So it's because Lisp is a bit more "consistent/seamless", so it's easier to have code modify itself in meaningful ways?
How easy is it typically to modify such a program to do something different?
Say you have a messaging server for inter module communications, and you want to convert it to a chat server for inter human communications;). Without glossing over the details and new special cases e.g. protocol changes for human specific requirements.
Would it be easy for an average lisp programmer to understand what was done, if the conversion was done with a bunch of macros? Or would that be the equivalent of getting an average perl programmer to understand one of those perl one-liners?
OK, if you were asked to write a DHCP _server_ in Lisp that stores leases and logic in an RDBMS, what would be the steps? Which Lisp site would you visit first, which modules to consider etc? Or would you just write everything yourself in less than 30 minutes:).
Being not so smart and lazy, I prefer using code written by smarter or more hardworking people, so I currently use perl and CPAN;).
Something like: #!/usr/bin/perl -w use strict; my $fh; my $size=1000000; my $buf="\000" x $size; my $fname='00zeroes.000';
open($fh,"> $fname") or die "cannot open file: $!";
while(1) {
last unless (print $fh $buf);
}
close ($fh);
unlink("$fname") or die "\a\a\aUnable to unlink $fname: $!";
print "\a\aDone\n!";
--- You probably might want to do a sync. But I'm not too bothered.
Those same attitudes work well when you want high quality stuff.
My friend who used to work for a Japanese car company was sent to Japan for training for a few months.
Whilst he was there he found that it was common for staff to be rearranging and tidying their desks etc at 9pm just so everything is nicely in place for efficient work the next day.
In the normal office area there were circles on the floor to mark where wastebaskets should be. Lots of little details like that.
I doubt the staff of the local office here have quite that commitment to work, attention to detail, discipline etc.
Which is probably why the local build quality is unlikely to be as good as that in Japan. Though I heard it is still better than other locally available brands.
"Gadzooks, man, what are you using, dd | gzip? Get yourself a program that just copies the used sectors into your image file!"
knoppix (using the noswap boot option). I only do the zeroing sometimes. It's no big deal to me - nowadays most of my main drive is used anyway.
The specific programs I often use for making drive images are: dd, lzop, sometimes ssh, sometimes netcat, sometimes split and smb/cifs mount.
gzip is a bit too slow on most hardware I have or encounter. I can usually get at least 30MB/sec with lzop, sometimes even through SSH connections (note: using RC4 tends to be faster than the default though it's not as secure).
Yep. I use a perl script on my Windows machine. For me it's not important that every last bit of unused space is zeroed - it's just to make things compress better.
But before you do such things it may be a good idea to do a full file system check, to fix your drive. If you think your file system is not in a good state, don't do it at all, just make the image.
For these sort of things I prefer to use lzop instead of gzip for compression even though it doesn't compress as well it is much faster.
But given a fast enough CPU (or a slow enough drive;) ), gzip should be able to compress at the drive's max throughput.
I'd say it again because so many people don't seem to get it:
1) Pick good defaults 2) Make easy/common things easy and quick 3) Make hard/advanced things possible
1) is hard to get right. But it is really important - you want most normal people to use basically the same UI and find it's ok.
The Gnome team seem to either not understand 2) and 3), or are unable to achieve 2) or 3). This is not a good sign for them.
It is a bit like fax compression:
You pick a good starting default dictionary, and you arrange things so that the most common stuff requires the fewest symbols/bits.
(each bit is a decision).
But you also allow _most_ other things to be represented, but still require fewer steps than specifying everything bit by bit.
Only very few things should require more bits to do - e.g. choosing "hacker mode" and then specifying everything bit by bit.
At least in the printing case mentioned, Gnome apparently requires you to jump straight to "hacker mode" if you want to do something more than "simple".
People have limited time, knowledge and resources, that is why compression is required.
If the Gnome people are actually removing functionality from their stuff then they're doing things wrong.
They should: 1) Pick good defaults. 2) Make easy things easy. 3) Make hard/advanced things possible[1].
So far they allegedly are removing 3).
