No they wouldn't. Reliability is critical to a working system. The problem with bleeding edge over older patched systems is that the former CAN be unreliable. It takes time to find and patch the flaws.
Printable View
No they wouldn't. Reliability is critical to a working system. The problem with bleeding edge over older patched systems is that the former CAN be unreliable. It takes time to find and patch the flaws.
patches are more likely to break older systems, especially ones with dozens of patches already, then you are to find a new system that doesn't work out of the box.Quote:
Originally Posted by RandBlade
Microsoft has done that a few times, corrupting older XP installs because patch 24542 didn't play well with the way patch 25 was installed several years prior.
It depends upon the system. Plus MS isn't concentrating solely on XP.
For my previous company the IT system being down for 10 minutes could cause hundreds of pounds of damage - tens of thousands probably nationally. When you want a system that's flawless then you're in no hurry to rush a change. And those that changed first took the hits.
no system from beginning to end can concentrate on one OS.Quote:
Originally Posted by RandBlade
testing is always important and even then sometimes things happen. Didn't Visa go down earlier this year cause of a bad patch?Quote:
For my previous company the IT system being down for 10 minutes could cause hundreds of pounds of damage - tens of thousands probably nationally. When you want a system that's flawless then you're in no hurry to rush a change. And those that changed first took the hits.
but there is a huge difference between rushing into the latest and greatest and using that as an excuse to stick to something like DOS when its 2012.
But both work. And what do you mean exactly with "reliability"? There are several different viewpoints to that. Reliability for the end user? For the sys admins? For the CEOs/CIOs? The list goes on and can even be sub-divided for each and every group. Is it data consistency? Backups? Speed of requests? Amount of features? Speed of the UI? ...Quote:
Originally Posted by RandBlade
You're just waking up to the fact that your statement was over-simplistic and a nonsensical talking point without any true meaning. That's why you have to backpedal so furiously right now.
By the way: Who exactly was speaking of "bleeding edge"? Oh, wait, you were.
I'm not backpedalling.
But you know what, I give up.
There is absolutely zero reason to want to keep a tried and stress tested system.
Reliability is not part of measuring whether a system works as designed or not.
Nobody would reasonably consider downtime to be not working properly.
The sole reason to keep a working system maintained rather than change is stubbornness.
FYI if you're using in-house software then any new system will by definition be bleeding edge.
EDIT: I guess I'm just totally insane to be considering whether a system is 100% reliable to be part of whether a system works or not. A system not working when you look at it sideways still is a working system according to Khen so I must just be bonkers.
As another illustrative example I might add that every single update to our county's much-maligned and ancient electronic health record system potentially puts lives at risk :o
A more serious concern than "just" losing hundreds of pounds for every minute you're offline and an excellent point well worth making Aimless :up:
To be fair there are a number of problems that put lives at risk if you DON'T update, so you're damned if you do, damned if you don't, even though the enemy you know is better than the one you don't. The risk may be worth taking if the upgrade is intended to make the system more resilient and less likely to fail catastrophically, or promises great long-term productivity gains. It just has to be rolled out in a decently planned way, with support from and preparation of users. We have contingency plans for when the systems get borked. I imagine it's actually a little harder for businesses to make these ultimately worthwhile leaps of faith than it is for us.Quote:
Originally Posted by RandBlade
Absolutely I agree.
Our system was so robust that I would put the software at virtually 100% reliable under the most stressful of circumstances - and reliability under stress is one of the primary measures of whether a system works or not. Virtually every failure we ever had (and they could do some major financial damage even though we're not talking safety/loss of life) was due to hardware-issues or issues outside our control (internet being lost due to a gas leak down the road being the most memorable in nearly 8 years with the company). The other advantage of having an older tried-and-tested method is that when faults occur the contingency plans are in place and people know instinctively what to do to put it right.
So stressful dealing with a "new" crisis.
Right, Rand. Let me give you another example: You have a reporting system that works, however, it will take bloody ages to generate its report which is why its done as a batch job overnight.
New algorithms, new databases, new hardware comes around and with an upgrade you now might obtain your reports in near-realtime.
But, since the old system works, you forgo the upgrade.
That's good business sense right there, hmmh? Geeze, even the aerospace industry is less conservative than you.
