Re:Sounds like a good idea, but..
on
The Dot in .mars
·
· Score: 1
We've been in this situation before - and indeed we still are in some ways. How do you, for example, communicate with a ship at sea whose only access is via satellites that they may or may not be able to reach, depending on the weather? The answer is typically uucp. Admittedly, this doesn't allow for streaming media or web surfing, but email and file transfer work. I suspect that the most likely configuration would be a uucp gateway on each end, with some automation being used to keep servers in sync. So if I am on Mars, and create a VR tour with my handheld camera, I can have it sent to a server on Earth for those on Earth who want to see it. Trying to serve that to Earth from Mars (or vice versa) would be a pain in the neck.
Re:Isn't .mars a bit of an Ameriocentric name?
on
The Dot in .mars
·
· Score: 1
If I get there first, I'm calling it Ares. Mostly just to annoy all of the people who would have to rewrite the textbooks.
What if you came into the situation where you could use something you may have coded privately, for a company project?
Has happened to me more than once. In any case where I have to sign over rights to previous work if used on a project, I inform the company that I will sign such an agreement, but that I will bring no previous code into the project, and will simply reimplement everything I need on their dime (and at my rate, that's not cheap).
Some companies choose to be reasonable, and some pay me more. Either way, I'm happy.
I wrote Mr. Clapp yesterday, after the original mention of Pioneer. His reply indicates that Pioneer is still a going (and apparently growing) concern, though updating the web page to reflect any of that is pretty low on the priority list.
Of course, Pioneer seems to be dying of neglect as far as I can tell. Sad, really, since it was a great idea. Their original idea was even better.
Currently, they fit rockets using hydrogen and oxygen, and only carry enough LOX to get up to the tanker, plus a reserve for powering back to a landing. Originally, they planned to fit two air-breathing engines, plus the rockets. The air breathers would use kerosene fuel, which could also be used (with LOX taken on at altitude) with the rockets. This means that they could land more safely during an abort, because no LOX would be on board. The extra dead weight of the engines once the vehicle started running on rockets would be an issue, as would the lower energy density of the fuel. However, the extra safety and simplicity would be worth it in my opinion.
Once this kind of system gets going, it will be possible to build larger rockets which could actually get orbital (rather than using an expendable upper stage - the heat protection has to be there regardless since the vehicle leaves the atmosphere). Original versions would be great "4 hour delivery anywhere in the world" vehicles, as well as LEO launchers. Later versions would be a dirt cheap way to get payloads to orbit.
In my case, I saved up a month and a half's worth of outlay as cash, in case there was a between-contract lull. In retrospect, I should have saved up three months.
I then put my resume on monster.com and computerjobs.com, and waited for calls. They came in a flood. (I am not programming, but doing systems architecture and administration, but there are certainly a lot of programming contracts out there as well.)
When the right call came in, I negotiated a rate and time period, then quit my job. Sadly, I did not get the actual contract until after I had quit, so the terms were not great. I have since built and periodically update an addenda document. Among other things, it states that notwithstanding any other provisions of this contract, I own certain intellectual property (which I list). Generally, it assigns usage rights to the customer for whatever I bring to them, and ownership of anything that I develop on time billable to the customer. In addition, it has a set of terms related to startup, termination, pay periods and lag time, non-compete terms (I refuse to sign non-competes that would limit my ability to do business in any way) and the like.
I have to use this (as a basis for negotiation) in virtually every contract negotiation that I am in. Some customers will not sign up to your minimal terms, and in general it is not worth the job in that case, because they are likely to either try and rip you off, or to be too beureaucratic to stand working for. (Keep at least three months worth of outlays in cash in the bank.)
Some customers will attempt to impose additional terms on you after the contract has started. In my addenda, additional terms not agreed on before the first day of actual billing to the customer are not binding, and termination of the contract on the basis on violation of such additional terms not agreed on prior to initiation of billing requires full payment of the remaining amount due on the contract, with assumptions about number of hours worked and the like. A customer who will not sign this will usually nonetheless sign something reasonable about adding binding terms after the initiation of the contract.
Over time, you will get to know people doing the same work you are, and informal networks of referrels develop. Eventually, this will become your primary means of getting jobs if you are any good.