I used to recall colleagues having to go to the CLI to configure even basic GUI stuff in gnome. When I said the defaults sucked they agreed but they also told me "but you can configure everything". That misses the point of "1) Pick good defaults". If you pick a good enough default most people won't bother to change it (or ask others to change it for them), it won't annoy them that much.
And that makes support calls easier since most people will be having their UI in the same state - only the advanced users will change things and they may not need support.
[1] Sure you can probably configure stuff by editing a text file somewhere, but if such cases are _increasing_ then gnome might as well shutdown as a project. People who like that should be using one of those "minimalist GUIs".
Sure MySQL has supported transactions for years, but tell me what is the best way to load 20GB of data into innodb tables?
I tried a fair number of recommended ways, but still got stuck with the disk writing at 2MB/sec even at the _start_ (which is not even the actual rate data is flowing into the DB). Not even got to the "slow down due to big index" part. Yes I did set the sync thing to 0 (sync max of once a second, instead of on every commit).
I was seriously considering putting the innodb log files on a ramdisk;).
I didn't wrap everything in one huge transaction, because according to the docs if "stuff" happens it takes 30 times longer to roll back everything. So if you are 3 hours into inserting something, it'll take 90 hours to rollback... There are workarounds, but I've looked at the workaround, and "Thanks but no thanks".
In the end I loaded the data into MyISAM tables instead (disable keys then enable keys).
With MyISAM if you want to add an index, MySQL makes an entire copy of the table. The manual doesn't say that Innodb is different, but I haven't tested it.
Fortunately it's what I'd call a small site.
Imagine if you had terabytes of data and to break the news to your boss, "yes sir, we need to double our storage capacity", "Yes sir, our data is only growing at 200GB a year, but we still need to add 2 more TBs _now_ in order to add an index".
By the way by default MyISAM tables are limited to 4GB... Talk about thinking ahead.
Whereas with innodb, say you have a huge ibdata1 file, how do you shrink it once you're done with some tables? While you can have tables in ibd files, it's still something new - the previous reported bugs sure don't give me a lot of confidence in it.
I use postgresql at home. One day i should go load up 20GB of data into it and see how long it takes.
"grow brains in vats and hook them up to the computers (roughly)."
We already have working human brains, and we are getting better at hooking them up to computers.
If they are talking about growing brains that they can easily "dehumanize" so they can be enslaved, I think that's not a good direction to head in. Even though people keep saying "tech XYZ" is inevitable, I believe there are many paths to different directions/destinations that one can take, and we would do best to avoid certain paths where possible.
Has google recently done something similar to Cyc in the machine translation area? They got lots of computers to go through lots of data to automatically find connections.
I've no experience in AI, so I may be wrong about this:it seems most of these AI/expert systems go by connections and contexts/groups/sets. Except I guess Cyc - which apparently has support for metaphors?
If that is true, I don't see how those sort of architectures would make it easy to derive new metaphors and analogies. It is easy for those systems to memorize existing ones and their correct answers, but that's about Google or even Teoma's level. They cannot derive new links easily on their own.
I was thinking to get to the next level (there are other levels I thought of one day but I've forgotten) would be having support for something that's closer to a "vector" than a "connection/link" or list of groups and patterns.
That would make it easier for an AI to understand "horse is to grass" like "car is to gasoline". And when you ask the AI what else is to grass in that manner, it'll be easy for it to give you "cow, goat" etc.
Whereas if you tell an AI, "horse is to grass" like "car is to road", and then you ask the AI, what else is like that to "railway track". The AI will be able to tell you.
And then if you tell the AI "krill is to balleen whale" like "grass is to cow", even if it doesn't know what krill is or even what a balleen whale is, it will still be able to learn something about both. And when you tell the AI that a whale is something that only lives in the sea, it should be able to realize that krills are most likely to live in the sea.
Maybe there's an AI "theory of relativity" - few absolute references - mostly relative;).
Some forms of humour seem to trigger "rewarding" of a sudden rapid reinforcement/addition/usage of extraordinary linkages ("Ah yes, that can mean that too").
