3 ms·
Why TCP/IP is on the way out
- timrosenblatt 12y agoHow would a switchover actually be implemented? Wouldn't this require being built into every consumer OS device? I'm not a network engineer, so if you are, please fill me in. I don't get the sense that this could be just swapped out at the data center level.
- leepowers 12y ago> What is a little difficult to grasp is what patented, arithmetical equations are being used and why they work. Which is why TCP/IP is going to be around for a long time yet, and very unlikely to be replace by a closed, proprietary alternative. The TCP/IP spec is open to all - if you want to build a network or device that talks TCP/IP there's the specification, have at it. The open nature of internet and web protocols helped make the modern internet possible. We should not be so quick to shut the door on such a rich and productive legacy. (Or, at least, refrain from writing hyperbolic headlines). That being said - a company selling an alternate, high-performance networking protocol could have a strong market. ISP routing over the last mile, educational/research or any other institution that operates a WAN may be a good customer.
- bediger4000 12y ago1. This article smells like a PR Hit from the startup that the "MIT, CalTech and Danish researchers" have put together. 2. A powerful push to substitute something with very distinct (and patented!) client-role and server-role would be just the thing to push the Internet Genie back in its bottle. Needing a license for a server would make good business sense (for Comcast, AT&T and the various national telecomms) and good "national security" sense for various governments and rogue agencies. Making that license pricey could put gatekeepers back in business, so mainstream media will definitely be in favor of network protocols that are proprietary and have a server role and a client role.