making the majority of the local populace afraid is likely to lead to serious negative feelings towards anyone that looks like they might be involved with that group. Once that negative feeling is in place it's hard to get rid of and it becomes easy to then convince the populace that it is time to invade. Lets face it that's what Bush and Blair want. A nice little crusade into the middle east.
I'm not saying that this has been put together by people in black helicopters - I'm sure the people stirring up problems are really just that stupid - but it's damn convenient if your long term plan is to invade the middle east.
... and choose the hammer. I've been putting screws in with a hammer for years and they all stay in, a hammer is also good for nails and I have found hitting two things enough often makes them stick together (especially if there are nails in them). So yes, standardize but on the hammer. Oh wait we were talking about programmign languages. Erm well yes I agree.
I realize that int eh long term it's not an either or but at the moment it could be argued it is because OOo doesn't have the resources for optimization.
You don't know much about testing it would seem. Non-critical updates appear on a daily basis - I am suggesting non-criticals are once ever six months after a QA process.
Yeah maybe I would but I consider myself to be a user of seriously memory sucking applications on a daily basis. At the moment I have FF, NetBeans and OOo running that are consuming just shy of 900m between them. The thing is though their working sets are actually quite small so most of that is off on disk. It's just not really that much of an issue.
I certainly wouldn't claim that the idea is all mine and Ubuntu is close to what I suggest but not quite there. I want a Debian style desktop distribution that is a little more stable than testing. I actually used testing for about 18 months and found it to be very stable but the constant updates were a pain (I suppose I could have just ignored them). I would much rather one big update every 6 months that has been through a defined QA process. Who knows perhaps we'll see it one day. I hope Debian starts to directly address this issue because I think it's the best managed distribution in town but by god is it hard to work with.
I ask myself this question: would I rather an application that was bloated and did everything I wanted or one that was super slim and only did 80% of what I wanted. I would go for the bloatware every single time because to me the extra $50 on ram isn't that great a cost especially if it saves me frustration and time everyday for a year because I don't have to keep switching between applications to get all the functionality I want. I would consider it was money very well spent.
I'm not saying for one minute that every piece of software should follow this rule but some, like OOo, should. Server applications like Apache should be optimized but they are working in a very different environment where making best use of all the resources is critical. Desktop applications though are generally living in an environment where only a tiny fraction of the resources the machine has to offer are actually being used.
The quickest way to spot bad science is by noticing that the press release comes before the journal paper. I would give this a 10% or less chance of actually working.
It's hard to say what the benifits are because we haven't really had the tools to play with yet. What I think history shows though is that given the tools we generally find something to do with them. I admit that we don't always find something good to do with them but that's part of research. I think in 5 to 10 years we will look back and wonder how we ever worked on 2D desktops. I wouldn't like to say what the final 3D interface will look like though - my guess would be that this is the first step towards an immersive VR interface.
How about this for a solution. Debain make a stable server version and a stable desktop version. The server version can be as stable (or even more so) as the current stable releases. The stable desktop version would be released every 6 to 12 months and be pretty damn stable but not as stable as the server version. That way you get the benifit of most of the stability on the desktop without falling horribly far behind. I'm currently tracking unstable because I don't want to fall too far behind but I would probably track "stable desktop" if I could be sure I would get regular well planned updates.
I don't really see much of a problem with OOo as it is. It seems to be developing at a fair pace and it is free (at least as in beer which is all I care about). Ok, so it uses Java, so what. I don't generally find Java slow but then I have a machine that is fairly up to date.
I think part of the problem here is that a good portion of the Linux community runs what most people would consider very old boxes. There is nothing wrong with that but I don't agree that we should hold back development to cater for it. I don't care if an application sucks 200MB of memory as long as it does what I want it to do. If I have a problem with it I'll stick in another GB of RAM to deal with it. There is a limit to this approach but we are no where near it yet.
I could see a tilted mouse working but full on vertical is a non-starter I think. My guess would be that vertial is to steep for the vast majority of people. Shaking hands is something that one does breifly and therefore I am willing to move my body into a less than optimal position. I don't find shaking hands particularly comfortable therefore I don't think I would find shakign hands all day with a mouse comfortable. Anyway, the big problem I see is that the mouse will tend to move away as you click. This makes sense as it has nothing to push against. A hand rest would solve that at teh expense of making the device clunky.
Slashdot Mirror
Thanks for a really interesting discussion. It's not often such well written technical material makes it's way onto /.. As a Java developer though I have read this thread and thanked my lucky stars that I will never ever have to deal with or know anything about these problems.
