A) Every patch required a reboot B) There was no centralized control of patching. This didn't come until WUS was released, and now WSUS makes Patch Tuesday irrelevant.
And it's pointless reasoning - because most enterprise customers still have a delay before pushing patches out of WSUS to the end users (at least several I've done work for have). So there's still a 24-168 hour window for a 0-day exploits to infect them, as the customers implement "testing" before pushing out patches.
And I'm will to bet that the vast majority of the Fortune 500 does this - that they aren't pushing out patches Tuesday morning or Tuesday night, but waiting until Wednesday or the weekend or a week later.
Windows Update and WSUS didn't exist back then... they do now, and that argument is mostly moot - enterprises can control patch dissemination at their leisure.
And no, they never fixed them whenever they found them, they fixed whatever they felt like fixing, as does every software vendor.
No, the idiots used to release product improvements in service packs and patches, and THAT caused a problem. They didn't constrain patches to simple fixes.
And WSUS makes their once-a-month policy moot anyway, because it puts upgrade power back in the hands of the site admins, and not WindowsUpdate.
And how do you price something that's one of a kind and unique? I like your thinking on this... Too bad I think we're fairly alone on this, and the majority of our fellow Citizens are happy coughing up their DNA if it'll capture kiddie-touchers and rapists.:-/
I'm with you. Particularly on the freedoms convicted felons lose, like the right to vote. Personally, I feel if you've done your time, spent your parole time cleanly, and clean up your act, you shouldn't have to pay a lifetime's penalty for one mistake or intentional act. Three strikes, maybe, but one offense? I'm not sure how I feel about three strikes either, I wonder how many people are in jail forever on over-trumped charges in three-strikes states?
OS/2 was in an odd position - at the time it was the "better Windows than Windows" but really needed 16-32MB of RAM to run, in an age when 1MB of memory was $40. Had it be 1997 instead of 1994, the world might be a very different place today...
There were plenty of native OS/2 apps floating around, even for this poor student at the time. And gaming wasn't an issue yet.:-)
There's no random selection, and no balance inherent in the system. Simply outspend your competition, and you're statistically more likely to get voted in - simply because you have more name recognition.
Here you introduce random selection via a jury-duty style lottery, and then all the candidates are on the same equal budgetary footing. The rest is the still the same.:-)
No, because you may break your workflow up into desktops, and each open App has different files needed for the task at hand.
I used to create a desktop for each project I was working on, and kept Word, Excel, Browser windows open on each desktop based on the project I was working on. It kept everything conceptually grouped.
A lot of it is because the filesystem bits needed to boot may already be cached. It also doesn't have to probe a lot of hardware, either. I've found Linux boots to be slower in Vmware though, so I'm not entirely sure what's going on.
I find that a little disingenuous. I was just speaking with a woman yesterday who was bitching about not being able to find her skype-compatible cellphone so she can talk to her contacts in Kenya.
I think the fact is that Internet is more of an enabler in emerging markets than we traditionally give it credit for, and applications like Skype, AIM and BT find ready users in this under-served market.
I remember being fingerprinted too, and it was a nice way to interact with the police without it being scary, but we got to take the fingerprint cards home to our parents, the schools/police didn't keep them.
If you're going to be kidnapped, before the age of instant computer retrieval, the best place to put these things is with the parents. Still is, in my opinion.
It is the touchpad, it's got crazy different sensitivity settings. Sometimes is VERY uber responsive, like his typing, and other times it's sluggish, skittish and in general pretty misbehaving. I think it's crap manufacturing, but maybe that's just me. It was a black friday $300 special.:-D
How much of a dork am I, I got a 14.4k internal Hayes modem for Christmas in 1993 to support my burgeoning desire to run a BBS. God, those were the good old days...
Which if Microsoft would ever get it's head of it's ass and properly support fork(), we could do! Seriously? IO completion ports, everything as a handle, and fork()!?!?
Raptor firewalls used to be based on Windows (or run on Windows and Solaris).
Wonder whatever happened to them... Lol
Huh? http://bugzilla.kernel.org/
The whole community was in an uproar because
A) Every patch required a reboot
B) There was no centralized control of patching. This didn't come until WUS was released, and now WSUS makes Patch Tuesday irrelevant.
And it's pointless reasoning - because most enterprise customers still have a delay before pushing patches out of WSUS to the end users (at least several I've done work for have). So there's still a 24-168 hour window for a 0-day exploits to infect them, as the customers implement "testing" before pushing out patches.
