I.                     Chapter 8 Major Markets--WordProcessors and Spreadsheets.

     A.       The Market for Spreadsheets

Today’s computer programmers take it for granted that the brass ring of programming is the “killer-app,” the next application that will take the market by storm. There have been a number of killer-apps in the history of the software industry—they’ve come and they’ve gone—but none has had more staying-power than the oldest killer-app of them all: the spreadsheet. Spreadsheets are still one of the mainstays of computer applications.

Spreadsheets allow the manipulation of numbers and formulas, and their conversion to charts. A feature of spreadsheets that might tend to lock in consumers is their use of macros—programs within the spreadsheet that allow the spreadsheet to perform certain repetitive tasks. Many of these macros prompt the user to enter specific data, allowing even less-skilled data-entry personnel to update, maintain, and make sophisticated use of data. For this reason, many users, particularly business users, are loath to switch to a spreadsheet that disables their old macros. This dependence on macros makes spreadsheets, more than most other programs, very susceptible path dependence. Of course, the cost of converting macros is a real cost that ought to weigh in any consideration of whether a switch is efficient.[1]

1.    The Evolution of the PC Spreadsheet Market

Credit for the invention of the spreadsheet goes to Dan Bricklin and Bob Frankston, who created VisiCalc for the Apple II. VisiCalc required 32k of memory to run. A chart from a “roundup” of spreadsheets in the 1982 Personal Computing magazine[2] lists eighteen spreadsheets. Most ran on either the Apple II or on machines with the CP/M operating system. VisiCalc ran on both, as well as the IBM PC, the Atari 800, the Commodore 8032, the Radio Shack TRS-80, and various HP hand calculators. VisiCalc had a list price of $250. The most expensive spreadsheet in the roundup was M.B.A, from Context Management Systems, which combined database, graphics, word processing, and communications functions for $695.

In January 1983, Lotus introduced Lotus 1-2-3 at a price of $495. It was immediately acknowledged to be a better product than VisiCalc. In December 1982 Gregg Williams wrote in Byte that 1-2-3 had “many more functions and commands than VisiCalc” and that 1-2-3 was “revolutionary instead of evolutionary.”[3] PC World called it “state of the art.”

In October 1983 PC World reported that 1-2-3 was outselling VisiCalc.[4] VisiCalc was removed from the market in 1985 after being purchased by Lotus. Users of VisiCalc were offered upgrades to 1-2-3.[5] Lotus would maintain a dominant market share for almost a decade. Unfortunately, we do not have detailed data on this early market. But from what we have pieced together, it is clear that in the early 1980s a superior product, Lotus 1-2-3, was able to wrest market share away from VisiCalc—and to do it very quickly.

By 1985 most spreadsheets were meant to work with the IBM PC. Lotus 1-2-3 was the market champ, with other contenders such as Computer Associates (Easy Planner; $195), Ashton-Tate (Framework; $695), Software Publishing (PFS Plan; $140), and IBM (PlannerCalc $80 and Multiplan 1.0; $250). Microsoft had Excel for the Mac (at $495) and Multiplan 2.0 ($195) for the PC. But many new spreadsheets were beginning to arrive on the scene. Some were clones of 1-2-3 at lower prices (VP Planner, The Twin, and VIP Professional). Others, such as Javelin and SuperCalc, offered improvements on 1-2-3. These alternatives elicited some praise, but the acclaim was not universal, and they did not make much of a splash in the market.

When Excel first appeared in 1985, it was offered only for the Macintosh. Jerry Pournelle, a well-known columnist for Byte (and science-fiction author), wrote (incorrectly but nonetheless prophetically): “Excel will make the Mac into a serious business machine.”[6] In late 1987 Microsoft ported Excel to the PC (running under Windows) and Borland introduced Quattro for DOS. Thus began a market struggle between Microsoft, Borland, and Lotus.

Reviews found Windows-based Excel to be the best product, with DOS-based Quattro in second place. 1-2-3's market share was unable to stand up to the assault of superior products. Excel's steadily gained on 1-2-3, eventually overtaking it. It has been firmly in first place ever since.

