Isn't it ironic, that the same market forces that are making it difficult to unseat Windows on the desktop (primarily, the network effect, along with the fact that people tend to stick with what they know) are preventing Windows from making it on the palmtop?
Microsoft doesn't know how to win over a market sector by making a better product. They only know how to steal a market sector by chaining a product to Office or Windows. That being impossible in the palmtop market, they don't stand a chance, unless they shipped a free Pocket PC with every copy of Windows. Hehe... "Well, sure it's portable, and you can take it anywhere, but it's an integrated part of the operating system and we can't just remove it!"
--
Two local loops, added distance
on
Homebrew S/ADSL
·
· Score: 4
Running short-haul modems (and now, point-to-point DSL equipment) over solid copper circuits has always been an interesting activity. This kind of thing has been getting done for decades in campus type situations, and for organizations that own multiple adjacent buildings.
The problem with deploying an ISP-style rollout of this technology is that the distance between the telco's central office and the ISP's POP now has to be added to the total length of every circuit. That can make a difference -- as anyone who lives on the outskirts of a town will tell you, with DSL every foot of cable counts!
I'd like to see this kind of thing tried with IDSL. IDSL uses 2B1Q encoding rather than the G.Lite technology typically used in an ADSL circuit -- it's basically an un-channelized ISDN line that terminates in a DSLAM instead of at the telco switch. The top speed it runs at is only 144kbps, but it can run nearly twice the distance that other DSL technologies can run -- and unlike ADSL and SDSL, the telco can extend IDSL circuits using U-loop repeaters. I'm considering ordering an IDSL circuit for my home.
Considering that the article we're currently commenting on discusses deployment of broadband in sparsely populated towns, it's likely that the extra distance will come in handy. I wonder if anyone is manufacturing peer IDSL modems?
This clearly demonstrates the problems associated with one entity having too much market share in any particular market. Any blacklist that bans AOL is shooting itself in the foot, because there's too much legitimate mail coming from the aol.com domain. For millions of people, AOL basically is the Internet. That's a problem. It demonstrates a problem we all know so well from the operating systems field: when one player has too much market share, they can basically act with reckless abandon. Everyone has to work with them or risk locking out their own customers, or potential customers.
Sure, but there are a couple of important differences between IOS and Windows.
For one thing, IOS just keeps going and going and going. Cisco routers have the kind of uptime we expect from our Linux systems.
IOS doesn't require constant upgrading. I've been installing routers for years, and some of those boxes have been running for that long without changing a thing. Set it and forget it.
When beating the KDE-vs-GNOME dead horse, there's something very important to keep in mind here. While most of us here are capable of switching between the two desktop environments, we are the Linux community. Most of the casual users being targeted here aren't part of the community, they're customers of the Linux industry.
Take a look at the commercial distributions and you'll find a nearly unanimous standardization on KDE - with RedHat being the obvious exception, of course, and Corel leading the charge.
Like it or not, while the Linux community is doing the parallel development thing, in the Linux industry, the race is pretty much over.
Yeah, I know about Eazel. Judging by the amount of hype they've generated, all I can say about them is: show me the code.
Now, the death of Netscape was sad, and it was due mostly to the fact that their playing field wasn't level
Excuse me?
NETSCAPE IS NOT DEAD! You sound like a ZDnet writer! Netscape is no longer an industry leader (thanks to their getting trampled on by the Beast of Redmond) but they're certainly NOT dead. They're still around, pouring immense resources into Mozilla, running a great portal, and will eventually become AOL's business brand. Whether you like them that way or not is irrelevant; they're still ALIVE either way.
Perhaps we should instead talk about Jamie Zawinski, and how the world would be better off if he were dead.
Where the mainframe excels is in its I/O channel. You can, for example, move gigabytes of data from one disk to another (or from disk to tape, or whatever) generating only one CPU interrupt. The channel is intelligent enough to do this kind of thing. That's why people tend to be surprised when they find out that the mainframe that just moved a few hundred million records around without breaking a sweat only has 64 MB of memory and a not-so-hot CPU.
