So, we know these four things, but what was Cairo? Was it going to be the next major version of Windows NT? Was it a testbed for things to be used in the future?
If you do a little bit of looking, you can find some references stating that Cairo was going to ship with a new UI, which later got recycled into Windows 95.
http://www.sigchi.org/chi96/proceedings ... ds_txt.htmExploration Phase
In this first phase we experimented with design directions and gathered initial user data. We began with a solid foundation for the visual design of the user interface by leveraging work done by the "Cairo" team. We inherited from them much of the fundamental UI and interaction design (the desktop, the "Tray", context menus, three-dimensional look and feel, etc.). We also collected data from product support about users' top twenty problems with Windows 3.1.
Figure 2 shows a prototype Windows 95 desktop design that we usability tested in January 1993. This design was based on Cairo and incorporated a first pass at fixing some of the known problems with Windows 3.1 (window management in particular).
http://technet.microsoft.com/en-us/libr ... 50186.aspxUser Interface
User interface differences between Windows 95 and Windows NT Workstation are short-term because the Windows 95 interface is based on early designs for Cairo, the next major release of Windows NT.
[...]
The Windows NT Workstation interface is based on the Windows 3.1 interface and probably won't get an updated shell until Cairo ships.
If you check out the Comes v. Microsoft anti-trust documents, specifically PX05542.pdf, there is an internal document published by Microsoft in 1993 about some of the goals to be met in Cairo - one of them is to create a new user interface. If you want to read it for yourself, check out PDF Page 21 (the printed page number says Page 18) of that document.
Finally, we presume Windows NT 3.5 Build 854 is a build of Cairo, since it contains ofs.sys. If we assume this to be true, then we see that there really was a new shell in store for Cairo, although I believe by that point it was referred to as the "Chicago Shell".
So, this shows that Cairo had a new UI that was being developed for it. But was Cairo just that? I think the end goal of Cairo was to be the next major version of Windows NT. Look back at that Technet article:
Right off, we see that maybe Cairo was meant to be the next major version of Windows NT. I didn't look too close at that antitrust document, but it seems to show there that Cairo was also going to be an OS.The Windows NT Workstation interface is based on the Windows 3.1 interface and probably won't get an updated shell until Cairo ships.
Thanks to Ken (you old-time beta collectors know who I'm talking about), I've gotten some screenshots of MSDN/Technet news from around 1993 which state that Cairo was going to be an OS:
The Technet article, the antitrust document and these screenshots all point to the fact that Cairo was going to be its own OS. If we assume that Windows NT 3.5 Build 854 is really a build of Cairo and not just a post-RTM of NT 3.5 with ofs.sys and the "Chicago Shell" in it, then this seems to further prove that Cairo was its own OS.
Finally, ofs.sys. We can confirm for sure that whatever Cairo was, it did have an object-oriented filesystem which later became WinFS. Cairo OFS is mentioned here in this ComputerWorld article, which describes a little bit of why Microsoft was developing it: http://www.computerworld.com/s/article/ ... d_to_Cairo
So, what was Cairo? Was it really just "a collection of abstract design goals"? From seeing all of the above, I personally believe that Cairo was going to be the next major version of Windows NT.
But what happened to Cairo? Whatever UI they designed later got recycled into Windows 95 and Cairo OFS turned into WinFS, which later failed. From what I've read, Cairo started in 1991 and was canceled in 1993.
Perhaps Longhorn was a shadow of Cairo... Trying to pull off too many things at once with too-new tools and inexperienced developers.