Why are they bothering? This is going to be greeted with all the enthusiasm of someone breaking wind in a swimming pool. It's great that Google have realized that it has people who want to use it's services that run linux but unless they are going to do the porting job properly I don't think they should do it at all.
Perhaps that's a little harsh but I don't want some clunky Windows app with a ton of Wine libs following it around cluttering up my system. Personally I find digiCam to be as good if not better than Picasa so I think I'll stick with that - certainly on Linux I feel that is the application they are competing against.
I certainly don't have a problem with the idea that we don't know everything. Perhaps my problem with dark matter is that it is reported in the press almost as if it is almost fact and yet in reality we haven't got any direct evidence. All we really have to go on is the fact that if dark matter didn't exist things, such as galaxies, wouldn't look like they do. If I had gone to my supervisor with an argument like that when I was doing chemistry he would probably have laughed himself stupid right before he sacked me. As for it not interacting with radiation thinking about it even if it did it could still be very hard to see. After all the universe is very big - you could easily hide something in it.
In the spirit of good science hence forth I am going believe that the FSM holds galaxies together with his noodly appendages. The reason the speed with which the universe is expanding is increasing is easly solved by saving that the universe is created by the FSM using lazy initialization. We gain the ability to see further faster so the FSM has to push on the edge of the universe harder hence making the universe expand faster. Simple really.
We had something similar happen to us. The bank were only calling to find out if we wanted some stupid over priced service or something but wouldn't tell us what it was unil they could verify who we were. We insisted on calling them back in order to verify they were in actual fact the bank and the guy on the other end started to get the hump. After arguing with him for about 10 minutes and explaining basic security ideas very slowly he gave up and told us his number. It wasn't a regular bank number so we called the fraud department who got back to us a couple of hours later to tell us it was indeed a bank employee. Apparently he wasn't meant to be asking security questions and had been given a serious talking to by the fraud office. I wondered if he was an employee that had gone bad. Just goes to show that you even have to be suspicious of the bank sometimes.
I'm sure that there are lots of people that are far more clued up on this than I am that can find holes in what I am about to say but I always felt like dark matter was a bit of a fudge because we don't understand what is happening.
My problem with dark matter is that it's almost as difficult to believe in as God. The only real proof we have is that the universe doesn't appear to move correctly without it. If that's as good as we can do then we might as well say God (or the FSM) is holding the universe together. To my mind it is a big leap from "the universe isn't moving as we expect" to "90% of the universe is made of something we can't see". Surely if the universe was full of this stuff we would be able to detect it because it would block radiation from distant galaxies - or is dark matter conveniently transparent?
You joke but here it is JOS. I think this is actually a midly interesting project - a whole OS (well most anyway) written in a very high level OO language.
But, what one word fits better. Yes, game isn't perfect, but it is close and everyone understands it so why change it?
For those wondering what the Aquarius was here is some linkage. I vaguely remember seeing one as a kid.
There has never been a machine as pretty at the TI-99 in it's brushed metal case. The keyboard was fantastic and, of course, it had a cartridge loading system so no waiting for tapes to load. I wish I had learnt more about it now.
After that I moved onto a BBC B and then a 386 (that was quite a leap).
Everyone knows that Emacs is better. The real reason most flamewars start is the because Java is fast and better than .NET.
I understand basically how this system will work but what I would like to know though is how they will spot machines that have cloned keys? Do the trusted devices phone home or something? If not then they would have to actually get hold of at least two actual devices that they suspect are cloned and extract the keys to compare them.
Seems to me it would take a long time for them to find cloned machines and therefore cloning might be worthwhile if it can be done cheaply enough (electron microscpoe not withstanding). Assume for a moment that an organised gang such as the mafia are doing the cloning. If they only produced 100 clones for every machine they bought the chances that the enforcement agencies would find two cloned machines from the same batch is pretty small. Revocation is only a really good threat if the criminal makes thousands of clones because in one fell swoop they could end the criminals business. If the criminal is a bit clever a revocation would only hit a small percentage of their business.
Without the phoning home part of the loop I struggle to see how the revocation side of this system will work very well.
To get even further from the original topic... the game that didn't work on 2000 but did work on 98 and XP probably used some horrbile hack that was introducted in 98 (or before). 2000 was M$es switch to the NT technology and it didn't 100% support the old 98 style technology. They improved backwards compatability in XP and probably fixed what support bug was stopping the game working in 2000. I wouldn't be supprised if the game stopped working again in Vista.
Perhaps I'm being a nut job conspiracy theorist type here but I could easily see this being one of the things that starts a major, perhaps world, war between primarily the middle east and the west. There has been a great deal of tension between these two groups for a long time which isn't helped by the middle east having something the west wants badly (oil).
