I have a dev server that is always on trunk. When trunk is ready to push up I tag it. I switch to stage and do a svn switch to the new tag. I check for bugs on stage, address them on the tag and commit, merging up to trunk. When all is ready I do a final svn switch on the live site.
Same thing happened to me. You, however, were lucky with only two days down time. I was down for over 6 weeks (I received nearly 1000 bounces from idiots blacklisting based on this) while SORBS took their lolly gagging free time responding to the ticket. Eventually they said "the IPs are marked for removal from the list, it will take effect in a few days." They weren't kidding, it took a week. Absolutely ridiculous set of processes over there. I can't even imagine how many other blocks of IPs they have wrongfully accused. I know I'll never use SORBS.
>>Another option is to hash against two very different algorithms, that even if both are partially insecure, the chances of being able to trick both are exponentially hig
Is it? Can you prove it? You shouldn't be able to because it simply isn't true. Collisions will exist.
The problem isn't that there aren't jobs. The problem is that there aren't enough smart people to fill the jobs. Companies have finally stopped hiring stupid people for jobs that require smart people. It can take months to fill a position with a talented engineer.
My view on this is a little harsh, but the basic idea holds true.
I think learning compilers and how they will take your code and mangle it into machine code is more important than learning assembly, specifically. Building your own compiler will require you to learn some assembly (or at least the notion of it) which is sufficient for this purpose.
If you're being interviewed by someone who only knows -his- "right" answer.. you probably don't want to work for that company. Think about it. If you're a talented engineer you'll have a choice in jobs.
This doesn't make any sense. Say for instance I patent some widgets. If some other company starts making some widgets, and I don't sue them, the patent becomes invalid. How would filing a patent anonymously make any sense. The anonymous company needs to go out and sue everyone (or at least threaten to sue) or the patent will go away.
Apache 2 with prefork + php works just fine. No problems whatsoever. I've been running this in production for several months now handling 35k requests/day.
On a dual processor FreeBSD 5.2.1 machine Apache 2 can compile from the ports (I had to tweak the portfile, though I think this has been fixed now) using worker threads. It even runs and handles all I could beat it with using apache bench ( ab(1) ) over fast ethernet LAN.
That same worker thread setup with PHP dumps core all over the place.
This library doesn't crack any DRM "stuff." Read the README. I use this software at home. I have a FreeBSD machine that is a PITA to get vmware running so that I can have some sort of iTunes app running on that machine. Instead, I use this libopendaap to connect to the share on my G4 machine and listen to my mp3s. Notice I said mp3s. It does not (read: will not and can not) play protected aac files, aka m4p). Last time I checked, the tunesbrowser that crazney also wrote to demo libopendaap won't even open up m4a files (aac, un protected files).
It is my understanding that as of current, in order to break the DRM you either need to burn to CD and rip back to some other format -or- you can use playfair to "strip" the drm from songs that -you- have purchased -legally-. Yes, you had to -purchase- the song first because you use -your- key to strip the drm.
Please, get the facts straight before ranting off on how these intelligent "crackers" are ruining iTunes for you.
I just graduated a few weeks ago with a CS degree from the UC system. 3.93 GPA. Been working in-industry for some 7 years holding down a few good internships. I've never worked retail or waited tables.
I had two offers straight out of school; I only interviewed at one company. The first offer came from my intern employer. 55k + 3 weeks vacation + benefits. The second offer came after an interview. Let's just say it was -well- worth leaving my internship. Both offers were with great companies that had great talent on their team. But as has been said here already, money talks. I'm in the SF bay area.
Experience is everything. I had built up my network of people resources over the past 7 years and when it came time to find a real job I tapped into just one person in the network. Employers will pay good money for talented engineers. Proving your talent isn't always easy.
Your scheme sum example runs fine for small summations. But for big numbers you'll blow the stack because you aren't tail recursive... a "better" solution to sum between a and b:
(define (sum a b)
(define (sum' mysum iter)
(if (> iter b)
mysum
(sum' (+ mysum iter) (+ iter 1))
Unless your background apps are actually doing something (man top(1) ) then they're not using CPU cycles and are thus not contributing to the heat situation. There are a number of reasons (process scheduling, memory usage) to consider closing background apps, however, on a modern operating system they are all moot.
