In this post (pretty sad when you have to link to different posts inside articles, eh?):-) I mentioned one thing that bothers me a bit, but doesn't turn me off totally (though it would others) -- why don't application developers have a standard way, on installation, to do the following in a cross-GUI/cross-distro/cross-platform (dare I say it?) way:
Know a good default location to put their application -- and no, home directories are *not* an option -- or maybe they could be if the app was being installed as a regular user:-)
Register their apps with whatever package managers are around for later removal or upgrades
Put their apps on a launch menu
Install MIME types and document icons for graphical file managers
I'm sure more could be contributed to this list. I encourage everyone to do so, maybe we can get some good ideas and good code going. (All hail open source!)
More standardized toolkits like GTK should help to take care of this. I'm finding myself using emacs keybindings more and more inside textboxes (i.e. Ctrl-E for end of line, Ctrl-A for beginning of line) because it is working in more and more apps.
Desktop standards v. networking standards
on
Gartner Slams Linux
·
· Score: 2
The problem is, that's not true. We have RFCs, just like everything else. Some stuff implemented on linux doesn't live up to the RFCs in question (dhcpd 2, for example, doesn't implement DHCPINFORM), but then again it also implements some stuff far better than mere standards-compliance (gnutar, bzip2, a whole gamut of decent shells).
This is a key point I missed commenting on earlier. It is very true that Linux and other open source OS's have a great track record for following networking and Internet standards, as well as some UNIX standards. (And I must admit I like reading some of Linux's manpages when it comes to having to deviate from a standard or explain why a standard is silly -- the attitude is very refreshing.):-) However, there are currently no good desktop standards, which I think was the thrust of the article.
For example, as far as filesystems go, there's the Filesystem Hierarchy Standard, but beyond the distribution-makers, who really follows it? (Don't even get me started on how StarOffice installs... groan.) Maybe that has something to do with the fact that Linux users are so grateful for certain kinds of applications that they don't even concern themselves with whether or not they follow established standards...?
Also consider that there are now countless ways of running your desktop. The unification efforts behind KDE and GNOME (and the nonexistent efforts between these guys and anyone else) are not anywhere near mature enough for me, as an application developer, to even say "ok, regardless of what environment Joe Blow has here, I want to register this mime type as mine, this icon, put this on the launch menu, etc." That's just as an important part of standards than sending the right bytes over your network wire.
Do I personally care so much about whether or not an application can install its own file associations and launch menu icons? Nah. I still like the power. But it sure would be nice.
MS Access and critical applications -- reliable?
on
Gartner Slams Linux
·
· Score: 1
...critical business applications are running on Access...
Perhaps offtopic, perhaps not -- but I'm surprised you can use Access for anything critical, unless it's got a real SQLish server (Oracle, PostgreSQL, whatever) behind it via ODBC. Is this the case?
Re:Gartner amazes me with correctness, for once
on
Gartner Slams Linux
·
· Score: 1
I agree. Of course you should, but programmers generally don't, or else they stuff their registry data under HKEY_LOCAL_MACHINE. (This is a generalization, folks, I know there are programmers out there more aware of how to use the registry correctly.) I think if Gartner is going to point out that a fault of Linux-based systems actually lies in its applications, then it is only fair to point out that a fault of NT (and 95/98) is the same.
Gartner amazes me with correctness, for once
on
Gartner Slams Linux
·
· Score: 5
It was a shock to me when I had to agree with the Gartner Group's analysis of anything. They have been so steadily wrong for so long that I had to double-check outside for any horsemen of the apocalypse.:-) Regardless of whether this month seems to be Microsoft's to launch attacks on Linux (see this article) -- it would be a logical conclusion that this is another Microsoft-paid opinion piece from the Gartner Group -- there are very good points here, and I think clarifications should be made.