Getting Linux onto the mainframe is a very important step, but it's only part of the picture. Facilities then need to be introduced to take advantage of the advanced I/O facilities that are available.
The Intel-PC world's attempt to imitate a mainframe's channel is Intelligent Input/Output (I2O). It does something similar, with intelligent peripherals designed to take the processing load off the main CPU. Mainframes have had this for decades. The Commodore 64 did, as well (I believe the article touched on this, actually). Now the PC world is finally catching up.
Alan Cox is working on getting I2O support worked into the Linux kernel. If the kernel interfaces for I2O are done with a sufficient level of abstraction, it is entirely possible that IBM could adapt them to use on an S/390 box as well.
I think what we're going to see in the future is the 'traditional unix-style software' developed with 'traditional unix-style tools'... and glitzy programs with lots of front-end eye candy developed with tools such as KDevelop.
Let's face it, we need the eye candy to attract the non-technical users and to get Linux running on demo machines in computer stores. The next great Linux desktop app will probably be written using something like KDevelop. The next great Linux server app will probably be written using vi or emacs.
His high-profile badmouthing of the Mozilla project is the primary reason for all of the death knells we keep hearing. And I personally think that the real reason for the Emacs/Xemacs split was a JWZ/RMS ego battle.
I personally hope that Microsoft and AOLscape end up settling down on nearly equal shares of the browser market. When two products have to go head-to-head for the long term, they've gotta constantly work on satisfying the customer. As a result, the customer wins. The perfect example is Coke and Pepsi.
I have very little respect for Jamie Zawinski. He seems to have a 'holier-than-thou' attitude about pretty much everything he touches. I'd guess that it was Zawinski's personality more than any technology or license issues that caused the Emacs split. Then his famous 'successful company' rant, which I read as 'anyone who joined Netscape after I did is a wannabe'.
But my biggest problem is the now-infamous rant he published upon leaving Netscape. He seemed determined to damage the company on his way out. I think he single-handedly did more to hurt public perception of Mozilla than any delay, any change of plans, any setbacks that the project itself ever had. Zawinski set the stage for the trade press to prematurely declare Mozilla a failure.
I truly hope that JWZ gets attacked by a rabid lizard. His personal vitriol is Mozilla's most significant liability, even though he's not involved anymore.
This, folks, is what happens when two or more companies are truly competing in a marketplace. Nobody can afford to get complacent, the technology (or whatever) advances at a rapid pace, and the prices are kept reasonable. The ultimate winner is the consumer.
This is the kind of thing you'd be seeing in software if Microsoft were not a monopoly. Linux has caused them to wake up a little, but if Microsoft had the kind of competition that Intel now faces, it would truly light a fire under their collective butts to deliver some real value to consumers.
I've thought about this problem a bit myself. One of the things we're working on for the Citadel BBS package is NNTP support directly in the server. Combine this with the text and web front ends that are already in place, and connectivity is universal.
Now I'm not saying that everyone should use Citadel, but putting support for popular message exchange protocols in all such products should be the norm. It allows for non-UseNet NNTP networks to emerge.
Yes, I said non-UseNet. While UseNet has a lot to offer (for example, lots of opportunities to make money fast and virtually unlimited pay-per-view pornography), there exist people who have a different idea of how a large distributed messaging network should be run. For example, there used to be a lot of people who preferred FidoNet over UseNet. And those of us on the Citadel network still enjoy the online company that's far more 'folksy' than UseNet.
Now here's the important point that I hope will get this post moderated up. As recently as five or ten years ago, it would have been foolish to connect your NNTP to anything other than UseNet, because many people only had UUCP connections, or some other low-grade means of connectivity to the outside world. But here and now, everyone's got TCP/IP to the open Internet. What this means is that multiple, smaller, NNTP networks is a real possibility. They can be operated by groups who have varying ideas of how it should be run, varying target audiences, etc. Since everyone has TCP/IP now, you simply point your newsreader at the network(s) you are interested in.
