Q+A with Ben about Web Browser History and More

My friend Todd Warren is co-teaching a class on Web Development at Ashesi University in Ghana right now (Fall 2020). He invited me and my friend Tim Krauskopf to spend 45 minutes on a Zoom video call today (9/3/2020) with the class (~120 students, mostly living in Ghana) to answer their questions about the history of the World Wide Web and much more.

The students submitted many more questions than we could possibly cover in 45 minutes, so I have taken the liberty of posting their questions here along with my answers.

In order to help the students prepare there questions — and in attempt to answer some of their questions ahead of time — I sent Todd the following information to pass on to his class…

Questions from Students

1. The Browser Wars
a. How Intense were the browser wars (Lily Ofori-Amanfo)
BEN: Intense. Netscape got started in April, 1994 (as Mosaic Communication Corp., but changed to Netscape to avoid copyright issues with “Mosaic”, which was the first GUI web browser built on Windows by the NCSA at U. of Illinois in Champaign-Urbana).
I started the “O’Hare” project in early October, 1994. (O’Hare is the airport adjacent to Chicago, IL, and since “Chicago” was the code name for what became Windows 95, and our web browser was the portal to the world of the Internet for Win95, “O’Hare” seemed an obvious code name).

During the 22 months I led the engineering effort for what became Internet Explorer, I worked 80-to-100 hours/week for 17 of those 22 months. We shipped IE versions 1 (6/1995), 2 (11/1995) and 3 (8/1996), and I also had time to build a new house with my wife Lisa, and our third child was born in 1995.
This book published in 2000 by my friends Prof. Cusumano and Prof. Yoffie tells the story mostly from the Netscape perspective: https://www.amazon.com/Competing-Internet-Time-Netscape-Microsoft/dp/0684863456

b. Were there any challenges with matching up to your competitors? (Stephen Etornam Yaw Torku)
BEN: The “Browsers Wars” quickly became a two-horse race: Microsoft and Netscape. When we released IE 3.0 in 8/1996, it was widely acknowledged to be the best web browser for Windows, and Windows had the largest share of personal operating systems at the time. IE 3.0 reached about 30% market share by the time IE 4.0 was released in 10/1997, and subsequent versions of continued to gain market share for several years.
I believe Microsoft beat Netscape because: a) singular focus on building worlds best browser, b) better software engineering practices, and c) better software engineers.

c. In your opinion, what were the functionalities that pushed Internet Explorer ahead of Netscape (other than the fact that it was free) and why has IE died out now? And how do you think rapid advances in technology affected your jobs? (Princess Adwoa Nyinaku Asante)
BEN: Internet Explorer was first to market with CSS, first to market with Dynamic HTML, and it was faster and less buggy than Netscape. Being free didn’t help IE versions 1.0 and 2.0 gain any market share, because those versions were not as capable as Netscape Navigator.
IE faded away because Steve Ballmer — as CEO of Microsoft — thought Windows was the most important product, and so he wanted IE to only work on Windows. Which created a market opening for other browsers to do a great job across platforms, and eventually Google Chrome eclipsed IE.
I was at Microsoft from 6/1985 to 8/1999, and computer technology advanced rapidly: RAM and hard disks kept getting cheaper and more capacious, CPUs got much faster, video screens got higher-resolution and more colors, we shifted from CRTs to LCDs, communications speeds got faster. But my job was to design and build the right products for the marketplace, building and leading a team to do that work (in my last few years, I didn’t write any code). And so my job changed from being an individual contributor — designing and coding solutions to parts of a product — to being a Project Leader and building teams of people to build entire features/products.

d. Since Spyglass was referenced as the technology base for IE3 and was considered groundbreaking, why did it get overshadowed during the browser war between Netscape and Microsoft? (Afua Fosua Ayiku)
BEN: Spyglass was in the business of licensing the original NCSA Mosaic source code to other companies and organizations. They realized quickly they didn’t have the money and people to try and compete with Netscape and Microsoft, so Spyglass quickly withdrew from the battle.