Gartner's piece states that Linux is to be avoided for business-productivity applications. Let's not forget what a "business-productivity" application is. It's Word, it's Excel, it's Access, it's PowerPoint. All of these are targeted at single-user applications. (Some might try to say that Access can be used for multiuser applications; let me tell you from experience that you can only get up to about five people before it really starts bombing out. Where I used to work, an Access-based application was totally corrupted by someone leaving their computer on overnight. I don't consider that a multiuser application.)
The problem is that there is really a pretty sad offering along the lines of single-user applications in Linux-based, and indeed other open source systems. I have a Linux workstation here and love it to death, but I'm an administrator and a developer. I have StarOffice for firing off memos, of course -- but there is simply no way I could effectively get the rest of the office to use Linux, even if I had the authority to send out a mandate from on high that Microsoft was to be abolished. (Now, perhaps I could get Macs in here...):-)
Where Linux as well as other UNIX clones and derivatives do excel is in multiuser applications. I don't care if you have figures showing that IIS performs better; I can do more and I can do it more effectively, and I can do it on an OS that was designed from the ground up to be shared among multiple users. To turn around a key point from that previous Microsoft piece (paraphrased: "Linux was not designed with a GUI in the core"), Windows NT was not designed with multiple users in mind. Its design is based on an OS that still really only can be effectively used by one person at a time. (Want proof? Go into \WINNT sometime and look at all the.INI files -- one person's settings easily override everyone's.)
Let's not also forget the key benefit of free or open source software. I can change it if I need to. I've done so on quite a few incidents, to fit my needs when the stock configurations didn't. My NT system on the other corner of my desk goes largely unused for several applications because I can't change its applications to do what I need them to.
What would it take to bring Linux to the desktop, or as Gartner puts it, the "business-productivity" market? Quite a bit. The latest round of GUI stuff is getting there but there are still so many key points to sweat out. Printing a memo off is still not a no-brainer on your typical Linux system unless it's been set up by someone with a clue. But in the meantime, Linux and other UNIX derivatives are what I and other administrators and engineers swear by for our desktops. We just can't get away from the power.:-)
It would seem to me that provided there are always other vendors of seed, then the free market would see the seed business moving to the vendors who didn't sell sterile seed. Monsanto would then have two choices: stop selling the sterile, or get out of the business.
It looks like the free market took care of this very well -- although the article's details are a bit sketchy, it appears that pressure from buyers led to the decision not to sell this seed.
Seems to me I heard of the QNX floppy a long time ago. Although it might not have had a webserver back then, it definitely had at least a browser, etc.
Still, it's worth checking out for those who haven't heard of it.
Intriguing idea. I had all but forgotten about gopher. IIRC, gopher's interactive capabilities were somewhat limited, however. Wasn't the only kind of input gopher could take a single search phrase, kind of like ISINDEX? It seems like that would be great for retrieval of information (give me this stock quote, look for news items about this term) but would preclude the capability for deeper interaction.
Gopher documents were also returned in plain text format. It seems like this would scale just about as well on a handheld as your typical/. article with 300 replies.:-) Either that, or the web designers will then be unleashed on the gopher servers, and we'll start seeing ``enhanced for Sprint/PCS'' gopher sites, all 10 characters wide:-)
Also, I wonder how secure all those old gopher servers are... most have obviously missed all the security brouhaha over the last few years.
Katz's comments seem to be in-line with what quite a few people are saying about Linux -- once it's running, it's really cool. But the installation is hell.
Linux certainly isn't alone in the confusing installation arena. I forget the article, and I'm too lazy to go find it, but a certain columnist just recently expressed how even as skilled as he was, he spent quite a bit of time installing Linux yet quite a lot more time installing Win2k.
It seems to me that in order to get the most desktop share, you really have to get preinstallation deals, plain and simple. OS installation is going to remain just plain difficult as long as we keep the unchangeable pieces of the computer as simple as possible. This is a good thing -- because the less that's unchangeable in the machine, the more can be innovated in software.
My wife is an extremely competent Mac user, but she doesn't do installs. I do it since I'm the one with the half a degree and voluminous experience (mind you, I didn't say I'm smart -- I've just been exposed to quite a bit):-).