I have @Home through Charter (originally through Marcus Cable). I have always been happy with their uptime - it's better than the T1s at many of the jobs I've had. I have always been happy with their tech support. They usually know what they are doing, always listen to me, and have never put me down (unlike several ISPs I've had in the past).
The only time that I've had a problem with them at all was when they recently changed out my old LANCity modem for a DOCSIS modem. The problem was that they didn't give me much notice before they did it. The tech wasn't well-versed in how non-Windows systems worked. (I let him play with my Mac for 10 minutes looking for the Network Neighborhood, and wouldn't let him go near my Linux boxes). Even so, 15 minutes after the tech got the modem hooked up, I was back up online.
Anybody else get the feeling that our representatives in Congress, Republican and Democrat alike, are more concerned with their petty squabbles and party lines than actually getting anything interesting done with the country?
By Jove, I think he's got it. If you are standing between the two parties, they look very different. In truth, though, they both want to expand the power of government (at the expense of citizens, which is the only way to do it) and they both want to determine what you do. The only difference is whether they want to put their hands down the front or the back of your pants before they squeeze. Check out the Libertarian Party for an alternative.
Imagine what we could achieve if we stopped pursuing the unattainable goal of security and used our funds to develop better user interfaces, quicker file systems, or more advanced multimedia technology.
I suspect that what would happen is that our systems would quickly become unusable as script kiddies hacked them. Some would come just to run up their seti@home scores, some to put up hotline or gnutella servers, and some just to erase your hard drive because they could. All of this has been done or attempted on my servers and those of my friends. All of it would be much easier for the kiddies if we didn't keep finding and closing holes.
I came onto the internet in 1988 (yes, '88 - not 1998), and at the time pretty much everything was open source routing, and there weren't firewalls on non-military sites, and the level of system cracking that occurred was not only low, but was usually done by people in a benign fashion. (I used to break into the college UNIX system and then mail the exploit as root to the admins, until they hired me to stop other people from doing the same thing.) I wish that we could go back to open source routing, which would eliminate the Internet's current huge vulnerability to outages. We can't, because then people would underbuy bandwidth knowing that they could mooch off of other people. I wish that we could get rid of proxies and firewalls. We can't, because then our systems would start being toppled one by one.
The investment in security of systems is no greater than, and no less important then, the investment in security of workplaces in general. At least in large companies, that investment is large in both areas.
Oh please, indeed. I make very good money being an over-30, highly qualified, very experienced IT worker. I have been a manager and have payed very good money for highly qualified, experienced IT workers (regardless of age and including H1B holders).
The reason that competent IT workers get paid so well is that there are so very few of them. I have seen dangerous system admins (by that I mean that they will do more harm to the systems than good) get $70K from unwitting employers. There is a lack of supply coupled with a high demand, meaning that prices are high (good for me, not good for my clients).
There need to be more H1Bs granted because it will allow more IT work to be done better. Even though that's bad for me in the short term (lowers my rates) it's good for me in the long term (strengthens the industry).
Do you think I could get away with patenting something like the mouse trap, if I made up a lot of marketroid language and never mentioned mice or trapping them?
(If you build a better mousetrap, someone will sue you.:-)
(2) linking any Internet web site, either directly or through a series of links, to any other Internet web site containing DeCSS.
So if I link to a site which links to a site which unbeknownst to me links to a site which contains the DeCSS code, I am (according to the MPAA) in violation of their IP rights? Oh, please!
I would only add to this that the real catch with W2K is that it attempts to own the network at the protocol level. You will find yourself having a problem, and the only solution is to migrate the DNS servers to W2K. Another problem a little later, and the only problem is to go to DHCP, and you'll have to do that with the servers on W2K. When you are trying to deploy LDAP-based apps/authentication, you will find problems that you cannot solve without creating other problems. I'm sure that there are others that I have not run across yet, but these three are big problems. The one hope you have of not letting W2K control the network eventually is to isolate it into its own network with only tenuous (SMTP, FTP, etc) links between the W2K net and anything else.
