I agree that apt is awesome. But dselect? dselect's lameness is the reason for apt's existence
dselect is based on apt. apt is the base of the package system. What you're thinking of is probably console-apt, known to most through apt-get, the central control binary.
apt is a frontend to dpkg.
Sorta. See, there's only one program in the whole bunch that actually knows the dpkg file format, and that's dpkg-deb. Apt is really a system for updating files. Any file. Apt can be easily adapted to any group of files that lend themselves to being packaged. You could use apt to get the latest Slashdot stories, if you wanted. apt-get update to grab the headlines, apt-get install some_headline to install the comments... &c.
The apt system is pretty complex. It was hard for me to get my mind around. Hell, I've been using Debian for around a year now, and I just recently 'got it'. Most people never have reason to learn, so they don't. ---
I think companies would be more likely to use an MPL-style license than GPL. Under the MPL, they can keep some parts of their product closed, while letting other parts be developed by the community. It's really perfect for them. And it obviously works, too. Look how many contributors Mozilla has.
We'll just have to see what kind of product they end up with. ---
I think they mean standards for work contracted to themself. I don't think they're about to regulate hobbyist programmers. Then again, if they consider programming to be dangerous, they might. After all, you need a license to fly a model plane. Then again, that's directly physically dangerous, so I doubt we need to fear licensing now. ---
If GPLv3 requires internally used software to be released, then people who don't like that will just continue using GPLv2. Perhaps they'll make a modification to it, stating that future versions of the GPL cannot be used to relicense it.
But basically, if people don't like the new license, they won't use it, assuming they're not sheep. Perhaps that assumption is a bad one, but I'll stick with it until I see good evidence to the contrary. ---
I don't see any indication of regulation in the report. None at all. What are you talking about?
"I'm just wondering if it's possible too big a risk to take?" How can doing contract work for another customer possible screw things up? Horrors, I've just agreed to a contract that requires me to license that particular work under *insert license here*... oh wait, sounds like a usual contract job. ---
The article seemed a bit hysterical in tone, given the actual content of the report. Goal 3 isn't "agreeing upon a single common licensing agreement", but rather includes "the use of common licenses should be encouraged". There's a huge difference.
I think we can all agree that using common licenses is generally good. Specifically, it's good when our license is compatible with someone else's license, since then we can share/merge code etc. The government encouraging this behaviour isn't to be feared. Horrors! The fed wants us to be good neighbors! Oh wait, that's the point of OSS, isn't it...
Also, the committee may not have RMS as a member, but it does have ESR, one of the other big TLAs. Why didn't ESR share this upcoming review with the OSS community? Who knows. Perhaps we should ask him. As to why the report didn't mention him, well, perhaps because it's not about the history of OSS? Sure, RMS will continue to have influence, but he's definitely not important in the scheme of a fedgov strategy. They're talking about ideas, not specifics. People are specifics.
The committee included people from many different places, such as Michael Tiemann from Redhat, Tim O'Reilly from, well, O'Reilly... people from NCSA, Microsoft, Collab.net, NASA, DOE, the EU (international concerns, international committee, if only somewhat), LANL, SGI, NSA, Intel, IBM, MITRE, NSF, and many other great acronyms.
So the committee surely isn't omitting any great group, except perhaps "the common man", if such exists in OSS. And many of the names are recognizable even to me, and I've only been interested in OSS for a couple years, and hardly involved. "All the usual suspects", as the article says. No fears here.
All in all, the report looks like a Good Thing for the OSS community, on the whole. ---
"Yeah, well, I got a new electron the other day, with the latest in quantum interfaces, so now I've got around 500 terrs of storage, more than enough to store my century of MP3s." ---
I haven't seen the actual kit, but the way it's described in the article, it contains no binaries of GPL code. All it has is a 'processor', something on a level with, say, Pascal to C converters, except this one converts between system calls and kernel hooks rather than languages. It also contains the code of two GPL'd drivers written by Becker.
All of this is fine. Users can then take this, plus the code for Linux drivers, and create binaries they can use with their Solaris x86. This is also perfectly within the GPL. The only problem is if someone starts distributing one of those binaries without the source code to it.