I would think that if new technology isn't providing enough stress for you, you're not pushing it hard enough. Sure, I can work at half of my ability all day long and have zero stress. Or I can work at 100% of my ability and have all of the stress that I thrive in.
However, new technology lets me accomplish more in the same amount of time when compared with old technology. How much more? Enough. Now buy me a new G5 please.
My iBook 12" has had its openfirmware modded by a few commands I found via Google and running whatever res I want on my 19" monitor with 1024x768 on the LCD isn't a problem.
The hacks are far from hairy too. If I remember correctly, it's all of 3-4 commands.
Almost every semi-reasonable cert out there (MC*, CCNA...) has guaranteed-pass training courses and a myriad of books to choose from in order to practially guarantee that, when used properly (as any high school grad should be able to do), you will pass the test.
All certs need a hands-on test like the CCIE. Or a person to person interview where the obvious numbnuts can be weeded out. Without this the cert is just another test of one's ability to memorize the answers that they received through some test training facility.
I once worked with an MCSE who, when I told him to do something in User Manager, asked me where it was. I couldn't believe it. But he was serious.
Cheating in the university system right now is absolutely ridiculous. As a computer science student at one of the University of California systems finest, I find that some large percentage of senior-level computer science students couldn't write code to save their lives. They have made it through all 4 years by cheating on their programming assignments.
Even more disturbing is the fact that these cheaters, when caught, get nothing more than a slap on the wrist. They are not kicked out of the school, the department or even the major. They are sent on their way with a note made on their "permanent record."
I'm quite sure that cheating is just as bad in other departments/majors as well. Something should be done about this.
Doesn't breaking RSA-576 mean that they have successfully found the prime factorization of -one- 576 bit number. It proves not that a 576 bit key is insecure, but simply that it is possible to factor it in less than a lifetime. Add to that the fact that it probably cost a whole lot more than the $10,000 prize money they'll get for factoring it and suddenly it just doesn't seem like a real threat anymore. Is your Yahoo! mail worth >$10k?
To show that these SDSC guys have computed 2 billion hashes and stored it in a table isn't anything particularly special. As people are finally pointing out on this post, it's a hash. There is no one-to-one relationship between a password and its hash.
Slashdot Mirror
Apache has no problem hiding those from the world.
I have a dev server that is always on trunk. When trunk is ready to push up I tag it. I switch to stage and do a svn switch to the new tag. I check for bugs on stage, address them on the tag and commit, merging up to trunk. When all is ready I do a final svn switch on the live site.
Same thing happened to me. You, however, were lucky with only two days down time. I was down for over 6 weeks (I received nearly 1000 bounces from idiots blacklisting based on this) while SORBS took their lolly gagging free time responding to the ticket. Eventually they said "the IPs are marked for removal from the list, it will take effect in a few days." They weren't kidding, it took a week. Absolutely ridiculous set of processes over there. I can't even imagine how many other blocks of IPs they have wrongfully accused. I know I'll never use SORBS.
Next week on slashdot: "How to get a cooler screensaver."
You forgot to mention its "Anti-Arabic" fonts.
>>Another option is to hash against two very different algorithms, that even if both are partially insecure, the chances of being able to trick both are exponentially hig Is it? Can you prove it? You shouldn't be able to because it simply isn't true. Collisions will exist.
I was surprised that it wasn't mentioned in the write-up that IronPort appliances use Brightmail.
The problem isn't that there aren't jobs. The problem is that there aren't enough smart people to fill the jobs. Companies have finally stopped hiring stupid people for jobs that require smart people. It can take months to fill a position with a talented engineer. My view on this is a little harsh, but the basic idea holds true.
I think learning compilers and how they will take your code and mangle it into machine code is more important than learning assembly, specifically. Building your own compiler will require you to learn some assembly (or at least the notion of it) which is sufficient for this purpose.
If you're being interviewed by someone who only knows -his- "right" answer.. you probably don't want to work for that company. Think about it. If you're a talented engineer you'll have a choice in jobs.
This doesn't make any sense. Say for instance I patent some widgets. If some other company starts making some widgets, and I don't sue them, the patent becomes invalid. How would filing a patent anonymously make any sense. The anonymous company needs to go out and sue everyone (or at least threaten to sue) or the patent will go away.
