The chroot-jail based hosting system I used had a number of problematic limitations: I couldn't install any software that needed to run as root (e.g. to open ports below 1024), I couldn't change configuration of a variety of services that would open security problems for the hosting provider (e.g. the mail server, I think they were using exim), users running processes on the same server that consumed too much CPU and or memory had too much negative impact on my server's performance.
Of those, only the last is relevant to FreeBSD jail setups. If I created a jail for you and gave you root, you would be root, full stop. The only things you could do would be install your own kernel (since only one kernel - that of the host OS - is running). We use them to virtualize multiple distinct systems on the same hardware, with the idea that the mailserver always runs under a light load and doesn't interfere with the database server hosted on the same machine.
There's a decent Wikipedia article on the subject, even if it kind of comes across like an advertisement. In short, it sounds like your hosting provider ran a bad server. Don't extrapolate their incompetence to the general state of the art.
VMWare ESX server *can* share memory and free disk space between instances. Multiple instances can even have the same page of physical memory mapped into them, so running multiple identical servers doesn't actually take more memory than it does on a single instance (although there are overheads).
Assuming a strategy like copy-on-write, I can understand how two instances started from the same configuration could begin with most of their memory shared. However, it seems like that would eventually become a tiny percentage of their actual address space as processes start and die, allocate memory and free it, etc. I mean, to the best of my knowledge, when Unix fork()s a process, it doesn't keep track of when it can later re-merge the address space of the parent and child. They may start as identical copies, but pretty soon their data segments will be completely different. In the case where each process is actually a virtualized system where the data segment is hugely bigger than the shared code, I'd think that would happen pretty quickly.
Also there's the problem of the service. In the above exemple, in the current states of affair, forcing to release code for something that is remotely ran, and never distributed, can seem rediculous. But once again the point is to protect end-user, from a potential future, where almost everything is rna remotely, and nobody has an access to anything because as those services aren't distributed, the requirements for code doesn't apply.
The thing is that users running applications remotely has been the norm for decades, and throughout the entire life of GPLv2 no one has claimed that users have the right to the source code for those applications. The idea of equating "execution" with "distribution" would have been preposterous. Huge chunks of infrastructure have been built around this arrangement. What magic switch was thrown that made everyone decide that web services are a new animal with different expectations?
You run virtualised web servers because 99.9% of all web servers are idle at any given time. So you put 100 on a server.
If you have a real need to run 100 separate Apache instances, then you'll want something much higher-level than VMWare. For us, that would be a FreeBSD jail, where each instance would get its own chrooted home directory and IP address. That way, you're not allocating resources to 100 little-used OS images; each shares from the same memory and hard drive pool. Jails are slightly limited in that I'd like a way to limit CPU and memory allocation, but in practical application this really works very well today.
But it was written after the rise of console apps. To the best of my knowledge, if I hack a GPLv2 text app and you SSH into my server to run it, I don't have to give you the modified source. I'm not distributing the app to you - I'm just letting you execute it. In what way is that essentially different from running an application via a web service?
As somebody who worked at various helpdesks for a few years as a phone monkey, this is SOP with any company.
As somebody who's bought stuff for a few years as a customer, I don't really give a rat's ass about your SOP. My warranty says that if it breaks, you'll fix it. Any conflicts between that legal document and your internal policy are your problem to deal with and completely uninteresting to me.
You don't need to learn every modern text editor separately, because they use keyboard shortcuts and controls that are consistent across them all, and elsewhere in the system, as others pointed out.
Since we're talking about Mac programs, sit down in front of any Mac and click on any control that allow you to enter text, whether single or multi-line. Press ^A to go to the begining of a line. Press ^E to go to the end.
Those are Emacs keybindings. Maybe it's not quite as nonstandard as you'd believe.
A native Aqua GUI text editors for OSX with all the features of TextMate released with a GPL license-- I'd love to see it.
After recently getting a Mac, Aquamacs was my first major addition. It's Aqua, GPL, and is pretty much guaranteed to support a superset of features of any other text editor.
I think the last few seasons have been karmic payback for anything we might have done wrong in the past.
Re:To advance, correct errors rather than drag it.
on
The Future of the PSP
·
· Score: 1
Hello? I said MORE not less. Either you increase the price with more features or remove features (hard to do since the PSP already has a game library that relies on all those features) to cut price.