Was there a particular target group that you built the browser for? Or what was one of the major mistakes that led to upgrading the MOSAIC browser? (Papa Kobina Orleans-Bosomtwe)
BEN: Like Windows 95 — which was the first consumer 32-bit operating system that reached a massive customer base — Internet Explorer was designed for everyone. [PKO-B: I don’t understand your second question. Please email me clarification.]

2. Current state of browsers

a. Ben – How do you feel about IE being shutdown for edge
BEN: Life is Change. I stopped working on IE in 8/1996. The subsequent teams who worked on IE helped grow its market share to ~95% by 2002, and then that share started to decline. It was down to about 50% by 2010, and IE and Chrome crossed in mid-2012 at 30% each.
As above, Microsoft made a strategic decision to stop investing in making the worlds best web browser, so this eventual decline was unavoidable.

b. Ben/Tim – I would love to know what drove the need for HTTP 2.0 from HTTP 1.1. Also, why is chrome developer tools preferred over edge developer tools? (Kweku Andoh Yamoah)
BEN: I wasn’t involved in any of the HTTP standards work, but from a quick read, HTTP 2 has a bunch of performance improvements over HTTP 1.1. This article seems to do a good job of describing the details without getting too geeky: https://medium.com/@factoryhr/http-2-the-difference-between-http-1-1-benefits-and-how-to-use-it-38094fa0e95b

3. Development of IE and Spyglass

a. If you could go back to the days of working on the Internet Explorer project, what would you do differently and why?
BEN: I’m very happy with how my team executed on IE 1.0, 2.0, and 3.0. We had 8 folks to start on IE 1.0, and I gave them 8 weeks to get IE 1.0 done (starting from early 1/1995). So we were ruthless about what features we added, and we had plenty of time to build a solid basic web browser to include in Windows 95. We “shipped” in 7/1995, and then I gave the IE 2.0 team — 4 senior software engineers — 4 months to build IE 2.0. We included a few key features — HTML tables and SSL in particular — and then some “fun” features like VRML, <blink>, <marquee>, and inline videos. IE 3.0 was the key release, and we were able to get CSS in, as well as JavaScript and Java to compete with Netscape, and VBScript (because, Microsoft).
So I wouldn’t change a thing!

Also why do you think the [IBM] System Application Architecture initiative didn’t work?
BEN: The short answer is that IBM’s SAA was a solution looking for a problem. A hammer looking for a nail. IBM had this fantasy of “write once/run anywhere” applications: they thought you could write a mainframe app and it would run on a PC, and vice versa. Later, Sun Microsystems had that fantasy for Java applications. That doesn’t work because different computers are just too different. Heck, it’s even hard to write a single application for iOS that works perfectly on both an iPhone (small screen) and iPad (larger screen).

a. How was the HTML tables, HTTP cookies and the crypto code for SSL integrated into Internet Explorer 2.0? (Richard Kafui Anatsui)
BEN: As above, IE 2.0 was a 4 month effort. I took 4 of my most senior software engineers, and they did the key work, including HTML tables and cookies. A new-hire SDE implemented cookies, and a summer intern from MIT implemented SSL, using a cryptography library that was available from the “crypto” group within Microsoft.

b. What were some of the obstacles you faced when it came to updating the software (from version 1.0 to 4.0) to meet the needs of customers? (also spyglass)
BEN: I only led the IE team for versions 1.0 through 3.0. I grew the team from 0 to 67 Software Design Engineers and Program Managers (another guy managed the testing team). See more at https://www.slivka.com/Case_Study_Internet_Explorer_1998-04-15.pdf

c. Without the internet (as we know it) at the time, how did you possibly navigate the development of such an intensive project?
BEN: You might as easily ask how human beings invented the sailing ship, the steam engine, the locomotive, the internal combustion engine, the automobile, the Wright Flyer, the transistor, and put a man on the moon in 1969!
Where there is a will, people find a way to accomplish their goals with the tools at hand!

e. What background research went into design for the MOSAIC browser ? (Ekow Woode)
BEN: A question for Tim!

f. How long did the team take, from idea inception to completion?
BEN: Asked and answered above for Internet Explorer versions 1.0 to 3.0.

