The fundamentals of good engineering have not changed much in the last twenty years. Dave Ricker was an engineer with a company I worked for as an apprentice Ironworker and again as a Journeyman Ironworker. The company was and still is involved with both new construction and the repair and strengthening of existing structures. He wrote a number of articles on the subject of structural steel design and has a well-earned reputation in the structural steel industry. Any article he wrote was based on his experience in design and fieldwork and the information is as valid now as it was when he wrote it.
The concept he is addressing is that you need to unload the existing structural member as much as practical to reduce or mitigate the chance of a structural failure and unexpected distortion during the welding of new steel to the existing steel. In most cases this would require the use of falsework and jacks to reduce the loads to the level suggested. Rarely is a main member in an existing structure not subject to dead loads. If there were no dead (or live) loads to contend with, there would be no need to have a structural member there to begin with. The allowable stress permitted by the design code and the magnitude of the load determines what size member is necessary. It is usually the practice to size the structural member so that it is loaded to near its allowable stress level. If steel was free, that might not be the case, but the reality is that no one wants to pay for larger structural members than necessary. Assuming the member is loaded to the maximum permitted means that it will have to be supported by temporary supports while reinforcement or replacement structure is added.
Best regards - Al