1.    Spreadsheet Quality

In the early and mid-1980s quality reviews rated 1-2-3’s closest competitor to be SuperCalc. PC Magazine, in its “Best of 1986” review had this to say: “If market dominance were based on rational criteria, Computer Associates' SuperCalc 4 would certainly replace 1-2-3 as the leading spreadsheet program. After all, it can do anything that 1-2-3 can do and adds some notable features of its own.”[7]

But although various spreadsheets had attributes that were sometimes considered superior to 1-2-3’s, there was no general consensus that any alternative was clearly the best. For example, in October 87 Michael Antonoff, in Personal Computing, said: “SuperCalc, VP-Planner, and Twin lack the elegance of 1-2-3 in links to applications.”[8]

But the quality of the competition began to ratchet up in late 1987. In a portentous statement, PC Magazine stated: “Microsoft Corp. has just unleashed a spreadsheet that makes 1-2-3 look like a rough draft.”[9] This reference was to Excel 1.0 for the PC, a port of its Macintosh program.

By 1988 the spreadsheet market had developed into a three-way fight among Lotus’s 1-2-3, Microsoft’s Excel, and Borland’s Quattro (Pro). Excel, created to run under a graphical operating system, was a late entry.  It was ported from Macintosh only after an early version of Windows became available. Borland and Lotus, on the other hand, did not produce Windows versions of their programs until after Windows 3.0 had proven itself as a successful operating system.

According to reviewers, Lotus 1-2-3 was falling behind the competition in terms of functionality and usability. According to one reviewer, “Excel offers a lot in the form of tantalizing features missing from the current version of 1-2-3.”[10] Another review called Quattro “a powerful spreadsheet with more features than 1-2-3 Release 2.01, yet fully compatible and at a better price.”[11]

These were not isolated opinions. Reviewers in general had a very high opinion of Excel in the late 1980s, and almost as high an opinion of Quattro. Below is a list of review opinions for Excel. This list of  reviews is not edited; it includes all the reviews in which the reviewer was willing to state an opinion. Clearly, Excel was thought to be the best spreadsheet.

 

Table 8.1: Excel Reviews

Magazine/Article

Date

Comments

InfoWorld: “Microsoft's PC Spreadsheet Sets New Standard” p. 41John Walkenbach

12/21/87

“Microsoft has a winner here” “ one of the year's most innovative products, more powerful and more forward looking than any other spreadsheet on the market”

PC Magazine: “A New Face for Spreadsheets” p 103 Jared Taylor

12/22/87

“feature for feature, Excel is far better than 1-2-3.” p. 111 “greater hardware demands only con”

InfoWorld: Software Review p. 55 John Walkenbach

1/11/88

“the industry's most customizable spreadsheet” “takes the industry standard and improves on it in the ways that matter most”

PC Magazine

1/12/88

AWARD FOR TECHNICAL EXCELLENCE p.176

PC World: “Excel: Should You Switch?” Ralph Soucie p. 108

Mar-88

“now a serious challenger- indeed, a better product has made its debut: Microsoft Excel.” p 108 “if you're training new spreadsheet users, Excel is definitely the product of choice.” p.115 for Release 2”

Byte: Rich Malloy p. 155 “Excel Extraordinaire”

Mar-88

New for IBM, “rare product that combines ease of use and exceptional power” p. 157

PC Magazine: First Looks p.33 Craig Stinson

1/31/89

“For well over a year, spreadsheet users who needed to generate gorgeous printed reports from their worksheets have had one product to rely on: Microsoft Excel.” p 33

PC Magazine: “Spreadsheet Analysis” Douglas Cobb & Stephen Cobb p139

3/28/89

“Microsoft's Excel is the most powerful spreadsheet on the market today.” p 139

PC Magazine: Spreadsheet Analysis p. 185 Craig Stinson

7/1/89

“About a year and a half ago, something happened. Microsoft introduced Excel, a revolutionary advance over SuperCalc, 1-2-3, and everything else in the field.” “more spreadsheet than 1-2-3” “more solid product than QuattroPro”

PC Magazine: “First Looks” p 33. Craig Stinson

10/11/90

“new Macintosh spreadsheet, just introduced for Windows and OS/2 markets is the most graphical of all spreadsheets.” “least compatible of major spreadsheet programs” “charting options unparalleled” “comes up short in the services it provides for routine spreadsheet work.” p. 35

Byte: “New Extra's for Excel,” Andrew Reinhardt p. 136

Feb-91

“unassailability of vision” p. 138, Highly Recommend

The reviewers’ main reservation about Excel was its need for powerful hardware, due to the memory requirements of its entirely graphical interface (see the December 87 review in PC Magazine). Microsoft’s election not to produce a DOS version of the program was something of a gamble. Windows’ success was far from assured until version 3.0, which was not available until 1990. To run Excel, DOS users had to load Windows, and then return to DOS for other applications. This had a dampening effect on sales. Also, many of Excel’s features worked best with a mouse, and at the time it was rare for PCs to be equipped with a mouse. Furthermore, as is the case with virtually all graphical software applications, Excel was slower at most tasks than the DOS competition, though it could show results that nongraphical (DOS) based applications could not. Some have guessed that reviewers preferred Excel only because they had modern and powerful equipment. In fact, though the reviews may have exhibited this bias to some degree, they nevertheless usually faulted Excel for its onerous hardware requirements.