Not Sun's problem. I mean, saying that this is `inviting infringement of the GPL' is like saying that a compiler does the same, because it allows the creation of binaries. Is my editor also inviting infringement, since I can create derivative works of GPL code, compile them, and distribute them without code? ---
MS also made various deals with vendors saying "We'll charge you hella lots for Windows if you ship your product with Netscape, but if you don't you get a discount. Capiche?" And the vendors did, and Netscape did not ship, and IE did gain much marketshare. So sayeth the logs of history. ---
Given as the Olympics doesn't matter, and this has provided 30 minutes of amusement (this and related links through this/. discussion) for me, I'd say that this is. ---
An MP3 does not have the capability to directly harm someone. A gun does.
You are obviously properly trained in the use of guns. But lets take someone who isn't properly trained in guns, and isn't properly trained in MP3s. They could harm someone with a gun, but not with an MP3. That's why guns aren't as safe as MP3s. ---
No, because it uses fewer bits for encryption. High enough to make near no difference for most people, but some people want the highest encryption possible, always. ---
I'm completely sure. I talked to other people. I got a message from here, then I sent a bunch, then I got another message, then she signed off.
I'm figuring it was an AIM bug myself. I had someone not receive a message again yesterday. They were disconnected from the network (physically), but not from AIM. ---
The server is stable enough, but there are no good clients yet. Gabber may be good, but I don't use Gnome. Jarl makes it rather difficult to create a new account.
There are, fortunately, a few libs, though they take some finding. Net::Jabber is in CPAN, but not on the sourceforge page. Jabberoo is available on its homepage, but not its sourceforge page nor as a download from www.jabber.org, ostensibly the development center. The dev doesn't appear to be terribly coordinated, though, judging by the traffic on the dev mailing list and the info available on the website.
There are a couple of Python modules available.. sorta.. but no documentation whatsoever.
There's Jabberbeans, for Java. I don't know anything about the state of that.
So is anyone actually using Jabber? What client are you using? ---
Last night, I was chatting on AIM with my girlfriend, when she stopped receiving messages. Well, ok, I'd been chatting earlier, and then she sent me a message, and I replied, but she never got it. I tried for an hour and 10 minutes, but she never got any messages. At that point she went to sleep.
This morning I was able to talk to her with Gaim, though.
Weird, eh? It looked to me like I was sending messages. No errors, but they just didn't get through.
You could also set up SSL telnet. It's probably not as secure as SSH, but mine does a 512-bit secure key exchange on connection (really need to make that be 1024). Not all clients support it, but if they do, you can do secure comm through telnet.
Of course, you could also just put ssh on port 23. If there's nothing there, nothing lost by putting ssh there.
Or you could do a one-time key thing with usernames/passwords then establish a secure connection after login if need be (I personally don't really care if someone reads my email/ICQ/AIM/IRC in transport, that being what I do through such connections anyways). Mmm, tunelling through telnet... ---
Who chose this guy? Oh wait, it's the folks who want Carnivore to get accepted, isn't it.
Shouldn't the people (and yes it should be people) who examine Carnivore be chosen by the people Carnivore is meant to examine? (no taxation without representation!;) I know I'd rather have hundreds of Open Sourcers examining it (even under NDA) than one guy chosen by them who used to work for DARPA, and thus obviously has the right attitude to be in the fed.
I wonder if I can find contact info for whoever's responsible... I doubt it, they're probably hiding like most people behind this sort of thing.
(random question: why doesn't work in the preview? Soon I'll know if it works in the comments too, but whether it works or not, it's a bug.) ---
Oooh, I HATE those sites that require you to have your browser 1000 pixels wide. Like some slashdot comment pages, even (why that's happening I don't know, but it's really annoying.)
I run in 1504x1128, but my browser windows are 800x800. I don't WANT to have a wider window. If you make me, I probably won't use your site as often. In fact, it has been driving me away from slashdot, which I used to spend far too much time on and now spend, well, a lot less. ---
Slashdot Mirror
Enjoy the pre-emptively created mirrors! Maybe they'll even be needed.
---
businesses sonet (OC3-OC48) services at very low rates.
Not so. After all, they need to pay their upstream provider for those bits.
---
I agree that apt is awesome. But dselect? dselect's lameness is the reason for apt's existence
dselect is based on apt. apt is the base of the package system. What you're thinking of is probably console-apt, known to most through apt-get, the central control binary.
apt is a frontend to dpkg.
