So what if NT can throw out static pages faster? I'm sticking with UNIXish systems because when I use them, I know I've got more power at my fingertips than NT's pop-up happy dungeon could ever give me. I can change the code on most systems... I can craft my own tools by combining existing ones... and I can find the answers to my questions on the Web rather than having to pay "per-incident" fees.
On top of it all, if anything does go wrong, 99.44% of the time I can fix it without standing in front of the server. NT doesn't have that capability. If a service needs to be upgraded I can do it without affecting anything else on the box.
UNIX is power because you can insert yourself at virtually any step along the way of any process. You want to do anything in NT, you must do it with whatever APIs MS thinks you need, or pay them to make more.
IF you think this is impossible, or even just unlikely, keep in mind that MS engineered Windows 3.x specifically to bomb out when running under DR-DOS in order to keep MS-DOS sales up.
I would have to say that the operating system itself is very minimalistic; in its strictest sense it would be the pieces that insulate the code from the unpleasant job of being aware of the hardware.
On top of that we have everything else: GUIs, toolkits, APIs for "system services", etc; and these are the platform that an application developer expects to be in place so he can say "xxx application runs on The Foo Platform(TM)".
I recently read Neal Stephenson's In the Beginning was the Command Line (good read, I wonder if that link is a legal copy?) and can't help but think about the probability that without support from Windows' home edition (whenever and whatever it may be), the IA-64 won't become a commodity chip and may be just as out of reach for Joe Hacker as the rest of the chip world (with the notable exception of PPC, of course.)
I'd hate to see that happen. I would love to ditch bloat on my CPUs as much as I loved ditching bloat in my OS:-)
Perhaps I did misspeak myself, at least in the use of a catchy title.:-) My concern is that making it open source might not be enough to resuscitate it. If it's in a decrepit state, it might need a large team of paid engineers to bring it to the point where it can be used, praised, and worked on by all kinds of people -- just like Mozilla is today.
Let me try again with a definition of what I call the "Mozilla Effect": open source in an of itself is not the magic bullet by which excessively large chunks of previously proprietary code can achieve greatness; it's just a building block. You need to infuse a project with a large team of knowledgeable people with oodles of time on their hands to achieve the next step.
I'm beginning to wonder about ESR. His quest to get any and all software on the face of the planet open sourced is noble, indeed. (I would love to see it.) But I think his credibility is beginning to be damaged every time he tells someone that ejecting an aging codebase into the 'net is going to magically resuscitate it.
To be fair, I haven't read the proposal yet, and I certainly don't have any idea what state the code is in. But I'd wager that it's in a similar state to the Communicator code when Mozilla was first launched: lots and lots of interesting code, but you can't really compile it into much.
Mozilla required almost complete rewriting with extensive corporate support to start shaping up as the awesome browser it will be. (One or two more Milestones, and I'll be happy to retire my well-worn copy of Communicator.) I wonder what AmigaOS will need, and if there is sufficient support to bring AmigaOS into the next millenium.
You know, this very thing was an idea from BOFH after it was picked up as a column in Network Week. Sadly, Network Week has folded, but you can still get the articles off Google if you try. I was lucky enough to grab archives from 1995-1998 before it shut down.
When I read this, I was astonished -- momentarily. Then I realized this was exactly the sort of nonsense that would arise from the cry for more campaign finance regulations. I'm sure there are plenty of clueless folk out there who would rather call this a "loophole" than a bad ruling.
The solution is simple: throw out all the ridiculous regulations on campaign financing and simply require full disclosure. If the FEC (which would probably lobby against this, since it would threaten their existence) decided it wanted to classify your site as a political expenditure, all that would be required of you is to have your name there.
Warning: this isn't totally related to Caldera; but Caldera and Nick's experiences with it illuminate the issues herein.
From the article (emphasis mine):
I discovered later that Caldera includes Pine in the/col/contrib/RPMS directory, whence I eventually installed it. But first I tried to install Pine from my Mandrake 6.1 CD-ROM. The Mandrake Pine RPM complained that my ncurses libraries were too old. I tried to upgrade the ncurses libraries from the Mandrake CD-ROM, but that RPM refused to work because it couldn't find other libraries upon which it depends.
