Mosaic was basically written without bandwidth considerations in mind, designed for a really fast network at a time when its developers had the NSF T3 backbone (45 Mbps).

There are two parts: (a) Mosaic was originally written without regard to bandwidth limits; (b) it was written with the NSF T3 backbone (45 Mbps) in mind. True: accounts by Mosaic developers or NCSA beyond Andreessen's own say (other developers, NCSA documentation, histories) confirm that early Mosaic design ignored bandwidth constraints and assumed high-speed networks such as the NSFNET T3 backbone, and records confirm that backbone ran at about 45 Mbps. Mostly true: the substance is confirmed but a detail is off, e.g. the developers assumed fast campus or Ethernet links rather than the T3 backbone specifically, or the backbone speed is slightly misstated. Mixed: only one part is corroborated. Mostly false: the evidence shows bandwidth was considered in the original design, though the claim has some basis. False: contemporaneous evidence shows Mosaic was designed with bandwidth limits as an explicit concern, or the backbone facts are wrong.

Set before the research began.

1. E1 NSFNET: A Partnership for High-Speed Networking | LivingInternet LivingInternet · 24 November 2020

   “Finally, the remaining T1 nodes would be upgraded to T3, increasing the total number of backbone nodes on the NSFNET from 13 to 16, all running at 45 Mbps.”

2. E2 NCSA Reflects on 35 Years as a Supercomputing Powerhouse National Center for Supercomputing Applications (NCSA) · 2 February 2022

   “NCSA Mosaic, written by Marc Andreessen and Eric Bina, working in NCSA’s Software Development Group, brought the experience to life with integrated pictures, movies and other elements, making it much easier to develop and interact with online content.”

3. E3 bkmosnav.rvw Vancouver Institute for Research into User Security (book review by Robert M. Slade) · 1 December 1994

   “He mentions the frustration involved with trying to use Mosaic with a”

4. E4 NSFNET: A Partnership for High-Speed Networking | LivingInternet LivingInternet · 24 November 2020

   “Finally, the remaining T1 nodes would be upgraded to T3, increasing the total number of backbone nodes on the NSFNET from 13 to 16, all running at 45 Mbps.”

5. E5 NSFNET: A Partnership for High-Speed Networking | LivingInternet LivingInternet · 24 November 2020

   “The last of the sixteen T3 sites was completed in the fall of 1991, over the Thanksgiving holiday, and production traffic was phased in.”

6. E6 NCSA Reflects on 35 Years as a Supercomputing Powerhouse National Center for Supercomputing Applications (NCSA) · 2 February 2022

   “While browsers weren’t new, NCSA Mosaic, written by Marc Andreessen and Eric Bina, working in NCSA’s Software Development Group, brought the experience to life with integrated pictures, movies and other elements”

7. E7 Internet Ascendant, Part 2: Going Private and Going Public technicshistory.com · 22 October 2020

   “So, when NSF asked Merit to upgrade the backbone from 1.5 Mbps T1 lines to 45Mbps T3, they took the opportunity to propose to Wolff a new contractual arrangement.”

8. E8 The building of the Internet - Jeffrey A. Hart Telecommunications Policy (Hart, Reed and Bar; PDF hosted by Indiana University) · November 1992

   “(bringing trunk transmission speeds up to 45 Mbps) in 1991-92 through”

9. E9 THE BIG BANG - stewart.sdsu.edu stewart.sdsu.edu (book chapter excerpt on Mosaic and Netscape)

   “By 1992, when the superfast T3 network was launched as the successor backbone for the Internet,”

10. E10 Internet Ascendant, Part 2: Going Private and Going Public technicshistory.com · 22 October 2020

    “T3 was a new frontier in networking – no prior experience or equipment existed for digital networks of this bandwidth, and so the companies argued that more private investment would be needed,”

Independent histories confirm that the NSFNET backbone was upgraded to 45 Mbps T3 in 1991–92 (E1, E5, E7, E8). E9 notes that the T3 network was launched as the successor backbone for the Internet by 1992. However, none of the evidence says how Mosaic was designed: nothing shows its developers ignored bandwidth or had the T3 backbone in mind. NCSA's retrospective names Andreessen and Bina as the authors but says nothing about bandwidth (E2, E6), and the review fragment in E3 is cut off too early to show anything about the design. Only the backbone speed is corroborated, not the claim about Mosaic's design, so the claim cannot be settled.