Well, there are a couple of benefits to using Hurd:
1) Hack value. It's fun to play.
2) Secure infrastructure that permits to user to do quite reasonable things.
In the Hurd, any user can create a secure chroot jail. Any user can mount a remote file system off of their home directory. Any user could mount a file (ISO or whatever) as a filesystem.
3) Translators
The concept of a translator is the "Everything is a file" taken a step further. If you imbed a program in an inode, it can then deal with filesystem calls to that inode as it sees fit.
This means that things like the Linux "proc" file system are easy to write, and could trivially show things like SQL data, or an FTP-fs.
Device drivers can either be colocated in the kernel (Which they generally are now, similar to Linux), or can be emulated in user space (Like an implementation of/dev/random that I've seen based on EGD)
This means that implementing PPPoE is going to be very simple for us, since we just need to write a simple translator the reads from the Ethernet card, and pushes those packets into the IP stack.
These are just beginning sort of examples of what's possible. Other clever stuff is doable if you're willing to think a little bit outside of the typical "unix" box.
Most ideas that come out of life aren't worth persuing. The great part is to see that some people do attempt them, and provide a nice learning experience for the rest of us.
In the corporate world, you have to rely on yourself and your coworkers for your scope of experience.
In the Free Software world, I have Torvalds, Cox, Stallman, Raymond, etc. as part of my scope of experience. That's really powerful.
I suspect you weren't looking at the latest "Milestone", since it *is* called Mozilla 0.8.
If you're looking at M18, it was truly a bad product.
The only time I need to use NS4.7 now is when I want to view source on a generated page. Mozilla's view source polls the server for the HTML, but doesn't bother to supply POST parameters. (In most cases a second instance of the POST being sent would be bad anyway).
I heard an urban legend once about a microsoft seminar where they were announcing they they had a fully complient Korn shell available with their operating system. A gentleman in the audience stood up to tell them in exactly what ways it wasn't compliant and promptly got into an argument with the Microsoft VP. Apparently the person in the audience was David Korn. Is this true?
I used a flexpro keyboard for years and loved it. One of my coworkers still uses one at his home. I tend to destroy about 2 keyboards a year, and it not only helped my wrists, but it prevented me from pounding on the keyboard.
Having done quite a bit of CSS-P, it's actually easier to do once you bend your mind to it. Table driven sites are an endless mess of tags inside your HTML code making it quite unreadable. Instead, just wrap the whole thing in a DIV tag, and use the positioning code to place it appropriately.
This means that you can hand the page to some junior HTMLer and know that the page will still have a chance of validating and looking right when you get it back. You can also simply hand it to a secretary with a comment of "Don't touch the angle brackets", and they can easily update the text on a page and have it work.
The trickey bit is writing your page in a way that makes sense when it's flat. Because of the positioning code, The DIVs can appear in any order. When I'm working on a site like this, I try to look at the page in lynx. When I'm happy with it there, I'll move the elements around the page and check it with Mozilla. Lynx never even sees the positioning code.
I suspect by the time IE6 and Mozilla 1.0 come out, that the worst of the positioning bugs will have gone away so that it's actually cross platform usable.
I've always thought that Hurd had an interesting potential to become a good gaming machine. Don't plan on using network? Unload the driver! That way you literally only have loaded exactly what you need to play.
That was *my* box you guys did in a year ago... =) I changed the address as quickly as I could to save my poor development system.
For an update, apache still compiles nicely on it, and recently many people got X running.
As to the Hurd web pages, I've taken over development on them and you can generally expect there to be updates to them starting this weekend. I'd appreciate any assistance people want to give. (web-hurd at gnu.org)
Another good place to go is http://www.debian.org/ports/hurd/
If you haven't heard this before, then you haven't been reading slashdot for long. This type of fear mongering is quite common when people talk about IPv6. The *recommended* way to generate an IPv6 address is through your MAC address. You're still welcome to assign them by hand if you so choose. Also, almost every Ethernet NIC can have its MAC address overridden.
The poster apparently hasn't been following slashdot either...
I disagree. It's useful to ask him something if you want a well-thought out *moral* response to a problem.
