High school is tough. Its is going to be awful for everyone (basically).
In other words, high school is (basically) not unlike a part-time prison. But you would know that if you took time to read the actual article, which is focused on the wrongs of the education system rather than on assumed guilt of nerdsm populars, or 'hormones'.
From my semi-forgotten cryptography course I seem to recall that ultimately, the encryption can be unbreakable if the message length is shorter or equal to key length, so with megabit-order keylengths short messages should be pretty secure. Would anyone with better understanding of the subject elaborate on that?
Non-existent specs, poorly understood assumptions for certain calculations - what a nightmare for any professional software developer!
Yes, we have these nightmares routinely. As the time passes you get used to that, and learn to extract and document as much domain knowledge and requirements of customers as possible.
Linux as a desktop OS still requires far more Unix knowledge than most people want to deal with.
My wife must be a Linux geek then, since after I have set up Debian she had no problems using it for routine non-techy tasks (edit texts, browse the web and play solitaire).
Mind you, most PCs come with operating system pre-installed by equipment manufacturer. In case with Linux this would mean that you need not teach user how to install it, nor how configure X11/compile necessary kernel modules/whatever. Thus you can have reasonably pre-set Linux system that Joe Average can use out of the box.
Not enough vendors sell Linux preinstalled? Well, yes, but it has nothing to do with Linux being (un)ready for the destop. Contrary to popular belief, installing and configuring Windows on random hardware for an average user is a task next to impossible.
Perhaps the problem arises because we have so many passwords to remember. My solution is to have one password for most of my accounts, which I share with nobody.
FWIW, my solution is to use a unique password per account, and store them with KeyRing on my Palm. It has flexible password generator, Linux desktop conduit and it stores its database with 3DES-encryption. ATM I have about 70 different passwords to various accounts, and have no problems managing them.
It saved me from lots of trouble when Sourceforge password database was compromised.
All of the improvements you've suggested (e.g., tail recursion) are done not by a human (who can make mistakes) but rather by a JIT.
Even if there actually exists a JIT complier that does tail call elimination, there is absolutely no (zero, none) support for closures and lexical scope. Ditto for multimethods. Ditto for multiple inheritance. It is not that JVM simply lacks support for them, you have to fight JVM architecture to have them.
I know that some of the best language designers work for Sun (Steele, Gosling among them). But I also know that support for other languages wasn't among their design goals. JVM does a fair job being a virtual machine for Java, but sucks for other lanugages.
Miguel de Icaza point to the advantage of being able to compile from multiple languages.
As a person who've got a feel of writing JVM-targeted compilers, I'd like to notice that it makes extremely poor target for other languages. JVM was designed explicitly for Java, without any other language in mind. Thus, writing translators from other languages takes certain number of convoluted tricks.
If your source language has closures, true lexical scoping, multimethods or multiple inheritance, JVM is clearly a suboptimal vehicle, unless you want to bend over a lot. From performance standpoint, its stack-oriented machine isn't optimal either. JVM architecture also leaves no easy ways to implement proper tail-recursion.
CLR is likely a much better target, but even that one, designed for interoperability from scratch, has some rough edges for non-mainstream programming languages.
he's someone that has Vision and can seek it out (even if we might not agree with his Vision, he's definitely got it!)
The most bloody and atrocious governers in the history were of that kind. Alexander the Great, Napoleon, Lenin, Stalin, Hitler, Mao all had their Visions.
When it comes to electing a US president, I (as a foreigner) will prefer any greedy, lying, corporate-sponsored puppet over one with grand plans regarding the world.
Funny or not, but the worst thing that gathering data from these devices, unlike from numberplates, can be trivially automated.
Police checkpoints that may glance on your numberplate is one thing, but tight country-wide network of sensors on earch crossroad makes a perfect Big Brother.
If you want it sorted, add ORDER BY to your SQL and you're done.
I don't have an RDBMS on my Palm, yet it somehow manages to sort my records. Magic happens?
I've written apps that maybe 60% were pure GUI code, just creating widgets and adding them to forms, etc.
I've written such applications too. OTOH, I was lucky to participate in projects doing things that were hardly ever done before, let alone mainstreamed, so noone made Wizards, invented Patterns and wrote Libraries that we could reuse for that kind of work.
If all you do is glueing GUIs for DBs - fine. World needs this kind of work. But people around you write GIS, mobile applications, distributed systems and ORDER BY functions for your favorite DBMS.
You'll get a lot of geeks sneering that a text editor is the only way to write code, but that is an obsolete way of working.
For anything but GUI drawing, good old text editors still beat all these point-and-click thingy.