It seems to me that the best method of getting copyright to work while still distributing material in electronic form would be to distribute it, essentially, twice. The first distribution would be the sample - a thumbnailed image, a few chapters of a book, a low-rate MP3 sample, time-limited software, etc. The second distribution would be the full book, the full-size image, the high-rate MP3 and so forth. The first would be available for free, and the second for a small fee - equivalent to the "real world value" (such as $1 per song, $0.10 and up per image, $5-$50 or so for a book). That way, you can check out a work before you purchase it, can compensate the copyright holder fairly, and can ensure that there is not a disincentive towards digitizing and indexing content.
As a consultant (system administration, not programming), I understand exactly what he means. I too am basically a prostitute. What I mean by that is that I will work for horrible companies (as long as they don't piss me off) in bad conditions (as long as they are safe) and on futile projects (as long as they are legal). I do not care what the conditions are as long as my price is met. I put in a lot of hours, and I am very careful to always give a good product to my customers. If they fold and I have to get another gig, as long as the reason that they folded has nothing to do with me, it doesn't bother me in the least. In other words, I'll do anything if you meet my price. It is a contractual relationship - I deliver what I promise and you pay what you promise - rather than a paternal relationship - you give us anything we ask and we'll take care of you.
I basically agree with you, but in reality you are also paying for their maintenance of an IP address space. Now, it would be nice if they would resell, say, a/28 or larger to their customers, and let their customers manage their own bandwidth on that. That said, there is really no requirement for @home to sell IP blocks, and they have chosen not to do so. They do not want to be just a pipe, even if every single customer of theirs whom I know (including myself) wants them to be just a pipe.
That was written with all the lucidity and purpose of a labrador retriever on acid. Actually, though, the lab would have written something more entertaining.
Anyone should be able to create a TLD, if they can set up a root-level name server for that TLD, prove that they can operate that root-level name server properly, prove that they can provide a 100% reliable connection to that root-level name server, and prove that they have a reasonable potential market for that TLD. (For example,.gnu is probably too narrow for a TLD, while.oss is probably sufficiently broad.)
Each TLD owner (and there should be exactly one owner per TLD) should be required to impose (or not) and enforce restrictions on the nature of owners of domain in their TLD. For example,.com addresses should not be given out to entities not legally registered as corporations, partnerships, proprietorships or the like. This would be more likely if there were one owner per TLD, and they were legally responsible for ensuring that domains they issue conform to the guidelines under which the TLD was created.
The existing TLDs should be destroyed as meaningless, and recreated under the above guidelines..net would still be useful if limited to organizations which exist to provide network connectivity (ISPs, telecom companies) or services (ASPs, registrars)..com probably needs to be broken into several domains, by either geography or the type of for-profit entity..edu needs to apply to more than just post-secondary institutions, and probably needs to be broken down geographically.
The number of domains owned by a given entity should be limited.
Each legal entity capable of issuing and enforcing trademarks should have a domain within an appropriate TLD for trademarks. For example,.tm.us for trademarks issued by the US Patent and Trademark Office. Then you could register etoys.tm.us, and there would be no possibility of confusion with etoy.org.fr. Then, refuse to allow anyone to register domains in the.tm.* domains except for the responsible trademark office.
These changes would, collectively, greatly increase the utility of the namespace in today's environment (as opposed to the pre-commercial environment in which the in-use namespace was conceived) and reduce confusion and lawsuits (as well as cybersquatting, if the limitations on the number of domains was done well). Of course, it will never happen, since it would require a big renaming. Maybe 10 years ago it could have been done, but a second Great Renaming now is probably not possible.
Having just finished a project where I was designing a method to integrate Win2K into an existing mixed-platform UNIX/NT environment, let me add a few things.
When attempting to integrate Active Directory with existing LDAP directories, MicroSoft's position is that Active Directory is LDAP. Technically true, since LDAP is an access protocol, but MS is monkeying with the system deliberately in order to prevent data synchronization unless you use not only Active Directory, but also MS's recently-acquired meta-directory (formerly Zoomit Via). Their directory can accomodate LDAP clients, but adds a lot of extensions and doesn't replicate well with systems that don't extend LDAP in ways not permitted by the standards.
When attempting to replace NIS (for scalability reasons), and attempting to get to a single authentication method for UNIX and Win2K, the only real answer is to use kerberos from the UNIX boxen with AD as the KDC, or to use Services for UNIX (an MS product), which will allow you to use AD as your NIS server. This of course won't work if you want to use someone else's KDC (since Win2K needs ACL information in the auth_data field of the kerberos cert) or if DCE is part of your product mix. MicroSoft's position is that it will work. Again, it will do so until you consider the real world.
