The appearance is only skin deep. Creating a theme that looks "good"? That's easy, get some graphic designers together with a usability safety inspector.
Writing a complete framework with rich, well thought-out object libraries? Now that is a feat. GNUStep is a lurker project that is getting close to hitting critical mass. They've got the hard stuff done that others are still swinging at but not quite hitting.
No, the GNUStep people have been much more concerned with laying sewer lines, roadways, electrical grids, water, gas, etc. When they get around to picking the color for their street signs, it'll be good.
In case you didn't notice it, it would appear that the Gregory Benford cited in the article is the same Gregory Benford who writes some very good hard sci-fi.
People like Benford who do many things well are inspiring.
Now, go check out the Galactic Center series. (Someone else can find the link to Amazon, Barnes and Noble, or your local book seller.)
...but I agree with your general assertion. I think that while KDE is fine, and Gnome has its merits, the Open Source community is making a strategic blunder by not capitalizing on the OpenStep API specs. Many, many man-hours and lots of money went into developing a system of great beauty.
Still, you have to remember that it's not The Open Source Community(tm)'s job to think strategically. Every programmer is entitled to scratch his/her own itch. Because of this, many blind alleys terminate in piles of dead projects. However, the ethos of "explore every facet of every problem domain" has also produced some wonderful and unique software.
Still, if world domination with FOSS was your prime objective, you could do a lot worse than BSD/GNUStep. Somebody else has already proven the technological model.:-)
Thanks Hubert! You and the rest of the NetBSD team rock.
I'm not sure that NetBSD is the #1 technical innovation in the last 25 years, but it's pretty damn nice.
I'm happily running NetBSD 2.0 with my Netgear WG311T wireless card, and I was pleasantly surprised at how seemless the whole process was. (Finding the right card was the trick).
In any case, NetBSD never fails to impress me with its professional polish.
It is very much like the ports collection (build from source or install from binary package).
However, as I pointed out in my original post, Pkgsrc is unique in that it supports so many platforms with one source tree--including some *cough* problem platforms where building my favorite Open Source utilities is a major pain in the ass.
It sounds like after visiting their web site, you have a better idea of why pkgsrc is so useful. I encourage you to give it a try!
-Peter
Ah. Blissful clean architecture.
on
NetBSD 2.0 Released
·
· Score: 5, Informative
NetBSD is _the_ most underrated free OS project.
Do not be distracted by the fact that it can run on most every architecture. This is only a side effect of an uncompromisingly elegant design and clean implementation.
NetBSD is quite performant on modern hardware. It keeps pace with other operating systems in most areas, and exceeds in others. Remember, NetBSD was probably the first 64-bit clean open source operating system. It had USB support before Linux. It had IPv6 before... well... anybody.
NetBSD makes a great all around OS. NetBSD tends to be willing to break with tradition where others aren't. Proof is in things like its re-engineering of the BSD init system. It's so simply correct, that I can barely remember the traditional BSD inits. Hence, FreeBSD (and OpenBSD?) have adopted it.
So, run. Don't walk. Download, install, and enjoy.
-Peter
P.S. NetBSD's pkgsrc is only thing that comes close to a truly cross platform package management/build system. It supports Irix, Solaris, NetBSD, Linux, OpenBSD, FreeBSD, OS X, and (to a lesser degree) AIX. I'm sure I'm leaving out a few.
Personal rapid transit is already being tested in some areas. It seems this solves many of the problems with the "smart car" plus gives you other gains like:
Fewer vehicles need to be built (most people don't spend all day in the car)
Computer route optimization (get there quicker).
Potentially adds economy of scale to energy consumption (each vehicle doesn't need a power plant
Potentially simpler to avoid crashes than with traditional road-based systems.
I don't know, these are only a few benefits I can think of off the top of my head. Perhaps others can think of some more.
Of course the biggest cost is in the ultra light rail infrastructure. But, how much do we spend maintaining roads, traffic lights, etc? It seems like an elevated light rail infrastructure could be made very cost effective.
Sure, it would probably start in metro areas, but it could spread to more rural areas (where I live, we soak the feds for highway dollars) and probably save money here too.
Plus, people would be less stressed from commutes. Pedestrians could once again take over city streets. And fewer senseless deaths would have to happen. Oh, yeah, and then there's the whole ecology thing.
Right. There is WebDAV and SMB. There is also NFS. Unfortunatly, there is no SSH as best as I can tell. Thus spoke Mac Help:
You can also connect to SMB/CIFS, NFS, FTP, and WebDAV servers running on Mac OS X Server, AppleShare, UNIX, Linux, Novell NetWare, Windows NT, Windows 2000, and Windows XP servers.