Each of us needs to have a set of values to work from, and decide when it's appropriate to push these values. I choose to write Free Software during my evenings and weekends, and I choose to volunteer to the GNU project as a sysadmin because the work that they do supports my values. However, in order to pay the bills, I develop proprietary software that is covered by IP patents. I make a concious choice to do this, and it's important that I recognise that this is very much a choice issue. At my work, I have convinced my employers a number of times to support Free software instead of writing proprietary because it was a better business case.
I value the advice that RMS has given me in the past, even though I may have chosen not to follow it. It allows me an outside perspective on a problem, and frequently has led me to what I've felt is a clean compromise.
The Lesser GPL is also used when there's a non-Free library that does the same thing. Glibc needed to be under this license historically since it was being positioned as a replacement native libcs on various Unix systems. If it were not under the LGPL now, it would be quickly replaced. However, a number of packages like libregex are under the GPL license, because it gives a competitive advantage to GPL software. GNU developers are encouraged to put as much as we can under the GPL to strengthen the movement.
The testing suite that I have been most impressed with is the one in use for GCC. IIRC, when a bug is reported, they write a testcase for the bug, *then* fix it. The goal for a release then is 0 regressions.
Because they produce snapshots on a regular basis, everyone who downloads them is encouraged to submit the test results to this database so that regular information can be collected:
http://gcc.gnu.org/testresults/
I was reading "Extreme Programming Explained" which seemed to also have a good philosophy (although I haven't gotten very good at it yet), which is to *first* write a test case for the objects that you're designing, and *then* write the code. When the tests pass, the code is complete. This way the automated test suite can be run many times a day, and you can track regressions that way.
Slashdot Mirror
Well, there are a couple of benefits to using Hurd:
/dev/random that I've seen based on EGD)
1) Hack value. It's fun to play.
2) Secure infrastructure that permits to user to do quite reasonable things.
In the Hurd, any user can create a secure chroot jail. Any user can mount a remote file system off of their home directory. Any user could mount a file (ISO or whatever) as a filesystem.
3) Translators
The concept of a translator is the "Everything is a file" taken a step further. If you imbed a program in an inode, it can then deal with filesystem calls to that inode as it sees fit.
This means that things like the Linux "proc" file system are easy to write, and could trivially show things like SQL data, or an FTP-fs.
Device drivers can either be colocated in the kernel (Which they generally are now, similar to Linux), or can be emulated in user space (Like an implementation of
This means that implementing PPPoE is going to be very simple for us, since we just need to write a simple translator the reads from the Ethernet card, and pushes those packets into the IP stack.
These are just beginning sort of examples of what's possible. Other clever stuff is doable if you're willing to think a little bit outside of the typical "unix" box.
Ummm, our office admin got the winamp plugin running. She's pretty bright, but not exactly technical.
If you still need an XMMS plugin, upgrade to a newer version, it's built in now.
I think the preferred term is "Saint iGNUcious"...
http://www.stallman.org/saint.html
Most ideas that come out of life aren't worth persuing. The great part is to see that some people do attempt them, and provide a nice learning experience for the rest of us.
In the corporate world, you have to rely on yourself and your coworkers for your scope of experience.
In the Free Software world, I have Torvalds, Cox, Stallman, Raymond, etc. as part of my scope of experience. That's really powerful.
I suspect you weren't looking at the latest "Milestone", since it *is* called Mozilla 0.8.
If you're looking at M18, it was truly a bad product.
The only time I need to use NS4.7 now is when I want to view source on a generated page. Mozilla's view source polls the server for the HTML, but doesn't bother to supply POST parameters. (In most cases a second instance of the POST being sent would be bad anyway).
But I digress...
Why, when you can be forced?
=)
When I visited the FSF, the one Hurd box they had was frozen. =)
My Hurd box has had 3 or 4 months of uptime, so I have no idea what the hell the did to it.
The FSF production machines all use Debian now.
I use Blackdown's JRE on Linux with the nightlies, and it works great. I've been doing this since mid December or so
I heard an urban legend once about a microsoft seminar where they were announcing they they had a fully complient Korn shell available with their operating system. A gentleman in the audience stood up to tell them in exactly what ways it wasn't compliant and promptly got into an argument with the Microsoft VP. Apparently the person in the audience was David Korn. Is this true?