g. In the early days, you had a different vision of how the internet was going to change everything. What motivated you into actually pursuing these interests?
BEN: I did have a vision for how the Internet might change software, and I have always liked building things, and so it was natural for me during my time at Microsoft to try to build software to make the Internet better.

h. How did you go about your research in terms of what kind of programming languages would work best and what was needed to come out with the best web browser? (Dionne Naa Mooley Adoteye)
BEN: No research required. We were building a 32-bit Windows application that needed to run decently on Windows 95 on a PC with an 80386 CPU running at (say) 20 MHz with 4 Mb of RAM. So our only real choice was C, as C++ had a large runtime library (so requiring more RAM) and tended to produce slower programs. I think IE 4.0 introduced some C++ code.

Eugenia Edomteh Aku Anyawoe Why does Microsoft Internet Explorer have the fastest startup time (as compared to other browsers) on a Microsoft Windows OS? Has it always been the fastest since its creation?
BEN: I was certainly focused on IE 1.0 to 3.0 being fast in all aspects, including start-up time. Just testing IE vs. Edge (latest versions) on Windows 10 right now, Edge starts up faster than IE. Chrome seems to start about as quickly as Edge, but both Chrome and Edge are built on the same code base, so that is not a surprise.

4. Impact / Legacy / Did you expect…

Did you have any inclination the level of impact that this project/company would gain at the time? (Tobias Ekow Woode) Seyram Fui Hugh-Tamakloe
BEN: When I joined Microsoft in 6/1985, I was expecting that Microsoft would have more impact going forward than IBM (where I had worked 6/1982-6/1983 and had 364K employees) or the “mini-computer” companies. But it wasn’t until we shipped Windows 95 and IE 1.0 that I realized how much impact Microsoft would have.

Considering the architecture of the internet then and now, would you say the processes and protocols involved in the past made the use of the internet safer as compared to present day safety?
BEN: Evolutionary theory was developed in an attempt to explain biology and the diversity of life, but it certainly is applicable to innovation and technology. The things you could do on the Internet in 1994 were pretty amazing compared to the pre-World Wide Web era! And we have had advances in networking speed — wired and especially wireless — as well as programming languages, frameworks, cloud services, and so much more.
As for safety, there is a “spy vs. spy” aspect to that, because it depends upon people and tools. And hackers — both individuals but especially hackers employed by nation states — have become increasingly sophisticated.
So the Internet can never be 100% safe because that is at odds with human nature. The best we can do is try to make systems and software safer, and educate ourselves about best safety practices.

5. General / Life /Career

a. Why is it important to stay updated on the most recent technology trends and developments? How do you constantly adapt to changes in the world, client demand and technology? (Petra Mensah Abosi)
BEN: I haven’t kept up-to-date on all sorts of different technologies! But then I stopped working for a salary back in 6/2000. In my experience from 1978-1999, though, I was able to keep up with the advances in technology by reading and working. I think you have to pick and choose what you want to learn about. More recently, I’ve been spending a lot of time trying to learn more about life sciences, molecular biology, genetics, and biochemistry.

b. Hey Ben, you mentioned that you worked about 80-100 hours a week while you were the lead of the Internet Explorer project. How did you handle that and did it have any effects on your body? (Andrew Philip Kojo Krampah Duncan)
BEN: I was 34 when I started the IE team and 36 when we shipped IE 3.0. I guess I had a lot of energy, as not only was I working long ours, but I was helping my wife with our two boys, our daughter was born in 1995, we built a new home in 1996-1997, and I still had time to read the Seattle Times and Wall Street Journal daily, and The Economist and Business Week weekly! The only effect was that my wife couldn’t take me to movies or plays or classical music performances, as I would fall asleep almost immediately!