Also say "relations" are stored and arranged in "spaces". Due to incomplete knowledge they use a bit more "virtual space". But when suddenly there is an "insight" things can be arranged more "efficiently".
I'm talking out of my field of expertise, so I guess I could be totally wrong about it all.
Once you get your computer to talk to you _genuinely_intelligently_ will you be certain you'd understand how you did it, and how it works?
There are a few ways of getting nonhuman intelligence. But if it involves throwing stuff together and without really understanding it, I suggest you might as well go to a pet store.
Or just breed a more intelligent animal.
I suppose the advantage is you can make a 100% copy of your AI, you can't do that with a dog.
But maybe that might not be possible given some sort of AI designs.
The thing is - most of those cases the driver is talking about something fairly related to the situation he needs to concentrate on.
Not something totally unrelated.
The cop probably looking at the vehicle he's chasing, describing it, saying where it's going. I'd find that not so hard to do that myself.
He's not trying to think of whether his girlfriend's maroon skirt (gf: "Not the red one, _maroon_") will go fine with her new top, or whether what he says next will get him in trouble with her...
As for F1 drivers, they are drivers who are highly coordinated and can probably multitask and drive at highspeeds. At least the top ones should be able to practically drive around tracks in their sleep;).
Apparently when the F1 racers were made to race in go-karts years ago, Ayrton Senna apparently was driving whilst tweaking the fuel-air mix on his kart's engine at the same time.
Rally drivers might even better at these sort of situations.
That's as silly as asking why the best sculptor would want to work for a very very rich patron paying him to sculpt instead of the sculptor going off and starting his own company.
Right, and what happens to people/stuff that eat the meat with the myostatin blocker in it? It may be fairly safe for cows (which aren't going to live long anyway), but humans?
<rant> Nowadays too many countries and people are treating food as fuel.
As long as it meets a certain official grade, it's food. Even if it involves dubious stuff like feeding chicken shit and feathers to cows. This apparently was stopped recently in the US because of BSE scares - turns out many chickens get fed bits of cows, so they were afraid that bits of "chicken feed" would end up being swept up along with the shit and feathers and being eaten by cows. Cool huh? Always taking care of the public...
Ironic that so many of the US people are afraid of eating the "weird foods" in my country;). And they go eat a burger instead (apparently there was one who only ate peanut butter the entire visit).
With all of this stuff going, with "food" that gets processed, substituted etc, I'm wondering about all these extreme peanut allergies going around in the US/Canada. Is it due to media playing it up, or is it getting so common?
Maybe kids in other countries are dying in numbers of peanut allergies and it doesn't get reported much?
If not, perhaps it's all the weird stuff that goes on with commercial peanut butter. e.g. extract the peanut oil, replace it with "cheapest veg oil of the day", add various emulsifiers "cheapest of the day", blend, and serve to young child.
I won't be surprised if it turns out that sort of thing confuses some immune systems.
Recently they found that taking a gene from one species and inserting it into another didn't cause the target to necessarily produce the exact same stuff - something similar is made, but it was a bit more toxic than what the original species produced...
Sure GM has uses, but I think one should wait for the amount of greed involved to reduce a bit first before one starts opening the floodgates.
Just block it till most of the greedy people who are just in it for the money go look for something else to do.
I say we can wait. Most (if not all) of the proposed GM "solutions", are just solutions looking for a problem. Many of those problems either can be solved easily and cheaply by other means, or GM actually doesn't deal with the problem (mass starvation is mostly due to bad politics, rather than insufficient food production). </rant>
BTW, germlines can "change effectively" due to environment influences.
Do a google search on: epigenetics dutch winter agouti
"The finding is remarkable because it suggests that a pregnant mother's diet can affect her health in such a way that not only her children but her grandchildren (and possibly great-grandchildren, etc.) inherit the same health problems."
I would find it surprising that no species would allow their germlines to change due to life experiences.
While you wouldn't allow major changes, it might be good to allow "tweaks". Perhaps even tweaks that make future major changes more likely.
