You're ignoring his main point, that they are under no obligation to pay when they receive this 'bill.' So either way you go, it'll end up in court eventually. As it should.
Google only shares personal information with other companies or individuals outside of Google in the following limited circumstances: [...] We have a good faith belief that access, use, preservation or disclosure of such information is reasonably necessary to (a) satisfy any applicable law, regulation, legal process or enforceable governmental request, (b) enforce applicable Terms of Service, including investigation of potential violations thereof, (c) detect, prevent, or otherwise address fraud, security or technical issues, or (d) protect against imminent harm to the rights, property or safety of Google, its users or the public as required or permitted by law.
This really shouldn't be all that crippling for the individuals involved. It appears that it was a corporation. The corporation is therefore liable, not the individuals involved. Corp goes bankrupt, liquidates, and everyone goes on with their lives. It's not a financial death sentence for the officers, etc.
It is generally possible to negotiate much better rates for high-bandwidth downstream companies than upstream. This means that ISPs, if they've negotiated properly, should be getting cheaper rates/mbit than a someone with the same bandwidth commitment upstream.
The only time PostgreSQL will be slower, generally, is in a situation where you have frequent connections/disconnections. Connection pooling greatly benefits pgsql. However, now there's pgBouncer which alleviates this.
In general, as far as query planning and such, it will outperform MySQL.
Dell XPS 410 (gaming) is eerily silent. I didn't expect it, it was quite a shock. It uses a BTX motherboard and large fans, which are probably the reasons for it, but it is quieter than my PowerMac G4, my new iMac and my macbook pro. Pretty crazy.
OS X and BSD should absolutely not be counted as Linux installations, because they're not Linux (nor are they even close)! I would imagine that including those as part of Linux installs would be quite insulting to the many people who have worked on all of those projects, including the Linux kernel.
From a developer perspective, MacPorts is a very nice system. Creating a port for MacPorts is much easier than doing so for FreeBSD, an RPM for RedHat, or a deb I've found. They've done a great job at removing some of the tedium from creating packages.
BTW - MacPorts also has a cool feature that will make a.pkg or.mpkg (the graphical installers) for any port you wish. Nice for redistribution. It'll package dependencies too.
FTOS is NetBSD-based, but it doesn't allow shell access like JunOS does. At least, not that I've seen. Juniper and Force10 both make fantastic equipment in their respective segments. The Juniper CLI, though, I think is the best around. For example, I wish they all used 'less' as their pager:-)
Hybrid drives sound like a much easier solution for most people. Let's say we had a hybrid drive with 4GB of solid-state storage (can't be that far off, can it?), the benefit here is that the drivers or hardware handles the tiering of your storage/data for you. If you're looking for performance and you have much more than 64GB of data, for example, the hybrid drive I think will do better than the solid-state/separate hard drive combo in the long run. I know I'd much rather have an LRU cache handled for me rather than me trying to create something similar by trying to manually place data on the correct storage tier.
The other option, of course, is to have a filesystem that does this for you. Give it your fast storage and your slow storage with costs assigned to each and have it manage your storage like a cache. This is what would make the 64GB storage+large hard drive a superior option in my mind mind. There are filesystems that already do this (Sun has one, for example), but I think it may be time for Microsoft and/or Apple to implement something like that for the home user as well.
Python's syntax is prettier to most people, but its library is more inconsistent. Ruby has the most consistent standard library that I know. I've been using it since 2000.
I'll also add that a rootkit isn't traditionally an exploit at all. It's usually a backdoor or series of backdoors (sometimes as well as tools/kernel mods to hide itself) that allow an attacker further or continued access to a system that he has exploited in some other way. So it's not what you use to get in the first time, but the times following.
That's like saying: "Oh, this is just a race condition, nothing new here . .." These are incredibly broad classes of exploits. And what he's talking about, while a race condition, is almost its own class of exploit because of how broad its scope is.
He mentioned the libc calls already. Those calls will use whatever method is appropriate for obtaining user information. On a default installation, this will be reading/etc/passwd.
Slashdot Mirror
For those that haven't seen it, djb threw up some information regarding this problem and various options a few years ago.
http://cr.yp.to/djbdns/forgery.html
There's a surprising number of providers that don't do egress source filtering. I definitely wouldn't rely on other peoples' security.
Not just now in California. It's been that way for a while. The judge just reaffirmed what everyone already knew.
You're ignoring his main point, that they are under no obligation to pay when they receive this 'bill.' So either way you go, it'll end up in court eventually. As it should.