I could go on, but I think that the point is made. MS has made it so that if you implement Win2K, you will also turn control of DNS, DHCP, LDAP, Kerberos, NIS and a number of other products over to MS, because Win2K will not work and play well with others. What scares me is what happens when Windows 2004 comes out and redefines name services, address assignment and the like. Do MS's enterprise customers then have to roll over and take it, because the cost of pulling out Windows will be higher than the cost of surrender?
You're assuming all that software that is missing on BeOS would be ported from MacOS-PPC. The big one, MS Office, would almost certainly *not* be ported, as an x86 MacOS would be directly competing with Windows.
Part of the beauty of MacOS X is that you can insert compatibility layers into it. For example, you can run MacOS 9 in a window, for software not written entirely to Carbon, and can run Carbonized software and take advantage of the memory management and multitasking in MacOS X.
I do not see running existing MacOS apps on Intel (Carbon or otherwise), but I could easily see software written for MacOS X being compiled fat to run on both PowerPC and Intel. Plug in a compatibility layer for Windows (maybe using the SoftWindows code as a start point) and you would be able to run Windows s/w much like the current Mac software runs on MacOS X for PowerPC.
This would certainly be a direct threat to Windows, and as a result I suspect that Apple's decision to do this or not would largely depend on the outcome of the anti-trust trial.
SCO sucks, and always has. They were in the right place at the right time, and that carried them for a while. The problem is that now they have to compete against other options, and most UNIX admins will resign rather than work on SCO (unless they are in a small market). We have SCO on our ISPs, and it is treated like an application (if the system runs out of swap, that's treated like a bug, and a new release is made so that that won't happen again). We use SCO primarily due to inertia, and we treat it this way because otherwise we'd have to work on it rather than letting the developers do it.
The problem with your analogy is that computers are used, at least outside of the office, by choice. Inside the office, they are still used by choice, just not necessarily by your choice. Compliance with the law, on the other hand, is compelled by the use of armed force.
ALSO, it has to have a REAL preferences menu, where you can disable ANY HTML type, (blink comes to mind), and ANY feature in the browser, and can be completly controled from the keyboard if need be.
This is my number 1 feature wish for a web browser! I would really like to be able to turn off the ability of web pages to control the Netscape window sizing/position, the BLINK tag (and a few others) and so forth. Turning off JavaScript would then be equivalent to turning off a tag; same with Java itself.
Slashdot Mirror
We've been in this situation before - and indeed we still are in some ways. How do you, for example, communicate with a ship at sea whose only access is via satellites that they may or may not be able to reach, depending on the weather? The answer is typically uucp. Admittedly, this doesn't allow for streaming media or web surfing, but email and file transfer work. I suspect that the most likely configuration would be a uucp gateway on each end, with some automation being used to keep servers in sync. So if I am on Mars, and create a VR tour with my handheld camera, I can have it sent to a server on Earth for those on Earth who want to see it. Trying to serve that to Earth from Mars (or vice versa) would be a pain in the neck.
If I get there first, I'm calling it Ares. Mostly just to annoy all of the people who would have to rewrite the textbooks.
Has happened to me more than once. In any case where I have to sign over rights to previous work if used on a project, I inform the company that I will sign such an agreement, but that I will bring no previous code into the project, and will simply reimplement everything I need on their dime (and at my rate, that's not cheap).
Some companies choose to be reasonable, and some pay me more. Either way, I'm happy.
-jeff
I wrote Mr. Clapp yesterday, after the original mention of Pioneer. His reply indicates that Pioneer is still a going (and apparently growing) concern, though updating the web page to reflect any of that is pretty low on the priority list.
-jeff
Of course, Pioneer seems to be dying of neglect as far as I can tell. Sad, really, since it was a great idea. Their original idea was even better.