I think the general feeling of the west is that the middle east is somewhere that we unfortunately have to deal with because they have oil. No sane person is really willing to go to war over oil yet because the middle east is playing ball with us. It isn't a happy relationship but one that both parties are happy enough to just live with.
A few gangs of thugs smashing up the local town centre and making the majority of the local populace afraid is likely to lead to serious negative feelings towards anyone that looks like they might be involved with that group. Once that negative feeling is in place it's hard to get rid of and it becomes easy to then convince the populace that it is time to invade. Lets face it that's what Bush and Blair want. A nice little crusade into the middle east.
I'm not saying that this has been put together by people in black helicopters - I'm sure the people stirring up problems are really just that stupid - but it's damn convenient if your long term plan is to invade the middle east.
... and choose the hammer. I've been putting screws in with a hammer for years and they all stay in, a hammer is also good for nails and I have found hitting two things enough often makes them stick together (especially if there are nails in them). So yes, standardize but on the hammer. Oh wait we were talking about programmign languages. Erm well yes I agree.
I realize that int eh long term it's not an either or but at the moment it could be argued it is because OOo doesn't have the resources for optimization.
You don't know much about testing it would seem. Non-critical updates appear on a daily basis - I am suggesting non-criticals are once ever six months after a QA process.
Yeah maybe I would but I consider myself to be a user of seriously memory sucking applications on a daily basis. At the moment I have FF, NetBeans and OOo running that are consuming just shy of 900m between them. The thing is though their working sets are actually quite small so most of that is off on disk. It's just not really that much of an issue.
I certainly wouldn't claim that the idea is all mine and Ubuntu is close to what I suggest but not quite there. I want a Debian style desktop distribution that is a little more stable than testing. I actually used testing for about 18 months and found it to be very stable but the constant updates were a pain (I suppose I could have just ignored them). I would much rather one big update every 6 months that has been through a defined QA process. Who knows perhaps we'll see it one day. I hope Debian starts to directly address this issue because I think it's the best managed distribution in town but by god is it hard to work with.
I ask myself this question: would I rather an application that was bloated and did everything I wanted or one that was super slim and only did 80% of what I wanted. I would go for the bloatware every single time because to me the extra $50 on ram isn't that great a cost especially if it saves me frustration and time everyday for a year because I don't have to keep switching between applications to get all the functionality I want. I would consider it was money very well spent.
I'm not saying for one minute that every piece of software should follow this rule but some, like OOo, should. Server applications like Apache should be optimized but they are working in a very different environment where making best use of all the resources is critical. Desktop applications though are generally living in an environment where only a tiny fraction of the resources the machine has to offer are actually being used.
The quickest way to spot bad science is by noticing that the press release comes before the journal paper. I would give this a 10% or less chance of actually working.
It's hard to say what the benifits are because we haven't really had the tools to play with yet. What I think history shows though is that given the tools we generally find something to do with them. I admit that we don't always find something good to do with them but that's part of research. I think in 5 to 10 years we will look back and wonder how we ever worked on 2D desktops. I wouldn't like to say what the final 3D interface will look like though - my guess would be that this is the first step towards an immersive VR interface.
How about this for a solution. Debain make a stable server version and a stable desktop version. The server version can be as stable (or even more so) as the current stable releases. The stable desktop version would be released every 6 to 12 months and be pretty damn stable but not as stable as the server version. That way you get the benifit of most of the stability on the desktop without falling horribly far behind. I'm currently tracking unstable because I don't want to fall too far behind but I would probably track "stable desktop" if I could be sure I would get regular well planned updates.
I don't really see much of a problem with OOo as it is. It seems to be developing at a fair pace and it is free (at least as in beer which is all I care about). Ok, so it uses Java, so what. I don't generally find Java slow but then I have a machine that is fairly up to date.
I think part of the problem here is that a good portion of the Linux community runs what most people would consider very old boxes. There is nothing wrong with that but I don't agree that we should hold back development to cater for it. I don't care if an application sucks 200MB of memory as long as it does what I want it to do. If I have a problem with it I'll stick in another GB of RAM to deal with it. There is a limit to this approach but we are no where near it yet.
I could see a tilted mouse working but full on vertical is a non-starter I think. My guess would be that vertial is to steep for the vast majority of people. Shaking hands is something that one does breifly and therefore I am willing to move my body into a less than optimal position. I don't find shaking hands particularly comfortable therefore I don't think I would find shakign hands all day with a mouse comfortable. Anyway, the big problem I see is that the mouse will tend to move away as you click. This makes sense as it has nothing to push against. A hand rest would solve that at teh expense of making the device clunky.
There you go ruining a perfectly good piece of "stuff wot I just made up" with facts.