It’s time we face reality, my friends: We’re not rocket scientists

During the development of Windows 95, it was common for team members to pay visits to other teams to touch base and let them know what's been happening on the Windows 95 side of the project.

It was during one of these informal visits that the one of my colleagues reported that he saw that one of the members of the partner team had a Gary Larson cartoon from The Far Side depicting a group of scientists studying a multi-stage rocket ship they just assembled, but the stages are connected all crooked. One of the scientists says, "It's time we face reality, my friends. … We're not exactly rocket scientists."

The comic was "enhanced" a bit by the partner team. They added a sign on the wall of the laboratory that says Windows 95 Development, and the stages of the rocket are alternately labeled 16-bit and 32-bit. The graffiti were clearly poking fun at Windows 95's attempt to straddle the 16-bit and 32-bit worlds.

The Windows 95 team knew how to take a joke, and for a time, they adopted "Hey, we're not rocket scientists" as a catch phrase.

Following up on that article from 2010: Microsoft ran a free seminar on Windows 95 development for Macintosh programmers at the 1995 MacWorld Expo. Upon successful completion, participants received T-shirts with the slogan "Windows 95 sucks less."

Comments (13)
  1. Joshua says:

    What's funny about the tee shirt is W95 did indeed appear to suck less at the time but when I fire up the VMs with the surviving useful or interesting software from the era everything runs better on 3.1.

  2. Not a scientist at all says:

    "On a larger scale from James' legendary exploits, we also have Microsoft's epic product launches. The launch of Windows 95 was a kind of coup de gráce in this regard…"

    Did coup de gráce regard OS/2?

  3. xpclient says:

    My father-in-law is an actual rocket scientist but he finds rocket science less complex than complex pieces of software. :)

  4. Antonio 'Grijan' says:

    @Not a scientist at all: the same text explains that at the same show Windows 3.1 was launched, in April 1992, IBM launched OS/2 2.0, too. That might have been OS/2's coup de gráce :-) .

    Microsoft got its share of bad press for mixing 16 and 32 bit code in Windows 95. IBM did the same in OS/2 2.0 (which was slower than '95, too), and somehow didn't get such criticism. Of course, people settled that when they bought Windows and not OS/2.

  5. Zan Lynx' says:

    Software may have run better on 3.1 but it crashed better on Windows 95. That might sound funny but you know what I mean.

    One software freeze in 3.1 would lock up the whole thing and leave you trying ctrl-alt-delete. Usually you had to reboot. Luckily that didn't take long.

    In 95 a crash would freeze up the program but you could kill it and keep going.

  6. Range Safety Officer (not really) says:

    For some reason this post reminds me of the rocket test launch sequence from the movie The Right Stuff (

  7. Yuhong Bao says:

    On the 16-bit code in OS/2, it reminds me of this:

    To add to what the article says, the original OS/2 2.0 had more 16-bit code than later versions. IBM ported the graphics subsystem for example in 2.1, which was released right around the time when the original Pentium was released.

  8. GWO says:

    "In 95 a crash would freeze up the program but you could kill it and keep going."

    Well, that's what the marketing material would have you believe.  The reality was somewhat different.  Mind you, the marketing campaign also told me I'd never see the egg-timer cursor again… that promise has nearly been delivered on 20 years later.

  9. Azarien says:

    Does the current 32-bit layer in 64-bit Windows suck less than Windows 95's 16/32 bit hybrid?

  10. Jim Lamb says:

    I was working at Symantec at this point and I remember the "Windows 95 sucks less" motto well. We were working to support Windows 95 "from day one" and it was definitely interesting. I remember being curious as to why the shell extensibility interfaces all took ANSI rather than Unicode strings. Little did I know just how long that transition would ultimately take.

  11. Mike Dimmick says:

    @Azarien: There are no 32-bit kernel components in 64-bit Windows. The whole stack is 64-bit drivers up to user mode. In user mode, system processes are all 64-bit. 32-bit user processes cannot load 64-bit DLLs and vice versa. The whole Win32 stack is present as 32-bit libraries, which are the same binaries which ship in the 32-bit except for components which call into kernel mode. Instead, these switch the processor into 64-bit long mode, then call into the WOW64 support libraries (wow64.dll for kernel components, wow64win.dll for calls into win32k.sys, and wow64cpu.dll for CPU emulation).

  12. Yuhong Bao says:

    Also don't forget the Win16Mutex that OS/2 never had, though it did have the SIQ.

Comments are closed.