I'm curious what will happen to the CDE/Motif portions of Solaris. Surely they'd have to be released in source form too, but I doubt the OSF will go for it...?
No, it's not a comparable situation. If Apple were to GPL Mac OS X, they would still be the principal developers for, at the very least, quite some time (look at how Mozilla started) and would probably not make money from it. OTOH, Dell is not a principal developer of either RH or Windows.
Well-formedness in HTML, in addition to adherence to the HTML 4.0 Strict specification (or HTML 2.0, for that matter), would make handheld browsing a reality. Of course it doesn't mean that it would be comfortable to read paragraphs worth on a 4x10 display, or whatever they are right now (-: (Keep in mind that just being standards-based doesn't mean your pages won't bomb out on handheld devices, speech synthesizers, or whatever; you have to keep the spirit in mind too.)
Those "visionaries" who thought that the Web would only be useful on a computer screen, and therefore opted to abuse HTML as much as possible, are to blame for the fact that you literally can't pull up anything on a device smaller than a checkbook.
I encourage everyone to consider the benefits of well-formedness and adherence to standards when they structure their HTML. The current CSS specs, along with nice browsers like Mozilla, can provide you with ways to still look pretty. If I hadn't accidentally deleted my web site I could demonstrate:-)
As I mentioned in a previous article, I do think that licensing probably did indeed have something to do with it. Certainly the GPL is unfriendly to people who depend on sales of licenses for their bottom line.
How would you ensure profitability by just building the hardware? Obviously, the goals of free software are going to clash head-on with you if you try to make your hardware work "best" with the software that you're developing in the open.
This is a genuine question and I'm wondering what the answer is, it's not rhetorical flamebait:-)
Part of that article struck me -- the apparent disdain of the Linux community for Apple's choice of BSD to base Mac OS X on. But having been a past and present Linux user professionally, and an exclusive FreeBSD user personally, I think I can see where Apple is looking.
Any professional OS that touts ease of configurability must have a consistent interface to everything underneath. The more I delve into Linux, the more I get confused by the different ways everything has to be done. I think this is a symptom of the "hack it till it just works" mentality.
I don't exclusively use FreeBSD because of its stability and security (although that was certainly the case when I switched a few years ago; Linux boxes were giving me and others unexplainable random problems left and right that an installation of FreeBSD remedied handily.) I continue to use FreeBSD today because things are done right and consistently. That might mean waiting a little longer for equivalent functionality, but I can trust said functionality.
The Linux systems I use seem to be able to do more, but I can't always find the correct way to do it. Perhaps that's more due to a lack of documentation, but when I do find the right way, it seems to be a hack that really doesn't fit into any consistent way of doing things. For example, getting my ATAPI CD-RW drive to work meant inserting a LILO flag. It seems to me that functionality should have been found elsewhere, perhaps in a kernel config file or by unloading one module and loading another. FreeBSD doesn't yet have IDE-SCSI support, but I bet when they do, it will be able to be enabled without rebooting and should also be easy to set up.
All in all, I can see the building of an organized, easily-setup, easily-administrated OS a lot more clearly on top of a BSD kernel. The tests of time have showed BSD to do well in those areas. I think Linux will get there, maybe on the next iteration of the kernel. 2.2 unloaded a lot of nasty old baggage that was around in 2.0, I have faith that 2.4 will be even better.
Of course, the rest of the story probably involves things like the GPL. Apple and other large companies who derive a huge bottom line from software licenses, I think, will continue to shy away from truly free licenses for some time. I think that is truly a shame for the forces of free software.
P.S. I have tried to be constructive here instead of being flamebait. I would appreciate replies that are in the same manner.
When I previewed that previous comment, it looked just great... but when I posted it, I guess the lt and gt entities got re-parsed:-) Here is paragraph #3 without entities:
It meant using elements for what they were intended for. It meant never using a table for anything other than tabular data. It meant using EM when I wanted emphasis and it meant using CODE when I wanted to mark up code fragments (mind you, I'm stuck using TT right now because/.'s HTML filter doesn't permit CODE!) It took some fiddling.
