High level languages like Ruby, Python, or even Java are strongly recommended for all new projects.
How about a high level, compiled language with static typing like Ocaml. More speed, more protection, and it's been officially certified as "The programming tool of choice for discriminating hackers".
And compiled high level language programs by definition can't be as fast as assembly language. But at some point we all decided that ease of programming and maintainance was more important.
As processors get faster and faster the time it takes to do the context switch will become less and less important.
I agree. College is a good idea for many reasons. Not the least of which is to become a better educated person. You are not going to learn much in college that will make you a better system administrator, but you will learn a lot that will make you a more rounded person.
If you can afford it, go to college. Take intro courses in things you're interested in, go to cultural events, talk with professors, get into midnight philisophical conversations with your friends. And as someone else pointed out, it'll be the best dating pool that you'll ever have in your life.
It's a great language, and transparently compiles to assembly on both windows and linux. Fast, modular, with all the functional fun that the kids want these days.
You're a brave man to post this on slash dot. You probably had a pretty good idea to the response you would get. So I'll be brave too.:)
I happen to agree with what you're saying. In the same way that we don't have one model of car that's right for everyone, we don't have one operating system that fills all niches. Different systems have been created with different design goals in mind.
Anyone who reads slashdot probably spends at least 60 hours a week in front of a computer, and has good working knowlege as to how one works. We build or manage systems (whether they be web sites, applications, databases, or networks). To do our jobs efficiently we need to be able to automate the work whereever possible, e.g.
Unix makes it easy to automate the complex tasks that we do every day. In addition, Unix was built from the ground up to be a multi-user OS, and has many network centric features that make it a pleasure to use as a server.
For better or worse, most of the world doesn't (and shouldn't) care. They use a computer for ad hoc projects. They keep an address book, write letters and proposals, surf the web, email, and play Diablo II (Well, I know we do that too).
Windows was designed with a clean, consistent user interface. Everything (including the applications written for it) is graphics centric. This makes doing one off operations a breeze, but makes it much more difficult to automate. So we hate it. But Microsoft can live with that.
As far as stability goes, it's all relative. I use Windows NT as a client at work, and I find that it rarely crashes. I typically power it down about once a week. If things start to act a little funky then I'll shut it down and restart. I've lost more work from power outages than I have from lockups.
All that said, Windows is a bad fit as a server operating system. To extent the car metaphore, it's like taking your Honda Accord offroading. Wrong tool for the job.
Just to be fair, the l0pht article is really a different point of view. Their argument is not that you shouldn't announce a vulnerablitity (or even code an example), but that if you do, you should announce the solution at the same time.
They are commenting on security venders that announce a problem, then want you to buy their product for the fix.
Many of these sites are not trying to lock you in
on
Web Site "Lock-In"
·
· Score: 2
On many of these sites, the back buttons don't work because they are redirecting you to another page in their site when you first arrive.
Why? So they can offer different content depending on what browser you are using, and what plug-ins you have installed. If you have the flash plug in, they'll redirect you to a flash splash page... If you're using an older browser, they'll redirect you to a frames free page, etc.
You might find this behavior annoying, but it's hardly sinister. People in this thread have already pointed out that it's easy to overcome the limitation. It's kinda funny that slashdot is making a big deal about it.
Protected memory and preemptive multitasking were old technologies in 1984. Doing them again wouldn't have changed much.
I suppose it really depends on whether you're looking to bring a new paradigm into the world, or make a good operating system. I just had to laugh as I read this, because as a programmer, protected memory and real multitasking are not features I would dismiss so lightly.
But your point is well taken... Thinking back, the Next system really covers both bases beautifully. But the Next was a commercial failure. Probably because PCs and Macs were too entrenched at that point.
It's really a pity. We could have skipped a decade of baby step operating systems upgrades, and ended up about where we are now.
Interesting thought. Maybe rather than account sharing, they are checking for account establishment.
They can already limit the damage from account sharing: I read in one of the comments that only one client could beam music from a given account at a given time.
But this would be easy to get around if I could create 10 accounts with my collection (each with a different hotmail address) and give each of my friends a different account:)
Your posting makes me wonder if you read the article very carefully. From the text:
Of the approximately 2500 sectors requested by the server, only 100 or so were requested 2 or more times and no track was requested more than 3 times. The server appears to be sampling the disk purely at random.
This is not client side security by any definition of the term. You are transmitting random sectors from the disk to the server for verification, not a checksum, not a thumbs up or thumbs down, but the raw data. The only way you can fake this is to have a complete copy of the CD in your possession.
They've really done a good job with this. Kudos to MP3.com for thinking it out.
First a disclaimer, I'm a starcraft addict, and I still play 2-3 games a day online. I've been playing pretty much since the game came out. I've switched to other games for a bit, and enjoyed them, but I just keep going back to SC.
