The Early Days of Networking and Domain Names

In the early days of networking and the internet, things were done differently in the UK compared to other parts of the world. The thing that was peculiar about the UK was the way they organized domain names. Instead of using the traditional reverse order like most other countries, the UK would use a different format, such as uk.ac.nottingham.computer-science. This meant that if you were contacting someone in Nottingham, you'd have to know their full address and then flip it around to get it right.

This created a lot of complexity in mail routing tables, which were already complicated enough. Users would often pick up business cards or receive addresses in email that weren't actually formatted correctly, so the routing tables had to be sorted out manually. This caused problems when countries like Czechoslovakia joined and got their own top-level domain (cs). Suddenly, you could pass addresses either way - they were considered part of czechoslovakia or part of the UK, which added even more complexity to the mail system.

To deal with this, users would start chopping off parts of the address, such as cs or uk, and then try different combinations until it worked. This was a time-consuming process that required patience and persistence. The problem didn't end there - the way addresses were formatted also caused issues with what was called the "big endian versus little endian" issue. This referred to how integers (typically made up of four bytes) were sent over the internet, with either the largest byte first or the smallest byte first.

The Endian Problem

The endian problem was a real challenge in those days. It wasn't clear who would win out and what format would be used, so it was decided that it would be easier to use a hierarchical system. This meant that each country would sort out its own domain names, which made sense because there were often organizations within countries that could divide up the work.

However, this also created an international problem when it came to addressing governments and other international entities. It wasn't clear who should grant names below certain levels, such as edu or gov. In the end, it was decided that it would be easier to sort out domain names by country because there were usually organizations within those countries that could divide up the work.

The International Domain Name System

In 1985, the internet community came together to create an international system for managing domain names. This system used a hierarchical structure, with each level having its own set of rules and procedures. The system was designed to be flexible and adaptable, but it also created new challenges and complexities.

Today, the internet uses a system called DNS (Domain Name System), which is responsible for translating human-readable domain names into IP addresses that computers can understand. The DNS system is hierarchical, with each level having its own set of rules and procedures. However, the early days of networking and domain names were a time of great experimentation and innovation, and it's interesting to look back on those times and see how they shaped the internet as we know it today.

The Legacy of Early Domain Name Systems

In retrospect, the early days of domain name systems were a trial by fire. The UK's unique approach to naming domains caused problems that took years to resolve, but it also taught valuable lessons about the importance of clear communication and cooperation in the development of new technologies. Today, the internet uses a more standardized system, but the legacy of those early days can still be seen in the way we use domain names and IP addresses.

The use of hierarchical systems for managing domain names has also had an impact on other areas of computer science, such as file systems and network protocols. The idea of breaking down complex problems into smaller, more manageable parts is a fundamental principle of computer science, and it's still used today in many different contexts. In the end, the early days of networking and domain names may seem like a relic of the past, but they played an important role in shaping the internet as we know it today.

The Rise of International Cooperation

As the internet grew and developed, the need for international cooperation became increasingly clear. The early days of domain name systems were marked by a lack of coordination between different countries and organizations, which led to confusion and frustration. However, over time, this has changed, and there is now a much greater sense of cooperation and collaboration.

The creation of the DNS system was just one example of how international cooperation can lead to better outcomes for everyone involved. The use of standardized systems and protocols has also made it easier for people around the world to communicate and access information. Today, the internet is a global community, and it's exciting to think about what the future may hold.

The UK's Unique Approach

Despite its quirks and challenges, the UK's approach to domain names was not without its advantages. The use of a hierarchical system allowed for greater flexibility and adaptability, which made it easier to manage complex problems. This approach has been adopted by many other countries, including France and Greece, which have also developed their own top-level domains.

The legacy of the UK's unique approach can still be seen in the way we use domain names today. The use of hierarchical systems for managing domain names is a fundamental principle of computer science, and it continues to shape the development of new technologies. In the end, the early days of networking and domain names may seem like a relic of the past, but they played an important role in shaping the internet as we know it today.

The Future of Domain Names

As the internet continues to evolve and grow, there will undoubtedly be new challenges and opportunities for cooperation and innovation. The use of domain names is just one example of how technology can shape our daily lives, but it's also a reminder of the importance of clear communication and collaboration.

In the future, we can expect to see even more sophisticated systems for managing domain names and IP addresses. Advances in artificial intelligence and machine learning are likely to play a major role in this, as they make it possible to automate many tasks that were previously done by humans. However, there will also be new challenges and opportunities for cooperation and innovation.

The Internet Today

Today, the internet is a global community, and it's exciting to think about what the future may hold. The use of domain names is just one example of how technology can shape our daily lives, but it's also a reminder of the importance of clear communication and collaboration.