You're on the right track if you're saying that this technology wouldn't even be needed if we could convince people to stick to standards. But you also need one other thing -- well-formedness.
You can stick to HTML 4.0, even the "Strict" dialect (which I encourage everyone to do!), and still have pages that completely blow up when pulled up outside one of the Big Two. On the website that I accidentally deleted some time ago, I had struggled for some time to not just reach for HTML 4.0 compliance but for well-formedness.
It meant using elements for what they were intended for. It meant never using a table for anything other than tabular data. It meant using when I wanted emphasis and it meant using when I wanted to mark up code fragments (mind you, I'm stuck using right now because/.'s HTML filter doesn't permit !) It took some fiddling.
But I turned out with a set of pages that were not only easier to maintain, but CSS applied very cleanly to them, making them pretty and consistent-looking, and they rendered perfectly on any hand-held or speech-synthesizing device you could throw at me. My information was useful to everyone, and that was the best high of all.
I strongly encourage everyone to pursue well-formedness. The more important stuff that is well-formed instead of hacked, the better browsers we'll get, too!
The reason I'm always spewing out hardcopies is because online doco doesn't fit on the screen. Sure, I've got 1152x864 on a nice 17", but there just isn't enough room for your average HTML doco alongside a couple xterms.
Now, this isn't ALWAYS the case. There is standards-compliant documentation out there that will properly wrap to a narrow browser window -- but it's unfortunately far too rare. That and no matter whether you get the standards right or not, any graphical screenshots or tables you do are going to be pretty much unreadable at less than 400 pixels wide.
Besides, let's not forget that most paper doesn't come from the bad guys in Captain Planet tearing down rainforests just for the sheer orgasmic thrill of it. It comes from a combination of the reconstituted scraps from the last paper runs and tree farms, among other things.
It seems to be slowing but I think is still progressing. There are a few key areas that still need to be worked on. Lack of a good IPC library, for example, is standing in the way of a working port of PostgreSQL.
Lots of working stuff is ported to Cygwin, though. My favorite site for that stuff is the Cygwin Porting Project.
Another, even better option -- Cygnus' Cygwin. Quite a bit of commonly-used open source software is already ported (most GNU stuff compiles out of the box, but there's lots of other stuff full of Linuxisms that needed to be ported better.) It has the added advantage of being able to run on 95/98 as well, although with substantially fewer features (like file security.)
In addition, Cygwin seems to still be actively maintained. EGCS is available for it among other things (to be fair, the same guy that ports EGCS to Cygwin also does it for UWIN.)
That was the case before, but may not be entirely correct now. I am still looking for more information from ICANN but they seem to be concentrating on the political rather than the technical.
Nowadays there are alternative registrars that actually seem to take you through the whole process. I gather that NSI is still on the backend but you do not actually become an NSI customer anymore. Unfortunately I haven't had an opportunity to try anybody out yet, I have more important things to do like reading Slashdot (-:
Maybe we can get someone who works for one of those registrars to clear things up here?
(P.S. I will... ummm... not be dropping by NSI anytime soon with the explosives that I don't have.:-)
Yet another lovely step back in time by the Clinton administration. I wonder if any of the candidates for the next presidential election have gone on record for crypto policy.
The primary reason that the concept key escrow absolutely petrifies me is that the to be useful, the keys need to travel in one form or another from their central repository (which I would hope would be as tightly locked up as the NSA) to the law enforcement agency responsible for unlocking the message. With the repeated demonstrations by the U.S. Government that they don't understand crypto, what's even going to guarantee the safety of my key (and therefore my data) in transit?
Don't make me hand over my keys. I have them because they protect me. And you can bet that if key escrow becomes a requirement, I will not surrender my stock of open-source crypto software, but only begin to use it more.