I'm not alone. Last time I checked there were over a million internet starcraft games getting played every day on Battle.net alone. This doesn't count the hundreds of Free Starcraft Game Servers that are hosted across the world.
In addition, while StarCraft has some amazing game design, and nice graphics, it is not exactly technically cutting edge. Because of this, it might be significantly easier to port.
Think about format conversion. If you can decrypt a DVD video stream, you can recompress it for distribution, in the same way people rip mp3s off a CD. There will be a substantial loss in quality, but video streaming is getting better all the time.
This is too true. You're better off getting your science news from "17" magazine than Wired. It can get pretty funny sometimes, though.
My favorite article of theirs was the one on personal rocketships that a small aerospace company was developing. Yes, in the next 10 years , you too could have your own space ship. The design they were showing had rocked powered roters! Very slick. Oh, and they were looking for investers...
Most of the flash sites I've seen seem to have no idea about design from the perspective of usability.
For example, why is it that many of these sites have navigation systems in which you can only see where a particular button leads when you move your mouse over it. In one example, The graphical buttons had been passed through a blur filter so the text on them was illegible until you floated above them, then they snapped into focus. I was slapping my head at the stupidity of it all.
God have mercy on us all, and not let the flash designers get their hands on mainstream applications.
"Randy Mills impressed me that he may also be brilliant. He talks off the top of his head in a way that other scientists can't. But that doesn't mean he's right. I think his results are right, but doesn't mean his theory is right," Miles said.
I'm usually suspicious of people that are glib about technical information. It often means they're more interested in selling you on something than in getting the facts correct. This guy definitely has something to sell.
Slashdot Mirror
It should be a crime to teach people C/C++.
High level languages like Ruby, Python, or even Java are strongly recommended for all new projects.
How about a high level, compiled language with static typing like Ocaml. More speed, more protection, and it's been officially certified as "The programming tool of choice for discriminating hackers".
Ocaml
And compiled high level language programs by definition can't be as fast as assembly language. But at some point we all decided that ease of programming and maintainance was more important.
As processors get faster and faster the time it takes to do the context switch will become less and less important.
Actually, I would be very pleased if I were the Klez developer.
I agree. College is a good idea for many reasons. Not the least of which is to become a better educated person. You are not going to learn much in college that will make you a better system administrator, but you will learn a lot that will make you a more rounded person.
If you can afford it, go to college. Take intro courses in things you're interested in, go to cultural events, talk with professors, get into midnight philisophical conversations with your friends. And as someone else pointed out, it'll be the best dating pool that you'll ever have in your life.
You can star in a porn movie though!
Have you worked with ocaml?
www.ocaml.org.
It's a great language, and transparently compiles to assembly on both windows and linux. Fast, modular, with all the functional fun that the kids want these days.
The post about the article claims a threefold decrease in effective key length. This is much greater than a three fold increase in speed.
If factoring a number is proportional to the size of a number, then a number that would take 280 years to factor might take 46 days.
It also says that the technique might not be effective on keys as short as the ones currently used.. so the speedup may be theoretical.
Or a stuffed bun-bun doll, for the discerning geek.
An observation that I respect and agree with. But there are already many, many people coding websited in ASP/COM/SqlServer.
They have already sold their souls, and have nothing to lose.
nt
I guess I am misunderstanding you.
How do you indent to encode 170,000 possible characters in 2 bytes worth of data?
Are you suggesting that unicode use additional bytes, or is there already an "escape" code which allows multicharacter encoding.
Isn't multicharacter encoding what unicode was meant to eliminate?
Instead of getting fat at my computer while playing Quake III, I could be getting into shape.
Body by Carmack!
What was his high score at Pac Man?
You're a brave man to post this on slash dot. You probably had a pretty good idea to the response you would get. So I'll be brave too. :)
I happen to agree with what you're saying. In the same way that we don't have one model of car that's right for everyone, we don't have one operating system that fills all niches. Different systems have been created with different design goals in mind.
Anyone who reads slashdot probably spends at least 60 hours a week in front of a computer, and has good working knowlege as to how one works. We build or manage systems (whether they be web sites, applications, databases, or networks). To do our jobs efficiently we need to be able to automate the work whereever possible, e.g.
find . -name "*html" -exec grep "badaddress" {} \; -exec perl -p -i -e "s/baddress/goodaddress/g;" {} \;
Unix makes it easy to automate the complex tasks that we do every day. In addition, Unix was built from the ground up to be a multi-user OS, and has many network centric features that make it a pleasure to use as a server.
For better or worse, most of the world doesn't (and shouldn't) care. They use a computer for ad hoc projects. They keep an address book, write letters and proposals, surf the web, email, and play Diablo II (Well, I know we do that too).
