2/ If you would have told me in 1997 that I would still be talking about the paperclip in 2021, I would have LOLed and OMGed. I definitely want to go through reasoning and choices of the feature because it is super interesting. But the real story is not as much why, but how?
3/ In 1995, Office was half of Microsoft's profits. Through the release of Office 95 there were over 12 million licenses sold (about 100M Windows capable PCs had been sold since 1990). Still, there was competition from two billion-dollar companies, leaders in the DOS era.
4/ Yet, we faced an enormous problem with our software (and the PC in general). The Graphical OS was supposed to make the PC much easier, but still people really struggled. In fact, PCs seemed to be getting more difficult to use as more people used them.
5/ Buying Office also meant probably buying one or more giant books (700+ pages) on how to use Office or maybe taking a course offered by work or at a community college. To many, learning Office was like learning a whole new profession. And customer satisfaction was declining.
6/ We had to take risk. The way the Office team had historically worked was to take big risks -- bet on GUI, rewrite PowerPoint, reset the WinWord project. Doing the Assistant was a big bet and we did it without fear of failure, without a zillion meetings, without consternation.
7/ There's something magical about that time--when MS was willing to take core product risk in a significant way. We took it seriously doing more testing and validation than we had ever done. It was kind of crazy. It's cliché, but testing went very well😬
8/ In the post you'll find the story of the technology and creating the feature. It is super interesting. The result and how that perception changed over years amazes me: from ridicule, to shame, to kitsch, and ultimately to "wow so ahead of its time".
9/ I think a lot about how special the company was that enabled risk-taking like this and doing so without fear of failure, fear of repurcussions, or marks on your performance record. This post has several embedded videos and background materials. FUN!
10/ We in Office (and then Windows) would continue to make big bets. It gets easier over time to avoid doing so. Please subscribe and join us on the journey of the PC. …rdcoresoftware.learningbyshipping.com
PS/ Conan was very tough to watch (even today--those cheers!) but then later when lolcat "memes" became a thing c. 2007 (anyone recall icanhascheeseburger?) this Clippy meme (using Office 2000 version) kind of made me smile.
PPS/ It is true that sometimes people snuck Easter eggs into the code so they could add credits. When I was an exec I had no, no official knowledge, of such activity. However, these are the credits for the Assistant feature. Office 97 RTM on Windows 98, no VM :) #soundon
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Study of Microsoft employees shows how remote work puts productivity and innovation at risk geekwire.com/2021/study-mic… via @GeekWire // This is a paper out from a large group of researchers at Microsoft (and others). I have many thoughts on this. 1/
2/ My intention is not to comment on research per se but on how it might be misapplied. Studies of electronic communication in orgs--have been put forth ever since MSFT introduced email. At best this is telemetry and at worst it can be used to imply causality.
3/ I have no doubt that this research accurately captures the flow of information using digital tools around the company for over 60K people. That's a huge amount of work and analysis. Kudos. The challenge from the outset is that it conflates that flow with "collaboration".
With early success in a product there's often a strong desire (or rush) to "make it a platform". Having an app is great and making it a platform is better.
How Microsoft and Apple worked *together* to create Macintosh is a huge lesson on building a platform. 1/
2/ Apple saw the value of having VisiCalc on Apple ][ and IBM saw that for PCs with 1-2-3.
The common thread is that platforms benefitted from a third party betting their future business on the platform. It was existential for the platform to have companies doing that.
3/ Conversely, third parties came to realize that betting their business on a platform can create a stronger relationship--an influencer relationship--with the platform. Bill Gates saw that potential with Macintosh.
I know it is difficult to believe, but there was a time when key tech leaders and influencers of the world were dead set against the graphical interface.
In 1985, less than a year after Macintosh was unveiled the naysayers were out in full force... 1/
2/ About 2M Apple ][ had been sold in total. About 3.5M IBM PCs (8086). About 8M Atari, TRS-80, C64 all combined. This was early. ~20M computers sold, worldwide, total.
Dr Dobbs, InfoWorld, Byte magazines were supreme. We're in "Halt and Catch Fire" S1. Joe MacMillan reads IW.
3/ If a hobbyist magazine printed a story you didn't like you probably just ranted at your user group meeting thursday night.
If it really bugged you, then you'd write a letter to the editor. Maybe they would print it a few weeks later. Then a few thousand people would see it.
Incompatible Files, Slipping, Office 97 RTM— new post in “Hardcore Software” …rdcoresoftware.learningbyshipping.com/p/045-incompat… // “Reviews” were a key part of the early days of the PC era. In the context of shipping Office 97, this post looks at how reviews were changing as the industry matured. 1/11
2/ From the earliest days through Windows 95, personal computer reviews were primarily done by “tech enthusiasts” and aimed at same. Basically hobbyists reviewed products for hobbyists. That was the industry. Here’s BYTE giving Office 97 ⭐️ ⭐️ ⭐️ ⭐️ ⭐️ across the board.
3/ Such enthusiast outlets were our key constituency as most all sales were driven through retail channel and most retail customers were buying magazines like these by the pound (at the peak, an issue of BYTE or PCMag was hundreds of pages, mostly ads). So victory.
"Making the Laptop Commonplace" - NY Times 1985. A great example of challenges in forecasting progress in technologies when improvements are happing at exponential rates. Article asks what happened to all those predictions from last year (!) that laptops would be a big deal? 1/
2/ "People don't want to lug a computer with them to the beach or on a train to while away hours they would rather spend reading the sports or business section of the newspaper."
What if the laptop were the place to read those!
3/ "Right now laptops cost considerably more than the equivalent desktop computer."
What if the need to innovate in laptops not only made components smaller, but cheaper, and at the same time consume less energy? Then who wouldn't want a laptop *instead* of a desktop?
The “gyrations” through Apple’s beta cycle since WWDC when it comes to Safari UX are fascinating—some might even say not very “Apple-like”. Changing is fine. But where the design seems to be ending up is suboptimal for a platform, IMO. Here’s why, but not why one might think /1
2/ Moving address bar up/down and/or having a separate address bar are rational design choices that people will (vigorously) debate. I have an opinion too.
What one has come to expect from Apple is a “point of view” in product design expressed through “the way it should work.”
3/ There was a lot of feedback about the early design and implementation. It was rough. That started the feedback loop from developers (those are who use the beta) that it was never going to be right. Developers of all people should know it could change. But momentum gained…