The billions of $$$ in venture capital for random web stuff has dropped off, so the excitement has waned a little.
In order to move to the next level, an order of magnitude more bandwidth is needed. Canberra (Australia's capital) is getting fibre optic to the curb, and it is expected that local businesses will offer things like video on demand (the holy grail?). It will come, things will move forward but it's going to take a bit more time. (Patience folks!).
I really have to disagree because I know that there are good languages out there - languages that don't have any major flaws. Languages that support closures, and OO and functional programming and do so with a simple syntax. Languages that CAN be a jack of all trades and solve 98% of the programming tasks are out there. Why complicate things with this "choose a language for a task" when there are languages that ARE good at all tasks. Perl is not it. Neither is C or C++. Java is closer but not it either. TCL is obviously not it.
Something like a combination of Scheme with Common Lisp object system IS one such a language. It's a good choice for everything from text processing, web pages, GUI, artificial intelligence, math, interactive, anything. Is Scheme the one true language? Not exactly, there may be better ones out there - ML or Haskell perhaps I'm not familiar with them. But why mess around with an abortion like Perl? Seriously folks, there's not a damned thing to be learned from this monster. I don't believe in all this relativism. Computer language research is mature enough that we can make some absolute stamements here!
Harnessing.1% of the ocean's energy sounds like an impossible task. Just as impossible as harnessing.1% of the Sun's energy hitting the earth or.1% of the wind. How many wave stations, solar panels, windmills would that take?
.5 MW is not very much power in the scheme of things, and I'd be interested to know how much it cost to build. I'd also like to know how much of the coastline they fouled up with their generators. And while renewable energy is good in theory I'd like to know how much energy was used manufacturing the equipment. One often finds the amount used is getting up there with the amount it will generate over many years. Is this just a solution for 400 remote homes?
I think it would be a big mistake to lump tech and internet stocks into the one basket. Budding.com businesses have discovered that people like their web sites free. OTOH technology is still high growth and hot. Look at the growth rates of big technology companies like computer, chip and software companies and they way outstrip their low-tech equivilent. Technology is not dead, and neither is the nasdaq.
I'm using a Data Hand keyboard. I would recommend it, because it really does exercise your fingers in various directions which gets rid of RSI. It's quite easy to learn, unlike chord keyboards, and the design overall is quite practical. Disadvantages are price, that it's fairly bulky and that the mouse is operated with the keys (which isn't that bad actually, but one does find oneself reaching for the mouse as well occasionally.) Would probably rate it best overall for practicality and anti-rsi ability, but not perfect.
I don't know what all this nonsense about web scripting languages is. Any "safe" programming language that you would use for programming in the large is likely to be a good choice for web programming. You wouldn't write general purpose code in PHP would you? Then don't use it for web pages. You would use Scheme/Smalltalk/Eiffel/Lisp for general purpose programming? Then why not use it for web scripting too. For my money Java is the pick of the 4 reviewed because it's a semi-decent programming language, but no need to use it just because everyone else is.
Well you'd have to wear body suits and bumping into each other would be simulated. Which is to be expected in a virtual world. That way you can't tell who is real and who isn't.
Re:Similar to FreeBSD jail()
on
User Mode Linux
·
· Score: 1
Do they make an export version of FreeBSD with a
gaol() system call?
All the "quality assurance" systems I've seen
implemented in companies for software development
have been useless at best, and a total lie at
worst. Ultimately the quality comes down to
the developers and their dedication to writing
automated test cases and documentation. You can't
measure that, but natural selection assures that
good software will prevail.
Why not wait for the courts to make a final ruling. The defence have got quite a strong case
to say that there is no copyright infringement/piracy occuring. Read the defence brief and wait for the courts to make a final ruling before making accusations.
Also, RMS's use of the phrase "information wants to be free", DOES have a moral foundation. Not one that you have to agree with, but a soundly thought out one none-the-less.
His reasoning is that since Information can be free, (and wants to, in the water sense), it is immoral for it not to be free. Reason being that it deprives millions of people of the use of it which is a larger moral consideration than the one person who may or may not be deprived of some royalties.
It's the same thinking that causes intellectual property to expire - that the good of society is an overriding moral principle.
So "information wants to be free", is a literary device - sure, but underlying it is an important point: that information can be free, that people like it to be free, and there is a perfectly defensible moral stance that it should be free. You might disagree in the end, but don't say it's not a valid point of view.
The whole free software movement came into being because of various beliefs, some moral, some practical regarding licencing. For you to now try and tell us to grow up and forget about licencing is ludicrous at best.