The reviews for Quattro were not quite as enthusiastic as those for Excel, but were still quite positive:

Table 8.2: Quattro Reviews

Magazine

Date

Comments

Byte: “Double Threats to Lotus 1-2-3” Diana Gabaldon p. 167

Jun-88

“all of Lotus's main features & extras at a lower price” “both have advantages over 1-2-3”

InfoWorld: “Spreadsheet Matches 1-2-3, Even Passes It in Some Areas” p. 50 John Walkenbach

8/22/88

 “multilevel Undo and Re-do, and background re-calculation(feature shared with Excel)”

Byte: Rich Malloy p. 111

Nov-89

“advantages of Excel with 1-2-3 compatibility.” p. 112

PC Magazine: First Looks p. 33 Craig Stinson

12/26/89

“noticeably slower times than competition.” p. 35

PC Magazine

1/16/90

“this is the top dog spreadsheet” p.98

Personal Computing: Joseph Devlin p. 145

Apr-90

“head and shoulders above Lotus” “requires add-ins to get Lotus to this level”

PC World: “Seven Sensible Spreadsheets” Richard Scoville p. 116

Apr-90

Quattro Pro 1.0 BEST BUY- “determined to do 1-2-3 2.2 better in every critical area, and it achieves that goal in spades.” p. 119

Byte: Andrew Reinhardt p. 156

Nov-90

“Quattro Pro is probably your best answer” p. 157

These statements strongly imply that by the late 1980s 1-2-3 had fallen behind its major competition. This is confirmed in the opinions of magazine readers.

Figure 8.1 shows the rankings of spreadsheets by readers of PC World. They were asked to pick the leading spreadsheet . We expect that readers will base their ratings on their experiences, the market shares of the products, and some idea of quality as indicated in magazine articles. These rankings, therefore, are indicators of neither market share, nor quality, but some combination of both.

<Fig 8.1>

The dominance of Lotus is clearly seen in the mid- to late 1980s, but in 1988, with the introduction of Excel and Quattro, Lotus’s dominance began to erode. The introduction of Windows in 1990 accelerated this decline. By 1992 Excel had surpassed 1-2-3 and Quattro had reached parity. By the time PC World’s reader rankings ended, in 1993, the die was cast: Lotus was doomed.


 Lotus’s fate is clear in Figure 8.2, which gives each spreadsheet’s number of annual “wins” (the number of times a product wins a comparison review or is declared to be the best product). Over the ten-year period shown Excel was the clear winner, although Quattro also managed a fair share of wins between 1989 and 1994. The remarkable feature of this chart, however, is that over the entire ten-year period, Lotus 1-2-3 just barely avoids a shutout, managing but a single win.[12]

<Fig 8.2>

Figure 8.3, which presents results from reviews that numerically score the three spreadsheets, shows the relative performance more precisely. Lotus’s poor performance is easily seen, as is Excel’s dominance after two initial poor showings. Note that the horizontal axis is scaled in chronological order and not by date. In other words, the distance on the axis between any two (or more) points has no meaning.[13]


<Fig 8.3>


One seeming inconsistency is the poor initial showing of Excel in 1988 and 1989, years when (as shown in Table 8.1) Excel was garnering rave reviews. The reason for this inconsistency is buried in the two negative review of Excel that appeared in Personal Computing. The deciding factor in Personal Computing’s quality ratings was speed.  Weight on this criterion was common at the time (since background recalculation was not yet standard and the operator had to wait at the keyboard for the spreadsheet to finish calculating before entering more data). Excel, of course, was exclusively a GUI-based application. And as we noted above in our discussion of the Macintosh versus PC, graphical products are much slower than text-based products. Listen to the authors of the review (from the September 1988 issue):

Our weighted average tended to punish Microsoft Excel for its lethargic performance; it rated only 1.3 on Overall Performance. [The others averaged about 7.] Don’t forget to look at the Overall Practicality rating, where it ranked first by more than half a point at 6.6. Where features are concerned, no other spreadsheet available today can beat Excel.