On a vaguely related note: I'm not sure if I recall correctly, but apparently when some slugs are trained, and they get eaten by another slug, some of the eaten slug's memories seem to get transferred.
People should stop focusing on the _last_ things keeping them from becoming a police state, and start focusing on the _first_ things.
Starting with very dubious electronic voting machines and who you vote as leaders.
Once you get too many of the wrong people in power, they can change all that stuff very quickly. Look at the Patriot Act, and all the recent crappy laws with dangerous long term consequences.
If citizens keep sticking their heads in the sand (or erm troughs of junk food?), the leaders can basically do what they want with impunity.
Even if you don't allow tracking now, Mr Evil Dictator can always turn it back on, once he's in power.
So the main thing is to never allow Mr Evil Dictator a chance to get power in the first place.
It is quite scary and sad that history has proven that many people will actually be willing to listen to some evil person and give him the power. These people will willingly kill anybody - even their relatives or parents/children just because "it's their job" or the supreme leader told them to.
Slashdot Mirror
So it's because Lisp is a bit more "consistent/seamless", so it's easier to have code modify itself in meaningful ways?
;). Without glossing over the details and new special cases e.g. protocol changes for human specific requirements.
:).
;).
How easy is it typically to modify such a program to do something different?
Say you have a messaging server for inter module communications, and you want to convert it to a chat server for inter human communications
Would it be easy for an average lisp programmer to understand what was done, if the conversion was done with a bunch of macros? Or would that be the equivalent of getting an average perl programmer to understand one of those perl one-liners?
OK, if you were asked to write a DHCP _server_ in Lisp that stores leases and logic in an RDBMS, what would be the steps? Which Lisp site would you visit first, which modules to consider etc? Or would you just write everything yourself in less than 30 minutes
Being not so smart and lazy, I prefer using code written by smarter or more hardworking people, so I currently use perl and CPAN
But can't you write code that writes code in Perl and many other scripting languages too?
:).
There are also potential security issues whenever you start treating data as code, or let external data significantly influence code generation.
I have looked but have not found any guidelines for doing such things securely in lisp.
On a related note, I have crashed a forth webserver on my first try
Just curious: how many good mathematicians use the Lisp style notation to work with their equations?
I know there's at least one who likes Lisp and has used it for info theory stuff.
But if the Lisp notation is so good wouldn't it be better for mathematicians? Or is there a reason why it doesn't work well in that environment?
Something like:
#!/usr/bin/perl -w
use strict;
my $fh;
my $size=1000000;
my $buf="\000" x $size;
my $fname='00zeroes.000';
open($fh,"> $fname") or die "cannot open file: $!";
while(1) {
last unless (print $fh $buf);
}
close ($fh);
unlink("$fname") or die "\a\a\aUnable to unlink $fname: $!";
print "\a\aDone\n!";
---
You probably might want to do a sync. But I'm not too bothered.
Instead of a while(1) you could put a limit.
Those same attitudes work well when you want high quality stuff.
My friend who used to work for a Japanese car company was sent to Japan for training for a few months.
Whilst he was there he found that it was common for staff to be rearranging and tidying their desks etc at 9pm just so everything is nicely in place for efficient work the next day.
In the normal office area there were circles on the floor to mark where wastebaskets should be. Lots of little details like that.
I doubt the staff of the local office here have quite that commitment to work, attention to detail, discipline etc.
Which is probably why the local build quality is unlikely to be as good as that in Japan. Though I heard it is still better than other locally available brands.
It'll be unlikely you'd save that much space unless most of your files are all one or two blocks in size.
Who said anything about zeroing stuff for security?
"Gadzooks, man, what are you using, dd | gzip? Get yourself a program that just copies the used sectors into your image file!"
knoppix (using the noswap boot option). I only do the zeroing sometimes. It's no big deal to me - nowadays most of my main drive is used anyway.