And I'm will to bet that the vast majority of the Fortune 500 does this - that they aren't pushing out patches Tuesday morning or Tuesday night, but waiting until Wednesday or the weekend or a week later.
Which buys them: nothing.
Windows Update and WSUS didn't exist back then... they do now, and that argument is mostly moot - enterprises can control patch dissemination at their leisure.
And no, they never fixed them whenever they found them, they fixed whatever they felt like fixing, as does every software vendor.
No, the idiots used to release product improvements in service packs and patches, and THAT caused a problem. They didn't constrain patches to simple fixes.
And WSUS makes their once-a-month policy moot anyway, because it puts upgrade power back in the hands of the site admins, and not WindowsUpdate.
It isn't like there aren't already boatloads of lifeforms on earth that hibernate for months, years or decades... Solved problem. :-)
And how do you price something that's one of a kind and unique? I like your thinking on this... Too bad I think we're fairly alone on this, and the majority of our fellow Citizens are happy coughing up their DNA if it'll capture kiddie-touchers and rapists. :-/
I'm with you. Particularly on the freedoms convicted felons lose, like the right to vote. Personally, I feel if you've done your time, spent your parole time cleanly, and clean up your act, you shouldn't have to pay a lifetime's penalty for one mistake or intentional act. Three strikes, maybe, but one offense? I'm not sure how I feel about three strikes either, I wonder how many people are in jail forever on over-trumped charges in three-strikes states?
OS/2 was in an odd position - at the time it was the "better Windows than Windows" but really needed 16-32MB of RAM to run, in an age when 1MB of memory was $40. Had it be 1997 instead of 1994, the world might be a very different place today...
:-)
There were plenty of native OS/2 apps floating around, even for this poor student at the time. And gaming wasn't an issue yet.
They tried this experiment already, and it failed. The clone makers ate Apple for breakfast.
Yet we "force" people into Jury Duty.
There's no random selection, and no balance inherent in the system. Simply outspend your competition, and you're statistically more likely to get voted in - simply because you have more name recognition.
:-)
Here you introduce random selection via a jury-duty style lottery, and then all the candidates are on the same equal budgetary footing. The rest is the still the same.
No, because you may break your workflow up into desktops, and each open App has different files needed for the task at hand.
I used to create a desktop for each project I was working on, and kept Word, Excel, Browser windows open on each desktop based on the project I was working on. It kept everything conceptually grouped.
VLC, f, c (crop) [repeat until proper aspect ratio]. :-)
A lot of it is because the filesystem bits needed to boot may already be cached. It also doesn't have to probe a lot of hardware, either. I've found Linux boots to be slower in Vmware though, so I'm not entirely sure what's going on.
I find that a little disingenuous. I was just speaking with a woman yesterday who was bitching about not being able to find her skype-compatible cellphone so she can talk to her contacts in Kenya.
I think the fact is that Internet is more of an enabler in emerging markets than we traditionally give it credit for, and applications like Skype, AIM and BT find ready users in this under-served market.
I remember being fingerprinted too, and it was a nice way to interact with the police without it being scary, but we got to take the fingerprint cards home to our parents, the schools/police didn't keep them.
If you're going to be kidnapped, before the age of instant computer retrieval, the best place to put these things is with the parents. Still is, in my opinion.
It is the touchpad, it's got crazy different sensitivity settings. Sometimes is VERY uber responsive, like his typing, and other times it's sluggish, skittish and in general pretty misbehaving. I think it's crap manufacturing, but maybe that's just me. It was a black friday $300 special. :-D
How much of a dork am I, I got a 14.4k internal Hayes modem for Christmas in 1993 to support my burgeoning desire to run a BBS. God, those were the good old days...
Yeah, the Compaq C712NR does the same thing... maybe you got the HP version of my laptop? It's frakkin' annoying.
What did you use for clustered storage, if I may ask?
Aside from WiFi, the networking world is really down to three major ethernet chipsets, Broadcom, realtek and intel.
A far far cry from the 3Com, DEC, Intel, Mylex and more...
Which if Microsoft would ever get it's head of it's ass and properly support fork(), we could do! Seriously? IO completion ports, everything as a handle, and fork()!?!?
we've already partially eliminated this with versioned DLLs ( .so's )
/lib/liblib.so == /lib/liblib.so.4 /lib/liblib.so == /lib/liblib.so.5
Now if there were an easy way to say that for program X:
and for program Y:
We'd be all set.