Currently, they fit rockets using hydrogen and oxygen, and only carry enough LOX to get up to the tanker, plus a reserve for powering back to a landing. Originally, they planned to fit two air-breathing engines, plus the rockets. The air breathers would use kerosene fuel, which could also be used (with LOX taken on at altitude) with the rockets. This means that they could land more safely during an abort, because no LOX would be on board. The extra dead weight of the engines once the vehicle started running on rockets would be an issue, as would the lower energy density of the fuel. However, the extra safety and simplicity would be worth it in my opinion.
Once this kind of system gets going, it will be possible to build larger rockets which could actually get orbital (rather than using an expendable upper stage - the heat protection has to be there regardless since the vehicle leaves the atmosphere). Original versions would be great "4 hour delivery anywhere in the world" vehicles, as well as LEO launchers. Later versions would be a dirt cheap way to get payloads to orbit.
-jeff
In my case, I saved up a month and a half's worth of outlay as cash, in case there was a between-contract lull. In retrospect, I should have saved up three months.
I then put my resume on monster.com and computerjobs.com, and waited for calls. They came in a flood. (I am not programming, but doing systems architecture and administration, but there are certainly a lot of programming contracts out there as well.)
When the right call came in, I negotiated a rate and time period, then quit my job. Sadly, I did not get the actual contract until after I had quit, so the terms were not great. I have since built and periodically update an addenda document. Among other things, it states that notwithstanding any other provisions of this contract, I own certain intellectual property (which I list). Generally, it assigns usage rights to the customer for whatever I bring to them, and ownership of anything that I develop on time billable to the customer. In addition, it has a set of terms related to startup, termination, pay periods and lag time, non-compete terms (I refuse to sign non-competes that would limit my ability to do business in any way) and the like.
I have to use this (as a basis for negotiation) in virtually every contract negotiation that I am in. Some customers will not sign up to your minimal terms, and in general it is not worth the job in that case, because they are likely to either try and rip you off, or to be too beureaucratic to stand working for. (Keep at least three months worth of outlays in cash in the bank.)
Some customers will attempt to impose additional terms on you after the contract has started. In my addenda, additional terms not agreed on before the first day of actual billing to the customer are not binding, and termination of the contract on the basis on violation of such additional terms not agreed on prior to initiation of billing requires full payment of the remaining amount due on the contract, with assumptions about number of hours worked and the like. A customer who will not sign this will usually nonetheless sign something reasonable about adding binding terms after the initiation of the contract.
Over time, you will get to know people doing the same work you are, and informal networks of referrels develop. Eventually, this will become your primary means of getting jobs if you are any good.
-jeff
I have @Home through Charter (originally through Marcus Cable). I have always been happy with their uptime - it's better than the T1s at many of the jobs I've had. I have always been happy with their tech support. They usually know what they are doing, always listen to me, and have never put me down (unlike several ISPs I've had in the past).
The only time that I've had a problem with them at all was when they recently changed out my old LANCity modem for a DOCSIS modem. The problem was that they didn't give me much notice before they did it. The tech wasn't well-versed in how non-Windows systems worked. (I let him play with my Mac for 10 minutes looking for the Network Neighborhood, and wouldn't let him go near my Linux boxes). Even so, 15 minutes after the tech got the modem hooked up, I was back up online.
-jeff
By Jove, I think he's got it. If you are standing between the two parties, they look very different. In truth, though, they both want to expand the power of government (at the expense of citizens, which is the only way to do it) and they both want to determine what you do. The only difference is whether they want to put their hands down the front or the back of your pants before they squeeze. Check out the Libertarian Party for an alternative.
-jeff
I came onto the internet in 1988 (yes, '88 - not 1998), and at the time pretty much everything was open source routing, and there weren't firewalls on non-military sites, and the level of system cracking that occurred was not only low, but was usually done by people in a benign fashion. (I used to break into the college UNIX system and then mail the exploit as root to the admins, until they hired me to stop other people from doing the same thing.) I wish that we could go back to open source routing, which would eliminate the Internet's current huge vulnerability to outages. We can't, because then people would underbuy bandwidth knowing that they could mooch off of other people. I wish that we could get rid of proxies and firewalls. We can't, because then our systems would start being toppled one by one.
The investment in security of systems is no greater than, and no less important then, the investment in security of workplaces in general. At least in large companies, that investment is large in both areas.
Oh please, indeed. I make very good money being an over-30, highly qualified, very experienced IT worker. I have been a manager and have payed very good money for highly qualified, experienced IT workers (regardless of age and including H1B holders).