=)
I used a flexpro keyboard for years and loved it. One of my coworkers still uses one at his home. I tend to destroy about 2 keyboards a year, and it not only helped my wrists, but it prevented me from pounding on the keyboard.
It was also really cool for MS-Pinball. =)
I don't think I've ever heard 'debian' called small before...
Is '6a' considered odd or even? =)
Having done quite a bit of CSS-P, it's actually easier to do once you bend your mind to it. Table driven sites are an endless mess of tags inside your HTML code making it quite unreadable. Instead, just wrap the whole thing in a DIV tag, and use the positioning code to place it appropriately.
This means that you can hand the page to some junior HTMLer and know that the page will still have a chance of validating and looking right when you get it back. You can also simply hand it to a secretary with a comment of "Don't touch the angle brackets", and they can easily update the text on a page and have it work.
The trickey bit is writing your page in a way that makes sense when it's flat. Because of the positioning code, The DIVs can appear in any order. When I'm working on a site like this, I try to look at the page in lynx. When I'm happy with it there, I'll move the elements around the page and check it with Mozilla. Lynx never even sees the positioning code.
I suspect by the time IE6 and Mozilla 1.0 come out, that the worst of the positioning bugs will have gone away so that it's actually cross platform usable.
There's plenty of stuff for just about any level of programmer listed at http://www.gnu.org/help/help.html
As a newly converted Debian user, I *like* mocking my RPM-based friends who spend hours doing upgrades.
Don't hate me because I'm superiour.. =)
Nope, Jeff Bailey of GNU. You will be assimilated. =)
I've always thought that Hurd had an interesting potential to become a good gaming machine. Don't plan on using network? Unload the driver! That way you literally only have loaded exactly what you need to play.
That was *my* box you guys did in a year ago... =) I changed the address as quickly as I could to save my poor development system.
For an update, apache still compiles nicely on it, and recently many people got X running.
As to the Hurd web pages, I've taken over development on them and you can generally expect there to be updates to them starting this weekend. I'd appreciate any assistance people want to give. (web-hurd at gnu.org)
Another good place to go is http://www.debian.org/ports/hurd/
If you haven't heard this before, then you haven't been reading slashdot for long. This type of fear mongering is quite common when people talk about IPv6. The *recommended* way to generate an IPv6 address is through your MAC address. You're still welcome to assign them by hand if you so choose. Also, almost every Ethernet NIC can have its MAC address overridden.
The poster apparently hasn't been following slashdot either...
I disagree. It's useful to ask him something if you want a well-thought out *moral* response to a problem.
Each of us needs to have a set of values to work from, and decide when it's appropriate to push these values. I choose to write Free Software during my evenings and weekends, and I choose to volunteer to the GNU project as a sysadmin because the work that they do supports my values. However, in order to pay the bills, I develop proprietary software that is covered by IP patents. I make a concious choice to do this, and it's important that I recognise that this is very much a choice issue. At my work, I have convinced my employers a number of times to support Free software instead of writing proprietary because it was a better business case.
I value the advice that RMS has given me in the past, even though I may have chosen not to follow it. It allows me an outside perspective on a problem, and frequently has led me to what I've felt is a clean compromise.
The Lesser GPL is also used when there's a non-Free library that does the same thing. Glibc needed to be under this license historically since it was being positioned as a replacement native libcs on various Unix systems. If it were not under the LGPL now, it would be quickly replaced. However, a number of packages like libregex are under the GPL license, because it gives a competitive advantage to GPL software. GNU developers are encouraged to put as much as we can under the GPL to strengthen the movement.
Except that Debian already calls themselves Debian GNU/Linux (per http://www.debian.org/)
Because they produce snapshots on a regular basis, everyone who downloads them is encouraged to submit the test results to this database so that regular information can be collected: http://gcc.gnu.org/testresults/
I was reading "Extreme Programming Explained" which seemed to also have a good philosophy (although I haven't gotten very good at it yet), which is to *first* write a test case for the objects that you're designing, and *then* write the code. When the tests pass, the code is complete. This way the automated test suite can be run many times a day, and you can track regressions that way.
I'm rambling, so I'll stop now! =)
Works for me Sep 11/2000 1606 GMT
Hmmm... Little 'f'. That means 'free like beer' doesn't it? =)