SSH+SCP would be really nice. fish:// on the other hand, is shear brilliance. It uses Perl on the server side to do some things that are not possible with just SSH+SCP. Those are great fallbacks, but fish:// is innovative. But, I'd be happy with just SSH+SCP. As far as I can tell, it doesn't exist in OSX.
This brings me to another annoyance with OSX: It doesn't tell you when it doesn't know about a protocol. I can tell my OS X 10.3 machine to connect to a server. For a URL I type in "bogusprotocol://foo@foo.foo". The Finder tells me, "Connection Failed. No response from the server. Please try again."
WTF? I'd prefer something like, "You moron, you've just typed in a protocol name that doesn't exist." Please don't say, "Sorry, but we couldn't connect to this perfectly valid URL because the host wasn't available."
I'm a regular MacOS X user. And I love MacOS X, but there are some things that I miss about KDE. I try to follow KDE's progress even though it is not my desktop of choice these days.
The network transparency of KDE is brilliant. I'm not sure where the holdup for OSX is, but I would kill to be able to open a location with cmd-k, fish://user@myhost
I suspect that for Apple to add these bits would require some OS level work as well as some finder work. I hope they'd take that opportunity to update the finder to be a cocoa application. (As a side note, the Finder continues to bother me. My Mac savvy friends and I joke that the Finder, Mail.app, and Quicktime teams are Microsoft moles trying to take Apple down from the inside).
Anyone have any speculation as to why Apple hasn't already done some of the truly nifty network protocols? They've already got a finder view for FTP (which, unfortunately is dog-slow). Still, Apple has proven itself as a very agile software company. They've got a track record for adding features correctly and quickly, but the lack of an SSH handler is baffling to me.
Even in whole disk mode, this thing is scorching. I've not seen anything commercial or free that touches it in terms of speed. The whole thing is multithreaded. I use it for restoring PCs on a test bench after regression testing. In fs-aware mode it'll restore a ~5Gb file system in a matter of about 4-5 minutes.
Oh, yes. LoTR was the most influential book of the century. A whole industry of neon acrylic puff paint fantasy art would have never grown up without it.
Christ. I love LoTR as much as the next guy, but how does it really compare to A Day in the Life of Ivan Denisovich. I mean that book highlighted the abuses of the Soviet system so that 1984 didn't resemble the world portrayed in some other apparently less important book.
I like slackware because it has captured the flavor of BSD very well. It has a cohesiveness that is the result of Pat's immense knowledge and steady hand.
However, given that Slackware seems to be one of the prefered distros for BSD-o-philes, why not move its old-school BSD init style to be more in line with the other BSDs? The NetBSD rc scripts are not a huge leap for people familiar with Slackware's. That's because the new style is a logical evolution from the old. FreeBSD made the jump too. The new rc script style feels more "BSD" than the old feels to me now...
I'm not sure why everyone is so hopped up on these removal tools. It seems to me that after being infected with a worm that installs a back door, more people ought to look at reinstallation from known good media.
Biggest Windows vulnerability ever, again. How many times have we said that this year? At work, it's begining to feel a bit like a duck and cover drill.
Gnustepweb is a framework that is supposedly compatible with WebObjects.
The parent post has a really, really good point. GNUStep has oh, so much potential and it's getting close to ready.
Like it or lump it, Apple has produced the most cohesive *nix environment out there. They've got support from the important corporate software vendors. Vendors want to port to Linux, but damn, the myriad gui toolkits and serious lack of complete frameworks is daunting for commercial entities.
I know choice is good, but is Cocoa/Aqua that unexpressive to code in? The proliferation of apps for the Mac would seem to point to the contrary. Why must we reinvent the boring stuff (i.e. toolkits and frameworks) over and over? Couldn't we just adopt a proven successful model, run with it, then tweak where needed?
I just built GNUStep from NetBSD's excellent cross-platform package management/build system, pkgsrc. GNUStep is pretty cool. It's like a slightly primative, somewhat ugly Mac. Other than that, it's very, very similar. It's clear people are starting to write useful apps with it. It's got a finder-like app called GWorkspace. It's got a pretty decent mail application that runs on both MacOS and GNUStep.
Mac-on-Linux is a start. I've used it in the past, and it was quite snappy (on par with VMWare for x86). Unfortunately, it is only hosted under Linux. But for guest operating systems it claims to support Linux, MacOS 9, and MacOS X.