The reason that competent IT workers get paid so well is that there are so very few of them. I have seen dangerous system admins (by that I mean that they will do more harm to the systems than good) get $70K from unwitting employers. There is a lack of supply coupled with a high demand, meaning that prices are high (good for me, not good for my clients).
There need to be more H1Bs granted because it will allow more IT work to be done better. Even though that's bad for me in the short term (lowers my rates) it's good for me in the long term (strengthens the industry).
Do you think I could get away with patenting something like the mouse trap, if I made up a lot of marketroid language and never mentioned mice or trapping them?
:-)
(If you build a better mousetrap, someone will sue you.
-jeff
So if I link to a site which links to a site which unbeknownst to me links to a site which contains the DeCSS code, I am (according to the MPAA) in violation of their IP rights? Oh, please!
I would only add to this that the real catch with W2K is that it attempts to own the network at the protocol level. You will find yourself having a problem, and the only solution is to migrate the DNS servers to W2K. Another problem a little later, and the only problem is to go to DHCP, and you'll have to do that with the servers on W2K. When you are trying to deploy LDAP-based apps/authentication, you will find problems that you cannot solve without creating other problems. I'm sure that there are others that I have not run across yet, but these three are big problems. The one hope you have of not letting W2K control the network eventually is to isolate it into its own network with only tenuous (SMTP, FTP, etc) links between the W2K net and anything else.
-jeff
It seems to me that the best method of getting copyright to work while still distributing material in electronic form would be to distribute it, essentially, twice. The first distribution would be the sample - a thumbnailed image, a few chapters of a book, a low-rate MP3 sample, time-limited software, etc. The second distribution would be the full book, the full-size image, the high-rate MP3 and so forth. The first would be available for free, and the second for a small fee - equivalent to the "real world value" (such as $1 per song, $0.10 and up per image, $5-$50 or so for a book). That way, you can check out a work before you purchase it, can compensate the copyright holder fairly, and can ensure that there is not a disincentive towards digitizing and indexing content.
-jeff
As a consultant (system administration, not programming), I understand exactly what he means. I too am basically a prostitute. What I mean by that is that I will work for horrible companies (as long as they don't piss me off) in bad conditions (as long as they are safe) and on futile projects (as long as they are legal). I do not care what the conditions are as long as my price is met. I put in a lot of hours, and I am very careful to always give a good product to my customers. If they fold and I have to get another gig, as long as the reason that they folded has nothing to do with me, it doesn't bother me in the least. In other words, I'll do anything if you meet my price. It is a contractual relationship - I deliver what I promise and you pay what you promise - rather than a paternal relationship - you give us anything we ask and we'll take care of you.
-jeff
I basically agree with you, but in reality you are also paying for their maintenance of an IP address space. Now, it would be nice if they would resell, say, a /28 or larger to their customers, and let their customers manage their own bandwidth on that. That said, there is really no requirement for @home to sell IP blocks, and they have chosen not to do so. They do not want to be just a pipe, even if every single customer of theirs whom I know (including myself) wants them to be just a pipe.
That was written with all the lucidity and purpose of a labrador retriever on acid. Actually, though, the lab would have written something more entertaining.
-jeff
Anyone should be able to create a TLD, if they can set up a root-level name server for that TLD, prove that they can operate that root-level name server properly, prove that they can provide a 100% reliable connection to that root-level name server, and prove that they have a reasonable potential market for that TLD. (For example, .gnu is probably too narrow for a TLD, while .oss is probably sufficiently broad.)
Each TLD owner (and there should be exactly one owner per TLD) should be required to impose (or not) and enforce restrictions on the nature of owners of domain in their TLD. For example, .com addresses should not be given out to entities not legally registered as corporations, partnerships, proprietorships or the like. This would be more likely if there were one owner per TLD, and they were legally responsible for ensuring that domains they issue conform to the guidelines under which the TLD was created.
The existing TLDs should be destroyed as meaningless, and recreated under the above guidelines. .net would still be useful if limited to organizations which exist to provide network connectivity (ISPs, telecom companies) or services (ASPs, registrars). .com probably needs to be broken into several domains, by either geography or the type of for-profit entity. .edu needs to apply to more than just post-secondary institutions, and probably needs to be broken down geographically.