The specific programs I often use for making drive images are: dd, lzop, sometimes ssh, sometimes netcat, sometimes split and smb/cifs mount.
gzip is a bit too slow on most hardware I have or encounter. I can usually get at least 30MB/sec with lzop, sometimes even through SSH connections (note: using RC4 tends to be faster than the default though it's not as secure).
Yep. I use a perl script on my Windows machine. For me it's not important that every last bit of unused space is zeroed - it's just to make things compress better.
;) ), gzip should be able to compress at the drive's max throughput.
But before you do such things it may be a good idea to do a full file system check, to fix your drive. If you think your file system is not in a good state, don't do it at all, just make the image.
For these sort of things I prefer to use lzop instead of gzip for compression even though it doesn't compress as well it is much faster.
But given a fast enough CPU (or a slow enough drive
I do zero out unused blocks on some of my drives from time to time.
This is especially when I am about to make full image backups of my drives. If you zero out the unused regions the drive image compresses much better.
Otherwise you end up using space to backup up deleted data. In some cases you do want to do that, but not always.
I'd say it again because so many people don't seem to get it:
1) Pick good defaults
2) Make easy/common things easy and quick
3) Make hard/advanced things possible
1) is hard to get right. But it is really important - you want most normal people to use basically the same UI and find it's ok.
The Gnome team seem to either not understand 2) and 3), or are unable to achieve 2) or 3). This is not a good sign for them.
It is a bit like fax compression:
You pick a good starting default dictionary, and you arrange things so that the most common stuff requires the fewest symbols/bits.
(each bit is a decision).
But you also allow _most_ other things to be represented, but still require fewer steps than specifying everything bit by bit.
Only very few things should require more bits to do - e.g. choosing "hacker mode" and then specifying everything bit by bit.
At least in the printing case mentioned, Gnome apparently requires you to jump straight to "hacker mode" if you want to do something more than "simple".
People have limited time, knowledge and resources, that is why compression is required.
Well I saw that phrase here:s /2005-December/000395.html
http://lists.osdl.org/pipermail/desktop_architect
If the Gnome people are actually removing functionality from their stuff then they're doing things wrong.
They should:
1) Pick good defaults.
2) Make easy things easy.
3) Make hard/advanced things possible[1].
So far they allegedly are removing 3).
I used to recall colleagues having to go to the CLI to configure even basic GUI stuff in gnome. When I said the defaults sucked they agreed but they also told me "but you can configure everything". That misses the point of "1) Pick good defaults". If you pick a good enough default most people won't bother to change it (or ask others to change it for them), it won't annoy them that much.
And that makes support calls easier since most people will be having their UI in the same state - only the advanced users will change things and they may not need support.
[1] Sure you can probably configure stuff by editing a text file somewhere, but if such cases are _increasing_ then gnome might as well shutdown as a project. People who like that should be using one of those "minimalist GUIs".
Sure MySQL has supported transactions for years, but tell me what is the best way to load 20GB of data into innodb tables?
;).
I tried a fair number of recommended ways, but still got stuck with the disk writing at 2MB/sec even at the _start_ (which is not even the actual rate data is flowing into the DB). Not even got to the "slow down due to big index" part. Yes I did set the sync thing to 0 (sync max of once a second, instead of on every commit).
I was seriously considering putting the innodb log files on a ramdisk
I didn't wrap everything in one huge transaction, because according to the docs if "stuff" happens it takes 30 times longer to roll back everything. So if you are 3 hours into inserting something, it'll take 90 hours to rollback... There are workarounds, but I've looked at the workaround, and "Thanks but no thanks".
In the end I loaded the data into MyISAM tables instead (disable keys then enable keys).
With MyISAM if you want to add an index, MySQL makes an entire copy of the table. The manual doesn't say that Innodb is different, but I haven't tested it.
Fortunately it's what I'd call a small site.
Imagine if you had terabytes of data and to break the news to your boss, "yes sir, we need to double our storage capacity", "Yes sir, our data is only growing at 200GB a year, but we still need to add 2 more TBs _now_ in order to add an index".
By the way by default MyISAM tables are limited to 4GB... Talk about thinking ahead.