I've occasionally had the desire to do some sandboxed work on my Mac (I use VMWare for the PC all the time), but I can't bring myself to install Linux on my Powerbook. Removing FreeBSD and installing Linux on my PC (for VMWare) was hard enough.
But, if you can live with Linux as your host OS on your Mac, give MOL a shot.
I really like NetBSD as a desktop OS. NetBSD's clean approach means that they often get cool new drivers and features before anyone else, including linux. Cases in point: IPv6 and USB Visor support, IIRC.
Pkgsrc is a nice evolution of the ports. It has some neat additional features like a security audit.
I've said it before: NetBSD gets a bum reputation as being only for obscure hardware. Not so! People alway make the connection that FreeBSD==server, OpenBSD==security and NetBSD==suitable for toaster. I'd posit that NetBSD should be considered for desktops.
So, give it a shot on your desktops. I think you'll like it.
Seriously, there are times when these scripting systems can't do the heavy lifting of a "real" scripting language. I've often thought that Python might be an ideal embedded scripting language for an installer, especially with Mark Hammond's excellent Windows Extensions.
Has anyone used this NSIS/Python package? I suppose the only thing stopping me from trying at this point is my own laziness. Alas, this plugin requires that you track your own Python module dependencies.
Would it be possible to use pkgsrc as the main package management system on a Linux box, say, Slackware? What I mean is, forget Slackware's package management system altogether and replace it with NetBSD's pkgsrc.
NetBSD has pre-built binaries for just what you describe! Check out http://www.pkgsrc.org for details.
They've got bootstrap source that will compile on a lot, plus binary packages for a bunch of operating systems including Slackware (And Darwin. And Debian. And Irix. And Solaris:-) )
I'm not sure of the state of all the packages on all the different platforms, but my guess is that it works similarly to on NetBSD.
One of the cool side effects of their correct and clean implementation is that you can do interesting things like build embedded NetBSD from a Windows workstation. The cross-compile support is quite simply the most complete.
Slashdot Mirror
The appearance is only skin deep. Creating a theme that looks "good"? That's easy, get some graphic designers together with a usability safety inspector.
Writing a complete framework with rich, well thought-out object libraries? Now that is a feat. GNUStep is a lurker project that is getting close to hitting critical mass. They've got the hard stuff done that others are still swinging at but not quite hitting.
No, the GNUStep people have been much more concerned with laying sewer lines, roadways, electrical grids, water, gas, etc. When they get around to picking the color for their street signs, it'll be good.
Some work is already going into theming.
Now that GNUStep is getting really close to being complete, I hope they look at Cairo as a base for doing something similar to Quartz.
-Peter
In case you didn't notice it, it would appear that the Gregory Benford cited in the article is the same Gregory Benford who writes some very good hard sci-fi.
People like Benford who do many things well are inspiring.
Now, go check out the Galactic Center series. (Someone else can find the link to Amazon, Barnes and Noble, or your local book seller.)
-Peter
...but I agree with your general assertion. I think that while KDE is fine, and Gnome has its merits, the Open Source community is making a strategic blunder by not capitalizing on the OpenStep API specs. Many, many man-hours and lots of money went into developing a system of great beauty.
:-)
Still, you have to remember that it's not The Open Source Community(tm)'s job to think strategically. Every programmer is entitled to scratch his/her own itch. Because of this, many blind alleys terminate in piles of dead projects. However, the ethos of "explore every facet of every problem domain" has also produced some wonderful and unique software.
Still, if world domination with FOSS was your prime objective, you could do a lot worse than BSD/GNUStep. Somebody else has already proven the technological model.
-Peter
Thanks Hubert! You and the rest of the NetBSD team rock.
I'm not sure that NetBSD is the #1 technical innovation in the last 25 years, but it's pretty damn nice.
I'm happily running NetBSD 2.0 with my Netgear WG311T wireless card, and I was pleasantly surprised at how seemless the whole process was. (Finding the right card was the trick).
In any case, NetBSD never fails to impress me with its professional polish.
-Peter
It is very much like the ports collection (build from source or install from binary package).
However, as I pointed out in my original post, Pkgsrc is unique in that it supports so many platforms with one source tree--including some *cough* problem platforms where building my favorite Open Source utilities is a major pain in the ass.
It sounds like after visiting their web site, you have a better idea of why pkgsrc is so useful. I encourage you to give it a try!
-Peter
NetBSD is _the_ most underrated free OS project.
Do not be distracted by the fact that it can run on most every architecture. This is only a side effect of an uncompromisingly elegant design and clean implementation.