And yet the DS with two screens and a touch sensor costs half as much and still turns a profit on each unit, which is especially sweet since they're outselling the overpriced PSP by a wide margin.
Either Nintendo has a warp gate to a magical cheap manufacturing planet, or Sony just can't shake its desire to build over-specced equipment out of expensive, proprietary parts.
They have been removed from KinderStart's search engine, in violation of their first amendment right to free speech!
Note that KinderStart isn't too proud to show Google Ads, though.
Frankly, if I were in Google's ad department, I'd be easing things like "meet sexy preteens!" into that particular rotation. After all, the site is supposed to be all about kids, right? Then surely the hyper-protective parents who'd visit would want to find more of them. KinderStart.com would be a perfect portal domain for everyone who loves children, and maybe Google could bury the hatchet by helping them along that path.
Imagine if the US News Media had to abide by such rules
I can easily imagine a system where the news media had to report on issues that the government wanted it to report on. Freedom of the press is for those commie "glasnost" whiners.
You can't remove any copyright, patent, or atribution notices. Kind of like the dreaded BSD advertising clause, in that if someone puts "Parts written by 1337 h4xx0rz" in the output of the program, you have to leave it there. Repeat ad nauseum for every contributor that jumps on the bandwagon, and things could get... unaesthetic.
They use almost the exact same patent control system as the GPLv3. If a program contains patented code, you're granted permission to use those patents to execute it. If you sue one of the patent holders for violations of your own patent, that permission is revoked. I think this is called the "please don't eat me, IBM!" clause. Seriously, though, this needs to be pointed out every single time some Microsoft shill attacks the GPLv3. You can dislike v3, but you can't really call it anti-business when the world's largest software vendor implemented parts of it in their own license.
You can't remove any copyright, patent, or atribution notices. Kind of like the dreaded BSD advertising clause, in that if someone puts "Parts written by 1337 h4xx0rz" in the output of the program, you have to leave it there. Repeat ad nauseum for every contributor that jumps on the bandwagon, and things could get... unaesthetic.
They use almost the exact same patent control system as the GPLv3. If a program contains patented code, you're granted permission to use those patents to execute it. If you sue one of the patent holders for violations of your own patent, that permission is revoked. I think this is called the "please don't eat me, IBM!" clause. Seriously, though, this needs to be pointed out every single time some Microsoft shill attacks the GPLv3. You can dislike v3, but you can't really call it anti-business when the world's largest software vendor implemented parts of it in their own license.
What he was saying, though, is that this isn't some nebulous group doing bad things on its own volition. Rather than RIAA, we should just say "the major record companies" since that's who we all know we're talking about. The distinction is that it reminds everyone of exactly who we should be angry with, regardless of what they might decide to change their front organization's name to in the future.
Who justified piracy? I don't advocate it, but I don't think college kids need to be bankrupted because they gave an MP3 to a friend. Furthermore, you seemed to miss the part about having received replies. I don't usually bother replying to emails that I'm just going to delete, but maybe that's just me.
Here's the letter I wrote to the president of UNL, the chancellor, each of the regents, and the CIO (Mr. Weir):
College is the time for many children to grow into adults by learning to
make their own decicions, often for the first time in their lives. Access
to information about those decisions is important, and the ability to get it
anonymously is critical. No boy wants his friends to know that he was
contemplating suicide. No girls wants to be seen asking about sexually
transmitted diseases. Unless they know that their questions can't be
tracked back to them, most kids won't ask them.
Walter Weir's IT policies contribute to the atmosphere of open learning that
a university, of all places, should strive to attain. Now, some group from
Hollywood wants UNL to overhaul its computer network for the explicit
purpose of destroying that, simply to serve ends that the school has no real
reason to care about. As a computer scientist and a Nebraska taxpayer, I
have these additional problems with their request:
1) I am not interested in seeing my money used to persecute kids for trading
songs, much as you and I traded tapes with our friends when we were younger.
2) IP addresses are traceable to computers, not people. If two or more kids
share a computer, who gets the cease-and-desist notice? The RIAA has a
history of doing asinine things like suing dead grandmothers; yes, that
really happened. I'd much rather see UNL say that their requests can't be
answered than to get involved in such foolish and expensive unpleasantry.
