It’s fun to read the history of the struggles creating these products that seemed so relatively polished at the time.
Having said that, I still write with my fingers hovered over Ctrl+S from the long-term trauma caused by having lost one too many high school essays to Word 6.0.
Also, there are two guys where I work who still use slash shortcuts and @ formulas in Excel!
This is funny. You just reminded my of something. Many years ago I saw somebody using Excel by creating formulas using the @ symbol. I asked him about it and he said he was “old school”.
I thought it might have been from Lotus 1-2-3 (which I had never used). But apparently it was part of early Excel? Was this used mostly on the DOS version?
Nice! Going way back to VisiCalc built-in functions used the @ to start with to differentiate from labels/ranges. Everyone kept that compatibility. You can see this in the original visicalc manual like this one for the IBM PC version http://toastytech.com/manuals/VisiCalc%201.1.pdf
I remember being told that Lotus had over 100 developers for 123 for Windows and that is why it was late, whereas the core Excel team was just 12 developers. Imagine the overhead of having 100 developers communicating.
Those numbers sound about right :-) Your comment really made me think about how often the number of developers on a project made it into the press or commentary. It was such a sign. And how often the really huge numbers (OS/2) were attached to projects in bad shape.
This is a great chapter. As written, you would not know that, say WordPerfect had shown a really good Windows product in the May 1990 COMDEX, which worked a lot like their already-shipping Mac product. In fact, seeing that and other Mac products "ported to Windows" at that COMDEX was a watershed for many of us in the valley; the half-decade discussion about which GUI platform would win was over for me once I said that, and I said so in my trip report to the Autodesk founders, who were still in charge at the time. With that event, I and my business cohorts turned off our NeXT machines and Macintoshes, and dedicated ourselves to Windows, after failed attempts to use it since 1986.
I licensed the file conversion tech from Mastersoft in Scottsdale for a database publishing product, and I think Microsoft did too, yes? Kent Mueller, the CEO at the time told me so. To me this was a huge feature for Word, because Word for Windows did a better job bringing WordPerfect DOS files forward than WordPerfect for Windows. I always thought that was an advantage for MSFT. Probably too much detail for you to get into.
WordPerfect for Windows finally shipped in 1992 (5.0) and was reviewed fairly negatively for both usability and quality reasons. The product stuck to the keystroke command structure which made it awkward to use with Windows. It even required one to exit Windows to install it (I set these all up during the course of writing). Similarly 1-2-3 was not particularly well-received and also stuck to its slash-command heritage. Neither really recovered from that though both continued for a long time.
I tend to think both were in the mode of serving the broader cross-platform market which optimized for their app at the center and that would have worked at the idea of Windows not become a central point in the broader customer perception.
I went to a 1-2-3-for-Windows demo, I think right before the release, run by a few Excel PMs. It was in the bldg 5 cafeteria, maybe? The PMs were giddy with delight. It barely ran. They had been building up this enormous list of glitches, and were just itching to get out on the road to their big clients. There was some corp-wide dictum, that we should all just sit down and shut up about 1-2-3, not go crowing about it. It seems silly now - to whom would we crow? - but that was early antitrust times, and mgmt was worried that we might come across as, mmm shall we say, arrogant.
RIght. Well, the shipping delay - nearly 2 years - of WP fits your mental model of incumbents being concerned about releasing on and enabling a platform from a company that has its own products on it.
Maybe they were too busy getting the NeXT version out the year before :-)
This is a great chapter, and how the OPU and DAD solved the integration problem is super interesting. In particular, the emphasis on partnerships and collaboration, and the deliberate rejection of viewing interface points as dependencies holding back the success of individual components, are very reminiscent of the solutions that the large civil and mechanical infrastructure projects I work in put in place to solve the interface and integration challenges between individual project parties that are always a headline risk to the success of the overall project.
I hope you're able to dive into this in more detail in subsequent chapters.
It’s fun to read the history of the struggles creating these products that seemed so relatively polished at the time.