Writing and adjusting your code is faster with text editor (unless you type with two fingers).
Non-boilerplate coding can't be done with point-and-click interface, be it UML, RAD or whatelse. Programming is not about changing superclasses and adding member variables: at some point you have to implement actual algorithms. At this point you have to resort to text editor and all the glory of CASE tools fades, since when you actually do want to change superclass you have to move your hands off the keyboard to mouse, swith to different window, and often you are not allowed to change CASE-tool-controlled parts of code by hand. I've yet to see any evidence that a CASE user beats competent developer with editor in terms of performance.
Those thinking of pointy-clicky interfaces being a magic wand should go and try writing bubblesort with mouse.
Now if only computer manufacturers could make equipment even remotely this sturdy.
They can, and they indeed do make it when necessary, say for your neighbourhood's nuclear power plant, or for a space probe like Pioneer 10.
Problem is, you don't need such hardware endurance even if you could afford it. How soon will you want a newer video card to your supermegareliable PC?
Well since that leaves you with only the atom bomb, the telephone, cotton gin and the laser..
Some inventions cannot be attributed to a single nation or person; laser is one of them.
'"Basic work in quantum electronics leads to the inventions of resonator and amplifier based on maser-laser theory", Townes, A.Prskhorov and N.Bason of Lebedev Institute in Moscow were awarded together the Nobel Prize of Physics of that year.'
Slashdot Mirror
And you call me a piece of work.
And rightly so. Rather than nitpicking, take courage to admit that you didn't read the article and had truly no idea what the fsck it was about.
High school is tough. Its is going to be awful for everyone (basically).
In other words, high school is (basically) not unlike a part-time prison. But you would know that if you took time to read the actual article, which is focused on the wrongs of the education system rather than on assumed guilt of nerdsm populars, or 'hormones'.
The test was obviously run on dual-processor systems.
From my semi-forgotten cryptography course I seem to recall that ultimately, the encryption can be unbreakable if the message length is shorter or equal to key length, so with megabit-order keylengths short messages should be pretty secure. Would anyone with better understanding of the subject elaborate on that?
Non-existent specs, poorly understood assumptions for certain calculations - what a nightmare for any professional software developer!
Yes, we have these nightmares routinely. As the time passes you get used to that, and learn to extract and document as much domain knowledge and requirements of customers as possible.
Linux as a desktop OS still requires far more Unix knowledge than most people want to deal with.
My wife must be a Linux geek then, since after I have set up Debian she had no problems using it for routine non-techy tasks (edit texts, browse the web and play solitaire).
Mind you, most PCs come with operating system pre-installed by equipment manufacturer. In case with Linux this would mean that you need not teach user how to install it, nor how configure X11/compile necessary kernel modules/whatever. Thus you can have reasonably pre-set Linux system that Joe Average can use out of the box.
Not enough vendors sell Linux preinstalled? Well, yes, but it has nothing to do with Linux being (un)ready for the destop. Contrary to popular belief, installing and configuring Windows on random hardware for an average user is a task next to impossible.
Perhaps the problem arises because we have so many passwords to remember. My solution is to have one password for most of my accounts, which I share with nobody.
FWIW, my solution is to use a unique password per account, and store them with KeyRing on my Palm. It has flexible password generator, Linux desktop conduit and it stores its database with 3DES-encryption. ATM I have about 70 different passwords to various accounts, and have no problems managing them.
It saved me from lots of trouble when Sourceforge password database was compromised.
All of the improvements you've suggested (e.g., tail recursion) are done not by a human (who can make mistakes) but rather by a JIT.
Even if there actually exists a JIT complier that does tail call elimination, there is absolutely no (zero, none) support for closures and lexical scope. Ditto for multimethods. Ditto for multiple inheritance. It is not that JVM simply lacks support for them, you have to fight JVM architecture to have them.
I know that some of the best language designers work for Sun (Steele, Gosling among them). But I also know that support for other languages wasn't among their design goals. JVM does a fair job being a virtual machine for Java, but sucks for other lanugages.
Miguel de Icaza point to the advantage of being able to compile from multiple languages.
As a person who've got a feel of writing JVM-targeted compilers, I'd like to notice that it makes extremely poor target for other languages. JVM was designed explicitly for Java, without any other language in mind. Thus, writing translators from other languages takes certain number of convoluted tricks.
If your source language has closures, true lexical scoping, multimethods or multiple inheritance, JVM is clearly a suboptimal vehicle, unless you want to bend over a lot. From performance standpoint, its stack-oriented machine isn't optimal either. JVM architecture also leaves no easy ways to implement proper tail-recursion.