You obviously don't know what "bleeding edge" is. Anything newer than what you're using is "bleeding edge"? What idiotic definition is that supposed to be? Good lord, let someone like you never be in charge of IT.Quote:
Originally Posted by RandBlade
I can only imagine it: You're upgrading from old 286 with Windows 3.11 to Pentium 2 with Windows 98 - and you'll call that "bleeding edge".
That'a actually not a point for your argument and you know it.Quote:
Originally Posted by RandBlade
Wait, I thought your system was so reliable that it never failed?Quote:
Originally Posted by RandBlade
Not to mention that I'd be somewhat suspicious of the reliability of a system if someone told me that "people know instinctively what to do". Because that means to me that faults happen so often that people have it fully memorized. Or that the faults are so trivial that a trained monkey could do the right thing.
I think he was specifically talking about software developed in-house :o
Same can be said for software. "Bleeding edge" means "partially untested and under heavy development".Quote:
Originally Posted by Aimless
If you're using business software in production that falls under that definition then you're an idiot. That's what stuff like testing is for (unit tests, functional tests, ...) so that, by the time you deploy from development/testing to production, you have software which behaves the way you want it to. Of course, if your in-house software team can't code its way out of a wet paper napkin, then you indeed have a problem.
I really want to jump in here, but I'm no longer clear on what the argument is. Khen, can you spell out your conclusion? I don't think you're saying "always upgrade", right? Rand doesn't seem to be saying "never upgrade" either. There are pro's and con's to upgrading a businesses software, and they need to be weighed when the decision is made. The resulting decision is also very dependent on the business itself, and can vary.
Quote:
Originally Posted by Wraith
His statement was an absolute and as such, absolutely wrong. He obviously won't admit that.Quote:
Originally Posted by Khendraja'aro
It may originally have been absolute, but I think Randblade has accepted that there are a number of drawbacks to failing to upgrade and anyone who doesn't have their head totally up their ass understands that businesses should not leap in and upgrade the second a new shiny toy becomes available. The decision to upgrade is essentially a trade off between competing needs, like Wraith said.
So where actually is the disagreement?
I never said an absolute Khen. Everyone else seems to have decent reading comprehension and able to understand the basic message. Don't rush into flawed upgrades, have a reason to make a change, make sure it's done sensibly ... All basic common business sense.
Windows 8 for $15.
The registration form takes just about any information. So back up your computer and give it a shot.
Hmm.. I paid 30 euros and still think it' s worth it. Browsing is significantly faster.
I bought it. The assistant didn't have any problems with my installed programs and games, only gave me one warning about Intel's USB 3.0 controller. I'll see how big of a problem that will be in about 30 minutes.
It will revert to 2.0 functionality without the correct drivers. And that's about it.
Well that was a god damn bitch. It failed to install twice from my Windows 7 install. So I reinstalled windows 7, reinstalled my wireless drivers, then attempted to download windows 8. That finally installed. Then I discovered that Windows 8 has even stricter driver enforcement, no more warnings about unsigned driver. It straight up denies you the ability to install them. You have to restart 8 into a special "ignore driver signature" mode to install most older or custom drivers. Of course I had use google on my netbook to figure that out.
I think I'm liking what I see so far however. I'm getting a lot more use out of my right mouse button it seems. All the menus I want, from the device manager to simply opening up a flash drive, seem to require right clicking in the bottom left. Found that by accident. With such a radical change in how an OS operates, they did a seriously shitty job at bundling in some easy to find help. But I guess that's job security for me.
Now to figure out how to close programs without having to return to the new desktop first, then figure out how to turn this bad boy off.
I've only installed it on a work machine, so I'm not sure what you mean: what's the problem with closing programs? You're not talking about windows store apps, are you?
I haven't installed anything on it yet, but everything I've opened so far seems to sit in the background when I reaccess the new metro desktop via the bottom left. You can see these windows and programs by tracing with your mouse from the top left down that edge. I don't see a close button, something that would do the same as the X in 3.1-windows 7 while in the program (such as IE right now). As of now the only way I see to close programs is to enter the metro desktop by hovering the bottom left, clicking on it when it pops up, then tracing from the top left down, then right clicking on one of the open programs, then selecting close.Quote:
Originally Posted by Wraith
I'm sure there has to be a more straightforward process, but I haven't found it yet. My easiest way to get around is heavy use of the right click, but for an OS that's built around touch, I don't see that being the "right" way.