3) Again, the current system works and I see no reason to change it to
benefit one outside group with dubious interests. Computer networks are
hard to build, and harder to build well. Mr. Weir's department has done a
fine job and he should not be made to enact its destruction.
Just say no to the RIAA. We have a system that serves us - the citizens,
taxpayers, and students of Nebraska - very well. UNL's network is meant for
the education and personal growth of its users. It is not meant to be the
unpaid police force for an outside party with no concern for our needs.
I got several replies of agreement, and I think that the school will be holding its ground.
I have written huge DBMS applications both native 100% FoxPro (and Foxbase before that) and with SQL backends for 20 years. I get blazingly fast performance even with tables with over a billion records and over a TB in size.
The most impressive part of your claim is that you fit a billion rows and a TB of data into a table that
only supports 1 billion rows and 2GB of data. Congratulations! I bow to your superior ability to exceed the boundaries of software design.
Unless, of course, you were referring to database (not FoxPro) tables that size, in which case I fail to see how FoxPro's (lack of) performance even enters into the equation. Are you really claiming that FoxPro's database library bindings are faster than other languages'?
We picked PostgreSQL for speed and for its generally better track record for protecting data. We haven't really picked a widget toolkit yet, although we now have some Python+wx apps in production. I'm personally hoping that QT4's Python bindings will be easy to develop with.
Slashdot Mirror
Of those, only the last is relevant to FreeBSD jail setups. If I created a jail for you and gave you root, you would be root, full stop. The only things you could do would be install your own kernel (since only one kernel - that of the host OS - is running). We use them to virtualize multiple distinct systems on the same hardware, with the idea that the mailserver always runs under a light load and doesn't interfere with the database server hosted on the same machine.
There's a decent Wikipedia article on the subject, even if it kind of comes across like an advertisement. In short, it sounds like your hosting provider ran a bad server. Don't extrapolate their incompetence to the general state of the art.
Assuming a strategy like copy-on-write, I can understand how two instances started from the same configuration could begin with most of their memory shared. However, it seems like that would eventually become a tiny percentage of their actual address space as processes start and die, allocate memory and free it, etc. I mean, to the best of my knowledge, when Unix fork()s a process, it doesn't keep track of when it can later re-merge the address space of the parent and child. They may start as identical copies, but pretty soon their data segments will be completely different. In the case where each process is actually a virtualized system where the data segment is hugely bigger than the shared code, I'd think that would happen pretty quickly.
The thing is that users running applications remotely has been the norm for decades, and throughout the entire life of GPLv2 no one has claimed that users have the right to the source code for those applications. The idea of equating "execution" with "distribution" would have been preposterous. Huge chunks of infrastructure have been built around this arrangement. What magic switch was thrown that made everyone decide that web services are a new animal with different expectations?
That's mostly true. Aren't good cross-platform toolkits spiffy?
Because QT 3 isn't available under GPL for Windows or Mac, while QT 4 is. Next question?
If you have a real need to run 100 separate Apache instances, then you'll want something much higher-level than VMWare. For us, that would be a FreeBSD jail, where each instance would get its own chrooted home directory and IP address. That way, you're not allocating resources to 100 little-used OS images; each shares from the same memory and hard drive pool. Jails are slightly limited in that I'd like a way to limit CPU and memory allocation, but in practical application this really works very well today.
But it was written after the rise of console apps. To the best of my knowledge, if I hack a GPLv2 text app and you SSH into my server to run it, I don't have to give you the modified source. I'm not distributing the app to you - I'm just letting you execute it. In what way is that essentially different from running an application via a web service?
As somebody who's bought stuff for a few years as a customer, I don't really give a rat's ass about your SOP. My warranty says that if it breaks, you'll fix it. Any conflicts between that legal document and your internal policy are your problem to deal with and completely uninteresting to me.
Since we're talking about Mac programs, sit down in front of any Mac and click on any control that allow you to enter text, whether single or multi-line. Press ^A to go to the begining of a line. Press ^E to go to the end.
Those are Emacs keybindings. Maybe it's not quite as nonstandard as you'd believe.
After recently getting a Mac, Aquamacs was my first major addition. It's Aqua, GPL, and is pretty much guaranteed to support a superset of features of any other text editor.
I think the last few seasons have been karmic payback for anything we might have done wrong in the past.