As for what "the corporate world" thinks about licencing or Linux in general, most of us really don't care that much, nor do we care about your big 5 consulting firm or its report.
If all that means free software gets a bad rap in your report so be it, but I think that reflects badly on your report rather than free software.
What's wrong with just using rpm or dpkg? Not invented here syndrome? If there's something wrong with rpm or dpkg, why don't they just contribute to these existing projects instead of reinventing the wheel yet again?
Umm, possibly. But you are really constrained by a C programming mindset. Either use a better interface (like the scheme interface for example) or borrow its good ideas. e.g. pass a linked list of arguments in your callbacks. Of course lack of garbage collection makes it a pain. Did I mention not to use C??
Umm. If KDE and Gnome were both using the same component model with the same widget set (so that themes work across both) and the same UI model, then there wouldn't be any KDE/Gnome divide then would there??? What you're basically saying is that if they were the same, with some pretend difference (i.e. the name "KDE" vs "GNOME"), then all would be well. You're virtually agreeing with me in a round about way.
But let me disagree with you. The only possible reason for different efforts is if the UIs are DIFFERENT. If one or the other actually explores new and advanced technologies for UIs and actually advances the state of the art, that is a justification for separate effort. If you are going to have common UI standards, then what pray tell is the point of duplicating effort?
I think he makes a very good point, and it goes totally against the "competition is good" crowd. Having both Gnome and KDE competing in the same space is about as cool as the US mobile phone standards he laments.
When people pick up a Linux app, they want it to conform to standards and interoperate fully in their environment. Yes sure you can run KDE and Gnome apps under each others environment, and Motif and Openlook and Athena and Tk at the same time. But they won't work together well like say MS-Office works together and the user won't get a consistent interface.
Having both KDE and Gnome is bad for Linux. It duplicates effort. It means some of your favourite apps probably won't integrate properly. It divides an already small segment of the market.
And.... it won't last. One or the other will die. It may take some years, but it will happen. I'll personally be glad when that happens for the reasons above, no matter which is the winner.
Personally I think Gnome will win because it is more commercial friendly with the licence, it is more inclusive with the language choices, more standards inclusive with Corba, and more practical without the C++ binary compatibility problems.
Either way, the sooner one side or the other gives up, the better off we'll all be. If one side has some better idea than the other, then joining forces gives the combination of both lots of good ideas, which is supposed to be the whole point of free software.
Yeah um, a garbage collected app that only runs for a few seconds will simply not collect the garbage. In other words, it's memory reclamation scheme is simply not to bother. Hardly a fair or realistic comparison. Just shows again that there are lies, damned lies and....
No, Americans are not a car-owning car-driving because of population density. Americans are car centric because they decided to. Many cities have the same or even more sparse population densities with strong viable public transport. The story I remember is that the car and oil companies in America conspired to buy out and bankrupt public transport and lobby government to build big highways. Once you build highways you attract cars and reduce the viability of public transport. That might not be so bad except that cars are an expensive, inefficient, poluting form of transport. If you work out the area devoted to cars in an American city (roads, garages, car parks etc) it is actually a lot more than that devoted to residential dwellings! Now if you removed all the roads, made the city more compact as a result and installed public transport you'd have a much nicer environment to live in.
The fact that an app you wanted to use had a lot of (Gnome) dependancies is GOOD. Get this through your head. It means that the author did not reinvent the wheel, he used existing infrastructure. It means that the app is likely to be much more standards compliant out of the box because it uses common code. It means the app has less code and is more maintainable. I know downloading some megabytes is a pain, but get with the program and just do it.
It's funny, I feel the same way about windows. I know how Unix systems work, I feel comparitively lost on a Windows system. It feels much less logical to _me_.
What I'd say is that your definition of enterprise-ready is suitable for a very large enterprise. I would say Postgresql would score very well on your general criteria, but it's not ready for enterprises with 300GB databases.
BTW who is supporting SQL 99? I'll bet no-one. Actually postgresql is starting to accumulate SQL-99 features, I'll bet Oracle hasn't moved far down that road.
I don't see the logic there. Even RMS has no
objection to selling software for money. The question is whether they can be REdistributed free. If so, then at least they're consistent.
That idea's got a lot of merit, but sadly - that's just not the UNIX way. Take the PATH for example. If every application had its own directory or whatever, your PATH variable would be a million lines long. So you'd have to think of a whole new infrastructure for executing apps.
Slashdot Mirror
The billions of $$$ in venture capital for random web stuff has dropped off, so the excitement has waned a little.