Is slashdot turning into digg? Whether or not you agree with the parent, he/she isn't a troll..
Unfortunately . . . .
Google only shares personal information with other companies or individuals outside of Google in the following limited circumstances:
[...]
We have a good faith belief that access, use, preservation or disclosure of such information is reasonably necessary to (a) satisfy any applicable law, regulation, legal process or enforceable governmental request, (b) enforce applicable Terms of Service, including investigation of potential violations thereof, (c) detect, prevent, or otherwise address fraud, security or technical issues, or (d) protect against imminent harm to the rights, property or safety of Google, its users or the public as required or permitted by law.
^D? Wrong control code..
It does the latter. Also look at freebsd-update written by the same guy and also included in the base system.
except fast.. apparently
This really shouldn't be all that crippling for the individuals involved. It appears that it was a corporation. The corporation is therefore liable, not the individuals involved. Corp goes bankrupt, liquidates, and everyone goes on with their lives. It's not a financial death sentence for the officers, etc.
It is generally possible to negotiate much better rates for high-bandwidth downstream companies than upstream. This means that ISPs, if they've negotiated properly, should be getting cheaper rates/mbit than a someone with the same bandwidth commitment upstream.
Details?! Is it specific to certain developments that sign on? The site is seems of devoid of that kind of information.
The only time PostgreSQL will be slower, generally, is in a situation where you have frequent connections/disconnections. Connection pooling greatly benefits pgsql. However, now there's pgBouncer which alleviates this.
In general, as far as query planning and such, it will outperform MySQL.
You should consider using schemas as opposed to entirely separate databases.
Dell XPS 410 (gaming) is eerily silent. I didn't expect it, it was quite a shock. It uses a BTX motherboard and large fans, which are probably the reasons for it, but it is quieter than my PowerMac G4, my new iMac and my macbook pro. Pretty crazy.
OS X and BSD should absolutely not be counted as Linux installations, because they're not Linux (nor are they even close)! I would imagine that including those as part of Linux installs would be quite insulting to the many people who have worked on all of those projects, including the Linux kernel.
From a developer perspective, MacPorts is a very nice system. Creating a port for MacPorts is much easier than doing so for FreeBSD, an RPM for RedHat, or a deb I've found. They've done a great job at removing some of the tedium from creating packages.
.pkg or .mpkg (the graphical installers) for any port you wish. Nice for redistribution. It'll package dependencies too.
BTW - MacPorts also has a cool feature that will make a
FTOS is NetBSD-based, but it doesn't allow shell access like JunOS does. At least, not that I've seen. Juniper and Force10 both make fantastic equipment in their respective segments. The Juniper CLI, though, I think is the best around. For example, I wish they all used 'less' as their pager :-)
Hybrid drives sound like a much easier solution for most people. Let's say we had a hybrid drive with 4GB of solid-state storage (can't be that far off, can it?), the benefit here is that the drivers or hardware handles the tiering of your storage/data for you. If you're looking for performance and you have much more than 64GB of data, for example, the hybrid drive I think will do better than the solid-state/separate hard drive combo in the long run. I know I'd much rather have an LRU cache handled for me rather than me trying to create something similar by trying to manually place data on the correct storage tier.
The other option, of course, is to have a filesystem that does this for you. Give it your fast storage and your slow storage with costs assigned to each and have it manage your storage like a cache. This is what would make the 64GB storage+large hard drive a superior option in my mind mind. There are filesystems that already do this (Sun has one, for example), but I think it may be time for Microsoft and/or Apple to implement something like that for the home user as well.
Python's syntax is prettier to most people, but its library is more inconsistent. Ruby has the most consistent standard library that I know. I've been using it since 2000.
I'll also add that a rootkit isn't traditionally an exploit at all. It's usually a backdoor or series of backdoors (sometimes as well as tools/kernel mods to hide itself) that allow an attacker further or continued access to a system that he has exploited in some other way. So it's not what you use to get in the first time, but the times following.
That's like saying: "Oh, this is just a race condition, nothing new here . . ." These are incredibly broad classes of exploits. And what he's talking about, while a race condition, is almost its own class of exploit because of how broad its scope is.
For what it's worth, Dungeon Runners (a semi-free MMO) is like this and can be somewhat amusing for a few minutes :P
/etc/passwd can be read by any user
He mentioned the libc calls already. Those calls will use whatever method is appropriate for obtaining user information. On a default installation, this will be reading /etc/passwd.