Yes, I really do think that 'one big UseNet' is an obsolete idea, and that's why it's deteriorated so much.
This is what happens when Internet protocols are developed in a closed manner and hoarded by their creators. If the Napster authors had published the protocol as an RFC and then brought their software to market, it would be far more difficult for them to get sued by the RIAA, and they wouldn't have to sue other parties that implemented the protocol.
This is the Instant Message fiasco all over again (a fiasco which still isn't resolved). Is this to be the way all new Internet technologies develop? To be brought to life in a proprietary coccoon, remaining that way until Linux users want to access them and are forced to reverse engineer?
Perhaps the biggest danger of attempting widespread deployment of a closed protocol is that it has a good chance of getting eclipsed by a competing closed protocol foisted by a bigger company (such as Microsoft or America Online).
The biggest technological drawback of the Napster protocol is that it depends on Napster's servers. This is the same problem that ICQ and AIM have. Any openness-minded developer would have designed it to run in a truly distributed fashion. Yes, the actual data transfer is peer-to-peer, but Napster could have been developed to do the peer locator without central servers as well. The first thought that comes to mind would be running the whole thing over IRC.
Well, those are my thoughts. I certainly hope this type of thing doesn't become the norm.
If there's one thing we have learned from the Year 2000 debacle, it is that people keep old equipment and technology around for FAR longer than even its originators ever would expect. The relation here? Simple: this copy protection mechanism is going to confuse the naive decoders in older vintage CD players.
Alienate the non computer literate CD buying public and they alienate their most loyal and honest customers.
Meanwhile, in the CD-ROM drive universe, workarounds will quickly be implemented. Since a Red Book CD is not encrypted, all they really have to distort is the 'encapsulation' of the data. Since the digital audio itself is recorded on the disc as plaintext... this really is a losing battle for the record company any way you slice it.
The more things change the more they stay the same
on
AOL Nation
·
· Score: 1
AOL/TW and Microsoft will keep each other in check. Internet standards will have to remain open in order for the Big Two to try to cross-lure each other's users.
With open standards, anyone with a server can continue to be a 'netcaster'. To put it in non-Internet terminology: AOL/TW, Microsoft, MCI/WorldCom, etc. are the big newspapers, while sites like Slashdot, UNCENSORED!, and The Onion are the 'zines.
I have to say, Mozilla is looking very nice these days. Can't wait to see it installed as the default browser in AOL 6.0 (are you listening, Steve?). It's quite usable on Linux now, though I still can't specify the font and font size I want to view pages with.
I can't imagine Mozilla immediately killing IE due to the 'it came with the computer' factor, but hopefully it'll at least keep the browser market somewhat balanced.
I've been hearing "BBS's are dead" for the past four or five years, and it's just not true. I've run one myself for nearly twelve years now. It's called UNCENSORED! BBS, and it runs on the Citadel software. You can telnet in, you can dial in, or you can use a browser-based front end.
It's a shame that dialup-only BBS's have, for the most part, faded away. Getting a BBS onto the Internet can really be a problem, but as DSL and Cable lower those obstacles, I think there will be more BBS's surfacing.
Please visit my BBS and see what you think. Those of us already in the community really enjoy it.
There's nothing about the GPL that would make NDS access for Linux impossible, because integrated access to a directory tree wouldn't go in the kernel. The proper place for it is in libc, which is LGPL'd. This is where NIS client-side code went, and it's where NDS client-side code would logically be placed. Problem solved.
Now that I think about it, I think it would make sense if the glibc were modified to allow plug-ins to change the behavior of various library calls. For example, on a vanilla glibc you call getpwnam() and it checks in/etc/passwd, but with an NDS plug-in installed, it would check the directory tree instead.
Actually, the real law of Zawinski goes something like this:
Any software package touched by Jamie Zawinski will get f**ked up until it no longer works properly, then he will make a big public show out of bailing out of the project.