And yet the DS with two screens and a touch sensor costs half as much and still turns a profit on each unit, which is especially sweet since they're outselling the overpriced PSP by a wide margin.
Either Nintendo has a warp gate to a magical cheap manufacturing planet, or Sony just can't shake its desire to build over-specced equipment out of expensive, proprietary parts.
Pretty much, yeah.
Note that KinderStart isn't too proud to show Google Ads, though.
Frankly, if I were in Google's ad department, I'd be easing things like "meet sexy preteens!" into that particular rotation. After all, the site is supposed to be all about kids, right? Then surely the hyper-protective parents who'd visit would want to find more of them. KinderStart.com would be a perfect portal domain for everyone who loves children, and maybe Google could bury the hatchet by helping them along that path.
I can easily imagine a system where the news media had to report on issues that the government wanted it to report on. Freedom of the press is for those commie "glasnost" whiners.
I noticed two main things in that license text:
You can't remove any copyright, patent, or atribution notices. Kind of like the dreaded BSD advertising clause, in that if someone puts "Parts written by 1337 h4xx0rz" in the output of the program, you have to leave it there. Repeat ad nauseum for every contributor that jumps on the bandwagon, and things could get... unaesthetic.
They use almost the exact same patent control system as the GPLv3. If a program contains patented code, you're granted permission to use those patents to execute it. If you sue one of the patent holders for violations of your own patent, that permission is revoked. I think this is called the "please don't eat me, IBM!" clause. Seriously, though, this needs to be pointed out every single time some Microsoft shill attacks the GPLv3. You can dislike v3, but you can't really call it anti-business when the world's largest software vendor implemented parts of it in their own license.
I noticed two main things in that license text:
You can't remove any copyright, patent, or atribution notices. Kind of like the dreaded BSD advertising clause, in that if someone puts "Parts written by 1337 h4xx0rz" in the output of the program, you have to leave it there. Repeat ad nauseum for every contributor that jumps on the bandwagon, and things could get... unaesthetic.
They use almost the exact same patent control system as the GPLv3. If a program contains patented code, you're granted permission to use those patents to execute it. If you sue one of the patent holders for violations of your own patent, that permission is revoked. I think this is called the "please don't eat me, IBM!" clause. Seriously, though, this needs to be pointed out every single time some Microsoft shill attacks the GPLv3. You can dislike v3, but you can't really call it anti-business when the world's largest software vendor implemented parts of it in their own license.
Right now, I'm working around that by paying for satellite radio. I still hear enough new music that I actually like to keep me interested.
The day that ClearChannel overtakes Sirius' programming, I'm smashing my receiver with a hammer. No, I'm not exaggerating.
I'm a planning type. Implementation details like that are for the junior programmers.
What he was saying, though, is that this isn't some nebulous group doing bad things on its own volition. Rather than RIAA, we should just say "the major record companies" since that's who we all know we're talking about. The distinction is that it reminds everyone of exactly who we should be angry with, regardless of what they might decide to change their front organization's name to in the future.
Who justified piracy? I don't advocate it, but I don't think college kids need to be bankrupted because they gave an MP3 to a friend. Furthermore, you seemed to miss the part about having received replies. I don't usually bother replying to emails that I'm just going to delete, but maybe that's just me.
Here's the letter I wrote to the president of UNL, the chancellor, each of the regents, and the CIO (Mr. Weir):
I got several replies of agreement, and I think that the school will be holding its ground.
GO HUSKERS!
Yep, I already forwarded the link on to coworkers. :-)
The most impressive part of your claim is that you fit a billion rows and a TB of data into a table that only supports 1 billion rows and 2GB of data. Congratulations! I bow to your superior ability to exceed the boundaries of software design.
Unless, of course, you were referring to database (not FoxPro) tables that size, in which case I fail to see how FoxPro's (lack of) performance even enters into the equation. Are you really claiming that FoxPro's database library bindings are faster than other languages'?
We picked PostgreSQL for speed and for its generally better track record for protecting data. We haven't really picked a widget toolkit yet, although we now have some Python+wx apps in production. I'm personally hoping that QT4's Python bindings will be easy to develop with.
Oh, I know, and I do the same thing all the time. But please understand that it's a moral imperative to point these things out; nothing personal.