The RPM file was mistaken -- the libraries were there. It just couldn't find them. This is a common problem you run into when you try to mix and match RPM installations from different distributions. It's a relatively easy problem to solve (you can use an RPM option like --nodeps, for example), but it can be frustrating if you don't know what you're doing. And it doesn't help to improve Linux's image as far as perceived compatibility between distributions goes.
Obviously anyone can come up with quite a large number of reasons why this particular situation exists. My personal favorite is the oft-cited "reckless abandon" with which incompatible versions of libraries are created. Honestly, though, it really doesn't matter. The package management system should make this transparent.
RPM seems to have become the crown prince of package management because it (a) has most of its sh*t together and (b) by virtue of being in what is widely regarded as the single most popular Linux-based distro out there. Many have even argued that it's cross-platform -- i.e. given RPM packages compiled for any system other than Linux, it will work on said system. (And to a degree, it does.)
Unfortunately, even with all the standards in the universe, there just plain are too many things that differ between system A and system B. Red Hat and Mandrake have not had to deal with this on too large of a scale because Mandrake is a direct derivative (and I am operating from hearsay here -- I have not yet had the opportunity to use Mandrake) of Red Hat. But as more and more distributions out there use RPM, we will start to discover more and more inconsistencies that perhaps RPM just isn't up to.
Ah, but this is the world of open source.:-) We can change these things, either by modifying RPM, another system, or by creating a new one altogether with bits and pieces.
An ideal package system would be able to read some sort of platform spec file (or perhaps query an interactive spec program?) and determine just where libraries are supposed to be, for starters. It would ask the same spec where to get those libraries if they weren't around. (You could always do the Microsoft thing and include the libs you needed with your package, but I thought we were trying to get *away* from waste):-) Along those lines, library packages -- or any others pulled in automatically as dependencies -- could also be marked as "dependency-only" and removed or possibly archived when no longer depended on.
One thing RPM does well, though, and I wish more open source projects did -- it maintains the ability to run on the command line, in batch mode. Graphical programs run on top of it. We must never lose that delineation.
What makes or breaks an OS? A short trip down at least the recent history of failed computers and those that stick around for the long run should show you.
It doesn't matter how wonderful your operating system is. It doesn't even really matter how stable, or how many features it has, or anything. No company, not even one as big as Microsoft, can create an OS and then create every application that will ever be needed for it. This requires inspiring a development community to develop for your OS. I submit that this is the reason that lesser-known, though probably superior, operating systems have failed and others have grown up -- by enticing or turning away developers.
Microsoft has virtually overlooked one part of the development community for quite some time now; the public domain/freeware/open source part. Admittedly, open source for Microsoft platforms is still next to nonexistent; most software with an open source license is also ported to UN*Xish platforms; but until Linux started taking off, the PD/freeware crowd was still going strong on Windows.
Now recently, many companies who previously sold software for Microsoft platforms almost exclusively have announced their intention to bring their applications to Linux. I firmly believe this is largely due to the precursor loss of the PD/free/open crowd. I know I've seen this pattern before. My best guess is that it follows the learning pattern of a programmer; whichever OS and development tools you can get your hands on without spending a fortune, you start to learn to develop for -- later in life, when you want to make money for developing, you still feel most comfortable on that platform.
In any event, Microsoft now sees too late the pending loss of their developer base, and is taking the completely incorrect route to recapture it. MainSoft's product will probably address the high-end, large-scale, mucho dinero market quite well. However, it won't do anything for the PD/free/open community, and as such it won't be anything more than a temporary win. If Microsoft was really interested in recapturing developer share -- which they absolutely must do to survive, they need to relax their ever-tightening grip and open everything up.
If I were running Microsoft, I would open up MainWin. Make it completely open source. I don't think it would be good for the developer community at all but it would definitely insure a win.
If alot of windows programs start flowing into *nix it could cause interest in open source to taper off.