Whereas with innodb, say you have a huge ibdata1 file, how do you shrink it once you're done with some tables? While you can have tables in ibd files, it's still something new - the previous reported bugs sure don't give me a lot of confidence in it.
I use postgresql at home. One day i should go load up 20GB of data into it and see how long it takes.
What are you running that locks you into Apache on Windows?
Why can't you use IIS? Or switch to Apache on Linux/*BSD?
The current IIS so far has had a better security track record than Apache.
If I can help it, I wouldn't use windows for servers though.
"grow brains in vats and hook them up to the computers (roughly)."
;).
We already have working human brains, and we are getting better at hooking them up to computers.
If they are talking about growing brains that they can easily "dehumanize" so they can be enslaved, I think that's not a good direction to head in. Even though people keep saying "tech XYZ" is inevitable, I believe there are many paths to different directions/destinations that one can take, and we would do best to avoid certain paths where possible.
Has google recently done something similar to Cyc in the machine translation area? They got lots of computers to go through lots of data to automatically find connections.
I've no experience in AI, so I may be wrong about this:it seems most of these AI/expert systems go by connections and contexts/groups/sets. Except I guess Cyc - which apparently has support for metaphors?
If that is true, I don't see how those sort of architectures would make it easy to derive new metaphors and analogies. It is easy for those systems to memorize existing ones and their correct answers, but that's about Google or even Teoma's level. They cannot derive new links easily on their own.
I was thinking to get to the next level (there are other levels I thought of one day but I've forgotten) would be having support for something that's closer to a "vector" than a "connection/link" or list of groups and patterns.
That would make it easier for an AI to understand "horse is to grass" like "car is to gasoline". And when you ask the AI what else is to grass in that manner, it'll be easy for it to give you "cow, goat" etc.
Whereas if you tell an AI, "horse is to grass" like "car is to road", and then you ask the AI, what else is like that to "railway track". The AI will be able to tell you.
And then if you tell the AI "krill is to balleen whale" like "grass is to cow", even if it doesn't know what krill is or even what a balleen whale is, it will still be able to learn something about both. And when you tell the AI that a whale is something that only lives in the sea, it should be able to realize that krills are most likely to live in the sea.
Maybe there's an AI "theory of relativity" - few absolute references - mostly relative
Some forms of humour seem to trigger "rewarding" of a sudden rapid reinforcement/addition/usage of extraordinary linkages ("Ah yes, that can mean that too").
Also say "relations" are stored and arranged in "spaces". Due to incomplete knowledge they use a bit more "virtual space". But when suddenly there is an "insight" things can be arranged more "efficiently".
I'm talking out of my field of expertise, so I guess I could be totally wrong about it all.
Just curious about this:
Once you get your computer to talk to you _genuinely_intelligently_ will you be certain you'd understand how you did it, and how it works?
There are a few ways of getting nonhuman intelligence. But if it involves throwing stuff together and without really understanding it, I suggest you might as well go to a pet store.
Or just breed a more intelligent animal.
I suppose the advantage is you can make a 100% copy of your AI, you can't do that with a dog.
But maybe that might not be possible given some sort of AI designs.
The thing is - most of those cases the driver is talking about something fairly related to the situation he needs to concentrate on.
;).
Not something totally unrelated.
The cop probably looking at the vehicle he's chasing, describing it, saying where it's going. I'd find that not so hard to do that myself.
He's not trying to think of whether his girlfriend's maroon skirt (gf: "Not the red one, _maroon_") will go fine with her new top, or whether what he says next will get him in trouble with her...
As for F1 drivers, they are drivers who are highly coordinated and can probably multitask and drive at highspeeds. At least the top ones should be able to practically drive around tracks in their sleep
Apparently when the F1 racers were made to race in go-karts years ago, Ayrton Senna apparently was driving whilst tweaking the fuel-air mix on his kart's engine at the same time.
Rally drivers might even better at these sort of situations.
That's as silly as asking why the best sculptor would want to work for a very very rich patron paying him to sculpt instead of the sculptor going off and starting his own company.