Sorry, I just don't have any respect for Jamie Zawinski. He is personally responsible for a lot of the brain damage in XEmacs and Netscape, and he walked out of Netscape like a big crybaby when things weren't going his way. This person is not an asset to the community. He should go work for Microsoft.
Dave is absolutely correct: Mozilla is probably the single most important project for the future of free software. Nearly all technology development for the next decade will be tied to the web in some way, and it's absolutely vital that web technology be kept open.
Hopefully, AOL realizes this. If Microsoft ends up controlling the web, it's only a matter of time before AOL is reduced to insignifance for most purposes. Perhaps they don't realize just how urgent it is, though. AOL needs to make the Netscape Client Engineering Group a very high priority, and get Mozilla into the AOL client as soon as possible. This alone will shift the browser market away from Microsoft in a huge way. Yes, I know about the bundling deal, and I don't think it's worth it.
We do need to focus on more than just the browser, though. While Mozilla is absolutely the most important, we still need to have a diverse array of software available, to give the Linux platform some value, both on the desktop and server side. I personally am working on a replacement for MS Exchange and hopefully will be able to hook up with the developers of some of the better Outlook clones, in order to offer a nice end-to-end integrated solution. Mozilla tie-in? Absolutely. Everything's gotta work with the Web, and I've already got a good web-based front end in place.
Heed Dave's call and spread the word. This is very important.
Even the external DSL equipment isn't fully standardized yet. The upcoming standard is called g.Lite, but not every DSL equipment manufacturer is compliant yet. Give it some time.
Once that happens, I really see no reason why a DSL card couldn't be treated like an Ethernet card by the kernel. It doesn't need all the weird circuit-switch code that ISDN does... just initialize the card, download the configuration from the DSLAM, and bind an IP address.
Unfortunately, I think internal DSL cards aren't really going to make a big splash. All of the manufacturers are going to want to do USB instead; it's easier than trying to provide tech support to a bunch of home users who would otherwise have to open up their computers to plug in an internal DSL card, or even an internal Ethernet card to connect to an external DSL router.
Slashdot Mirror
Isn't it ironic, that the same market forces that are making it difficult to unseat Windows on the desktop (primarily, the network effect, along with the fact that people tend to stick with what they know) are preventing Windows from making it on the palmtop?
... "Well, sure it's portable, and you can take it anywhere, but it's an integrated part of the operating system and we can't just remove it!"
Microsoft doesn't know how to win over a market sector by making a better product. They only know how to steal a market sector by chaining a product to Office or Windows. That being impossible in the palmtop market, they don't stand a chance, unless they shipped a free Pocket PC with every copy of Windows. Hehe
--
Running short-haul modems (and now, point-to-point DSL equipment) over solid copper circuits has always been an interesting activity. This kind of thing has been getting done for decades in campus type situations, and for organizations that own multiple adjacent buildings.
The problem with deploying an ISP-style rollout of this technology is that the distance between the telco's central office and the ISP's POP now has to be added to the total length of every circuit. That can make a difference -- as anyone who lives on the outskirts of a town will tell you, with DSL every foot of cable counts!
I'd like to see this kind of thing tried with IDSL. IDSL uses 2B1Q encoding rather than the G.Lite technology typically used in an ADSL circuit -- it's basically an un-channelized ISDN line that terminates in a DSLAM instead of at the telco switch. The top speed it runs at is only 144kbps, but it can run nearly twice the distance that other DSL technologies can run -- and unlike ADSL and SDSL, the telco can extend IDSL circuits using U-loop repeaters. I'm considering ordering an IDSL circuit for my home.
Considering that the article we're currently commenting on discusses deployment of broadband in sparsely populated towns, it's likely that the extra distance will come in handy. I wonder if anyone is manufacturing peer IDSL modems?
--
This clearly demonstrates the problems associated with one entity having too much market share in any particular market. Any blacklist that bans AOL is shooting itself in the foot, because there's too much legitimate mail coming from the aol.com domain. For millions of people, AOL basically is the Internet. That's a problem. It demonstrates a problem we all know so well from the operating systems field: when one player has too much market share, they can basically act with reckless abandon. Everyone has to work with them or risk locking out their own customers, or potential customers.