Why exactly do you need to close them? For what reason?
They'll be closed automatically if you actually need the tiny bit of RAM they occupy.
alt+f4?
^ What he said, which was why I was asking. You're supposed to actually need to close Metro/London/Whatever-its-called-now apps anymore.Quote:
Originally Posted by Khendraja'aro
Of course it doesn't work well with iTunes.Quote:
Originally Posted by Wraith
I advise people planning to upgrade who have iTunes on their machine to act as follows.
Before the upgrade;
1. remove everything Apple from your PC. (also withdraw authorisation)
2. install windows 8
3. download iTunes
4. run it as administrator
5. when the finish screen shows do NOT click finish
6. attach your iPhone and wait untill your PC has recognized your iPhone
7. click finish for the iTunes installation
... iTunes will recognise your iPhone and will synchronise it. If you're Lucky it will also automatically pick up all your audio and other files like in the previous setting.
Well, that's probably not so much due to Windows but more a thing of Apple writing pretty shitty Windows applications.
During an earlier install of iTunes (when I still had a gifted iPod), the installer asked me three times for admin rights. And the app itself is slow as molasses.
Why not?Quote:
Originally Posted by Khendraja'aro
Security, efficiency, because you're finished ... I can think of many reasons.
It's not like it's hard to do so on other OS's (Windows/iOS not sure re:Linux). On iOS you double press home, hold icon, click x. Click x on all now if you want to close all.
Not as easy as Windows of course where you just press the x in the top right corner, but that's because touch OSes are normally stripped down compared to the open functionality of PCs.
Of those, only security is at all valid, and if you're really relying on an application being closed to provide security...er, well, don't do that.
Store apps only have 3 seconds of system resources from the time you switch out of them until they're put on ice. If you don't switch back within two minutes of switching out, they can be closed at any point by the OS. The system's designed so that you don't actually need to close store apps, ever.
Because I'm sure if I didn't emphasize this someone would get the wrong idea, the above applies only to store apps, and not to legacy desktop applications.
True, this is due to Apple's crappy/evil nature. Just thought the above might be of use to people who use apple portables in combination with a PC.Quote:
Originally Posted by Khendraja'aro
"Security"? If you have a program running which has the admin rights to read other program's data in RAM (because programs without elevated rights can't access memory outside what Windows gave them - they'll crash if they try!), then you've got a whole lot of other security problems.Quote:
Originally Posted by RandBlade
Efficiency? Flushing a program out of RAM is way faster than restarting the program from scratch.
That's actually why Windows 7 preloads applications you start often into RAM as well. You just don't notice that.
Not that type of security, but if I share a PC with others in the office and I'm working on a document I close the document before I walk away ...
Wow. That's your "security"? You close the document?Quote:
Originally Posted by RandBlade
Well, then I have a brute-force hack you should consider (and it's a really nifty one!):
Open Word, click on the top-most entry in "Recent Documents" and... done. Haxx0r3d your security in about 5 seconds.
Again not that type of security, though perhaps it was the wrong word. Try privacy.
How many million program's do you have running when you Alt+Tab or has that bit of functionality bitten the dust?
Privacy? Right. Again, I can see within mere seconds what you last did on a particular computer. Regardless of whether you closed your programs or not.
You either log out / shut down or you lock the screen. And that functionality still exists.
And "millions"? Hyperbole much? Again?
I don't run Windows 8, but usually programs like Word have the facility to close open documents without closing the program itself.
Any programs that you can run on your Windows 7 or earlier machine will work the same way on Windows 8. You can close Word documents, Word itself, etc. and not be able to alt+tab into them any more. Nothing's changed here.
The change is in new Metro apps. They don't need to be explicitly closed anymore. They'll close when they're supposed to without you. Ideally, for most apps, there isn't supposed to be much of a difference between cold booting them and switching into them while they're running anyways. You just need to get used to the idea that closing metro apps is something you "have" to do; you don't anymore. They won't eat your system resources while they're not the focus, and you won't gain any 'security' advantage by closing them.
Wraith, is there are difference between a Metro app and a regular app other than the underlying architecture? What's the deal with them?