> He's long espoused the virtues of free trade. > He says that he supported Nafta and that for 12 > years he's subscribed to The Economist, a > hymnal in the free trade church. But now he's > questioning core beliefs.
now, i'm not saying that he was wrong and now he's right, or viceversa, but i want to say this:
next time around, think more carefully before putting your "faith" in something...
I think you are 100% right. I would like to add something though.
Every time i read the weekly discussion about how ready is linux for the desktop i can't help wondering why WINE almost never comes up in the discussion.
It is apparent to me that other than a nice and easy desktop (which is almost there), and seamless installation and hardware management (almost there as well) in the end what every OS has to do is TO RUN SOME APPS.
Since as you pointed out very well the user won't change the application he is using unless he's constrained to do so, then in order to take over the desktop, linux simply HAS TO RUN WINDOWS APPS !!
(Of course then they will be ported natively later, once linux has gained its place on the desktop)
Another thing, i can almost already read some reply along the lines: "if you use win32 programs you are better off using windows", now, i think these people fail to realize how important would be getting linux on the desktop market before DRM locks down everybody for good. Some chances simply don't come twice.
Well i have to say that i'd rather use proprietary software that runs on an open OS than the other way around.
While i agree that in theory, if the source of an app is open you can recompile it to run on another OS, in reality you need to count on compilers, librares, system calls, API, services...
This means that if some of the above is proprietary it can change, cease to comply to standards or simply cease to exist and you can't do a damn thing about that.
Conversely if the platform is open then its better for developers, since they will always have access to the underlying code, and ultimately it's better for the user since he has a choice among proprietary and open applications.
The number of resources, particularly human resources, that an organization is able to use is necessarily bounded, at some point.
Sometimes too many cooks spoil the broth, and this is especially true in design processes.
So if you have company A and B, each with 500 employees, sometimes it's not true that an imaginary company C resulting from the merger of A and B creates a better product because there are more amployees working on the same problem. The reason for a merger is often that by eliminating competition you could easily enforce higher prices on the market WITHOUT having to improve your production.
So, in the end, there IS a waste of resources but is really in the interest of evolution of the product, because it allows to explore possibilities using resources that you couldn't use in many other ways.
I would say that the problem with the capitalistic way of production is not really the waste of resources, but the inefficiency that comes from things like long lasting patents, closed source, industrial secrets, because it FORCES designers to reinvent the same wheel many times, instead of leaving them the CHOICE of reinventing the wheel AFTER having a good look at the existing wheels.
But then, the film does take place in the future. Is Zalewski surprised to see unpatched SSH servers running in the year AD 2199? "It's not that uncommon for people to run the old distribution," he says.
comment: the film does take place in 2199 but inside the matrix the year must be close to 1999 so i guess it's quite ok to see the unpatched SSH around...
i'd like to ask a question that has not, to my knowledge, been asked yet:
why, excactly, the communication was lost so abruptly at 7:59 AM ??
was it an electrical type of fault, perhaps some communication subsystem started to burn ? perhaps the antenna ? if so, how could a termal failure propagate itself through the communication system ?
or, rather, the shuttle itself began rotating so everything started to burn, and the communication was lost after that ?
given that shuttle attitude was "almost" normal, at the moment in which the communication was lost, it seems more an electrical type of failure.
a related question is if it is known (by the many videos) at what time exactly the shuttle was broken in two pieces...
giampy Premature Optimization is The Root of All Evil - D.Knuth
"The hard part is actually getting enough people to understand the issues and join the boycott. It's not simple..."
yeah, i agree it's not simple but it's not impossible either
I think we just need to state clearly three or four FACTS, (like for one, the price of a CD and how much actually goes to the artist).
Once a movement is started, people will join,
and sooner or later the media have to pay some attention even if they don't want to. And if the message is based on facts and if it is simple enough, then it can hardly be folded or devised.
If my opinion counts, i think
it takes some time to RTFM and
learn how to solve problems with a
completely new operating system...
Furthermore, that time is often very
fragmented since one typically has to
learn at home, and be SURE he is able
to work effectively on the new system
before he can switch at work !!
That being said, i have red hat 8 at
home and i plan to switch at work
within the next few months.