--
- Dump the top 100 Microsoft execs feet-first into a wood chipper
- Destroy all existing copies of all Microsoft products
- Send anyone who, for some reason, likes Microsoft products to 're-education' camps
- Transform the former Microsoft headquarters into a giant paintball arena
There you have it. Very easy. Have fun, DOJ!--
Sure, but there are a couple of important differences between IOS and Windows.
For one thing, IOS just keeps going and going and going. Cisco routers have the kind of uptime we expect from our Linux systems.
IOS doesn't require constant upgrading. I've been installing routers for years, and some of those boxes have been running for that long without changing a thing. Set it and forget it.
--
When beating the KDE-vs-GNOME dead horse, there's something very important to keep in mind here. While most of us here are capable of switching between the two desktop environments, we are the Linux community. Most of the casual users being targeted here aren't part of the community, they're customers of the Linux industry.
Take a look at the commercial distributions and you'll find a nearly unanimous standardization on KDE - with RedHat being the obvious exception, of course, and Corel leading the charge.
Like it or not, while the Linux community is doing the parallel development thing, in the Linux industry, the race is pretty much over.
Yeah, I know about Eazel. Judging by the amount of hype they've generated, all I can say about them is: show me the code.
--
Excuse me?
NETSCAPE IS NOT DEAD! You sound like a ZDnet writer! Netscape is no longer an industry leader (thanks to their getting trampled on by the Beast of Redmond) but they're certainly NOT dead. They're still around, pouring immense resources into Mozilla, running a great portal, and will eventually become AOL's business brand. Whether you like them that way or not is irrelevant; they're still ALIVE either way.
Perhaps we should instead talk about Jamie Zawinski, and how the world would be better off if he were dead.
--
They needed to do this. Chris Carter has to show the Lone Gunmen prominently because of the rumored spinoff featuring them.
--
Where the mainframe excels is in its I/O channel. You can, for example, move gigabytes of data from one disk to another (or from disk to tape, or whatever) generating only one CPU interrupt. The channel is intelligent enough to do this kind of thing. That's why people tend to be surprised when they find out that the mainframe that just moved a few hundred million records around without breaking a sweat only has 64 MB of memory and a not-so-hot CPU.
Getting Linux onto the mainframe is a very important step, but it's only part of the picture. Facilities then need to be introduced to take advantage of the advanced I/O facilities that are available.
The Intel-PC world's attempt to imitate a mainframe's channel is Intelligent Input/Output (I2O). It does something similar, with intelligent peripherals designed to take the processing load off the main CPU. Mainframes have had this for decades. The Commodore 64 did, as well (I believe the article touched on this, actually). Now the PC world is finally catching up.
Alan Cox is working on getting I2O support worked into the Linux kernel. If the kernel interfaces for I2O are done with a sufficient level of abstraction, it is entirely possible that IBM could adapt them to use on an S/390 box as well.
--
I think what we're going to see in the future is the 'traditional unix-style software' developed with 'traditional unix-style tools' ... and glitzy programs with lots of front-end eye candy developed with tools such as KDevelop.
Let's face it, we need the eye candy to attract the non-technical users and to get Linux running on demo machines in computer stores. The next great Linux desktop app will probably be written using something like KDevelop. The next great Linux server app will probably be written using vi or emacs.
--
His high-profile badmouthing of the Mozilla project is the primary reason for all of the death knells we keep hearing. And I personally think that the real reason for the Emacs/Xemacs split was a JWZ/RMS ego battle.
Truly, this guy is not worth our attention.
--
I personally hope that Microsoft and AOLscape end up settling down on nearly equal shares of the browser market. When two products have to go head-to-head for the long term, they've gotta constantly work on satisfying the customer. As a result, the customer wins. The perfect example is Coke and Pepsi.