I hadn't even thought of this point. If the licensing of MainWin was done in a certain way, it could actually end up being a license violation to make a MainWin-hosted program open source. Kind of like (and I know I risk making an apples-to-oranges comparison here, but bear with me) QT's "free" edition forces GPL, and to that end any GPL'd library won't let you link proprietary stuff to it.
This would seem to be quite in line with the tagline on their website that makes me more nervous than anything:
Develop on Windows, run on UNIX
If this self-titled conspiracy theory is to be believed, here's a perfect reason to do it -- to drag potentially drifting developers back to Win32 with the promise of cross-platform binary compatibility.
All we can do it wait and see, I suppose. Oh well, I prefer programming in UN*X any day:-)
Re:Or run the OpenBSD hppa port ? Free, downloadab
on
PuffinFest at ALS
·
· Score: 2
I was under the impression (from the page) that OpenBSD/hppa isn't really "usable" yet either. Correct me if I'm wrong...
I don't call what O2K saves as "XML". Show me the DTD and the style sheets so I can edit it elsewhere. Show me the specifications so other software can read it. Then I'll call it XML.
What bothered me most about outtakes from that interview appearing in the article was the following quote from Mr. Gates:
"What we do is create tools like a word processor that lets people express their ideas and we're not at all involved in how they choose to use it.
"The thing we're involved with has democratised communications and anyone can publish on the internet. You don't have to own a TV channel or a newspaper."
This is bunk. Any software, and especially Microsoft's, has a keen influence on "how" it is used. I would go so far to say that even posts on/. carry a certain weight from the/. software, though not as much baggage as something like Word. The capabilities and the automation of software influence greatly what you can say just as the language you write in. The earlier comment about MHz vs. Mhz is a small example; I would submit that Word's set of templates, rules about documents, and "autocorrection" lends its influence to the output greatly.
I am at least glad to hear he said "The thing we're involved with has democratised..." instead of "We have democratised...". Although the Internet probably would have had a slightly different face without the influence of Microsoft and Internet Explorer, it would certainly still be democratized! I hate to sound crochety, but I feel it was much better before Windows had TCP/IP capability:-)
Prob. not usable yet, but get HP-UX 10.20 gratis
on
PuffinFest at ALS
·
· Score: 2
I haven't checked up on the port in a few months, but I doubt it's anywhere near as usable as your typical setup. It should pick up steam soon though.
What I would like to tell everyone who has an opportunity to pick up a 9000/7xx machine is to go ahead and do it, because even if it comes with HP-UX 9.xx, you can get an ugprade to HP-UX 10.20 gratis from HP as part of their Y2K program. I "rescued" a 720 from an engineering firm who was "upgrading" to NT and have been happily using it for quite some time now at least as a glorified X terminal for my FreeBSD box. My copy of 10.20 is in the mail.:-) In any event, you can have a very usable system -- and quite possibly learn a bit about other UNIXes (which is a very good thing, IMO) while you wait for the Puffin Group to bring their work to a more usable state.
Grab them while you can! They're really nice boxes.
Those tiny, underpowered "fuel-efficient" cars don't sell well in the States. Maybe it's because our land mass is so much larger, so we have so much farther to go.:-) When I went to get a car recently, I started out thinking about saving some money, but the efficient models I tried just cramped me to no end.
MTV2 comes on my DirecTV, and every time I've gone through the guide they're showing music videos.
If there wasn't so much marketing power behind the name "MTV", I would suggest they drop that moniker altogether and start calling themselves something else, since they really don't do much with music anymore.
I like Button1/Button2 myself. It makes a lot of sense. Unfortunately it leads to two problems for me (EEMMV):
At work, I have a two-button mouse (I really must get that replaced.) At home, I have several machines with three-button mice. I find myself, when at home, trying to paste by clicking the left and right buttons simultaneously. That doesn't work plus it screws up whatever I was trying to copy and paste.:-)
After having used the Windows/Macintosh select-then-copy metaphor for so long, I find myself selecting something then going over to paste it where text already exists. Selecting that text to delete it kills my selection process and I have to go back. This is just something I have to train myself to do, though -- clear the destination before trying to move the text.