The number of domains owned by a given entity should be limited.
Each legal entity capable of issuing and enforcing trademarks should have a domain within an appropriate TLD for trademarks. For example, .tm.us for trademarks issued by the US Patent and Trademark Office. Then you could register etoys.tm.us, and there would be no possibility of confusion with etoy.org.fr. Then, refuse to allow anyone to register domains in the .tm.* domains except for the responsible trademark office.
These changes would, collectively, greatly increase the utility of the namespace in today's environment (as opposed to the pre-commercial environment in which the in-use namespace was conceived) and reduce confusion and lawsuits (as well as cybersquatting, if the limitations on the number of domains was done well). Of course, it will never happen, since it would require a big renaming. Maybe 10 years ago it could have been done, but a second Great Renaming now is probably not possible.
Having just finished a project where I was designing a method to integrate Win2K into an existing mixed-platform UNIX/NT environment, let me add a few things.
When attempting to integrate Active Directory with existing LDAP directories, MicroSoft's position is that Active Directory is LDAP. Technically true, since LDAP is an access protocol, but MS is monkeying with the system deliberately in order to prevent data synchronization unless you use not only Active Directory, but also MS's recently-acquired meta-directory (formerly Zoomit Via). Their directory can accomodate LDAP clients, but adds a lot of extensions and doesn't replicate well with systems that don't extend LDAP in ways not permitted by the standards.
When attempting to replace NIS (for scalability reasons), and attempting to get to a single authentication method for UNIX and Win2K, the only real answer is to use kerberos from the UNIX boxen with AD as the KDC, or to use Services for UNIX (an MS product), which will allow you to use AD as your NIS server. This of course won't work if you want to use someone else's KDC (since Win2K needs ACL information in the auth_data field of the kerberos cert) or if DCE is part of your product mix. MicroSoft's position is that it will work. Again, it will do so until you consider the real world.
I could go on, but I think that the point is made. MS has made it so that if you implement Win2K, you will also turn control of DNS, DHCP, LDAP, Kerberos, NIS and a number of other products over to MS, because Win2K will not work and play well with others. What scares me is what happens when Windows 2004 comes out and redefines name services, address assignment and the like. Do MS's enterprise customers then have to roll over and take it, because the cost of pulling out Windows will be higher than the cost of surrender?
You're assuming all that software that is missing on BeOS would be ported from MacOS-PPC. The big one, MS Office, would almost certainly *not* be ported, as an x86 MacOS would be directly competing with Windows.
Part of the beauty of MacOS X is that you can insert compatibility layers into it. For example, you can run MacOS 9 in a window, for software not written entirely to Carbon, and can run Carbonized software and take advantage of the memory management and multitasking in MacOS X.
I do not see running existing MacOS apps on Intel (Carbon or otherwise), but I could easily see software written for MacOS X being compiled fat to run on both PowerPC and Intel. Plug in a compatibility layer for Windows (maybe using the SoftWindows code as a start point) and you would be able to run Windows s/w much like the current Mac software runs on MacOS X for PowerPC.
This would certainly be a direct threat to Windows, and as a result I suspect that Apple's decision to do this or not would largely depend on the outcome of the anti-trust trial.
SCO sucks, and always has. They were in the right place at the right time, and that carried them for a while. The problem is that now they have to compete against other options, and most UNIX admins will resign rather than work on SCO (unless they are in a small market). We have SCO on our ISPs, and it is treated like an application (if the system runs out of swap, that's treated like a bug, and a new release is made so that that won't happen again). We use SCO primarily due to inertia, and we treat it this way because otherwise we'd have to work on it rather than letting the developers do it.
This will be more addictive than crack.
The problem with your analogy is that computers are used, at least outside of the office, by choice. Inside the office, they are still used by choice, just not necessarily by your choice. Compliance with the law, on the other hand, is compelled by the use of armed force.
This is my number 1 feature wish for a web browser! I would really like to be able to turn off the ability of web pages to control the Netscape window sizing/position, the BLINK tag (and a few others) and so forth. Turning off JavaScript would then be equivalent to turning off a tag; same with Java itself.
You have a probably logic error in line 3. :-)