--
I have very little respect for Jamie Zawinski. He seems to have a 'holier-than-thou' attitude about pretty much everything he touches. I'd guess that it was Zawinski's personality more than any technology or license issues that caused the Emacs split. Then his famous 'successful company' rant, which I read as 'anyone who joined Netscape after I did is a wannabe'.
But my biggest problem is the now-infamous rant he published upon leaving Netscape. He seemed determined to damage the company on his way out. I think he single-handedly did more to hurt public perception of Mozilla than any delay, any change of plans, any setbacks that the project itself ever had. Zawinski set the stage for the trade press to prematurely declare Mozilla a failure.
I truly hope that JWZ gets attacked by a rabid lizard. His personal vitriol is Mozilla's most significant liability, even though he's not involved anymore.
This, folks, is what happens when two or more companies are truly competing in a marketplace. Nobody can afford to get complacent, the technology (or whatever) advances at a rapid pace, and the prices are kept reasonable. The ultimate winner is the consumer.
This is the kind of thing you'd be seeing in software if Microsoft were not a monopoly. Linux has caused them to wake up a little, but if Microsoft had the kind of competition that Intel now faces, it would truly light a fire under their collective butts to deliver some real value to consumers.
I've thought about this problem a bit myself. One of the things we're working on for the Citadel BBS package is NNTP support directly in the server. Combine this with the text and web front ends that are already in place, and connectivity is universal.
Now I'm not saying that everyone should use Citadel, but putting support for popular message exchange protocols in all such products should be the norm. It allows for non-UseNet NNTP networks to emerge.
Yes, I said non-UseNet. While UseNet has a lot to offer (for example, lots of opportunities to make money fast and virtually unlimited pay-per-view pornography), there exist people who have a different idea of how a large distributed messaging network should be run. For example, there used to be a lot of people who preferred FidoNet over UseNet. And those of us on the Citadel network still enjoy the online company that's far more 'folksy' than UseNet.
Now here's the important point that I hope will get this post moderated up. As recently as five or ten years ago, it would have been foolish to connect your NNTP to anything other than UseNet, because many people only had UUCP connections, or some other low-grade means of connectivity to the outside world. But here and now, everyone's got TCP/IP to the open Internet. What this means is that multiple, smaller, NNTP networks is a real possibility. They can be operated by groups who have varying ideas of how it should be run, varying target audiences, etc. Since everyone has TCP/IP now, you simply point your newsreader at the network(s) you are interested in.
Yes, I really do think that 'one big UseNet' is an obsolete idea, and that's why it's deteriorated so much.
This is what happens when Internet protocols are developed in a closed manner and hoarded by their creators. If the Napster authors had published the protocol as an RFC and then brought their software to market, it would be far more difficult for them to get sued by the RIAA, and they wouldn't have to sue other parties that implemented the protocol.
This is the Instant Message fiasco all over again (a fiasco which still isn't resolved). Is this to be the way all new Internet technologies develop? To be brought to life in a proprietary coccoon, remaining that way until Linux users want to access them and are forced to reverse engineer?
Perhaps the biggest danger of attempting widespread deployment of a closed protocol is that it has a good chance of getting eclipsed by a competing closed protocol foisted by a bigger company (such as Microsoft or America Online).
The biggest technological drawback of the Napster protocol is that it depends on Napster's servers. This is the same problem that ICQ and AIM have. Any openness-minded developer would have designed it to run in a truly distributed fashion. Yes, the actual data transfer is peer-to-peer, but Napster could have been developed to do the peer locator without central servers as well. The first thought that comes to mind would be running the whole thing over IRC.
Well, those are my thoughts. I certainly hope this type of thing doesn't become the norm.
If there's one thing we have learned from the Year 2000 debacle, it is that people keep old equipment and technology around for FAR longer than even its originators ever would expect. The relation here? Simple: this copy protection mechanism is going to confuse the naive decoders in older vintage CD players.
Alienate the non computer literate CD buying public and they alienate their most loyal and honest customers.