Sorta. See, there's only one program in the whole bunch that actually knows the dpkg file format, and that's dpkg-deb. Apt is really a system for updating files. Any file. Apt can be easily adapted to any group of files that lend themselves to being packaged. You could use apt to get the latest Slashdot stories, if you wanted. apt-get update to grab the headlines, apt-get install some_headline to install the comments... &c.
The apt system is pretty complex. It was hard for me to get my mind around. Hell, I've been using Debian for around a year now, and I just recently 'got it'. Most people never have reason to learn, so they don't.
---
I think companies would be more likely to use an MPL-style license than GPL. Under the MPL, they can keep some parts of their product closed, while letting other parts be developed by the community. It's really perfect for them. And it obviously works, too. Look how many contributors Mozilla has.
We'll just have to see what kind of product they end up with.
---
I think they mean standards for work contracted to themself. I don't think they're about to regulate hobbyist programmers. Then again, if they consider programming to be dangerous, they might. After all, you need a license to fly a model plane. Then again, that's directly physically dangerous, so I doubt we need to fear licensing now.
---
Yes. In fact, it says so on the webpage that there are provisions for the insertion of corrective lenses (into the product).
---
If GPLv3 requires internally used software to be released, then people who don't like that will just continue using GPLv2. Perhaps they'll make a modification to it, stating that future versions of the GPL cannot be used to relicense it.
But basically, if people don't like the new license, they won't use it, assuming they're not sheep. Perhaps that assumption is a bad one, but I'll stick with it until I see good evidence to the contrary.
---
I don't see any indication of regulation in the report. None at all. What are you talking about?
"I'm just wondering if it's possible too big a risk to take?" How can doing contract work for another customer possible screw things up? Horrors, I've just agreed to a contract that requires me to license that particular work under *insert license here*... oh wait, sounds like a usual contract job.
---
Yeah, the article is broken. It's got the wrong URL on the link to the PDF.
---
The article seemed a bit hysterical in tone, given the actual content of the report. Goal 3 isn't "agreeing upon a single common licensing agreement", but rather includes "the use of common licenses should be encouraged". There's a huge difference.
I think we can all agree that using common licenses is generally good. Specifically, it's good when our license is compatible with someone else's license, since then we can share/merge code etc. The government encouraging this behaviour isn't to be feared. Horrors! The fed wants us to be good neighbors! Oh wait, that's the point of OSS, isn't it...
Also, the committee may not have RMS as a member, but it does have ESR, one of the other big TLAs. Why didn't ESR share this upcoming review with the OSS community? Who knows. Perhaps we should ask him. As to why the report didn't mention him, well, perhaps because it's not about the history of OSS? Sure, RMS will continue to have influence, but he's definitely not important in the scheme of a fedgov strategy. They're talking about ideas, not specifics. People are specifics.
The committee included people from many different places, such as Michael Tiemann from Redhat, Tim O'Reilly from, well, O'Reilly... people from NCSA, Microsoft, Collab.net, NASA, DOE, the EU (international concerns, international committee, if only somewhat), LANL, SGI, NSA, Intel, IBM, MITRE, NSF, and many other great acronyms.
So the committee surely isn't omitting any great group, except perhaps "the common man", if such exists in OSS. And many of the names are recognizable even to me, and I've only been interested in OSS for a couple years, and hardly involved. "All the usual suspects", as the article says. No fears here.
All in all, the report looks like a Good Thing for the OSS community, on the whole.
---
Yes, but they aren't storing analog information.
---
"Yeah, well, I got a new electron the other day, with the latest in quantum interfaces, so now I've got around 500 terrs of storage, more than enough to store my century of MP3s."
---
... does not exist, from the sounds of it.
I haven't seen the actual kit, but the way it's described in the article, it contains no binaries of GPL code. All it has is a 'processor', something on a level with, say, Pascal to C converters, except this one converts between system calls and kernel hooks rather than languages. It also contains the code of two GPL'd drivers written by Becker.
All of this is fine. Users can then take this, plus the code for Linux drivers, and create binaries they can use with their Solaris x86. This is also perfectly within the GPL. The only problem is if someone starts distributing one of those binaries without the source code to it.