There has to be a catch—and there is. Unlike the other five spreadsheets reviewed and benchmarked here, Excel is graphics based instead of character based (it runs under Microsoft Windows). That means a lot more bits and bytes have to be processed. In five of our seven tests, Excel came in last by a wide margin.

The message is clear: Excel was clearly the leading spreadsheet in terms of capabilities. It should have been easy to predict the outcome:  Once the hardware had caught up to the software (and Windows itself improved) there would be no serious challengers to Excel as long as Excel continued to outperform its competition.

2.    The Role of Price

This brings us to another consideration: price. If there are important differences in the prices of competing products, a lower-quality and lower-priced product might have a larger market share than a higher-quality higher-priced product. Ford sells more Tauruses and generates higher revenues than Mercedes-Benz does with its mid-sized sedan, though Mercedes would normally be considered to be of substantially higher quality.

The magazine review articles, which generally report list prices, give an account of how list prices for spreadsheets varied across time and package, but this information does not explain much about how the spreadsheet market evolved. For one thing, in any one year, the list prices for competing products don’t differ by very much. As Figure 8.4 shows, Lotus normally charged in the vicinity of $500 for 1-2-3, and that was about what Microsoft charged for Excel. Although Borland initially (1988, 1989) charged considerably less for its early version of Quattro, its more advanced product, Quattro Pro (introduced in 1990) was listed at just about the same price as the other two products.  But the magazine review articles contain list price. And as we shall see shortly, an examination of actual prices tells quite a different story.

<Fig 8.4>


There are many reasons that list prices don’t tell much about the market. First, list prices do not include upgrade purchases, which are less expensive than first-time purchases. Second, list prices fail to account for units sold in office suites. Third, list prices fail to account for the units sold to OEMs, which carry a far lower price.


If, then, we want to look at prices, it is far more informative to look at average prices received by the manufacturer (Figure 8.5).  The history of average prices clearly shows Borland’s price-discount strategy which was far less apparent looking at retail prices.  It shows that Lotus kept its prices similar to Excel’s even in the face of the latter’s increasing market share and superior reviews. Lotus began to undercut Microsoft’s price significantly only in 1996, well after it had fallen below Excel in market share.

<Fig 8.5>

The big story in Figure 8.5, however, is the stunning decline in prices over time. By 1997, the typical price (received by the vendor) for a spreadsheet had fallen to approximately $50, a fall of over 80 percent from the typical price in 1988.

3.    Changes in Market Share: Analysis

The history of price and quality in the spreadsheet market indicates that Lotus 1-2-3 should have lost market share and market dominance. Prices were roughly equivalent for all major competitors, but Lotus 1-2-3 lost its ranking as the top-ranked product. The winner should have been Excel. Is this what happened? Absolutely. 


Figure 8.6 shows market share in the spreadsheet market based on units sold over time. It shows 1-2-3 to be decisively losing its dominant position to Excel. The evidence is even more powerful if one examines market shares based on revenues (Figure 8.7).

<Fig 8.6>

<Fig 8.7>

These results support the view that markets choose better products, but it is also true that Excel’s edge in quality was reinforced by several additional factors. Not only was Excel the best Windows spreadsheet, but it was also the first Windows spreadsheet. It was part of the first office suite, and it was also part of the best office suite. These reinforcing factors were strategic gambles by Microsoft.  They paid off handsomely, but that doesn’t mean that Microsoft was “leveraging.”  It means that Microsoft was smart.

Furthermore, these results do not provide any empirical confirmation for the idea that the spreadsheet market was “tipping,” a concept that has been adopted rather uncritically by many economic commentators. If network effects (or increasing returns) are important drivers in a market, then when one firm achieves a sufficiently large market share, consumers ought to flock to that firm to take advantage of network effects. We saw a theoretical illustration of this in chapter 3 in our model of multiple equilibria. In that model, any movement away from the unstable equilibrium  would lead to a single victorious firm. Within the limited contest of those models, when a market moves away from that equilibrium point,  network effects would become self-reinforcing, accelerating changes in market shares.


In a tipping market, when a product has a small market share, network effects should work against an increase in market share. But if market share does pass some break-out point, however, network effects should work to accelerate further market share increases.[14] This implies that we ought to see a pronounced upturn, where plots of market share against time (as in Figures 8.6 and 8.7 ) are flatter to the left of the break-out point, and steeper to the right. Similarly, there ought to be a pronounced downturn for a product with a market share that is declining.