Can't do that (was Re:And Windows NT is turn...)
on
Linux Turns 8
·
· Score: 1
No, you can't make that comparison. NT is still loaded with code that's been around since DOS, and OS's with Linux as the kernel are generally loaded with code from GNU that's been around a similar amount of time.
Unless you want to compare the NT kernel with the Linux kernel...:-) (But who can separate the NT kernel from anything... heck, you can't even pull it away from the GUI!)
Slashdot Mirror
Perhaps you should amend "hope it works" to the build process as well... I have compiled one too many things to have blind faith in that combination.
Well, why not make this into that forum? :-)
In this post (pretty sad when you have to link to different posts inside articles, eh?) :-) I mentioned one thing that bothers me a bit, but doesn't turn me off totally (though it would others) -- why don't application developers have a standard way, on installation, to do the following in a cross-GUI/cross-distro/cross-platform (dare I say it?) way:
Know a good default location to put their application -- and no, home directories are *not* an option -- or maybe they could be if the app was being installed as a regular user :-)
Register their apps with whatever package managers are around for later removal or upgrades
Put their apps on a launch menu
Install MIME types and document icons for graphical file managers
I'm sure more could be contributed to this list. I encourage everyone to do so, maybe we can get some good ideas and good code going. (All hail open source!)
More standardized toolkits like GTK should help to take care of this. I'm finding myself using emacs keybindings more and more inside textboxes (i.e. Ctrl-E for end of line, Ctrl-A for beginning of line) because it is working in more and more apps.
This is a key point I missed commenting on earlier. It is very true that Linux and other open source OS's have a great track record for following networking and Internet standards, as well as some UNIX standards. (And I must admit I like reading some of Linux's manpages when it comes to having to deviate from a standard or explain why a standard is silly -- the attitude is very refreshing.) :-) However, there are currently no good desktop standards, which I think was the thrust of the article.
For example, as far as filesystems go, there's the Filesystem Hierarchy Standard, but beyond the distribution-makers, who really follows it? (Don't even get me started on how StarOffice installs... groan.) Maybe that has something to do with the fact that Linux users are so grateful for certain kinds of applications that they don't even concern themselves with whether or not they follow established standards...?
Also consider that there are now countless ways of running your desktop. The unification efforts behind KDE and GNOME (and the nonexistent efforts between these guys and anyone else) are not anywhere near mature enough for me, as an application developer, to even say "ok, regardless of what environment Joe Blow has here, I want to register this mime type as mine, this icon, put this on the launch menu, etc." That's just as an important part of standards than sending the right bytes over your network wire.
Do I personally care so much about whether or not an application can install its own file associations and launch menu icons? Nah. I still like the power. But it sure would be nice.
Perhaps offtopic, perhaps not -- but I'm surprised you can use Access for anything critical, unless it's got a real SQLish server (Oracle, PostgreSQL, whatever) behind it via ODBC. Is this the case?
I agree. Of course you should, but programmers generally don't, or else they stuff their registry data under HKEY_LOCAL_MACHINE. (This is a generalization, folks, I know there are programmers out there more aware of how to use the registry correctly.) I think if Gartner is going to point out that a fault of Linux-based systems actually lies in its applications, then it is only fair to point out that a fault of NT (and 95/98) is the same.
It was a shock to me when I had to agree with the Gartner Group's analysis of anything. They have been so steadily wrong for so long that I had to double-check outside for any horsemen of the apocalypse. :-) Regardless of whether this month seems to be Microsoft's to launch attacks on Linux (see this article) -- it would be a logical conclusion that this is another Microsoft-paid opinion piece from the Gartner Group -- there are very good points here, and I think clarifications should be made.
Gartner's piece states that Linux is to be avoided for business-productivity applications. Let's not forget what a "business-productivity" application is. It's Word, it's Excel, it's Access, it's PowerPoint. All of these are targeted at single-user applications. (Some might try to say that Access can be used for multiuser applications; let me tell you from experience that you can only get up to about five people before it really starts bombing out. Where I used to work, an Access-based application was totally corrupted by someone leaving their computer on overnight. I don't consider that a multiuser application.)