Not Sun's problem. I mean, saying that this is `inviting infringement of the GPL' is like saying that a compiler does the same, because it allows the creation of binaries. Is my editor also inviting infringement, since I can create derivative works of GPL code, compile them, and distribute them without code?
---
MS also made various deals with vendors saying "We'll charge you hella lots for Windows if you ship your product with Netscape, but if you don't you get a discount. Capiche?" And the vendors did, and Netscape did not ship, and IE did gain much marketshare. So sayeth the logs of history.
---
Yesterday, when that update was posted on k5, that paragraph was actually there. And damnit, I wanna know! Mmm, fresh knowlege.
---
Given as the Olympics doesn't matter, and this has provided 30 minutes of amusement (this and related links through this /. discussion) for me, I'd say that this is.
---
What's the name of that Samba browser? I haven't seen anything good in that area...
---
An MP3 does not have the capability to directly harm someone. A gun does.
You are obviously properly trained in the use of guns. But lets take someone who isn't properly trained in guns, and isn't properly trained in MP3s. They could harm someone with a gun, but not with an MP3. That's why guns aren't as safe as MP3s.
---
No, because it uses fewer bits for encryption. High enough to make near no difference for most people, but some people want the highest encryption possible, always.
---
I'm completely sure. I talked to other people. I got a message from here, then I sent a bunch, then I got another message, then she signed off.
I'm figuring it was an AIM bug myself. I had someone not receive a message again yesterday. They were disconnected from the network (physically), but not from AIM.
---
The server is stable enough, but there are no good clients yet. Gabber may be good, but I don't use Gnome. Jarl makes it rather difficult to create a new account.
There are, fortunately, a few libs, though they take some finding. Net::Jabber is in CPAN, but not on the sourceforge page. Jabberoo is available on its homepage, but not its sourceforge page nor as a download from www.jabber.org, ostensibly the development center. The dev doesn't appear to be terribly coordinated, though, judging by the traffic on the dev mailing list and the info available on the website.
There are a couple of Python modules available.. sorta.. but no documentation whatsoever.
There's Jabberbeans, for Java. I don't know anything about the state of that.
So is anyone actually using Jabber? What client are you using?
---
Last night, I was chatting on AIM with my girlfriend, when she stopped receiving messages. Well, ok, I'd been chatting earlier, and then she sent me a message, and I replied, but she never got it. I tried for an hour and 10 minutes, but she never got any messages. At that point she went to sleep.
This morning I was able to talk to her with Gaim, though.
Weird, eh? It looked to me like I was sending messages. No errors, but they just didn't get through.
I'm using v0.9.20.
---
You could also set up SSL telnet. It's probably not as secure as SSH, but mine does a 512-bit secure key exchange on connection (really need to make that be 1024). Not all clients support it, but if they do, you can do secure comm through telnet.
Of course, you could also just put ssh on port 23. If there's nothing there, nothing lost by putting ssh there.
Or you could do a one-time key thing with usernames/passwords then establish a secure connection after login if need be (I personally don't really care if someone reads my email/ICQ/AIM/IRC in transport, that being what I do through such connections anyways). Mmm, tunelling through telnet...
---
Who chose this guy? Oh wait, it's the folks who want Carnivore to get accepted, isn't it.
Shouldn't the people (and yes it should be people) who examine Carnivore be chosen by the people Carnivore is meant to examine? (no taxation without representation! ;) I know I'd rather have hundreds of Open Sourcers examining it (even under NDA) than one guy chosen by them who used to work for DARPA, and thus obviously has the right attitude to be in the fed.
I wonder if I can find contact info for whoever's responsible... I doubt it, they're probably hiding like most people behind this sort of thing. (random question: why doesn't work in the preview? Soon I'll know if it works in the comments too, but whether it works or not, it's a bug.)
---
Oooh, I HATE those sites that require you to have your browser 1000 pixels wide. Like some slashdot comment pages, even (why that's happening I don't know, but it's really annoying.)
I run in 1504x1128, but my browser windows are 800x800. I don't WANT to have a wider window. If you make me, I probably won't use your site as often. In fact, it has been driving me away from slashdot, which I used to spend far too much time on and now spend, well, a lot less.
---