This, of course, is not at all what we find in the spreadsheet market.  Instead what we find is a fairly steady increase in Excel’s market share and a fairly steady decrease in 1-2-3’s. This may mean that network effects are just not a very important factor in this market, or that their impact on market share is more complex than indicated in the concept of tipping.

There are some other issues worthy of examination. In a world of instant scalability, why did it take five years, and not one or two, for Lotus to be dethroned? We can suggest a possible answer. Users know that software products often leapfrog each other in capabilities. The costs of switching are high, and it does not normally make sense for a consumer to switch products every time one product exceeds the capabilities of another. Instead, a rational consumer will bide his time, confident that the next upgrade of a product that has fallen behind, but that has a history of technological innovation, will contain the features missing from the current version—and perhaps even add a few extra features. It appears to have taken two failed generations of Lotus 1-2-3 before the market was convinced that 1-2-3 was not going to catch up to Excel any time soon.

A second question is why Quattro never surpassed 1-2-3. Most reviews rated Quattro as superior to 1-2-3, though inferior to Excel. There are two reasons, however, why Quattro might not have surpassed an inferior 1-2-3. First, if a 1-2-3 user was planning to switch, it would make more sense to switch to the number 1 product than to the number 2 product. Quattro’s only advantage over Excel was its lower price, and this didn’t last. Second, because of instant scalability, Microsoft could meet the demand of all defecting Lotus users without having to raise its price. In markets without instant scalability, the number two firm likely would pick up some of the slack due to constraints on production for the leading firm.

Furthermore, between 1988 and 1995, when Quattro earned better reviews than Lotus, the gap between Lotus and Quattro did diminish—from a ratio of about 7:1 in units to a ratio of 2:1. Eventually, however, Quattro came out with a product that didn’t beat 1-2-3 in the reviews. Quattro for Windows never established a clear quality superiority over 1-2-3—and sometimes was rated worse.  After that, not surprisingly, Quattro failed to make any further gains. And of course, both products were losing market share to Microsoft’s Excel.

One wonders whether the Lotus hegemony over spreadsheets would have ended without the advent of Windows and office suites. Certainly some GUI was going to replace DOS, and spreadsheets were going to have to take advantage of it. If Lotus couldn’t produce a high-quality graphical spreadsheet, it was destined to lose the market to someone.

Microsoft would have been a strong candidate to dethrone Lotus in any graphical environment because of its experience in the Macintosh market. If the market had gone to Macintosh, where Excel was clearly the dominant product (as we demonstrate below), Lotus almost certainly would have lost. If OS/2 had predominated, there is every reason to believe that the OS/2 version of Excel would have replaced 1-2-3. And even if some third-party graphical operating system had prevailed on Intel-based machines, there is every reason to suspect that Microsoft would have produced a better product than Lotus because of its familiarity with graphical products for the Mac.

4.    Why Did Lotus Fail?

Lotus 1-2-3 failed because other competing products were judged to be superior in quality.  This is not surprising. One might wonder, however, why Lotus was unable to keep up.  Why didn’t it manage to produce a quality product that would have allowed it to retain its market leadership?

The internal reasons that Lotus was unable to produce a product that was as good as its competitors’ are beyond the strategic analyses of this book. But it’s clear that Lotus made some management blunders in positioning itself in the market as a whole.

One tremendous oversight was Lotus’ general disregard of the Macintosh market. Although the revenues from this market might have seemed relatively inconsequential, the experience in producing graphical spreadsheets in what was clearly the leading GUI of its time is something that Lotus should have valued.

Second, Lotus bogged itself down producing too many spreadsheet variations. Concerned that it would cede some market share in the 286 market (slower processors and smaller memory), Lotus continued to work on 2.x (text-based DOS) versions of 1-2-3, even though it was producing a quasi-graphical version (3.x) that also ran under DOS. The 3.x version of 1-2-3 was late to the market because the software engineers decided to translate it into the C language (to make it portable to other systems) which required considerable tweaking to keep it small enough to fit in the typical PC’s memory.[15] Meanwhile, Lotus was slow to create a version of 1-2-3 for Windows. Microsoft, of course, also produced multiple product lines, but it seemed better able to handle multiple versions of its products.[16]

In 1990 Lotus came out with Improv, a truly innovative spreadsheet that introduced what has come to be known as the pivot table. But Lotus squandered its innovative edge. There were two problems with Improv. First, it was introduced for the NeXT computer, a technologically oversophisticated (for ordinary users), impractical (the hard drive was replaced by a rewriteable CD-ROM), and expensive machine that failed to survive in the market. Second, Improv was a completely separate product, not a version of 1-2-3, and for all of Improv’s sophistication, it lacked some basic spreadsheet functions. If Lotus had chosen to include some of Improv’s features in 1-2-3, Lotus would undoubtedly have had something of a coup. But by the time Lotus managed to incorporate Improv’s innovations into 1-2-3, the competition had already caught up.