Apache 2 with prefork + php works just fine. No problems whatsoever. I've been running this in production for several months now handling 35k requests/day.
On a dual processor FreeBSD 5.2.1 machine Apache 2 can compile from the ports (I had to tweak the portfile, though I think this has been fixed now) using worker threads. It even runs and handles all I could beat it with using apache bench ( ab(1) ) over fast ethernet LAN.
That same worker thread setup with PHP dumps core all over the place.
This library doesn't crack any DRM "stuff." Read the README. I use this software at home. I have a FreeBSD machine that is a PITA to get vmware running so that I can have some sort of iTunes app running on that machine. Instead, I use this libopendaap to connect to the share on my G4 machine and listen to my mp3s. Notice I said mp3s. It does not (read: will not and can not) play protected aac files, aka m4p). Last time I checked, the tunesbrowser that crazney also wrote to demo libopendaap won't even open up m4a files (aac, un protected files).
It is my understanding that as of current, in order to break the DRM you either need to burn to CD and rip back to some other format -or- you can use playfair to "strip" the drm from songs that -you- have purchased -legally-. Yes, you had to -purchase- the song first because you use -your- key to strip the drm.
Please, get the facts straight before ranting off on how these intelligent "crackers" are ruining iTunes for you.
I just graduated a few weeks ago with a CS degree from the UC system. 3.93 GPA. Been working in-industry for some 7 years holding down a few good internships. I've never worked retail or waited tables.
I had two offers straight out of school; I only interviewed at one company. The first offer came from my intern employer. 55k + 3 weeks vacation + benefits. The second offer came after an interview. Let's just say it was -well- worth leaving my internship. Both offers were with great companies that had great talent on their team. But as has been said here already, money talks. I'm in the SF bay area.
Experience is everything. I had built up my network of people resources over the past 7 years and when it came time to find a real job I tapped into just one person in the network. Employers will pay good money for talented engineers. Proving your talent isn't always easy.
Why not build from source on machine 1. Then have machine 1 build a package to use on machines 2->n? Yahoo! Best of both worlds.
This would definitely be great in the interrogation room. Though then we'd have some big bitchfest in the YRO section wouldn't we?
(define (sum a b)
)(sum' 0 a) )
Unless your background apps are actually doing something (man top(1) ) then they're not using CPU cycles and are thus not contributing to the heat situation. There are a number of reasons (process scheduling, memory usage) to consider closing background apps, however, on a modern operating system they are all moot.
However, new technology lets me accomplish more in the same amount of time when compared with old technology. How much more? Enough. Now buy me a new G5 please.
My iBook 12" has had its openfirmware modded by a few commands I found via Google and running whatever res I want on my 19" monitor with 1024x768 on the LCD isn't a problem. The hacks are far from hairy too. If I remember correctly, it's all of 3-4 commands.
Almost every semi-reasonable cert out there (MC*, CCNA...) has guaranteed-pass training courses and a myriad of books to choose from in order to practially guarantee that, when used properly (as any high school grad should be able to do), you will pass the test.
All certs need a hands-on test like the CCIE. Or a person to person interview where the obvious numbnuts can be weeded out. Without this the cert is just another test of one's ability to memorize the answers that they received through some test training facility.
I once worked with an MCSE who, when I told him to do something in User Manager, asked me where it was. I couldn't believe it. But he was serious.
Even more disturbing is the fact that these cheaters, when caught, get nothing more than a slap on the wrist. They are not kicked out of the school, the department or even the major. They are sent on their way with a note made on their "permanent record."
I'm quite sure that cheating is just as bad in other departments/majors as well. Something should be done about this.
Doesn't breaking RSA-576 mean that they have successfully found the prime factorization of -one- 576 bit number. It proves not that a 576 bit key is insecure, but simply that it is possible to factor it in less than a lifetime. Add to that the fact that it probably cost a whole lot more than the $10,000 prize money they'll get for factoring it and suddenly it just doesn't seem like a real threat anymore. Is your Yahoo! mail worth >$10k? To show that these SDSC guys have computed 2 billion hashes and stored it in a table isn't anything particularly special. As people are finally pointing out on this post, it's a hash. There is no one-to-one relationship between a password and its hash.