I agree almost completely, complex systems are really hard BUT,
modularity, (layers are modules), has proven to be an extremely
effective way to manage complexity, and examples are all around us.
Of course there's no free lunch, and a certain decrease in performance
is the first price you pay when you _constrain_ your system to be modular.
( this is exaclty what happens in the 2-dimensional array example)
Anyway i think that what _really_ joel wants to tell in the article
is that when modularity works 99% of the times, people don't even
care to learn what happens inside the modules, so when something
goes wrong at a lower level they are completely lost.
( of course this _malfunction_ could be caused by bad
interface desing, lack of specifications, or can simply
manifest itself as an enourmous performance drop...)
Of course you can hire and mantain 10 experts for each darn
layer, but this may be feasible for an airline corporation
not for a small start-up company...
so i think the bottom line is that, even if modularity works
99% of the time, it is safe to hire people that know where to
look for if something does not seem to be working fine at
a lower level...
In the university where work as a research assistant, the majority of PhD students are from China and India. Chinese students invariably tend to be the best ones. It seems like by the time they come here they have already done A LOT of practical hitech research in their universities. Because of this, (and because most of them don't mind being paid 2k$/month or less) a lot of departments actually prefer to hire Chinese students for tech projects.
Many of them will go back to China once their studies are over. It is also worth considering that for each student that makes it to the US, maybe 100 will stay in China. And, as the article says: "Hundreds of universities with strong tech departments have been created."
Is this enough to say that China is headed towards becoming the place where hi tech is conceived and grown ? I think that, IF these government-funded policies will go on, it will be just a matter of time (maybe a couple of generations, maybe less...).
And btw, it seems to me that they are very inclined toward the sharing of knowledge and information,...and files:-)
Slashdot Mirror
I think in those times the relationship
with time was much different. Much less hectic.
The rhythms of work and life were much more
subjected to things like daylight, seasons and
stuff like that.
from the article:
...
> He's long espoused the virtues of free trade.
> He says that he supported Nafta and that for 12
> years he's subscribed to The Economist, a
> hymnal in the free trade church. But now he's
> questioning core beliefs.
now, i'm not saying that he was wrong and
now he's right, or viceversa, but i want to say this:
next time around, think more carefully before
putting your "faith" in something
giampy
> $400 million drop in unearned revenue
:))
call me old fashioned but shouldn't any
revenue be EARNED ???
I think you are 100% right. I would like to add something though.
Every time i read the weekly discussion about how ready is linux for the desktop i can't help wondering why WINE almost never comes up in the discussion.
It is apparent to me that other than a nice and easy desktop (which is almost there), and seamless installation and hardware management (almost there as well) in the end what every OS has to do is TO RUN SOME APPS.
Since as you pointed out very well the user won't change the application he is using unless he's constrained to do so, then in order to take over the desktop, linux simply HAS TO RUN WINDOWS APPS !!
(Of course then they will be ported natively later, once linux has gained its place on the desktop)
Another thing, i can almost already read some reply along the lines: "if you use win32 programs you are better off using windows", now, i think these people fail to realize how important would be getting linux on the desktop market before DRM locks down everybody for good. Some chances simply don't come twice.
Well i have to say that i'd rather use proprietary software that runs on an open OS than the other way around.
...
While i agree that in theory, if the source of an app is open you can recompile it to run on another OS, in reality you need to count on compilers, librares, system calls, API, services
This means that if some of the above is proprietary it can change, cease to comply to standards or simply cease to exist and you can't do a damn thing about that.
Conversely if the platform is open then its better for developers, since they will always have access to the underlying code, and ultimately it's better for the user since he has a choice among proprietary and open applications.
The number of resources, particularly human resources,
that an organization is able to use is necessarily bounded,
at some point.
Sometimes too many cooks spoil the broth,
and this is especially true in design processes.
So if you have company A and B, each with 500 employees,
sometimes it's not true that an imaginary company C resulting
from the merger of A and B creates a better product because
there are more amployees working on the same problem.
The reason for a merger is often that by eliminating
competition you could easily enforce higher prices
on the market WITHOUT having to improve your production.
So, in the end, there IS a waste of resources but
is really in the interest of evolution of the product,
because it allows to explore possibilities using
resources that you couldn't use in many other ways.