The problem is that there is really a pretty sad offering along the lines of single-user applications in Linux-based, and indeed other open source systems. I have a Linux workstation here and love it to death, but I'm an administrator and a developer. I have StarOffice for firing off memos, of course -- but there is simply no way I could effectively get the rest of the office to use Linux, even if I had the authority to send out a mandate from on high that Microsoft was to be abolished. (Now, perhaps I could get Macs in here...) :-)
Where Linux as well as other UNIX clones and derivatives do excel is in multiuser applications. I don't care if you have figures showing that IIS performs better; I can do more and I can do it more effectively, and I can do it on an OS that was designed from the ground up to be shared among multiple users. To turn around a key point from that previous Microsoft piece (paraphrased: "Linux was not designed with a GUI in the core"), Windows NT was not designed with multiple users in mind. Its design is based on an OS that still really only can be effectively used by one person at a time. (Want proof? Go into \WINNT sometime and look at all the .INI files -- one person's settings easily override everyone's.)
Let's not also forget the key benefit of free or open source software. I can change it if I need to. I've done so on quite a few incidents, to fit my needs when the stock configurations didn't. My NT system on the other corner of my desk goes largely unused for several applications because I can't change its applications to do what I need them to.
What would it take to bring Linux to the desktop, or as Gartner puts it, the "business-productivity" market? Quite a bit. The latest round of GUI stuff is getting there but there are still so many key points to sweat out. Printing a memo off is still not a no-brainer on your typical Linux system unless it's been set up by someone with a clue. But in the meantime, Linux and other UNIX derivatives are what I and other administrators and engineers swear by for our desktops. We just can't get away from the power. :-)
It would seem to me that provided there are always other vendors of seed, then the free market would see the seed business moving to the vendors who didn't sell sterile seed. Monsanto would then have two choices: stop selling the sterile, or get out of the business.
It looks like the free market took care of this very well -- although the article's details are a bit sketchy, it appears that pressure from buyers led to the decision not to sell this seed.
My question is -- how is this caused? The article from Wired is skimpy on the issue. Which is it?
If it's any of these, it's bad design and needs to be corrected and rolled out ASAP. If AOL does anything less, they're negligent.
Seems to me I heard of the QNX floppy a long time ago. Although it might not have had a webserver back then, it definitely had at least a browser, etc.
Still, it's worth checking out for those who haven't heard of it.
Intriguing idea. I had all but forgotten about gopher. IIRC, gopher's interactive capabilities were somewhat limited, however. Wasn't the only kind of input gopher could take a single search phrase, kind of like ISINDEX? It seems like that would be great for retrieval of information (give me this stock quote, look for news items about this term) but would preclude the capability for deeper interaction.
Gopher documents were also returned in plain text format. It seems like this would scale just about as well on a handheld as your typical /. article with 300 replies. :-) Either that, or the web designers will then be unleashed on the gopher servers, and we'll start seeing ``enhanced for Sprint/PCS'' gopher sites, all 10 characters wide :-)
Also, I wonder how secure all those old gopher servers are... most have obviously missed all the security brouhaha over the last few years.
Katz's comments seem to be in-line with what quite a few people are saying about Linux -- once it's running, it's really cool. But the installation is hell.
Linux certainly isn't alone in the confusing installation arena. I forget the article, and I'm too lazy to go find it, but a certain columnist just recently expressed how even as skilled as he was, he spent quite a bit of time installing Linux yet quite a lot more time installing Win2k.
It seems to me that in order to get the most desktop share, you really have to get preinstallation deals, plain and simple. OS installation is going to remain just plain difficult as long as we keep the unchangeable pieces of the computer as simple as possible. This is a good thing -- because the less that's unchangeable in the machine, the more can be innovated in software.
My wife is an extremely competent Mac user, but she doesn't do installs. I do it since I'm the one with the half a degree and voluminous experience (mind you, I didn't say I'm smart -- I've just been exposed to quite a bit) :-).
