According to http://cbmfiles.com/geos/geos-3.php the first disk came with

[....]the basic boot loader program, the kernal file which contains the actual GEOS operating system, the deskTop program which is the graphical user interface, and the Configure program which is used to set up your disk drives and ram disks. [...]

All that in 57k ... (well, it might be compressed off course, not sure how Wraptor works)

Amazing how much was actually included in the suite : http://en.wikipedia.org/wiki/GEOS_(8-bit_operating_system)
I remember blowing everybody's socks of @ school with the things you could produce using GeoPublish.... (with a /bit/ of patience =)

And yes, it was rather intense on the floppy drive as it used it for swapping. At the time I actually had 2 floppy drives, a C1541 and a DEC-ALPHA, the latter was slightly faster but 100 times as noisy =)

Good ol' times...

GEOS would boot from a 170k floppy on my C64.
It didn't have a web browser though.

Call it pre-emptive discussing, sooner or later it was going to end into the discussion anyway =)

As for 'making it easy for the user to be tricked': I wonder how hard it would be to convince people to install GreatestScreenSaverEver.rpm given a nice screenshot and some flattering text .. heck simply stating that it comes with a free would probably be sufficient.
If people just click OK or YES on every request they see then what can the OS do about it ? (assuming the user fully owns & manages the machine. Friend of mine set up his dads pc so he can only use what's there and not install anything new. Refuses to give the admin-account password. If his dad wants something installed, he has to call and ask to install program x. The good thing here is that the machine has been trouble-free ever since. the downside is that it takes some time every once in a while when he needs to log in remotely and scrutinize and install the software.; then again, probably less than when he has to re-image the machine every 6 months).

I don't want to come of as a windows-worshipper, far from it. But I agree that I like quite a bit of things about the OS and it gets on my nerves that people somehow always need to bash on it because their OS of choice is soooo much better. When asked what's better they then always come with the 'bible-list' of 'virusses, not open, how you'll have blue-screens every hour or so, how you need to reboot after changing tcp-ip settings, how you can only tweak things by changing things in the registry, etc, etc... none of which applies when using the OS with a bit of thought. (IMHO)

Back in the days I had an Amiga and it was fun to discuss with people who had wintels about the merits of the different systems. Usually the discussion would start with "but it's not compatible (Tm) !!' and end with, "multitasking on 3.11 sucks, wished I had those colors & my GUS sounds much better so if only I could load this driver high than we could try play game X with the mouse too, I hear DOS 6 will come with a better memmaker!'. Over time things changed and I had to agree that Win95 was actually rather nice looking but it took until Win2000 before I made the switch (Commodore was long dead by then). In each case, what I wanted to say was that in order to have a proper discussion with the PC-crowd I had to learn quite a bit of what that system looked like, operated like etc... and had the 'decency' to stay with the times as to not insult those I was talking to claiming they were still on DOS 5.0 with crappy CGA graphics and creative beeping.
Funnily, 'knowing the enemy' actually landed me a job in a Wintel-centric software company [although I spent the first years there without having a PC at home, 'just' my old trusty Amiga =)].

Most of what you say is true but
* servers may be available in vast numbers, they're hardly ever run/maintained by mom I'm betting most MS-servers live a virus-free life too.
* malware writers are used to writing malware for windows. Why throw away all that knowledge and experience and start over from scratch on another platform ?
* there IS plenty of malware out there for Andriod... you'll probably prefer to call them trojans and dismiss them as 'not real virusses' because they simply use the user as the attack vector (swipe here for free ponies!) and don't really do that much harm except for stealing user info etc... but hey, let's rather bring up SASSER again, after all it's only 9 years ago windows got hit by that one!

I somehow don't get why people always need come up with the 'you need to run as administrator on windows' argument. You don't need to, maybe a lot of people do, but you don't need to. Ever since the introduction of the UAC I see little difference between me running Win7 or me running Ubuntu; except that the UAC is quite a bit more user-friendly. In fact, after having to type my password for the umptieth time so I can change an ini file hoping to finally get my display at full resolution (old laptop) I started running Linux in 'administrator' mode.

Get with the times and stop bashing Windows with old arguments. I don't like Microsoft because of the way they run their business, but that doesn't make (all) their products bad. Let people simply use what they like and quit these 'religious arguments' about what OS is superior. If the OS lets you run all the software you /need/, then what is the problem ? You like Ubuntu & Gimp, fine. She likes OSX & PaintShop, fine. I like Windows and Paint, so leave me alone.