As we look to the future, there are many new opportunities for innovation and growth. The development of artificial intelligence and machine learning will continue to play a major role in shaping the internet, but it's also likely that other technologies, such as blockchain and quantum computing, will become more prominent.

"WEBVTTKind: captionsLanguage: enthere weren't that many people sending email so it was kind of which which way is it going to go um you know we'll go with the obviously clearly better way of starting with a big name and they seem to have gone this way each side said they'll see the light in due time and i sat in several meetings where they discussed this and said oh you're all wrong eventually just the weight of history settled it and we've all gone for this scheme the uk and the us disagreed in multiple ways so the internet initially went with so originally they just had dot harper so you'd be at stanford arbor joe at stanford harbor or something like that again that wasn't going to scale we couldn't all belong to the harper they split it up into several top level domains dot edu was for educational sites mostly universities dot mil for military because obviously this came out of the arpanet and there was military people dot net was generally for network providers dot com for commercial gov for the government hint for international organizations they were thinking things like the un and anything else dot org for any other organization that list has now grown they've added all sorts of other things to it but this is the way it kind of worked so you were at uh say berkeley.edu amazon.com you were split up in those sort of top-level domains in the uk they decided and actually across most of europe they decided they were going to go more hierarchical so there would be a uk domain say in france dot fr and then below the uk they had co for commercials and ac for academic community and then you you divide it up between that and for the longest time there was only uk ac basically in the uk there was there was like one commercial domain just to sort of prove it worked for the longest while however the bad thing for us is that um not only was this this sort of a dodge we could sort of live with that because you know knew to send all the uk stuff to the uk and any of these other things you'd send off mostly to the arpanet or the internet as it was beginning to come but the thing we did weirdly in the uk was to flip everything around so this would be uk dot ac dot not dot cs brother computer science at nottingham and if you're in the internet it would be the other way around cs dot not ac dot uk and they both had good reasons for doing this both of them said it was much more natural way to write the addresses you would always start with the most significant thing work your way down no you'd always have the most significant thing on the right hand side and work it down that way this caused us no end of issues in our mail routing tables because typically users would pick up business cards or they'd be sent addresses in email addresses not actually in the headers but oh you can contact me at in berkeley.edu so we would type in you know eric berkley.edu or whatever and we'd have to sort this mess out within the routing tables and mostly it was okay if it sort of ended with edu that was okay but there were a few palindromic addresses and we were really hit when checklist czechoslovakia i can hardly say that joined and they have the top level domain of cs so now you could actually pass this either way this could be somewhere in czechoslovakia or it could be somewhere in the uk so this added even more complexity to the mail tables which were already pretty complex so you take an address and you try and pass it both ways you'd say does it work this way if not you'd flip it around and say does it work that way if you got a hit that was all well and good you knew where to send it but typically you didn't have a complete list of every place in the known world so you'd start chopping bits off you'd chop off the cs and say do i know about nottingham in ac uk and if you didn't know about that you'd chop off the next bit and say well what about uk ac or just uk so you work your way back up but you have to do it each way so in actual fact you chop off the cs going this way and then you drop off the uk going the other way and see if that made sense and then you you'd start chopping off in the opposite directions and you keep working up until you found the longest possible match and i don't think we ever really ended up with completely palindropic addresses but that was more by luck than judgment really you know it was just crazy times you get these addresses you have to flip them around and we'd give waitings to some of them and sometimes you'd fill out your table more than you needed just to say no this place although it looks ambiguous no it really does need to go off to the internet and not to not to this way or that way i think this sort of formally died out somewhere in the 1990s uh but but it lasted for no end of time and most nearly everywhere else kind of ended up with the us version um and even if they had their own names like france or greece or or whatever their own top level domain they tended to stick with this one so it was specifically uk problem of trying to get all that sorted out really a problem we could have done without it's sort of like an endian problem isn't it oh absolutely it was called big endian versus little endian it's still the same on the internet do you if you're sending like an integer which is typically often composed of four bites do you send the biggest one first or the smallest one first uh luckily i think those are the two only options now but this this one was sort of a a problem around making i don't know why we didn't just well it wasn't really clear at the time who was going to win out and they came up with this in the internet and said this is going to win this is going to be this is the obvious way to do things no no it's much easier to do it hierarchically and in some there is some sympathy with that you know you you can sort of divulge this the uk and say right the uk you sort out what goes on below uk uh because this one's a bit more awkward if you're going to sort out who grants names below edu it's sort of an international thing same with govs and things like that so it's much easier to do it by country because you've usually got some body there that can divide it up and there should be 32 bits there now each of these bits has a number