Re:No language that I like better
on
What is Perl 6?
·
· Score: 1
If you're on a machine with Unix man pages, try man perlfunc. On the plus side, it works even if your internet connection dies. It also works when you're on an airplane, frantically trying to meet project deadlines;).
Typically it's unusual to see ``just a crash.'' Most programmes written in C and C++ crash due to buffer overflows, which frequently lead to running unsigned code. As a general rule, if a C or C++ code crashes, it is a fairly likely possibility to be able to run arbitrary code. Just because nobody's done it yet doesn't mean that it's not possible.
I've said it before and I'll say it again, take this gentleman's tech predictions with an extremely large grain of salt.
Re:Tempting....
on
Why Use GTK+?
·
· Score: 2, Informative
Sean Egan's Open Source Messaging Application Development: Building and Extending Gaim covers basic to intermediate GTK+ programming [with sockets, no less] quite nicely. He meticulously describes everything he does with numerous examples. He never skimps on the details and even gives insights into his own thinking on most problems.
It's a very, very good read, even for non-Gaim developers. On the plus side, you'll be supporting open-source development if you purchase it.
Will someone please mod this up? If you're using RDP to do nothing other than manage users and groups, restart services, or pretty much any typical Windows 2000 domain tasks, you're wasting tons of bandwidth.
For most everything you do, open up MMC and add a snap in for a remote computer. You can do everything remotely that you can do from the local ``Administrative Tools'' folder and it's a hell of a lot snappier [and less bandwidth-hungry] than a full RDP session.
A few of the real hole in the wall Mexican places around where I live [in the Los Angeles area] sell bottled Mexican sodas [include Coke] with all the labels in Spanish.
They use real sugar and it isn't a dollar a bottle.
Search Ebay or ask around for IBM Thinkpad laptops. I've been able to get my hands on many, all for less than 50$. Obviously you aren't going to run Fedora or Windows XP on these. However, Windows 98 or NT [or 2000 even if you've got one with 32 megs of RAM or more] should run just spiffy on one of these. Obviously there are also the dozens of tiny Linux distros like DamnSmallLinux that work like a charm on hardware like this.
The older Thinkpads are built rock-solid and are great for debugging machines. Dead batteries aside, the only problem I've ever had was a screen that died on a model with a known-flaky screen [avoid the 365XD if you can]. I'm not sure how the newer Thinkpads are. I've never really used them since I've had no need to upgrade.
How about using RFID to FIND the damn keyring. I know it won't allow you to pinpoint exact location, but narrowing down the location of my keys or anything else small, portable, and easily lost to a single room would be immensely helpful. A certain amount of radio direction finding could possibly be used to help even further.
Yeah, I know there exists technology to make a little siren go off on the key ring or whatever, but let's stay on topic here.
Saying that, when it comes to technology at least, he is speculative is something of an understatement. Take what he says with an extremely large grain of salt.
You can speak badly of the president all you want in any public forum. That falls under the prime First Amendment right: freedom of speech. However, it's a whole different ballgame if you start making direct threats to the safety of the president.
Of course, I'm assuming you're talking about the United States. Other countries with a president as head of state may not have the same freedom of speech clauses in their governing documents.
As soon as you have a mesh network that large, you have to look into rewriting the routing protocols. The current mix we have [i.e., mainly OSPF and BGP] just wouldn't possibly scale to that many nodes unless you start introducing supernodes. With supernodes comes centralisation, and with centralisation you have a target for the FCC or any similar organisation.
That's not to say it isn't much easier to change supernodes when all the links are wireless. Also, the solution of more efficient routing protocols seems a little more realistic. But then you have to have a router in every house..
It's a lovely thought, but it's still a fair amount of time away.
And here I am surrounded by trees on all sides with their RF-absorbing foliage..
The old and multiple times dead Ricochet used to have repeaters at every major intersection, with supernodes with a T1 [or something to the effect] interspersed.
It's a great idea, but look where it got Ricochet.
Plus GPL'd tools like uClibc and Busybox [quite the pair] make it really easy to develop and build a system that is efficient with both memory and storage. Last I heard, all of the offerings from MS on this behalf are behemoths.
When a single kilobyte makes a difference, you're going to want the system that's already compact to begin with.
Not to mention the fact that you've got a system with a large embedded following and a huge user support community. Slap a few dollars into most underfunded developers' hands and they'll listen to your requests keenly, to boot.
