I happened across an old (circa 1990) introduction to the internet, that described it as a "network of networks", and was amused, because at the time, this was an oft-heard description, and I probably haven't heard or seen it in print in the last three to five years. As amusingly antiquated as it sounds, it is exactly right. So I'll repeat it.
The internet is a network of networks. And those networks are heterogeneous, whether they are LANs running TCP/IP over Ethernet, or IPX over token ring and so on. And all these networks have different conventions for how individual hosts are recognized. The architects of the internet made it so that, rather transparently, any two hosts could exchange data, regardless of their network environs.
This required standards, and adherence to these standards by a large number of programmers, network engineers, and so on. A host connected to the internet must provide a set of standard services, and conform to standards regarding naming and addressing. Services defined by a standard must operate according to that standard for the whole cooperative house to stand. If I ping a host, I should not expect a ten-million packet response. If I ftp to a host, I should not be required to issue "MYNAME blah" instead of "USER blah" for authentication. At the most basic level, if I attempt to access a host, I should be sure that, assuming my software was written according to agreed-upon standards, that I can actually connect to it.
The internet became one of the largest, best and earliest examples of the kind of cooperation that has made linux (and other open source projects) happen and work. If these standards were built by the same people who are attempting to break these rules, such as Mr. e-, the internet still wouldn't exist, because every decision would require a judge, at least seven lawyers, and twelve appeals.
That the domain naming system is too weak to handle the crush of online business is manifestly clear. There are plenty of possible workable solutions. It seems a very poor idea, however, to break something that works in the name of a handful of people making money.
Actually, only x.com is illegal, according to the relevant RFCs. RFC 1123 states:
The syntax of a legal Internet host name was specified in RFC-952 [DNS:4]. One aspect of host name syntax is hereby changed: the restriction on the first character is relaxed to allow either a letter or a digit. Host software MUST support this more liberal syntax.
1. A "name" (Net, Host, Gateway, or Domain name) is a text string up to 24 characters drawn from the alphabet (A-Z), digits (0-9), minus sign (-), and period (.). Note that periods are only allowed when they serve to delimit components of "domain style names". (See RFC-921, "Domain Name System Implementation Schedule", for background). No blank or space characters are permitted as part of a name. No distinction is made between upper and lower case. The first character must be an alpha character. The last character must not be a minus sign or period. Single character names or nicknames are not allowed.
So any length between two and twenty-four characters is okay, and hp.com is fine. How x.com was registered is beyond me, and serves as an illustration of how network solutions never really was very good at its job.
The problem with CSS is that none of the major browsers correctly implement even CSS1, which has been a standard since 1997.
There are two general classes of problem:
1. Simple non-compliance. Take a look at the w3c style sheet "acid test" in whatever browser you are using. Unless it is M12, it will probably be wrong. I've tested this on:
IE5
Netscape 4.7
Opera 3.61
Amaya 2.4 (!)
Mozilla M12
all on NT.
2. Worse, at least NS 4.* has a implementation so buggy that styles often crash the browser.
One may say that browsers will be forced to follow, but try doing that now in a real-world commercial situation. "Sorry, dear client, but since we are big on style sheets, we are going to build your site such that only users of a pre-alpha browser will be able to properly view it. Oh, and yeah, about 30% of your visitors browsers are going to crash while visiting your site." "Why? oh, because we think that your web site should perform a public service, even if it means you lose business and visitors. But just you wait until the browsers catch up!"
CSS works a lot better in intranet setups. That's what I have our corporate intranet using, because we have influence over the browsers used on the network and understand that things are going to look somewhat different in IE/NS/Opera/M12. Plus, we understand (unlike clients, god bless them) that CSS will, eventually, work, and that we will benefit from having plenty of internal experience. But that won't be for a while.
Slashdot Mirror
I happened across an old (circa 1990) introduction to the internet, that described it as a "network of networks", and was amused, because at the time, this was an oft-heard description, and I probably haven't heard or seen it in print in the last three to five years. As amusingly antiquated as it sounds, it is exactly right. So I'll repeat it.
The internet is a network of networks. And those networks are heterogeneous, whether they are LANs running TCP/IP over Ethernet, or IPX over token ring and so on. And all these networks have different conventions for how individual hosts are recognized. The architects of the internet made it so that, rather transparently, any two hosts could exchange data, regardless of their network environs.
This required standards, and adherence to these standards by a large number of programmers, network engineers, and so on. A host connected to the internet must provide a set of standard services, and conform to standards regarding naming and addressing. Services defined by a standard must operate according to that standard for the whole cooperative house to stand. If I ping a host, I should not expect a ten-million packet response. If I ftp to a host, I should not be required to issue "MYNAME blah" instead of "USER blah" for authentication. At the most basic level, if I attempt to access a host, I should be sure that, assuming my software was written according to agreed-upon standards, that I can actually connect to it.
The internet became one of the largest, best and earliest examples of the kind of cooperation that has made linux (and other open source projects) happen and work. If these standards were built by the same people who are attempting to break these rules, such as Mr. e-, the internet still wouldn't exist, because every decision would require a judge, at least seven lawyers, and twelve appeals.
That the domain naming system is too weak to handle the crush of online business is manifestly clear. There are plenty of possible workable solutions. It seems a very poor idea, however, to break something that works in the name of a handful of people making money.
Actually, only x.com is illegal, according to the relevant RFCs. RFC 1123 states:
So, 2600.org is perfectly fine.
RFC 952 states:
So any length between two and twenty-four characters is okay, and hp.com is fine. How x.com was registered is beyond me, and serves as an illustration of how network solutions never really was very good at its job.
The problem with CSS is that none of the major browsers correctly implement even CSS1, which has been a standard since 1997.
There are two general classes of problem:
1. Simple non-compliance. Take a look at the w3c style sheet "acid test" in whatever browser you are using. Unless it is M12, it will probably be wrong. I've tested this on:
- IE5
- Netscape 4.7
- Opera 3.61
- Amaya 2.4 (!)
- Mozilla M12
all on NT.2. Worse, at least NS 4.* has a implementation so buggy that styles often crash the browser.
One may say that browsers will be forced to follow, but try doing that now in a real-world commercial situation. "Sorry, dear client, but since we are big on style sheets, we are going to build your site such that only users of a pre-alpha browser will be able to properly view it. Oh, and yeah, about 30% of your visitors browsers are going to crash while visiting your site."
"Why? oh, because we think that your web site should perform a public service, even if it means you lose business and visitors. But just you wait until the browsers catch up!"
CSS works a lot better in intranet setups. That's what I have our corporate intranet using, because we have influence over the browsers used on the network and understand that things are going to look somewhat different in IE/NS/Opera/M12. Plus, we understand (unlike clients, god bless them) that CSS will, eventually, work, and that we will benefit from having plenty of internal experience. But that won't be for a while.