I’ve got kinda sidetracked here waffling on about the history of dial-up BBS systems. What I really want to be concerned with is what made The Microsoft Network unique and interesting: the interface.
The big thing in Windows 95 was their new shell. They wanted everything to go through this. They had this vision of every object in the computer being represented as a shell object, so there would be a seamless intermix between files, documents, system components, you name it. They had this project called Cairo that was supposed to throw out that scruffy old file-based filesystem and bring in a shiny new Object Based File System instead. It never happened, so we’ll never know exactly how it might have turned out. But the brave lads at MS didn’t give up that easily and so the idea stayed on, admittedly without the tech to back it up, and the principles wormed their way into such glorious developments as The Microsoft Network.
And so The Microsoft Network wasn’t a program you loaded like CompuServe. It was part of the OS, with folder icons that looked just like real folders. It was a kind of version of the Web where you could browse online data the same way you browsed your file system. This is what made it cool.
I vaguely remember something about this project – and of course, the whole concept of integrating literally everything into the Windows 95 shell did see some adoption here and there in the early days of Windows 95. Looking at it from today’s perspective, I still kind of like it – Microsoft would try the concept again with Windows Phone 7, where services like Twitter and Facebook were supposed to be integrated into the operating system, without having to use crappy applications to access them.
That is still a good idea today.
This is basically Plan 9 plus a web-scale filesystem namespace (IPFS/IPNS?). The only issue would then be showing all those files, virtual and otherwise, graphically, but we’ve got decades of experience in creating graphical file browsers. You could even wrap the whole thing in a 3D fsn/fsv alike interface if you wanted to spend your day shouting “It’s a Unix system, I know this!” instead of tab-completing.
Windows is a remarkably poor choice for a basis for something like this, for exactly the same reason that PowerShell couldn’t simply be a Unix shell.
Edited 2018-08-29 22:31 UTC
That was the best part of that film. FINALLY someone on the island actually knew something!
Thankfully, structured data and Lisp machine like experience is so much better.
But not really popular.
For UNIX refugees on Windows.
All I remember about it was how on any computer I had to set-up, there was a need to remove this icon, which uselessly cluttered the desktop, along with a few other items.
Integrating Twitter and FB with the OS is just so, so wrong.
Most of the world don’t use those dens of iniquity so why force these things on them? Any OS that has this sort of feature would NEVER be used by me.
Operating Systems should be lean, mean and get out of the way of what is being done at the user level.
Back in the day, you could build a working RT-11 Kernel that was 3Kb in size. It never got in the way of what you were trying to do. {shows my age…}
Yes, this indeed shows your age. Computers were toys for experts that could barely get the simplest things done after spending 5 years of getting to know their tooling in great depth. If you put a regular person behind that 3 Kb kernel he would just walk away because he couldn’t do anything. Nowadays computers can do incredible things and regular people can operate them without much thinking
Facebook has peaked and is declining in use and importance, unless you want to later piss off remaining users by killing the OS support, you would end up having to support 500 different services and try to keep up with changes made to them.
MySpace is a great example of why it would be stupid to base your OS around popular services, it went from being all the rage to being irrelevant very quickly.
Facebooks user count might still be high, but googling it, i don’t seem to be the only one who find that overall activity and items shared/written has gone down a lot over the years. Other services are challenged as well, like LinkedIn used to be a good place for work related content, but has been totally overrun by people who think it is a sales channel.
The base services of a general purpose OS seems to have been rather stable for a long time, while the online service landscape still seems to be far from stable. Who knows, in a couple of years people might start to realize that most of it is just a waste of time and find something else to do, while most likely you still need to do office work, edit videos, do programming, etc, on a computer.
I already wrote why I think integrating social services into the OS shouldn’t happen, but let’s stay factual:
You seem to think that a general purpose OS runs on a computer (meaning PC). However more people are now using phones for their general purposes and they spend an awful lot of time in the “online service landscape” on FaceBooking, Youtubing, Twittering, Whatapping, Instagramming, et.
Which is part of the problem with malware.
Viruses/worms were quite (maybe more) common also in the olden days…
Malware is a much broader categorie than just viruses/worms which have almost disappeared because of appstores and built-in virusscanners and other OS-level-mechanisms.
Aye, that’s why I specifically didn’t use the term “malware” but the subcategories “viruses/worms” (which were all the rage back then – the point beeing that hard to use OSes didn’t stop it…)
Which of course by todays standards wasn’t much. Systems are much more complex now, much of it unnecessary and due to progressive development but still there.
IIRC the QNX4 microkernel was about 7-8KiB* in size however with other parts required externally. The QNX Neutrino (version 5+) started at 32KiB but included the external parts plus being more POSIX compatible.
(* Why couldn’t they standardize kiB instead of KiB? It’s a pain to remember)
I would rather say The Microsoft Network was the attempt by Microsoft to kill the up and coming internet and replace it by a proprietary network of theirs – with paid membership and all that later on…
Up to that point they owned the whole desktop and controlled whatever people used. Every time some new killer app started to rise they would squash it with their own solution. Sucessfully.
But the internet slipped through and they panicked. So they tried their usual approach. But luckily for us it did not work out for them.
+1
I shall add that OS/2’s workplace shell was much more advanced in its “object-orientated” integration
From [those days](http://www.legaline.com/column7.htm):
Something that fixed thread priorities, allowed 80bit compilation, and aimed for low jitter.
It’s great they didn’t succeed.
The incredible success of the web is exactly because it’s free form, not a stringent object definition with transitive dependencies, spurring innovation never seen before.
Free form admittedly have some downsides, but the web wouldn’t explode without it.
Funny how people kept trying with CORBA and SOAP.
Microservices
Well, “microservices” are supposed to be stateless, glorified HTTP requests, but of course needs a sexier name than what was already being done.
Integrating social media services into an OS is simply idiotic. Not only are they trends, they have absolutely nothing to do with being the middle-man between the user and the hardware. It’s bad enough that you can’t delete social media apps from cellphones these days without rooting the device. Crapware should be a choice, not shoved down users throats, and certainly not baked in.
I found out a while back that Atari had one running on tons of different 8bits back in the day, that eventually was upgraded to STs. https://www.atarimagazines.com/v5n10/ataribulletinboard.html
Also check out this coolness!
https://atariage.com/forums/topic/196354-ataris-plato-cartridge-ques…
If you’re interested, I’ve got what is pretty much the “Insider” story on MSN…since I was on the Forum & Content Management Team that helped put it together.
More to follow..
Give us more!