I'm curious what will happen to the CDE/Motif portions of Solaris. Surely they'd have to be released in source form too, but I doubt the OSF will go for it...?
Anyone heard about this little thorn?
No, it's not a comparable situation. If Apple were to GPL Mac OS X, they would still be the principal developers for, at the very least, quite some time (look at how Mozilla started) and would probably not make money from it. OTOH, Dell is not a principal developer of either RH or Windows.
Well-formedness in HTML, in addition to adherence to the HTML 4.0 Strict specification (or HTML 2.0, for that matter), would make handheld browsing a reality. Of course it doesn't mean that it would be comfortable to read paragraphs worth on a 4x10 display, or whatever they are right now (-: (Keep in mind that just being standards-based doesn't mean your pages won't bomb out on handheld devices, speech synthesizers, or whatever; you have to keep the spirit in mind too.)
Those "visionaries" who thought that the Web would only be useful on a computer screen, and therefore opted to abuse HTML as much as possible, are to blame for the fact that you literally can't pull up anything on a device smaller than a checkbook.
I encourage everyone to consider the benefits of well-formedness and adherence to standards when they structure their HTML. The current CSS specs, along with nice browsers like Mozilla, can provide you with ways to still look pretty. If I hadn't accidentally deleted my web site I could demonstrate :-)
As I mentioned in a previous article, I do think that licensing probably did indeed have something to do with it. Certainly the GPL is unfriendly to people who depend on sales of licenses for their bottom line.
How would you ensure profitability by just building the hardware? Obviously, the goals of free software are going to clash head-on with you if you try to make your hardware work "best" with the software that you're developing in the open.
This is a genuine question and I'm wondering what the answer is, it's not rhetorical flamebait :-)
Part of that article struck me -- the apparent disdain of the Linux community for Apple's choice of BSD to base Mac OS X on. But having been a past and present Linux user professionally, and an exclusive FreeBSD user personally, I think I can see where Apple is looking.
Any professional OS that touts ease of configurability must have a consistent interface to everything underneath. The more I delve into Linux, the more I get confused by the different ways everything has to be done. I think this is a symptom of the "hack it till it just works" mentality.
I don't exclusively use FreeBSD because of its stability and security (although that was certainly the case when I switched a few years ago; Linux boxes were giving me and others unexplainable random problems left and right that an installation of FreeBSD remedied handily.) I continue to use FreeBSD today because things are done right and consistently. That might mean waiting a little longer for equivalent functionality, but I can trust said functionality.
The Linux systems I use seem to be able to do more, but I can't always find the correct way to do it. Perhaps that's more due to a lack of documentation, but when I do find the right way, it seems to be a hack that really doesn't fit into any consistent way of doing things. For example, getting my ATAPI CD-RW drive to work meant inserting a LILO flag. It seems to me that functionality should have been found elsewhere, perhaps in a kernel config file or by unloading one module and loading another. FreeBSD doesn't yet have IDE-SCSI support, but I bet when they do, it will be able to be enabled without rebooting and should also be easy to set up.
All in all, I can see the building of an organized, easily-setup, easily-administrated OS a lot more clearly on top of a BSD kernel. The tests of time have showed BSD to do well in those areas. I think Linux will get there, maybe on the next iteration of the kernel. 2.2 unloaded a lot of nasty old baggage that was around in 2.0, I have faith that 2.4 will be even better.
Of course, the rest of the story probably involves things like the GPL. Apple and other large companies who derive a huge bottom line from software licenses, I think, will continue to shy away from truly free licenses for some time. I think that is truly a shame for the forces of free software.
P.S. I have tried to be constructive here instead of being flamebait. I would appreciate replies that are in the same manner.
When I previewed that previous comment, it looked just great... but when I posted it, I guess the lt and gt entities got re-parsed :-) Here is paragraph #3 without entities:
Sorry about that, folks!
You're on the right track if you're saying that this technology wouldn't even be needed if we could convince people to stick to standards. But you also need one other thing -- well-formedness.