In order to move to the next level, an order of magnitude more bandwidth is needed. Canberra (Australia's capital) is getting fibre optic to the curb, and it is expected that local businesses will offer things like video on demand (the holy grail?). It will come, things will move forward but it's going to take a bit more time. (Patience folks!).
I really have to disagree because I know that there are good languages out there - languages that don't have any major flaws. Languages that support closures, and OO and functional programming and do so with a simple syntax. Languages that CAN be a jack of all trades and solve 98% of the programming tasks are out there. Why complicate things with this "choose a language for a task" when there are languages that ARE good at all tasks. Perl is not it. Neither is C or C++. Java is closer but not it either. TCL is obviously not it.
Something like a combination of Scheme with Common Lisp object system IS one such a language. It's a good choice for everything from text processing, web pages, GUI, artificial intelligence, math, interactive, anything. Is Scheme the one true language? Not exactly, there may be better ones out there - ML or Haskell perhaps I'm not familiar with them. But why mess around with an abortion like Perl? Seriously folks, there's not a damned thing to be learned from this monster. I don't believe in all this relativism. Computer language research is mature enough that we can make some absolute stamements here!
Harnessing .1% of the ocean's energy sounds like an impossible task. Just as impossible as harnessing .1% of the Sun's energy hitting the earth or .1% of the wind. How many wave stations, solar panels, windmills would that take?
.5 MW is not very much power in the scheme of things, and I'd be interested to know how much it cost to build. I'd also like to know how much of the coastline they fouled up with their generators. And while renewable energy is good in theory I'd like to know how much energy was used manufacturing the equipment. One often finds the amount used is getting up there with the amount it will generate over many years. Is this just a solution for 400 remote homes?
Many questions, many more answers required.
I think it would be a big mistake to lump tech and internet stocks into the one basket. Budding .com businesses have discovered that people like their web sites free. OTOH technology is still high growth and hot. Look at the growth rates of big technology companies like computer, chip and software companies and they way outstrip their low-tech equivilent. Technology is not dead, and neither is the nasdaq.
I'm using a Data Hand keyboard. I would recommend it, because it really does exercise your fingers in various directions which gets rid of RSI. It's quite easy to learn, unlike chord keyboards, and the design overall is quite practical. Disadvantages are price, that it's fairly bulky and that the mouse is operated with the keys (which isn't that bad actually, but one does find oneself reaching for the mouse as well occasionally.) Would probably rate it best overall for practicality and anti-rsi ability, but not perfect.
I don't know what all this nonsense about web scripting languages is. Any "safe" programming language that you would use for programming in the large is likely to be a good choice for web programming. You wouldn't write general purpose code in PHP would you? Then don't use it for web pages. You would use Scheme/Smalltalk/Eiffel/Lisp for general purpose programming? Then why not use it for web scripting too. For my money Java is the pick of the 4 reviewed because it's a semi-decent programming language, but no need to use it just because everyone else is.
Well you'd have to wear body suits and bumping into each other would be simulated. Which is to be expected in a virtual world. That way you can't tell who is real and who isn't.
Do they make an export version of FreeBSD with a
gaol() system call?
All the "quality assurance" systems I've seen
implemented in companies for software development
have been useless at best, and a total lie at
worst. Ultimately the quality comes down to
the developers and their dedication to writing
automated test cases and documentation. You can't
measure that, but natural selection assures that
good software will prevail.
Why not wait for the courts to make a final ruling. The defence have got quite a strong case
to say that there is no copyright infringement/piracy occuring. Read the defence brief and wait for the courts to make a final ruling before making accusations.
"The authors of DeCSS have not been paid for their work"...
DeCSS is free software. The authors of DeCSS DON'T WANT TO BE PAID FOR THEIR WORK. Seriously, get a clue.
Also, RMS's use of the phrase "information wants to be free", DOES have a moral foundation. Not one that you have to agree with, but a soundly thought out one none-the-less.
His reasoning is that since Information can be free, (and wants to, in the water sense), it is immoral for it not to be free. Reason being that it deprives millions of people of the use of it which is a larger moral consideration than the one person who may or may not be deprived of some royalties.
It's the same thinking that causes intellectual property to expire - that the good of society is an overriding moral principle.
So "information wants to be free", is a literary device - sure, but underlying it is an important point: that information can be free, that people like it to be free, and there is a perfectly defensible moral stance that it should be free. You might disagree in the end, but don't say it's not a valid point of view.
you are a fool.
The whole free software movement came into being because of various beliefs, some moral, some practical regarding licencing. For you to now try and tell us to grow up and forget about licencing is ludicrous at best.
