It also encourages people to be able to speak freely without fear of persecution. Without anonymity it would be impossible for whistleblowers to out evil empire corps without losing their jobs and probably never being hired again.
I agree completely. Anonymity is a cornerstone of free society.
Without anonymity we wouldn't have vitriolic bloggers; we wouldn't have this fantastic forum of discourse where we can speak our minds and not worry about being smacked with a lawsuit (well, not including the video professor).
Of course, that doesn't mean we should encourage people writing inflammatory bullshit just for the sake of it, because they're trolls and know they can hide their bias behind the veil of anonymity.
Or, in other words, anonymity definitely has its uses, but that doesn't imply it should be encouraged.
not going to happen dude. perhaps if you had mentioned more trains, more mass transport, ok. but you are asking people to exert a lot of effort. you realize that, right? you want some office worker to bike 10 miles a day? really? are you serious?
While it's certainly not going to happen, it is difficult to conceive of any genuinely negative results should the average office worker start riding 10 miles to work.
One of the most onerous problems facing users of Windows are the 2GB per process and 4 GB total limits.(often only 3.5GB or so actually usable) This seems to be a corporate decision by Microsoft rater than an actual limitation of the hardware, as all of the modern processors will allow 38 bit addressing.
It's an engineering decision. Lots of drivers for cheap and nasty hardware break when PAE is enabled.
PAE can be enabled in server versions of Windows and all versions before XP. However, if you've got 4GB+ of RAM, you're better off using x64.
I believe the answer is due to poor OS design. As far as I know, linux will much more effectively use nearly all of your RAM before it tries to swap anything out to disk. Not so with windows.
Windows will "pre-emptively" swap pages to disk so that memory can be freed up more quickly (ie: without having to swap), if necessary.
Whether or not that is "bad design" depends on your objectives...
whilst many people would be capable of making the same deductions, many people are not.
I think the word you're after here is 'paranoia', not "deductions".
I don't see GitTorrent offering any meaningful advantage. In any managed software project, the "choke point" is the person or person who decides what goes into a given relaese, not anything technical.
Because the question asked was what systemic improvements regarding systems management were introduced in Vista, not what good (or bad) experiences have people had using it.
By all means, cherry pick points and ignore those which don't reflect favourably on your point of view.
I'm ignoring the points which are irrelevant, or grossly exaggerated by your obvious bias.
I am yet to see a "Microsoft" deployment plan that gives me half as many options as Arconis.
When was the last time you looked ?
I use it as a straight up imaging tool for identical laptops. 10 minutes and there is a new XP image ready to be put on the domain and into service (2.5 GB from a DVD). Vista's massive space requirements make this difficult..
8G for a default install. Heady stuff, indeed, when you can barely even find new machines with drives smaller than 160G.
The Vista installed DVD has about 2.5G of data on it. That should give you some idea as to why you should be looking at recommended deployment methods, rather than ones you have come up with in the past.
Yes I could move images to a network or USB drive but still, this tends to make it slower or less convenient then DVD plus increases my need for storage and reliable backups where as a DVD can be copied cheaply and easily as many times as needed.
If re-imaging from a network source is slower for you than a DVD, you have bigger problems to deal with.
The machines and AD are configured correctly. UAC is at best security theatre.
These two statements are in conflict. UAC provides exactly the functionality, for the same reasons, at the same times and in basically the same way, as other temporary privilege escalation functions on other OSes (eg: sudo). (Although it may well be that you think sudo is "security theatre" as well, but I doubt it.)
As I said (and I do hate repeating myself) users have already learned to instinctively click "allow" and not even bother reading what UAC has to say.
If your machines are in a Domain, users can't just "click allow" to dismiss a UAC prompt. They need to provide Administrator (or some other) credentials. Unless, of course, you've deliberately configured their systems otherwise.
Then they come and bother me as to why they need to do this 10 times an hour.
If it's happening that often (which is seriously doubt) then you should fix the root cause (which would almost certainly be some program trying to write data to system areas).
Its not that the irreplaceable data isn't backed up, its that they will need to waste valuable work time re-copying it, also the data users consider replaceable will not be moved to servers. You also fail to account for the "road warriors" that need to keep data on local storage because servers are not always available. Beyond this getting users to keep all data on the servers is like herding cats especially when you start dealing with multi-gigabyte data sets.
Actually, it's pretty easy. Firstly, you have the appropriate managerial body institute a policy that all data must be located on managed, centralised servers (with appropriate exceptions for laptop users, along with synchronisation facilities). Then you tell them that any data that isn't on a server is 100% their responsibility, and you won't even attempt to recover it if something goes wrong.
A professional what?
A professional systems administrator. You know, what this whole discussion is about.
Any good Sysadmin would know that you must always consider the end user [...]
From what you've said so far, you don't put much consideration into helping your end users.
So I say again, where is the improvement if its forced me to go back to the tool I already had.
Because, apparently, it makes better attempts to fix itself and allows the end user to also do some basic
Technically correct, but any malware that runs purely with user privileges is going to be pretty tame [...]
Just what is it you think the average piece of malware might want to do, that it can't do with regular user-level privileges ?
[...] - especially as it should be DEAD easy to find and kill (admittedly therefore some kind of "antivirus" software would actually be needed, but nowhere near the complex scale of current things)
The rather large assumption here is that the average user would even notice that something was wrong, let alone start looking for a cause.