Meanwhile, in the CD-ROM drive universe, workarounds will quickly be implemented. Since a Red Book CD is not encrypted, all they really have to distort is the 'encapsulation' of the data. Since the digital audio itself is recorded on the disc as plaintext... this really is a losing battle for the record company any way you slice it.
AOL/TW and Microsoft will keep each other in check. Internet standards will have to remain open in order for the Big Two to try to cross-lure each other's users.
With open standards, anyone with a server can continue to be a 'netcaster'. To put it in non-Internet terminology: AOL/TW, Microsoft, MCI/WorldCom, etc. are the big newspapers, while sites like Slashdot, UNCENSORED!, and The Onion are the 'zines.
I have to say, Mozilla is looking very nice these days. Can't wait to see it installed as the default browser in AOL 6.0 (are you listening, Steve?). It's quite usable on Linux now, though I still can't specify the font and font size I want to view pages with.
I can't imagine Mozilla immediately killing IE due to the 'it came with the computer' factor, but hopefully it'll at least keep the browser market somewhat balanced.
I've been hearing "BBS's are dead" for the past four or five years, and it's just not true. I've run one myself for nearly twelve years now. It's called UNCENSORED! BBS, and it runs on the Citadel software. You can telnet in, you can dial in, or you can use a browser-based front end.
It's a shame that dialup-only BBS's have, for the most part, faded away. Getting a BBS onto the Internet can really be a problem, but as DSL and Cable lower those obstacles, I think there will be more BBS's surfacing.
Please visit my BBS and see what you think. Those of us already in the community really enjoy it.
There's nothing about the GPL that would make NDS access for Linux impossible, because integrated access to a directory tree wouldn't go in the kernel. The proper place for it is in libc, which is LGPL'd. This is where NIS client-side code went, and it's where NDS client-side code would logically be placed. Problem solved.
/etc/passwd, but with an NDS plug-in installed, it would check the directory tree instead.
Now that I think about it, I think it would make sense if the glibc were modified to allow plug-ins to change the behavior of various library calls. For example, on a vanilla glibc you call getpwnam() and it checks in
- It's late to market, by a VERY long period of time.
- Open source developers aren't exactly lining up to help out with the project.
- In the meantime, a much better product is gaining significant market share.
Indeed, when you add up the facts, there can be no doubt: Windows 2000 is dead.Dave is absolutely correct: Mozilla is probably the single most important project for the future of free software. Nearly all technology development for the next decade will be tied to the web in some way, and it's absolutely vital that web technology be kept open.
Hopefully, AOL realizes this. If Microsoft ends up controlling the web, it's only a matter of time before AOL is reduced to insignifance for most purposes. Perhaps they don't realize just how urgent it is, though. AOL needs to make the Netscape Client Engineering Group a very high priority, and get Mozilla into the AOL client as soon as possible. This alone will shift the browser market away from Microsoft in a huge way. Yes, I know about the bundling deal, and I don't think it's worth it.
We do need to focus on more than just the browser, though. While Mozilla is absolutely the most important, we still need to have a diverse array of software available, to give the Linux platform some value, both on the desktop and server side. I personally am working on a replacement for MS Exchange and hopefully will be able to hook up with the developers of some of the better Outlook clones, in order to offer a nice end-to-end integrated solution. Mozilla tie-in? Absolutely. Everything's gotta work with the Web, and I've already got a good web-based front end in place.
Heed Dave's call and spread the word. This is very important.
Even the external DSL equipment isn't fully standardized yet. The upcoming standard is called g.Lite, but not every DSL equipment manufacturer is compliant yet. Give it some time.
... just initialize the card, download the configuration from the DSLAM, and bind an IP address.
Once that happens, I really see no reason why a DSL card couldn't be treated like an Ethernet card by the kernel. It doesn't need all the weird circuit-switch code that ISDN does
Unfortunately, I think internal DSL cards aren't really going to make a big splash. All of the manufacturers are going to want to do USB instead; it's easier than trying to provide tech support to a bunch of home users who would otherwise have to open up their computers to plug in an internal DSL card, or even an internal Ethernet card to connect to an external DSL router.