Re:Cain and VoIP Sniffing
on
VoIP Security
·
· Score: 0
While that may have been the case for older versions of SSH [I honestly don't know myself], modern versions of OpenSSH seem to have the client send the password in a whole packet. In others words, the client handles the input of the password and sends it over as one packet, instead of sending each character individually. I'm not sure if the parent was alluding to this fact or not, so I'm just clarifying.
Has anyone considered the implications of a DDoS involving a zombie army of machines with 100 mbit uplinks? This could spell disaster for just about everybody except those with the absolute fattest pipes. It takes an awful lot of hosts to swamp an OC3 now, but that's with hosts that rarely have a half megabit uplink, if that. It would be frighteningly easy to swamp the heavy links with a few 100 mbit links.
That is, of course, unless the bigger pipes grow at a rate proportional to the smaller ones. That also assumes symmetrical links for the home connections. Oh the irony of a 100 mbit / 128 kbit connection.
It only plays reggae since reggae usually uses only a few, simple chords, but it is still way cool. They even made it remote controllable. And of course, all the source is on their site.
A lot of the Slashdot effect is also due to a huge number of new sockets being opened for the HTTP requests. This can take out a mid-quality machine even with a very fat pipe.
I'm surprised that nobody has mentioned the Linux Logical Volume Manager subsystem. It has many of the features of RAID arrays [such as spanning across multiple drives] with the added flexibility of being able to dynamically add [and theoretically remove] drives.
Unfortunately, aside from RAID'ing the volumes or something similar, I haven't been able to find any information on making the system redundant.
Read about it more on TLDP. It's a very robust system that works well on both servers and desktops.
This sounds like basically the same principle of inductive coils in the mouse pad and mouse that Afroman used in his wireless mouse charger. At the end of the article, he says the mouse works fine on the pad without using batteries. It's an interesting [and funny] read, as is everything else on the site.
Slashdot Mirror
Bated breath, goddammit.
If you're on a machine with Unix man pages, try man perlfunc. On the plus side, it works even if your internet connection dies. It also works when you're on an airplane, frantically trying to meet project deadlines ;).
Typically it's unusual to see ``just a crash.'' Most programmes written in C and C++ crash due to buffer overflows, which frequently lead to running unsigned code. As a general rule, if a C or C++ code crashes, it is a fairly likely possibility to be able to run arbitrary code. Just because nobody's done it yet doesn't mean that it's not possible.
Quiet down old man.
Not only does he post outright lies about using a passive repeater in a really long wireless link, he didn't have the balls to admit that he was lying.
For cripe's sake, he even jumped on the Google bandwagon.
I've said it before and I'll say it again, take this gentleman's tech predictions with an extremely large grain of salt.
Sean Egan's Open Source Messaging Application Development: Building and Extending Gaim covers basic to intermediate GTK+ programming [with sockets, no less] quite nicely. He meticulously describes everything he does with numerous examples. He never skimps on the details and even gives insights into his own thinking on most problems.
It's a very, very good read, even for non-Gaim developers. On the plus side, you'll be supporting open-source development if you purchase it.
Will someone please mod this up? If you're using RDP to do nothing other than manage users and groups, restart services, or pretty much any typical Windows 2000 domain tasks, you're wasting tons of bandwidth.
For most everything you do, open up MMC and add a snap in for a remote computer. You can do everything remotely that you can do from the local ``Administrative Tools'' folder and it's a hell of a lot snappier [and less bandwidth-hungry] than a full RDP session.
Don't they teach this in MSCE or anything?
A few of the real hole in the wall Mexican places around where I live [in the Los Angeles area] sell bottled Mexican sodas [include Coke] with all the labels in Spanish.
They use real sugar and it isn't a dollar a bottle.
Search Ebay or ask around for IBM Thinkpad laptops. I've been able to get my hands on many, all for less than 50$. Obviously you aren't going to run Fedora or Windows XP on these. However, Windows 98 or NT [or 2000 even if you've got one with 32 megs of RAM or more] should run just spiffy on one of these. Obviously there are also the dozens of tiny Linux distros like DamnSmallLinux that work like a charm on hardware like this.
The older Thinkpads are built rock-solid and are great for debugging machines. Dead batteries aside, the only problem I've ever had was a screen that died on a model with a known-flaky screen [avoid the 365XD if you can]. I'm not sure how the newer Thinkpads are. I've never really used them since I've had no need to upgrade.