NetBSD is quite performant on modern hardware. It keeps pace with other operating systems in most areas, and exceeds in others. Remember, NetBSD was probably the first 64-bit clean open source operating system. It had USB support before Linux. It had IPv6 before... well... anybody.
NetBSD makes a great all around OS. NetBSD tends to be willing to break with tradition where others aren't. Proof is in things like its re-engineering of the BSD init system. It's so simply correct, that I can barely remember the traditional BSD inits. Hence, FreeBSD (and OpenBSD?) have adopted it.
So, run. Don't walk. Download, install, and enjoy.
-Peter
P.S. NetBSD's pkgsrc is only thing that comes close to a truly cross platform package management/build system. It supports Irix, Solaris, NetBSD, Linux, OpenBSD, FreeBSD, OS X, and (to a lesser degree) AIX. I'm sure I'm leaving out a few.
I don't know, these are only a few benefits I can think of off the top of my head. Perhaps others can think of some more.
Of course the biggest cost is in the ultra light rail infrastructure. But, how much do we spend maintaining roads, traffic lights, etc? It seems like an elevated light rail infrastructure could be made very cost effective.
Sure, it would probably start in metro areas, but it could spread to more rural areas (where I live, we soak the feds for highway dollars) and probably save money here too.
Plus, people would be less stressed from commutes. Pedestrians could once again take over city streets. And fewer senseless deaths would have to happen. Oh, yeah, and then there's the whole ecology thing.
-Peter
SSH+SCP would be really nice. fish:// on the other hand, is shear brilliance. It uses Perl on the server side to do some things that are not possible with just SSH+SCP. Those are great fallbacks, but fish:// is innovative. But, I'd be happy with just SSH+SCP. As far as I can tell, it doesn't exist in OSX.
This brings me to another annoyance with OSX: It doesn't tell you when it doesn't know about a protocol. I can tell my OS X 10.3 machine to connect to a server. For a URL I type in "bogusprotocol://foo@foo.foo". The Finder tells me, "Connection Failed. No response from the server. Please try again."
WTF? I'd prefer something like, "You moron, you've just typed in a protocol name that doesn't exist." Please don't say, "Sorry, but we couldn't connect to this perfectly valid URL because the host wasn't available."
-Peter
I'm a regular MacOS X user. And I love MacOS X, but there are some things that I miss about KDE. I try to follow KDE's progress even though it is not my desktop of choice these days.
The network transparency of KDE is brilliant. I'm not sure where the holdup for OSX is, but I would kill to be able to open a location with cmd-k, fish://user@myhost
I suspect that for Apple to add these bits would require some OS level work as well as some finder work. I hope they'd take that opportunity to update the finder to be a cocoa application. (As a side note, the Finder continues to bother me. My Mac savvy friends and I joke that the Finder, Mail.app, and Quicktime teams are Microsoft moles trying to take Apple down from the inside).
Anyone have any speculation as to why Apple hasn't already done some of the truly nifty network protocols? They've already got a finder view for FTP (which, unfortunately is dog-slow). Still, Apple has proven itself as a very agile software company. They've got a track record for adding features correctly and quickly, but the lack of an SSH handler is baffling to me.
-Peter
Sorry, I can't imagine the DMV allowing somebody to have that as a license plate. It was, in fact, a bumper sticker.
Sorry for the confusion.
-Peter
BU__ SH__!
-Peter
You have no idea what sort of trojan / backdoor / über nasty malware was loaded on each individual machine.
Your time is much better spent developing a network reimaging system so that your machines can be reverted to a known state relatively quickly.
-Peter
You should be able to get this to work on your gentoo live CDs. I'm not sure about it working in OSX:
http://www.emulab.net/software.php3
Even in whole disk mode, this thing is scorching. I've not seen anything commercial or free that touches it in terms of speed. The whole thing is multithreaded. I use it for restoring PCs on a test bench after regression testing. In fs-aware mode it'll restore a ~5Gb file system in a matter of about 4-5 minutes.
-Peter
Wow, this must be a first. Their website appears to be down and this was linked from the BSD section.
BSD trolls can put that in their respective pipes and smoke it.
-Peter
P.S. Did anyone download this? Does it work? Is it even remotely as good as VMWare?
Oh, yes. LoTR was the most influential book of the century. A whole industry of neon acrylic puff paint fantasy art would have never grown up without it.
Christ. I love LoTR as much as the next guy, but how does it really compare to A Day in the Life of Ivan Denisovich. I mean that book highlighted the abuses of the Soviet system so that 1984 didn't resemble the world portrayed in some other apparently less important book.
That being said, huzzah for the Entsez!
-Peter
Public libraries are one of the few public institutions we have that break down economic barriers to gaining knowledge.