(Boy, this is really getting off-topic fast, isn't it?):-)
Slashdot Mirror
So what if NT can throw out static pages faster? I'm sticking with UNIXish systems because when I use them, I know I've got more power at my fingertips than NT's pop-up happy dungeon could ever give me. I can change the code on most systems... I can craft my own tools by combining existing ones... and I can find the answers to my questions on the Web rather than having to pay "per-incident" fees.
On top of it all, if anything does go wrong, 99.44% of the time I can fix it without standing in front of the server. NT doesn't have that capability. If a service needs to be upgraded I can do it without affecting anything else on the box.
UNIX is power because you can insert yourself at virtually any step along the way of any process. You want to do anything in NT, you must do it with whatever APIs MS thinks you need, or pay them to make more.
I'll stick with the power, thank you.
IF you think this is impossible, or even just unlikely, keep in mind that MS engineered Windows 3.x specifically to bomb out when running under DR-DOS in order to keep MS-DOS sales up.
Normally I ignore flamebait like this, but this particularly bothers me.
Where is it written that...
Your most treasured "techies" work on your web page?
The Web is the be-all and end-all of any commercial venture?
A company even has to have a web page to be a useful part of society?
Yeesh.
The news icon pretty much sucks.
Why not a great, big, ominous obelisk, like in that much-hyped sci-fi movie we all hate to love?
I would have to say that the operating system itself is very minimalistic; in its strictest sense it would be the pieces that insulate the code from the unpleasant job of being aware of the hardware.
On top of that we have everything else: GUIs, toolkits, APIs for "system services", etc; and these are the platform that an application developer expects to be in place so he can say "xxx application runs on The Foo Platform(TM)".
I recently read Neal Stephenson's In the Beginning was the Command Line (good read, I wonder if that link is a legal copy?) and can't help but think about the probability that without support from Windows' home edition (whenever and whatever it may be), the IA-64 won't become a commodity chip and may be just as out of reach for Joe Hacker as the rest of the chip world (with the notable exception of PPC, of course.)
I'd hate to see that happen. I would love to ditch bloat on my CPUs as much as I loved ditching bloat in my OS :-)
Perhaps I did misspeak myself, at least in the use of a catchy title. :-) My concern is that making it open source might not be enough to resuscitate it. If it's in a decrepit state, it might need a large team of paid engineers to bring it to the point where it can be used, praised, and worked on by all kinds of people -- just like Mozilla is today.
Let me try again with a definition of what I call the "Mozilla Effect": open source in an of itself is not the magic bullet by which excessively large chunks of previously proprietary code can achieve greatness; it's just a building block. You need to infuse a project with a large team of knowledgeable people with oodles of time on their hands to achieve the next step.
HTH, and I'm sorry if I misspoke.
I'm beginning to wonder about ESR. His quest to get any and all software on the face of the planet open sourced is noble, indeed. (I would love to see it.) But I think his credibility is beginning to be damaged every time he tells someone that ejecting an aging codebase into the 'net is going to magically resuscitate it.
To be fair, I haven't read the proposal yet, and I certainly don't have any idea what state the code is in. But I'd wager that it's in a similar state to the Communicator code when Mozilla was first launched: lots and lots of interesting code, but you can't really compile it into much.
Mozilla required almost complete rewriting with extensive corporate support to start shaping up as the awesome browser it will be. (One or two more Milestones, and I'll be happy to retire my well-worn copy of Communicator.) I wonder what AmigaOS will need, and if there is sufficient support to bring AmigaOS into the next millenium.
It has most, but regrettably it trails off in mid-1997, to say nothing of 1998 or 1999. Personally, I would love to see BOFH return in print. Sigh.
But we're getting off-topic :-)
You know, this very thing was an idea from BOFH after it was picked up as a column in Network Week. Sadly, Network Week has folded, but you can still get the articles off Google if you try. I was lucky enough to grab archives from 1995-1998 before it shut down.