I would say that the problem with the capitalistic
way of production is not really the waste of resources,
but the inefficiency that comes from things like
long lasting patents, closed source, industrial secrets,
because it FORCES designers to reinvent the same wheel
many times, instead of leaving them the CHOICE of
reinventing the wheel AFTER having a good look at
the existing wheels.
that was funny actually ...
quote from the article:
...
But then, the film does take place in the future. Is Zalewski surprised to see unpatched SSH servers running in the year AD 2199? "It's not that uncommon for people to run the old distribution," he says.
comment:
the film does take place in 2199 but inside the matrix the year must be close to 1999 so i guess it's quite ok to see the unpatched SSH around
choice, the problem is choice !!!
This reminds me of how the romans used to test their bridges: they put the designer under the bridge while marching over it with the entire legion.
Of course, a bridge i a MUCH simpler thing than a program, but, hey, 2000 years, all the bridges are still there !!!
It is strange to me that no one
...
focused his attention on why exactly
the communication was lost at 7:59 am
in the middle of a "Roger and b..."
sure there was some heating and drag going on
in the left wing but nothing really dramatic
had happened to the shuttle yet
anyone with some more knowledge
would like to elaborate that ??
giampy
i'd like to ask a question that has not,
...
to my knowledge, been asked yet:
why, excactly, the communication was lost
so abruptly at 7:59 AM ??
was it an electrical type of fault,
perhaps some communication subsystem
started to burn ? perhaps the antenna ?
if so, how could a termal failure
propagate itself through the
communication system ?
or, rather, the shuttle itself began
rotating so everything started to burn,
and the communication was lost after that ?
given that shuttle attitude was "almost" normal,
at the moment in which the communication was lost,
it seems more an electrical type of failure.
a related question is if it is known (by the
many videos) at what time exactly the shuttle
was broken in two pieces
giampy
Premature Optimization is The Root of All Evil - D.Knuth
"The hard part is actually getting enough people to understand the issues and join the boycott. It's not simple ..."
yeah, i agree it's not simple but it's not impossible either
I think we just need to state clearly three or four FACTS, (like for one, the price of a CD and how much actually goes to the artist). Once a movement is started, people will join, and sooner or later the media have to pay some attention even if they don't want to. And if the message is based on facts and if it is simple enough, then it can hardly be folded or devised.
If my opinion counts, i think it takes some time to RTFM and learn how to solve problems with a completely new operating system
Furthermore, that time is often very fragmented since one typically has to learn at home, and be SURE he is able to work effectively on the new system before he can switch at work !!
That being said, i have red hat 8 at home and i plan to switch at work within the next few months.
I agree almost completely, complex systems are really hard BUT, modularity, (layers are modules), has proven to be an extremely effective way to manage complexity, and examples are all around us.
Of course there's no free lunch, and a certain decrease in performance is the first price you pay when you _constrain_ your system to be modular. ( this is exaclty what happens in the 2-dimensional array example)
Anyway i think that what _really_ joel wants to tell in the article is that when modularity works 99% of the times, people don't even care to learn what happens inside the modules, so when something goes wrong at a lower level they are completely lost. ( of course this _malfunction_ could be caused by bad interface desing, lack of specifications, or can simply manifest itself as an enourmous performance drop
Of course you can hire and mantain 10 experts for each darn layer, but this may be feasible for an airline corporation not for a small start-up company
so i think the bottom line is that, even if modularity works 99% of the time, it is safe to hire people that know where to look for if something does not seem to be working fine at a lower level
In the university where work as a research assistant, the majority of PhD students are from China and India. Chinese students invariably tend to be the best ones. It seems like by the time they come here they have already done A LOT of practical hitech research in their universities.
Because of this, (and because most of them don't mind being paid 2k$/month or less) a lot of departments actually prefer to hire Chinese students for tech projects.
Many of them will go back to China once their studies are over.
It is also worth considering that for each student that makes it to the US, maybe 100 will stay in China.
And, as the article says: "Hundreds of universities with strong tech departments have been created."
Is this enough to say that China is headed towards becoming the place where hi tech is conceived and grown ?
I think that, IF these government-funded policies will go on, it will be just a matter of time (maybe a couple of generations, maybe less
And btw, it seems to me that they are very inclined toward the sharing of knowledge and information,
giampy