You can stick to HTML 4.0, even the "Strict" dialect (which I encourage everyone to do!), and still have pages that completely blow up when pulled up outside one of the Big Two. On the website that I accidentally deleted some time ago, I had struggled for some time to not just reach for HTML 4.0 compliance but for well-formedness.
It meant using elements for what they were intended for. It meant never using a table for anything other than tabular data. It meant using when I wanted emphasis and it meant using when I wanted to mark up code fragments (mind you, I'm stuck using right now because /.'s HTML filter doesn't permit !) It took some fiddling.
But I turned out with a set of pages that were not only easier to maintain, but CSS applied very cleanly to them, making them pretty and consistent-looking, and they rendered perfectly on any hand-held or speech-synthesizing device you could throw at me. My information was useful to everyone, and that was the best high of all.
I strongly encourage everyone to pursue well-formedness. The more important stuff that is well-formed instead of hacked, the better browsers we'll get, too!
The reason I'm always spewing out hardcopies is because online doco doesn't fit on the screen. Sure, I've got 1152x864 on a nice 17", but there just isn't enough room for your average HTML doco alongside a couple xterms.
Now, this isn't ALWAYS the case. There is standards-compliant documentation out there that will properly wrap to a narrow browser window -- but it's unfortunately far too rare. That and no matter whether you get the standards right or not, any graphical screenshots or tables you do are going to be pretty much unreadable at less than 400 pixels wide.
Besides, let's not forget that most paper doesn't come from the bad guys in Captain Planet tearing down rainforests just for the sheer orgasmic thrill of it. It comes from a combination of the reconstituted scraps from the last paper runs and tree farms, among other things.
It seems to be slowing but I think is still progressing. There are a few key areas that still need to be worked on. Lack of a good IPC library, for example, is standing in the way of a working port of PostgreSQL.
Lots of working stuff is ported to Cygwin, though. My favorite site for that stuff is the Cygwin Porting Project.
Another, even better option -- Cygnus' Cygwin. Quite a bit of commonly-used open source software is already ported (most GNU stuff compiles out of the box, but there's lots of other stuff full of Linuxisms that needed to be ported better.) It has the added advantage of being able to run on 95/98 as well, although with substantially fewer features (like file security.)
In addition, Cygwin seems to still be actively maintained. EGCS is available for it among other things (to be fair, the same guy that ports EGCS to Cygwin also does it for UWIN.)
That was the case before, but may not be entirely correct now. I am still looking for more information from ICANN but they seem to be concentrating on the political rather than the technical.
Nowadays there are alternative registrars that actually seem to take you through the whole process. I gather that NSI is still on the backend but you do not actually become an NSI customer anymore. Unfortunately I haven't had an opportunity to try anybody out yet, I have more important things to do like reading Slashdot (-:
Maybe we can get someone who works for one of those registrars to clear things up here?
(P.S. I will... ummm... not be dropping by NSI anytime soon with the explosives that I don't have. :-)
Yet another lovely step back in time by the Clinton administration. I wonder if any of the candidates for the next presidential election have gone on record for crypto policy.
The primary reason that the concept key escrow absolutely petrifies me is that the to be useful, the keys need to travel in one form or another from their central repository (which I would hope would be as tightly locked up as the NSA) to the law enforcement agency responsible for unlocking the message. With the repeated demonstrations by the U.S. Government that they don't understand crypto, what's even going to guarantee the safety of my key (and therefore my data) in transit?
Don't make me hand over my keys. I have them because they protect me. And you can bet that if key escrow becomes a requirement, I will not surrender my stock of open-source crypto software, but only begin to use it more.
No, you can't make that comparison. NT is still loaded with code that's been around since DOS, and OS's with Linux as the kernel are generally loaded with code from GNU that's been around a similar amount of time.
Unless you want to compare the NT kernel with the Linux kernel... :-) (But who can separate the NT kernel from anything... heck, you can't even pull it away from the GUI!)