associated with it so this would be considered bit zero and this would be considered bit 31 and then we can count down so this is then bit 24 that's bit 23 16 and 15 and then thatthere weren't that many people sending email so it was kind of which which way is it going to go um you know we'll go with the obviously clearly better way of starting with a big name and they seem to have gone this way each side said they'll see the light in due time and i sat in several meetings where they discussed this and said oh you're all wrong eventually just the weight of history settled it and we've all gone for this scheme the uk and the us disagreed in multiple ways so the internet initially went with so originally they just had dot harper so you'd be at stanford arbor joe at stanford harbor or something like that again that wasn't going to scale we couldn't all belong to the harper they split it up into several top level domains dot edu was for educational sites mostly universities dot mil for military because obviously this came out of the arpanet and there was military people dot net was generally for network providers dot com for commercial gov for the government hint for international organizations they were thinking things like the un and anything else dot org for any other organization that list has now grown they've added all sorts of other things to it but this is the way it kind of worked so you were at uh say berkeley.edu amazon.com you were split up in those sort of top-level domains in the uk they decided and actually across most of europe they decided they were going to go more hierarchical so there would be a uk domain say in france dot fr and then below the uk they had co for commercials and ac for academic community and then you you divide it up between that and for the longest time there was only uk ac basically in the uk there was there was like one commercial domain just to sort of prove it worked for the longest while however the bad thing for us is that um not only was this this sort of a dodge we could sort of live with that because you know knew to send all the uk stuff to the uk and any of these other things you'd send off mostly to the arpanet or the internet as it was beginning to come but the thing we did weirdly in the uk was to flip everything around so this would be uk dot ac dot not dot cs brother computer science at nottingham and if you're in the internet it would be the other way around cs dot not ac dot uk and they both had good reasons for doing this both of them said it was much more natural way to write the addresses you would always start with the most significant thing work your way down no you'd always have the most significant thing on the right hand side and work it down that way this caused us no end of issues in our mail routing tables because typically users would pick up business cards or they'd be sent addresses in email addresses not actually in the headers but oh you can contact me at in berkeley.edu so we would type in you know eric berkley.edu or whatever and we'd have to sort this mess out within the routing tables and mostly it was okay if it sort of ended with edu that was okay but there were a few palindromic addresses and we were really hit when checklist czechoslovakia i can hardly say that joined and they have the top level domain of cs so now you could actually pass this either way this could be somewhere in czechoslovakia or it could be somewhere in the uk so this added even more complexity to the mail tables which were already pretty complex so you take an address and you try and pass it both ways you'd say does it work this way if not you'd flip it around and say does it work that way if you got a hit that was all well and good you knew where to send it but typically you didn't have a complete list of every place in the known world so you'd start chopping bits off you'd chop off the cs and say do i know about nottingham in ac uk and if you didn't know about that you'd chop off the next bit and say well what about uk ac or just uk so you work your way back up but you have to do it each way so in actual fact you chop off the cs going this way and then you drop off the uk going the other way and see if that made sense and then you you'd start chopping off in the opposite directions and you keep working up until you found the longest possible match and i don't think we ever really ended up with completely palindropic addresses but that was more by luck than judgment really you know it was just crazy times you get these addresses you have to flip them around and we'd give waitings to some of them and sometimes you'd fill out your table more than you needed just to say no this place although it looks ambiguous no it really does need to go off to the internet and not to not to this way or that way i think this sort of formally died out somewhere in the 1990s uh but but it lasted for no end of time and most nearly everywhere else kind of ended up with the us version um and even if they had their own names like france or greece or or whatever their own top level domain they tended to stick with this one so it was specifically uk problem of trying to get all that sorted out really a problem we could have done without it's sort of like an endian problem isn't it oh absolutely it was called big endian versus little endian it's still the same on the internet do you if you're sending like an integer which is typically often composed of four bites do you send the biggest one first or the smallest one first uh luckily i think those are the two only options now but this this one was sort of a a problem around making i don't know why we didn't just well it wasn't really clear at the time who was going to win out and they came up with this in the internet and said this is going to win this is going to be this is the obvious way to do things no no it's much easier to do it hierarchically and in some there is some sympathy with that you know you you can sort of divulge this the uk and say right the uk you sort out what goes on below uk uh because this one's a bit more awkward if you're going to sort out who grants names below edu it's sort of an international thing same with govs and things like that so it's much easier to do it by country because you've usually got some body there that can divide it up and there should be 32 bits there now each of these bits has a number associated with it so this would be considered bit zero and this would be considered bit 31 and then we can count down so this is then bit 24 that's bit 23 16 and 15 and then that\n"