CLR is likely a much better target, but even that one, designed for interoperability from scratch, has some rough edges for non-mainstream programming languages.
he's someone that has Vision and can seek it out (even if we might not agree with his Vision, he's definitely got it!)
The most bloody and atrocious governers in the history were of that kind. Alexander the Great, Napoleon, Lenin, Stalin, Hitler, Mao all had their Visions.
When it comes to electing a US president, I (as a foreigner) will prefer any greedy, lying, corporate-sponsored puppet over one with grand plans regarding the world.
USAF research department issued a new book: "100 new ways to cook a human".
Nobody was expecting the same memories; they were, however, expecting the same behavior patterns.
And I am happy with the result. Hope it will stop all this 'bad bloodline' and 'spoiled genes' crap habit of judging people by their ancestors.
Funny or not, but the worst thing that gathering data from these devices, unlike from numberplates, can be trivially automated.
Police checkpoints that may glance on your numberplate is one thing, but tight country-wide network of sensors on earch crossroad makes a perfect Big Brother.
If you want it sorted, add ORDER BY to your SQL and you're done.
I don't have an RDBMS on my Palm, yet it somehow manages to sort my records. Magic happens?
I've written apps that maybe 60% were pure GUI code, just creating widgets and adding them to forms, etc.
I've written such applications too. OTOH, I was lucky to participate in projects doing things that were hardly ever done before, let alone mainstreamed, so noone made Wizards, invented Patterns and wrote Libraries that we could reuse for that kind of work.
If all you do is glueing GUIs for DBs - fine. World needs this kind of work. But people around you write GIS, mobile applications, distributed systems and ORDER BY functions for your favorite DBMS.
Don't overgeneralize.
You'll get a lot of geeks sneering that a text editor is the only way to write code, but that is an obsolete way of working.
For anything but GUI drawing, good old text editors still beat all these point-and-click thingy.
Writing and adjusting your code is faster with text editor (unless you type with two fingers).
Non-boilerplate coding can't be done with point-and-click interface, be it UML, RAD or whatelse. Programming is not about changing superclasses and adding member variables: at some point you have to implement actual algorithms. At this point you have to resort to text editor and all the glory of CASE tools fades, since when you actually do want to change superclass you have to move your hands off the keyboard to mouse, swith to different window, and often you are not allowed to change CASE-tool-controlled parts of code by hand. I've yet to see any evidence that a CASE user beats competent developer with editor in terms of performance.
Those thinking of pointy-clicky interfaces being a magic wand should go and try writing bubblesort with mouse.
Also the reason we have 360 degrees in a circle despite the fact that radians make much more sense is because greek
Actually, the 360 degree system was innovated by Sumerian astronomers long before Greeks and Romans got any culture to talk of.
365 days a year is just a lucky coincidence.
No, it is not.
When can I get one that hovers in space, all on its own, with no cords?
Wireless LAN + battery inside + powerful enough fan on the floor underneath. Most likely it will also rotate.
At least if we take the common design mistakes as the metric.
'Poor email intergration' sounds pretty sophisticated compared to 'don't use the <blink> tag'.
I sicnerely hope it won't be wuftpd..
Now if only computer manufacturers could make equipment even remotely this sturdy.
They can, and they indeed do make it when necessary, say for your neighbourhood's nuclear power plant, or for a space probe like Pioneer 10.
Problem is, you don't need such hardware endurance even if you could afford it. How soon will you want a newer video card to your supermegareliable PC?
1. Invite a geek celebrity into the company
2. Ask geeks for support
3. ????
3. Profit
Watching and reading LotR lets me relax from all these (evil and not) problems, prospects and inventions surrounding us everywhere.
When I want to ponder on social problems, I take a copy of '1984'. Mixing it with a fairy tale would spoil the value both.
No matter how much Japan or France or Russia or China may want to, they simply cannot build an F-22 for a long time to come.
Well, in fact they can (at least France or Russia). Another issue is that they don't have stimuli (France) or money (Russia) to mass-produce them.
And yet another fun thing: Russia has no money for that now because it spent way too much money on military in past. Hint hint.
Well since that leaves you with only the atom bomb, the telephone, cotton gin and the laser..
Some inventions cannot be attributed to a single nation or person; laser is one of them.
'"Basic work in quantum electronics leads to the inventions of resonator and amplifier based on maser-laser theory", Townes, A.Prskhorov and N.Bason of Lebedev Institute in Moscow were awarded together the Nobel Prize of Physics of that year.'
..Borland buys Microsoft!