Improv’s failure should have been predictable. After all, even if Lotus couldn’t have been expected to know the extent to which the NeXT would flop, it should have noticed NeXT’s price, which was thousands more than competing computers, would keep it from mass market appeal any time soon. If Lotus wanted to try a new market, it should have tried the Macintosh market, which did not yet have an entry from Lotus, and which would have to wait two additional years for one.

Lotus also placed a wrong bet on OS/2, creating a spreadsheet for that failed operating system.[17] More seriously, the product that Lotus created for OS/2, and then also for Windows, was merely a translation of its 3.x product for DOS (using the “portable” C code). DOS products were not similar to products designed for GUIs, and Lotus should have known that a straight port of a DOS product would not turn heads in the GUI world. And even this product was delayed.

Lotus might have been able to survive several of these mishaps if it had done one or two things right. But it seemed to choose wrong at every turn.  By 1993 some reviewers were already penning obituaries:

Unfortunately, it looks as if Lotus is internally hog-tied by an infrastructure of 1-2-3 diehards the same way Apple was nearly strangled by Apple II loyalists.… The company's tombstone will read “1-2-3 and out!”

Lotus was the king of spreadsheets, and was rightly the leader, as it dominated the entire market. Disregarding Improv, nobody would call Lotus a leader in the field it pioneered anymore. People would cite Excel or Quattro Pro as the leading spreadsheet program.

The long-term winner in this battle is Microsoft, which can unload Windows plus Excel for such a low site-license price that few companies can resist. [18]



[1] Several programs created macro readers that were completely compatible with that of Lotus 1-2-3. Lotus sued several of these companies over the look and feel of their software, and won its cases against Paperback Software, Mosaic software, and forced Borland to remove its 100 percent compatible macro reader.

[2] April, page 62.

[3] Page 182

[4] PC World, October 1983, page 120.

[5] PC World, December 1985, page 221.

[6] Review Of Excel Demo At Comdex, Byte Magazine, September 1985, p. 347.

[7] p. 115, January 13, 1987, by M. David Stone

[8] Page 101.

[9] November 10, 1987, PC Magazine: First Looks” p. 33 - Jared Taylor -

[10] Michael Antonoff, December 1987 Personal Computing, p. 102.

[11] Mike Falkner, PC Magazine, December 12, 1987, First Looks” - Quattro: More Than 1-2-3 At Less than Half the Cost, p.33.

[12] We exclude the November  93 issue of InfoWorld that gave Lotus the highest marks, since Excel was not included in the comparison.

[13] Beginning with the October 1992 InfoWorld review, all reviewed programs are of Windows versions. Prior to that review, all are of DOS versions, except Excel.]

[14] We are using recent market shares to check for tipping which is consistent with our view that consumers update their forecasts based on what has happened lately. Alternatively, it is possible to use installed base as a measure of network effect, which would be more consistent with much of the previous literature. We will examine that issue in future research.

[15] “PC Spreadsheet Software” IDC 1991. From page 9: “The production of Lotus’s upgrade to 1-2-3 Release 2.1 proved to be a difficult undertaking for the company. In the end, Lotus shipped two products as possible upgrades to 2.1….Release 3.0’s production schedule stretched out as Lotus attempted to fit the product within reasonable memory restraints….Finally, Lotus went to an outside source, Rational Systems, and purchased DOS extender technology.”

[16] Microsoft also had several product lines: Excel for Macintosh, Excel for Windows, Excel for OS/2 (released in October  89), and Multiplan for DOS. Multiplan had been a significant product, placing number two among DOS spreadsheet in 1988 and fifth in 1989. Microsoft wisely let it go in favor of Excel.

[17] This was Lotus 1-2-3 3.0g, where the g stood for graphical. It was essentially the same as Lotus 3.0 for DOS, as was the initial version of 1-2-3 for Windows.

[18] Dvorak, John C. “Lotus, move forward or die!” PC Magazine: Mar 16, 1993, page 93.