Right, and what happens to people/stuff that eat the meat with the myostatin blocker in it? It may be fairly safe for cows (which aren't going to live long anyway), but humans?
;). And they go eat a burger instead (apparently there was one who only ate peanut butter the entire visit).
<rant>
Nowadays too many countries and people are treating food as fuel.
As long as it meets a certain official grade, it's food. Even if it involves dubious stuff like feeding chicken shit and feathers to cows. This apparently was stopped recently in the US because of BSE scares - turns out many chickens get fed bits of cows, so they were afraid that bits of "chicken feed" would end up being swept up along with the shit and feathers and being eaten by cows. Cool huh? Always taking care of the public...
Ironic that so many of the US people are afraid of eating the "weird foods" in my country
With all of this stuff going, with "food" that gets processed, substituted etc, I'm wondering about all these extreme peanut allergies going around in the US/Canada. Is it due to media playing it up, or is it getting so common?
Maybe kids in other countries are dying in numbers of peanut allergies and it doesn't get reported much?
If not, perhaps it's all the weird stuff that goes on with commercial peanut butter. e.g. extract the peanut oil, replace it with "cheapest veg oil of the day", add various emulsifiers "cheapest of the day", blend, and serve to young child.
I won't be surprised if it turns out that sort of thing confuses some immune systems.
Recently they found that taking a gene from one species and inserting it into another didn't cause the target to necessarily produce the exact same stuff - something similar is made, but it was a bit more toxic than what the original species produced...
Sure GM has uses, but I think one should wait for the amount of greed involved to reduce a bit first before one starts opening the floodgates.
Just block it till most of the greedy people who are just in it for the money go look for something else to do.
I say we can wait. Most (if not all) of the proposed GM "solutions", are just solutions looking for a problem. Many of those problems either can be solved easily and cheaply by other means, or GM actually doesn't deal with the problem (mass starvation is mostly due to bad politics, rather than insufficient food production).
</rant>
Yeah, but the article said the injected agent also made the "mighty" mice without myostatin grow 24% larger muscles.
So it's likely that the injected agent isn't just affecting the myostatin stuff.
Ah, but they are HP servers and not Compaq servers ;).
Seriously though, probably depends on which range of dates were involved.
Heh, I sure hope there's sound support.
;).
Might as well piss off the music and movie industry as well while they are at it
BTW, germlines can "change effectively" due to environment influences.
Do a google search on: epigenetics dutch winter agouti
"The finding is remarkable because it suggests that a pregnant mother's diet can affect her health in such a way that not only her children but her grandchildren (and possibly great-grandchildren, etc.) inherit the same health problems."
I would find it surprising that no species would allow their germlines to change due to life experiences.
While you wouldn't allow major changes, it might be good to allow "tweaks". Perhaps even tweaks that make future major changes more likely.
On a vaguely related note: I'm not sure if I recall correctly, but apparently when some slugs are trained, and they get eaten by another slug, some of the eaten slug's memories seem to get transferred.
Allegedly Winston Churchill said: "This is the sort of bloody nonsense up with which I will not put."
;).
Of course it may be someone else who said that..
English is funny: Think how the word "one" should be pronounced.
The myriad "ough" pronounciations...
And in the US it's common to drive on the parkway and park on the driveway
People should stop focusing on the _last_ things keeping them from becoming a police state, and start focusing on the _first_ things.
Starting with very dubious electronic voting machines and who you vote as leaders.
Once you get too many of the wrong people in power, they can change all that stuff very quickly. Look at the Patriot Act, and all the recent crappy laws with dangerous long term consequences.
If citizens keep sticking their heads in the sand (or erm troughs of junk food?), the leaders can basically do what they want with impunity.
Even if you don't allow tracking now, Mr Evil Dictator can always turn it back on, once he's in power.
So the main thing is to never allow Mr Evil Dictator a chance to get power in the first place.
It is quite scary and sad that history has proven that many people will actually be willing to listen to some evil person and give him the power. These people will willingly kill anybody - even their relatives or parents/children just because "it's their job" or the supreme leader told them to.