For example, did you know that throwing an exception is 10,000 times slower than a return statement in HP/UX 11?
Unless you were throwing exceptions to return values, this is not a fair comparison.
I would compare throwing exceptions to returning a value and testing for an error code every time you call a function regardless of whether an error occurred or not. If you have a deep call stack, you will also have to check for an error at multiple levels, manually propagating them.
The whole point of the internet is dumb network, smart nodes. If the end nodes aren't smart enough to deal with spam (99.9% is quite easy to identify) and viruses (hello MS, I'm talking to you), then that is the problem of the end nodes, not the network.
And if someone uses compromised computers on the net to DDOS your server, it is your problem. After all, your end node should be smart enough to deal with it, right?
Wrong! I think that the administrators of the compromised servers carry part of the blame. The same goes for the administrators for the open-relay servers.
No one really has perfected the art of accurately estimating projects.
Even if you would posess the magic tool to perfectly estimate project costs when no specifications are given:-) it would be worthless.
If you bid for a project, this formula will yield a large developement time and large costs. You would likely be outbid by companies using 'less than perfect' estimating tools, that can promise lower costs and developement time.
Re:How can you get a BIGGER back door than CRII?
on
Code Red III
·
· Score: 1
I'd like a web interface to control the compromised machine.
Ranking by how many users clicked on the link is meaningless... I usually try a lot of links, because I cannot decide just by the link if this is the corect site or not.
The problem is not with the language or the method but with the programmer. Many programmers today use a technique I like to call "Programming by tying". That is, they write an initial program. It does not work so they try changing an "i+1" to an "i+2". Works better now. "Let's try calling this function twice here." "OK, now its even better. The program is ready". Later when the program crashes with BSOD, their conclusion is: "Windows sucks!".
I believe the best introductory course in programming should use a pencil, a paper and a thick book on mathematics. Otherwise "people that can hardly tie their shoes will be able to write Web Browsers using languages like VB or Java"(1). What will happen to our salaries?:)
Slashdot Mirror
For example, did you know that throwing an exception is 10,000 times slower than a return statement in HP/UX 11?
Unless you were throwing exceptions to return values, this is not a fair comparison.
I would compare throwing exceptions to returning a value and testing for an error code every time you call a function regardless of whether an error occurred or not. If you have a deep call stack, you will also have to check for an error at multiple levels, manually propagating them.
Except that from the 2-nd year on, they will charge you a 'maintenance fee' on the account.
And if someone uses compromised computers on the net to DDOS your server, it is your problem. After all, your end node should be smart enough to deal with it, right?
Wrong! I think that the administrators of the compromised servers carry part of the blame. The same goes for the administrators for the open-relay servers.
--alex
No one really has perfected the art of accurately estimating projects. :-) it would be worthless.
If you bid for a project, this formula will yield a large developement time and large costs. You would likely be outbid by companies using 'less than perfect' estimating tools, that can promise lower costs and developement time.
Even if you would posess the magic tool to perfectly estimate project costs when no specifications are given
I'd like a web interface to control the compromised machine.
Ranking by how many users clicked on the link is meaningless... I usually try a lot of links, because I cannot decide just by the link if this is the corect site or not.
> You want to have every
:)
:)
> program crash on you ?
The problem is not with the language or the method but with the programmer. Many programmers today use a technique I like to call "Programming by tying". That is, they write an initial program. It does not work so they try changing an "i+1" to an "i+2". Works better now. "Let's try calling this function twice here." "OK, now its even better. The program is ready". Later when the program crashes with BSOD, their conclusion is: "Windows sucks!".
I believe the best introductory course in programming should use a pencil, a paper and a thick book on mathematics. Otherwise "people that can hardly tie their shoes will be able to write Web Browsers using languages like VB or Java"(1). What will happen to our salaries?
haral.
(1) see the INTERCAL documentation
It always surprises me how every topic on programming languages turns into a battle between Java and C++.
:)
I would like to remind you all that far more succesfull systems were built in languages other than Java and C++ than in Java and C++ taken together.
I would conclude that both Java and C++ are obsolete
Have a nice day,
alex.