Great, but does Debian deal with cross-platform issues? Would Debian function just as well running off the idiosyncracies of, say, a Solaris system?
When I read this, I was astonished -- momentarily. Then I realized this was exactly the sort of nonsense that would arise from the cry for more campaign finance regulations. I'm sure there are plenty of clueless folk out there who would rather call this a "loophole" than a bad ruling.
The solution is simple: throw out all the ridiculous regulations on campaign financing and simply require full disclosure. If the FEC (which would probably lobby against this, since it would threaten their existence) decided it wanted to classify your site as a political expenditure, all that would be required of you is to have your name there.
Warning: this isn't totally related to Caldera; but Caldera and Nick's experiences with it illuminate the issues herein.
From the article (emphasis mine):
Obviously anyone can come up with quite a large number of reasons why this particular situation exists. My personal favorite is the oft-cited "reckless abandon" with which incompatible versions of libraries are created. Honestly, though, it really doesn't matter. The package management system should make this transparent.
RPM seems to have become the crown prince of package management because it (a) has most of its sh*t together and (b) by virtue of being in what is widely regarded as the single most popular Linux-based distro out there. Many have even argued that it's cross-platform -- i.e. given RPM packages compiled for any system other than Linux, it will work on said system. (And to a degree, it does.)
Unfortunately, even with all the standards in the universe, there just plain are too many things that differ between system A and system B. Red Hat and Mandrake have not had to deal with this on too large of a scale because Mandrake is a direct derivative (and I am operating from hearsay here -- I have not yet had the opportunity to use Mandrake) of Red Hat. But as more and more distributions out there use RPM, we will start to discover more and more inconsistencies that perhaps RPM just isn't up to.
Ah, but this is the world of open source. :-) We can change these things, either by modifying RPM, another system, or by creating a new one altogether with bits and pieces.
An ideal package system would be able to read some sort of platform spec file (or perhaps query an interactive spec program?) and determine just where libraries are supposed to be, for starters. It would ask the same spec where to get those libraries if they weren't around. (You could always do the Microsoft thing and include the libs you needed with your package, but I thought we were trying to get *away* from waste) :-) Along those lines, library packages -- or any others pulled in automatically as dependencies -- could also be marked as "dependency-only" and removed or possibly archived when no longer depended on.
One thing RPM does well, though, and I wish more open source projects did -- it maintains the ability to run on the command line, in batch mode. Graphical programs run on top of it. We must never lose that delineation.
What makes or breaks an OS? A short trip down at least the recent history of failed computers and those that stick around for the long run should show you.
It doesn't matter how wonderful your operating system is. It doesn't even really matter how stable, or how many features it has, or anything. No company, not even one as big as Microsoft, can create an OS and then create every application that will ever be needed for it. This requires inspiring a development community to develop for your OS. I submit that this is the reason that lesser-known, though probably superior, operating systems have failed and others have grown up -- by enticing or turning away developers.
Microsoft has virtually overlooked one part of the development community for quite some time now; the public domain/freeware/open source part. Admittedly, open source for Microsoft platforms is still next to nonexistent; most software with an open source license is also ported to UN*Xish platforms; but until Linux started taking off, the PD/freeware crowd was still going strong on Windows.
Now recently, many companies who previously sold software for Microsoft platforms almost exclusively have announced their intention to bring their applications to Linux. I firmly believe this is largely due to the precursor loss of the PD/free/open crowd. I know I've seen this pattern before. My best guess is that it follows the learning pattern of a programmer; whichever OS and development tools you can get your hands on without spending a fortune, you start to learn to develop for -- later in life, when you want to make money for developing, you still feel most comfortable on that platform.
In any event, Microsoft now sees too late the pending loss of their developer base, and is taking the completely incorrect route to recapture it. MainSoft's product will probably address the high-end, large-scale, mucho dinero market quite well. However, it won't do anything for the PD/free/open community, and as such it won't be anything more than a temporary win. If Microsoft was really interested in recapturing developer share -- which they absolutely must do to survive, they need to relax their ever-tightening grip and open everything up.