Windows was designed with a clean, consistent user interface. Everything (including the applications written for it) is graphics centric. This makes doing one off operations a breeze, but makes it much more difficult to automate. So we hate it. But Microsoft can live with that.
As far as stability goes, it's all relative. I use Windows NT as a client at work, and I find that it rarely crashes. I typically power it down about once a week. If things start to act a little funky then I'll shut it down and restart. I've lost more work from power outages than I have from lockups.
All that said, Windows is a bad fit as a server operating system. To extent the car metaphore, it's like taking your Honda Accord offroading. Wrong tool for the job.
Just to be fair, the l0pht article is really a different point of view. Their argument is not that you shouldn't announce a vulnerablitity (or even code an example), but that if you do, you should announce the solution at the same time.
They are commenting on security venders that announce a problem, then want you to buy their product for the fix.
On many of these sites, the back buttons don't work because they are redirecting you to another page in their site when you first arrive.
Why? So they can offer different content depending on what browser you are using, and what plug-ins you have installed. If you have the flash plug in, they'll redirect you to a flash splash page... If you're using an older browser, they'll redirect you to a frames free page, etc.
You might find this behavior annoying, but it's hardly sinister. People in this thread have already pointed out that it's easy to overcome the limitation. It's kinda funny that slashdot is making a big deal about it.
That's right, nobody uses Windows anymore. It's practically unsupported. The entire marketplace has started using the BeOS.
What planet do you come from?
Protected memory and preemptive multitasking were old technologies in 1984. Doing them again wouldn't have changed much.
I suppose it really depends on whether you're looking to bring a new paradigm into the world, or make a good operating system. I just had to laugh as I read this, because as a programmer, protected memory and real multitasking are not features I would dismiss so lightly.
But your point is well taken... Thinking back, the Next system really covers both bases beautifully. But the Next was a commercial failure. Probably because PCs and Macs were too entrenched at that point.
It's really a pity. We could have skipped a decade of baby step operating systems upgrades, and ended up about where we are now.
Interesting thought. Maybe rather than account sharing, they are checking for account establishment.
:)
They can already limit the damage from account sharing: I read in one of the comments that only one client could beam music from a given account at a given time.
But this would be easy to get around if I could create 10 accounts with my collection (each with a different hotmail address) and give each of my friends a different account
Of the approximately 2500 sectors requested by the server, only 100 or so were requested 2 or more times and no track was requested more than 3 times. The server appears to be sampling the disk purely at random.
This is not client side security by any definition of the term. You are transmitting random sectors from the disk to the server for verification, not a checksum, not a thumbs up or thumbs down, but the raw data. The only way you can fake this is to have a complete copy of the CD in your possession.
They've really done a good job with this. Kudos to MP3.com for thinking it out.
Just a few thoughts...
First a disclaimer, I'm a starcraft addict, and I still play 2-3 games a day online. I've been playing pretty much since the game came out. I've switched to other games for a bit, and enjoyed them, but I just keep going back to SC.
I'm not alone. Last time I checked there were over a million internet starcraft games getting played every day on Battle.net alone. This doesn't count the hundreds of Free Starcraft Game Servers that are hosted across the world.
In addition, while StarCraft has some amazing game design, and nice graphics, it is not exactly technically cutting edge. Because of this, it might be significantly easier to port.
Piracy is a little bit of an issue.
Think about format conversion. If you can decrypt a DVD video stream, you can recompress it for distribution, in the same way people rip mp3s off a CD. There will be a substantial loss in quality, but video streaming is getting better all the time.
I suspect that people are doing this already.
This is too true. You're better off getting your science news from "17" magazine than Wired. It can get pretty funny sometimes, though.
My favorite article of theirs was the one on personal rocketships that a small aerospace company was developing. Yes, in the next 10 years , you too could have your own space ship. The design they were showing had rocked powered roters! Very slick. Oh, and they were looking for investers...
Um, no.
Most of the flash sites I've seen seem to have no idea about design from the perspective of usability.
For example, why is it that many of these sites have navigation systems in which you can only see where a particular button leads when you move your mouse over it. In one example, The graphical buttons had been passed through a blur filter so the text on them was illegible until you floated above them, then they snapped into focus. I was slapping my head at the stupidity of it all.
God have mercy on us all, and not let the flash designers get their hands on mainstream applications.
Here's the actual link to the
Wall Street Journal article.
"Randy Mills impressed me that he may also be brilliant. He talks off the
top of his head in a way that other scientists can't. But that doesn't mean
he's right. I think his results are right, but doesn't mean his theory is right,"
Miles said.
I'm usually suspicious of people that are glib about technical information. It often means they're more interested in selling you on something than in getting the facts correct. This guy definitely has something to sell.