6. Microsoft Strategy

a. How did IBM’s separation from Microsoft affect the whole organization and how did he feel about the split between the two companies. (Papa Kojo Ayensu Oseku-Afful)
BEN: Microsoft first licensed MS-DOS (aka PC-DOS) to IBM in 1980, and the IBM PC was launched in 8/1981. IBM also offered CPM from Digital Research, but at a higher price. Bill Gates and Steve Ballmer felt that Microsoft needed the endorsement of IBM for DOS, and so they worked hard to keep that commercial relationship alive.
The OS/2 effort started in the fall of 1985 as the Joint Development Agreement between IBM and Microsoft. IBM developed the PS/2 line of personal computers — an attempt to create a new line of proprietary hardware that could not be “cloned” by Compaq, PCs Limited (Dell), HP, and so many other “cloners”.
But the PS/2 hardware was very expensive, incompatible with the original IBM PC standards (especially the bus for add-in cards and the ROM BIOS), and early versions of OS/2 were not sufficiently compatible with exiting DOS applications.
I spent my first 5.5 years at Microsoft working on OS/2 (1.0, 1.1, 1.2, 1.3, and 2.0) — including building a prototype of “80386 OS/2” which used the virtual x86 mode of the Intel 80386 to be able to run four MS-DOS applications simultaneously (we could run Lotus 1-2-3, WordPerfect, Microsoft Word, and Microsoft MultiPlan).
So when the “divorce” from IBM happened in 1990, I was disappointed that all my work on OS/2 was history. But on the other hand, Microsoft now controlled its own destiny, and we could build a world-class operating system without any interference from IBM. So I was overall quite happy with that result.

b. Would you say your initial idea of content indexing the whole internet at the time would have become what Google is today or better if you had continued with that idea? (Perrin Nii Obodai Provencal)
BEN: I had the idea to create a search engine for the entire World Wide Web in late 1994 or early 1995. I event signed a license agreement for the Lycos web crawl database in mid-1995. But I didn’t have the personal bandwidth to spin up the search engine, and handed it off to the MSN folks, who were clueless about how to build a search engine. So that was a big failure.
If I had had the insight to hire a strong person to start the search engine effort within Microsoft, things could have been dramatically different. But that is a lot of “woulda coulda shoulda!”

If Microsoft Edge is the new default browser for windows and there will not be newer versions of Internet Explorer why has it not been discarded completely(it can still be found on windows computers)? Does it mean that the current version of IE still has some use or advantages over the Edge browser? (Nana Osei Somuah)
BEN: IEXPLORE.EXE remains in Windows because there may be many customers who depend upon that name for a web browser program. We call this “compatibility.” Interestingly, Windows 10 no longer automatically has the ability to run MS-DOS programs nor even 16-bit Windows programs (i.e., Windows programs written before Windows 95 shipped with support for 32-bit programs in 8/1995).

What makes Microsoft Edge different from Internet Explorer.
(Hephzibah Emereole)

  1. Based on your experience in the early pioneer of the world wide web, what are some of its areas that have remained constant over the years of the WWW improvement?
    BEN: Compatibility with older web pages have generally been maintained for several decades now. The JPEG and GIF formats from the 1980s and 1990s can still be rendered today, and most HTML from the 1990s and later still render correctly in 2020 in modern web browsers.
  2. What are the major requirements to consider when developing a browser apart from the interest of the company?
    BEN: Mobile phones have a much larger installed base compared to desktop and laptop PCs today. And that has been true for many years now. I think it would be foolish for any organization — let alone a for-profit company — to start a new web browser project in 2020.
  3. What are the key differences you have noticed in browser standards over the years of improvement?
    BEN: There have been many important improvements in web browser standards since 1994. The market — websites, web developer tools, consumers — have dictated winning vs. losing advancements.
  4. Why is the Google Chrome browser more popular than Internet Explorer (This is my opinion though)?
    BEN: As I’ve mentioned previously, Google aimed to build the world’s best web browser regardless of operating system. While Microsoft abandoned that goal in the early aughts.

Author: benslivka

19 start-ups, software, hardware, biotech, space launch, neurodiversity, learning, free markets, food, wine, cycling, walking, Seattle, Microsoft, Northwestern University, Garfield HS, DreamBox Learning, IBM, Amazon.

3 thoughts

  1. I’ve used you as an example to my coaching clients, both young and old; that they can be a ‘part of history’, not just read about it.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s