Think about that during the next mil levy.
-Peter
I like slackware because it has captured the flavor of BSD very well. It has a cohesiveness that is the result of Pat's immense knowledge and steady hand.
However, given that Slackware seems to be one of the prefered distros for BSD-o-philes, why not move its old-school BSD init style to be more in line with the other BSDs? The NetBSD rc scripts are not a huge leap for people familiar with Slackware's. That's because the new style is a logical evolution from the old. FreeBSD made the jump too. The new rc script style feels more "BSD" than the old feels to me now...
-Peter
I'm not sure why everyone is so hopped up on these removal tools. It seems to me that after being infected with a worm that installs a back door, more people ought to look at reinstallation from known good media.
Biggest Windows vulnerability ever, again. How many times have we said that this year? At work, it's begining to feel a bit like a duck and cover drill.
-Peter
Gnustepweb is a framework that is supposedly compatible with WebObjects.
The parent post has a really, really good point. GNUStep has oh, so much potential and it's getting close to ready.
Like it or lump it, Apple has produced the most cohesive *nix environment out there. They've got support from the important corporate software vendors. Vendors want to port to Linux, but damn, the myriad gui toolkits and serious lack of complete frameworks is daunting for commercial entities.
I know choice is good, but is Cocoa/Aqua that unexpressive to code in? The proliferation of apps for the Mac would seem to point to the contrary. Why must we reinvent the boring stuff (i.e. toolkits and frameworks) over and over? Couldn't we just adopt a proven successful model, run with it, then tweak where needed?
I just built GNUStep from NetBSD's excellent cross-platform package management/build system, pkgsrc. GNUStep is pretty cool. It's like a slightly primative, somewhat ugly Mac. Other than that, it's very, very similar. It's clear people are starting to write useful apps with it. It's got a finder-like app called GWorkspace. It's got a pretty decent mail application that runs on both MacOS and GNUStep.
-Peter
Mac-on-Linux is a start. I've used it in the past, and it was quite snappy (on par with VMWare for x86). Unfortunately, it is only hosted under Linux. But for guest operating systems it claims to support Linux, MacOS 9, and MacOS X.
I've occasionally had the desire to do some sandboxed work on my Mac (I use VMWare for the PC all the time), but I can't bring myself to install Linux on my Powerbook. Removing FreeBSD and installing Linux on my PC (for VMWare) was hard enough.
But, if you can live with Linux as your host OS on your Mac, give MOL a shot.
-Peter
My understanding that this _IS_ the guest OS port to the Xen architecture. Check it out at http://www.netbsd.org/Changes/#newportxen.
Now, who knows what operating systems can host the Xen virtual machine monitor.
-Peter
I really like NetBSD as a desktop OS. NetBSD's clean approach means that they often get cool new drivers and features before anyone else, including linux. Cases in point: IPv6 and USB Visor support, IIRC.
Pkgsrc is a nice evolution of the ports. It has some neat additional features like a security audit.
I've said it before: NetBSD gets a bum reputation as being only for obscure hardware. Not so! People alway make the connection that FreeBSD==server, OpenBSD==security and NetBSD==suitable for toaster. I'd posit that NetBSD should be considered for desktops.
So, give it a shot on your desktops. I think you'll like it.
-Peter
And now for something completely different.
It's...
Python scripting for NSIS.
Seriously, there are times when these scripting systems can't do the heavy lifting of a "real" scripting language. I've often thought that Python might be an ideal embedded scripting language for an installer, especially with Mark Hammond's excellent Windows Extensions.
Has anyone used this NSIS/Python package? I suppose the only thing stopping me from trying at this point is my own laziness. Alas, this plugin requires that you track your own Python module dependencies.
-Peter
This is great news. People need to list as a requirement for clients and servers XMPP compliance.
Everyone who uses iChat, stop what you are doing and go fill out a bug request form on Apple's developer site (http://developer.apple.com).
I'm going to go fill my request right now.
While you're at it, maybe you should request that they open a protocol plugin api to developers.
-Peter
NetBSD has pre-built binaries for just what you describe! Check out http://www.pkgsrc.org for details.
They've got bootstrap source that will compile on a lot, plus binary packages for a bunch of operating systems including Slackware (And Darwin. And Debian. And Irix. And Solaris
I'm not sure of the state of all the packages on all the different platforms, but my guess is that it works similarly to on NetBSD.
One of the cool side effects of their correct and clean implementation is that you can do interesting things like build embedded NetBSD from a Windows workstation. The cross-compile support is quite simply the most complete.
-Peter