How about using RFID to FIND the damn keyring. I know it won't allow you to pinpoint exact location, but narrowing down the location of my keys or anything else small, portable, and easily lost to a single room would be immensely helpful. A certain amount of radio direction finding could possibly be used to help even further.
Yeah, I know there exists technology to make a little siren go off on the key ring or whatever, but let's stay on topic here.
This is the man who brought us the mathematically impossible 6.5 mile 802.11 link with a passive repeater. The repeater that he never showed to anybody. He also shows us an idealistic world of a community cable and telephone company that nobody's ever seemed to find evidence of.
Saying that, when it comes to technology at least, he is speculative is something of an understatement. Take what he says with an extremely large grain of salt.
You can speak badly of the president all you want in any public forum. That falls under the prime First Amendment right: freedom of speech. However, it's a whole different ballgame if you start making direct threats to the safety of the president.
Of course, I'm assuming you're talking about the United States. Other countries with a president as head of state may not have the same freedom of speech clauses in their governing documents.
As soon as you have a mesh network that large, you have to look into rewriting the routing protocols. The current mix we have [i.e., mainly OSPF and BGP] just wouldn't possibly scale to that many nodes unless you start introducing supernodes. With supernodes comes centralisation, and with centralisation you have a target for the FCC or any similar organisation.
That's not to say it isn't much easier to change supernodes when all the links are wireless. Also, the solution of more efficient routing protocols seems a little more realistic. But then you have to have a router in every house..
It's a lovely thought, but it's still a fair amount of time away.
And here I am surrounded by trees on all sides with their RF-absorbing foliage..
The old and multiple times dead Ricochet used to have repeaters at every major intersection, with supernodes with a T1 [or something to the effect] interspersed.
It's a great idea, but look where it got Ricochet.
Plus GPL'd tools like uClibc and Busybox [quite the pair] make it really easy to develop and build a system that is efficient with both memory and storage. Last I heard, all of the offerings from MS on this behalf are behemoths.
When a single kilobyte makes a difference, you're going to want the system that's already compact to begin with.
Not to mention the fact that you've got a system with a large embedded following and a huge user support community. Slap a few dollars into most underfunded developers' hands and they'll listen to your requests keenly, to boot.
Last I heard, Ethereal can do the same thing.
The BeOS car never made it to the race; it had been abandoned by its creators. However, in retrospect, it probably would have won.
While that may have been the case for older versions of SSH [I honestly don't know myself], modern versions of OpenSSH seem to have the client send the password in a whole packet. In others words, the client handles the input of the password and sends it over as one packet, instead of sending each character individually. I'm not sure if the parent was alluding to this fact or not, so I'm just clarifying.
Has anyone considered the implications of a DDoS involving a zombie army of machines with 100 mbit uplinks? This could spell disaster for just about everybody except those with the absolute fattest pipes. It takes an awful lot of hosts to swamp an OC3 now, but that's with hosts that rarely have a half megabit uplink, if that. It would be frighteningly easy to swamp the heavy links with a few 100 mbit links.
That is, of course, unless the bigger pipes grow at a rate proportional to the smaller ones. That also assumes symmetrical links for the home connections. Oh the irony of a 100 mbit / 128 kbit connection.
Not quite as versatile, but making up for that with geek factor is:
Ukulele Mindstorms Robot
It only plays reggae since reggae usually uses only a few, simple chords, but it is still way cool. They even made it remote controllable. And of course, all the source is on their site.
A lot of the Slashdot effect is also due to a huge number of new sockets being opened for the HTTP requests. This can take out a mid-quality machine even with a very fat pipe.
I'm surprised that nobody has mentioned the Linux Logical Volume Manager subsystem. It has many of the features of RAID arrays [such as spanning across multiple drives] with the added flexibility of being able to dynamically add [and theoretically remove] drives.
Unfortunately, aside from RAID'ing the volumes or something similar, I haven't been able to find any information on making the system redundant.
Read about it more on TLDP. It's a very robust system that works well on both servers and desktops.
The previous text is the interface to XP's default shell, cmd.exe.
This sounds like basically the same principle of inductive coils in the mouse pad and mouse that Afroman used in his wireless mouse charger. At the end of the article, he says the mouse works fine on the pad without using batteries. It's an interesting [and funny] read, as is everything else on the site.
Furthermore, I believe we should place a ban on electricity.