As for what "the corporate world" thinks about licencing or Linux in general, most of us really don't care that much, nor do we care about your big 5 consulting firm or its report.
If all that means free software gets a bad rap in your report so be it, but I think that reflects badly on your report rather than free software.
What's wrong with just using rpm or dpkg? Not invented here syndrome? If there's something wrong with rpm or dpkg, why don't they just contribute to these existing projects instead of reinventing the wheel yet again?
Umm, possibly. But you are really constrained by a C programming mindset. Either use a better interface (like the scheme interface for example) or borrow its good ideas. e.g. pass a linked list of arguments in your callbacks. Of course lack of garbage collection makes it a pain. Did I mention not to use C??
Umm. If KDE and Gnome were both using the same component model with the same widget set (so that themes work across both) and the same UI model, then there wouldn't be any KDE/Gnome divide then would there??? What you're basically saying is that if they were the same, with some pretend difference (i.e. the name "KDE" vs "GNOME"), then all would be well. You're virtually agreeing with me in a round about way.
But let me disagree with you. The only possible reason for different efforts is if the UIs are DIFFERENT. If one or the other actually explores new and advanced technologies for UIs and actually advances the state of the art, that is a justification for separate effort. If you are going to have common UI standards, then what pray tell is the point of duplicating effort?
I think he makes a very good point, and it goes totally against the "competition is good" crowd. Having both Gnome and KDE competing in the same space is about as cool as the US mobile phone standards he laments.
When people pick up a Linux app, they want it to conform to standards and interoperate fully in their environment. Yes sure you can run KDE and Gnome apps under each others environment, and Motif and Openlook and Athena and Tk at the same time. But they won't work together well like say MS-Office works together and the user won't get a consistent interface.
Having both KDE and Gnome is bad for Linux. It duplicates effort. It means some of your favourite apps probably won't integrate properly. It divides an already small segment of the market.
And.... it won't last. One or the other will die. It may take some years, but it will happen. I'll personally be glad when that happens for the reasons above, no matter which is the winner.
Personally I think Gnome will win because it is more commercial friendly with the licence, it is more inclusive with the language choices, more standards inclusive with Corba, and more practical without the C++ binary compatibility problems.
Either way, the sooner one side or the other gives up, the better off we'll all be. If one side has some better idea than the other, then joining forces gives the combination of both lots of good ideas, which is supposed to be the whole point of free software.
Yeah um, a garbage collected app that only runs for a few seconds will simply not collect the garbage. In other words, it's memory reclamation scheme is simply not to bother. Hardly a fair or realistic comparison. Just shows again that there are lies, damned lies and....
No, Americans are not a car-owning car-driving because of population density. Americans are car centric because they decided to. Many cities have the same or even more sparse population densities with strong viable public transport. The story I remember is that the car and oil companies in America conspired to buy out and bankrupt public transport and lobby government to build big highways. Once you build highways you attract cars and reduce the viability of public transport. That might not be so bad except that cars are an expensive, inefficient, poluting form of transport. If you work out the area devoted to cars in an American city (roads, garages, car parks etc) it is actually a lot more than that devoted to residential dwellings! Now if you removed all the roads, made the city more compact as a result and installed public transport you'd have a much nicer environment to live in.
The fact that an app you wanted to use had a lot of (Gnome) dependancies is GOOD. Get this through your head. It means that the author did not reinvent the wheel, he used existing infrastructure. It means that the app is likely to be much more standards compliant out of the box because it uses common code. It means the app has less code and is more maintainable. I know downloading some megabytes is a pain, but get with the program and just do it.
It's funny, I feel the same way about windows. I know how Unix systems work, I feel comparitively lost on a Windows system. It feels much less logical to _me_.
The better of the two proprietry dbs in the test was Oracle. The lessor of the two was MS-SQL.
What I'd say is that your definition of enterprise-ready is suitable for a very large enterprise. I would say Postgresql would score very well on your general criteria, but it's not ready for enterprises with 300GB databases.
BTW who is supporting SQL 99? I'll bet no-one. Actually postgresql is starting to accumulate SQL-99 features, I'll bet Oracle hasn't moved far down that road.
I don't see the logic there. Even RMS has no
objection to selling software for money. The question is whether they can be REdistributed free. If so, then at least they're consistent.
That idea's got a lot of merit, but sadly - that's just not the UNIX way. Take the PATH for example. If every application had its own directory or whatever, your PATH variable would be a million lines long. So you'd have to think of a whole new infrastructure for executing apps.