(And don't come whining about how for every program on Windows there is a viable alternative on Linux. There isn't, and besides, for every program on Linux there is a viable alternative on Windows too and yes, often it's free too... although it might not be 'open', but that's true on Android too and I not heard people asking for the source of Angry Birds yet.)

Good question, was the Kernel of Ubuntu released in August 2001 built with or without those options ?
(oh wait...)

I really don't get you guys. Throwing around expensive terms and stuff but failing to do some basic reading upfront.
I'll help you : I'm assuming that by 'heap randomization' you're referring to http://en.wikipedia.org/wiki/Address_space_layout_randomization
You might want to read up on http://en.wikipedia.org/wiki/Address_space_layout_randomization#Linux and http://en.wikipedia.org/wiki/Address_space_layout_randomization#Microsoft_Windows

I'm not sure why you'd want to bring SMP into this discussion... (unless you mean there is somehting like SMP randomization which in that case I've no knowledge of)

Then do both, like I do... doubly-so.

* I use the search function to start quite a lot of things
but
* I also use the 10-ish icons on the start-menu that represent the things I've worked with lately/often (not sure what the logic is, it works). Added bonus is that (a lot of them) are able to show a sub-menu with the most recent documents I worked with for that given program; I even can pin those if I want. Genius!
* Old-school as I am I also revived the Quick-Launch toolbar and have like 20 icons of things that I work with most often.

Yep, there is some overlap between 2 & 3, but as far as usability goes I am quite happy with this setup.

PS: I don't like pinning stuff to my system-bar for some reason. Tried it, annoyed me and haven't done it ever since. Everyone's different I guess.

Off course said theory isn't busted. It's not like http://en.wikipedia.org/wiki/Linux_malware doesn't exist, it's simply hardly worth the effort to focus on Linux.

On top of that, the typical/average Linux-user is much better informed about security issues than is the typical/average Windows-user (**). So making a website to trick people into clicking/downloading/running something malicious is more likely to work in the latter case.
Switch mom&pop to Linux and before you know it they'll be clicking the same links and the botnets will live happy ever after.

Linux might never see the sheer volume of malware that exists for Windows because it's "late in the game" and because simply put both the developers and the users have learnt quite a bit over the years making it harder for viruses etc to propagate.

(**: Apple used to be 'virus-free' too. As its user base is growing (and dare I say dumbing down?) there is an uptake on the amount of malware too...)

I know, hence my 'for whole numbers of bedbugs' remark ... the egg-laying makes it even more challenging indeed ... wikipedia here I come =)

Who's to say they don't ?

Don't get me wrong, I do like a lot of the things Google did and does; not all of it, but quite a lot. Same goes for MS & Apple btw. But assuming either of these 3 would do something that hurts their own chart of accounts is simply fooling himself. I'm sure all three companies have plenty of talent around that would love to break/build/tweak and bring out stuff that might be world-changing; but in the end it's the bean-counters that decide what needs to be focused on.

Mathematically speaking I would think that it's impossible that all bugs will get caught eventually, no ?

Night 0 : 100 bedbugs run around

Night 1 : 100 * .77 bugs get caught, 100 * .23 remain
Night 2 : (100 * .23 ) * .77 bugs get caught, (100 * .23) *.23 remain ...
Night n : 100 * (.23 ^ n) bugs remain...

So you'll get an asymptote that borders on catching them all, but not ever really... Especially as we're not taking into account that the remaining bugs will probably multiply...

But I agree that for 'whole numbers of bedbugs' n should be smallish... might make a nice spreadsheet/graph to figure out, especially if you add variables like how long it takes for them to reproduce etc ...

So....
* if Google publishes something 'decent but much better on android' then it's Apple's fault.
* if MS publishes something decent but much better on WP8' then it's MS's fault.

What's with all the MS hate here anyway... if you don't like it don't buy it and just walk on. Do people get some kind of ego-boost out of bad-mouthing every single MS-product or decision ? Sjeezsss...

Actually, there is no spoon.

Actually, TCP is all about dropping packets; it's a major part of how it optimises throughput. Software on top of the network layer doesn't get told if (or how many) packets are dropped; it simply reports when it's ready for more.

I hadn't noticed it to be honest -shame on me- but off course you're right.

Then again I can see how marketing people wouldn't be able to sell 7/3 as being an improvement over 16/9.. hell, all the numbers went down, considerably! How can that be good ??

And sadly, 80% of people around (and I honestly think I'm being generous) looking for a new screen would think the same.

I've found books to be a great alternative...