Having said that, I still write with my fingers hovered over Ctrl+S from the long-term trauma caused by having lost one too many high school essays to Word 6.0.
Also, there are two guys where I work who still use slash shortcuts and @ formulas in Excel!
This is funny. You just reminded my of something. Many years ago I saw somebody using Excel by creating formulas using the @ symbol. I asked him about it and he said he was “old school”.
I thought it might have been from Lotus 1-2-3 (which I had never used). But apparently it was part of early Excel? Was this used mostly on the DOS version?
Nice! Going way back to VisiCalc built-in functions used the @ to start with to differentiate from labels/ranges. Everyone kept that compatibility. You can see this in the original visicalc manual like this one for the IBM PC version http://toastytech.com/manuals/VisiCalc%201.1.pdf
I remember being told that Lotus had over 100 developers for 123 for Windows and that is why it was late, whereas the core Excel team was just 12 developers. Imagine the overhead of having 100 developers communicating.
Those numbers sound about right :-) Your comment really made me think about how often the number of developers on a project made it into the press or commentary. It was such a sign. And how often the really huge numbers (OS/2) were attached to projects in bad shape.
I was visiting Microsoft, specifically the Excel team at the time, so those numbers were not from a press leak.
Oh sorry I was just riffing. I wasn’t accusing you of anything :-) I was thinking about other stories in addition to your comment. Sorry.
This is a great chapter. As written, you would not know that, say WordPerfect had shown a really good Windows product in the May 1990 COMDEX, which worked a lot like their already-shipping Mac product. In fact, seeing that and other Mac products "ported to Windows" at that COMDEX was a watershed for many of us in the valley; the half-decade discussion about which GUI platform would win was over for me once I said that, and I said so in my trip report to the Autodesk founders, who were still in charge at the time. With that event, I and my business cohorts turned off our NeXT machines and Macintoshes, and dedicated ourselves to Windows, after failed attempts to use it since 1986.
I licensed the file conversion tech from Mastersoft in Scottsdale for a database publishing product, and I think Microsoft did too, yes? Kent Mueller, the CEO at the time told me so. To me this was a huge feature for Word, because Word for Windows did a better job bringing WordPerfect DOS files forward than WordPerfect for Windows. I always thought that was an advantage for MSFT. Probably too much detail for you to get into.
WordPerfect for Windows finally shipped in 1992 (5.0) and was reviewed fairly negatively for both usability and quality reasons. The product stuck to the keystroke command structure which made it awkward to use with Windows. It even required one to exit Windows to install it (I set these all up during the course of writing). Similarly 1-2-3 was not particularly well-received and also stuck to its slash-command heritage. Neither really recovered from that though both continued for a long time.
I tend to think both were in the mode of serving the broader cross-platform market which optimized for their app at the center and that would have worked at the idea of Windows not become a central point in the broader customer perception.
I went to a 1-2-3-for-Windows demo, I think right before the release, run by a few Excel PMs. It was in the bldg 5 cafeteria, maybe? The PMs were giddy with delight. It barely ran. They had been building up this enormous list of glitches, and were just itching to get out on the road to their big clients. There was some corp-wide dictum, that we should all just sit down and shut up about 1-2-3, not go crowing about it. It seems silly now - to whom would we crow? - but that was early antitrust times, and mgmt was worried that we might come across as, mmm shall we say, arrogant.
RIght. Well, the shipping delay - nearly 2 years - of WP fits your mental model of incumbents being concerned about releasing on and enabling a platform from a company that has its own products on it.
Maybe they were too busy getting the NeXT version out the year before :-)
This is a great chapter, and how the OPU and DAD solved the integration problem is super interesting. In particular, the emphasis on partnerships and collaboration, and the deliberate rejection of viewing interface points as dependencies holding back the success of individual components, are very reminiscent of the solutions that the large civil and mechanical infrastructure projects I work in put in place to solve the interface and integration challenges between individual project parties that are always a headline risk to the success of the overall project.
I hope you're able to dive into this in more detail in subsequent chapters.
Thank you. Will be doing that and hopefully not too much (which is more what I am worried about!)