If I were running Microsoft, I would open up MainWin. Make it completely open source. I don't think it would be good for the developer community at all but it would definitely insure a win.
I hadn't even thought of this point. If the licensing of MainWin was done in a certain way, it could actually end up being a license violation to make a MainWin-hosted program open source. Kind of like (and I know I risk making an apples-to-oranges comparison here, but bear with me) QT's "free" edition forces GPL, and to that end any GPL'd library won't let you link proprietary stuff to it.
Definitely food for thought.
This would seem to be quite in line with the tagline on their website that makes me more nervous than anything:
If this self-titled conspiracy theory is to be believed, here's a perfect reason to do it -- to drag potentially drifting developers back to Win32 with the promise of cross-platform binary compatibility.
All we can do it wait and see, I suppose. Oh well, I prefer programming in UN*X any day :-)
I was under the impression (from the page) that OpenBSD/hppa isn't really "usable" yet either. Correct me if I'm wrong...
I don't call what O2K saves as "XML". Show me the DTD and the style sheets so I can edit it elsewhere. Show me the specifications so other software can read it. Then I'll call it XML.
What bothered me most about outtakes from that interview appearing in the article was the following quote from Mr. Gates:
This is bunk. Any software, and especially Microsoft's, has a keen influence on "how" it is used. I would go so far to say that even posts on /. carry a certain weight from the /. software, though not as much baggage as something like Word. The capabilities and the automation of software influence greatly what you can say just as the language you write in. The earlier comment about MHz vs. Mhz is a small example; I would submit that Word's set of templates, rules about documents, and "autocorrection" lends its influence to the output greatly.
I am at least glad to hear he said "The thing we're involved with has democratised..." instead of "We have democratised...". Although the Internet probably would have had a slightly different face without the influence of Microsoft and Internet Explorer, it would certainly still be democratized! I hate to sound crochety, but I feel it was much better before Windows had TCP/IP capability :-)
I haven't checked up on the port in a few months, but I doubt it's anywhere near as usable as your typical setup. It should pick up steam soon though.
What I would like to tell everyone who has an opportunity to pick up a 9000/7xx machine is to go ahead and do it, because even if it comes with HP-UX 9.xx, you can get an ugprade to HP-UX 10.20 gratis from HP as part of their Y2K program. I "rescued" a 720 from an engineering firm who was "upgrading" to NT and have been happily using it for quite some time now at least as a glorified X terminal for my FreeBSD box. My copy of 10.20 is in the mail. :-) In any event, you can have a very usable system -- and quite possibly learn a bit about other UNIXes (which is a very good thing, IMO) while you wait for the Puffin Group to bring their work to a more usable state.
Grab them while you can! They're really nice boxes.
Those tiny, underpowered "fuel-efficient" cars don't sell well in the States. Maybe it's because our land mass is so much larger, so we have so much farther to go. :-) When I went to get a car recently, I started out thinking about saving some money, but the efficient models I tried just cramped me to no end.
*sigh* Good news, if only I could get out there. Would it be too much to ask to hold one in Chicago or somewhere nearer the East Coast next year? :-)
There are people like this, but they're not hackers. They're generally called "script kiddies".
MTV2 comes on my DirecTV, and every time I've gone through the guide they're showing music videos.
If there wasn't so much marketing power behind the name "MTV", I would suggest they drop that moniker altogether and start calling themselves something else, since they really don't do much with music anymore.
I like Button1/Button2 myself. It makes a lot of sense. Unfortunately it leads to two problems for me (EEMMV):
At work, I have a two-button mouse (I really must get that replaced.) At home, I have several machines with three-button mice. I find myself, when at home, trying to paste by clicking the left and right buttons simultaneously. That doesn't work plus it screws up whatever I was trying to copy and paste. :-)
After having used the Windows/Macintosh select-then-copy metaphor for so long, I find myself selecting something then going over to paste it where text already exists. Selecting that text to delete it kills my selection process and I have to go back. This is just something I have to train myself to do, though -- clear the destination before trying to move the text.
(Boy, this is really getting off-topic fast, isn't it?) :-)