As two distinct protocols, TCP and UDP are amenable to comparison. I'm curious as a software architect, would choose TCP over UDP in these two specific cases.

The four TCP/IP abstraction layers have me baffled.

While developing a reliable transport protocol, early versions of the protocol discussed in class used a NAK message, sent by the receiver to the sender to indicate a packet was not correctly received. This kind of message, however, is not found in further protocols, like GBN, SR or even TCP. Why not? In other words, why don't these protocols provide a way to signal to the sender that the received message was received with an error

Is the preference for TCP over UDP due to the former's supposedly higher level of safety?

Here, the distinctions between "connectionless" and "connection-based" protocols are examined.

As two distinct protocols, TCP and UDP are amenable to comparison. I'm curious as to why you, as a software architect, would choose TCP over UDP in these two specific cases.

The four TCP/IP abstraction layers are as follows; please describe them in detail.

Consider connectionless protocols as an alternative to connection-based ones and analyze their benefits and drawbacks.A pause was observed in the conversation.

When designing Elementary data connection protocols, what kind of assumptions were made?

Consider the pros and cons of connectionless protocols as a potential substitute for connection-based protocols in a brief overview