Anecdotes are interesting, but ultimately irrelevant. The OP wanted to know what systemic improvements were available with Vista, not what good or bad experiences people have had.
Still not equal to Arconis. Only good if you do things the "Microsoft" way and the "Microsoft" way only.
As opposed to "the Arconis way" ?
Great way to make my users angry and frustrated. It's security theatre at best, all our current Vista users already igrore it and click allow by default.
Then maybe you should configure their machines properly. An Active Directory domain would probably be a good starting point.
This is not needed in an enterprise/SMB.
Having seen it successfully and productively used in same, I have to disagree.
I beg to differ, in fact I'd say it was a massive leap backwards (see my comments above). Microsoft assumes you are running a single large drive for Vista (stupid, because if Vista goes belly up it takes all your data with it).
Not as stupid as keeping irreplaceable data on client machines.
Many organisations including mine use a small system drive and large data drive. Yes you can redirect the CSC folder but the procedure is beyond the average user [...]
We're not talking about 'the average user', we're talking about professionals.
[...] and who's dumb idea was it to make the folder completely inaccessible?
Probably the same guy who was sick of getting support calls from people who tried to be clever by indiscriminately deleting files they didn't think were important. Of course, as evidenced by your Linux Live CD comments, you can never underestimate the ingenuity of fools.
Like taking out telnet so I cant even check for the ability to connect to a web server?
So change your standard system build to install it.
You mean like the "network and sharing centre" not being able to tell me that it cant get an address from DHCP?
I'm pretty sure it does, although you might need to try and 'repair' it first. Far easier to just run ipconfig from a commandline.
I've yet to see any improvement in the Event Viewer. No improvements here I'm afraid
Then you couldn't have looked very hard, since even a trivial glance would show the much larger range of event logs. It can now also send events back to a centralised logging host and use events as triggers for other tasks.
It sounds like you are trying to manage your Vista machines with knowledge, processes and tools that date from the days of Windows 95 workgroup networks.
I wish people would stop parroting this fallacy all the time. Market share has nothing to do with how easy it is to break into a system.
No-one ever suggests that it does.
If you have something like windows where security is bolted on after the fact, and OS that was never meant to be a multi-user OS connected to the internet (all these were added as features later on and done poorly) [...]
This is simply false.
UNIX on the other hand was designed from day one to be networked multi-user OS, and security and separation of concerns was there from beginning.
The very first versions of UNIX were not multiuser.
You really had put it on a system it did not belong on to see it do poorly. Case in point I just finally retired a K6 system with 256M of RAM that ran XP SP2 well enough to do 'the basics'. Web, mail, basic office stuff. Hell I even had Starcraft, UT, and at one point ran Dungeon Siege on it. (DS really pushed it to it's max limit but it did run.)
If all you want to do is "the basics", then a fast PC from 1999-2000 (1Ghz+ CPU) will suffice for Vista, although it will probably require some minor upgrades from its original configuration, like more RAM (to 1Gb) and a better video card (maybe $30).
And, yes, I have tried Vista on a PC like this, just for the hell of it - an 800Mhz P3 with 768Mb RAM and a 64M GeForce 5600 (my old DOS games box). It was marginally slower than my 1Ghz G4 iBook with 768Mb.
Maybe that is what MS likes to say but when I've seen real world systems, such as your example system above, have XP vs Vista the difference is huge. A ~1.6Ghz dual core with 2-3G of RAM will run noticeably faster with XP than with Vista.
Benchmarks say they're basically identical. It's true Vista requires (relatively) a lot more RAM, but IMHO it's worth it, given how cheap RAM is. My current workstation is a ~3-year-old 1.8Ghz dual-core Opteron with 4G RAM and it's fine. Plus, since it also has three 1920x1200 monitors attached and I typically have dozens of windows and tabs open, it's being driven much harder than the average Vista box would be.
Finally you should keep in mind that I'm referring to the cost of computers upon Vistas release. Looking back I am sure I did not make that clear enough but that is my point. Sure now it's easy to say that cost of what it takes to make a decent Vista machine is minimal but when the OS was released dual core PCs with more than 2G of RAM were not cheap.
When Vista was released, dual-core machines with 2G RAM started at about US$800, but within a matter of months they'd dropped to US$5-600. A fast (3Ghz+ or equivalent) single-core would also suffice, and they were around the same price, if not a bit cheaper.
That is in comparison to XP which, as I pointed out, could be installed on machines that were running 9x and do just fine.
"Running 9x" covers a massive range of hardware (5-6 years worth, during one of the fastest periods for performance improvement and price reduction in computing ever seen). A machine that was "running 9x" could be anything from a 100Mhz 486 with 32Mb RAM to a 1Ghz P3 with 1Gb RAM. XP will be fine on the latter, but unusable on the former.
And that doesn't happen in civilised countries, basically, ever (short of storms bringing down lines).
You might want to expand on "civilised countries" here, so we can tell how much of an ass you're being.
What kind of government would survive if it allowed 'brownouts'?
Electricity supply is privatised (typically with negative results) in many countries. Hardly something the Government has a huge amount of control over.
That's half-true. The power company MUST meet peak demand, as well the rest of the electrical system. You can't say to people "sorry, you won electricity tonight because your neighbor is consuming too much power". That's absurd and ridiculous.
Say what ? That's EXACTLY what happens - when power companies can't meet demand you get brownouts and blackouts (which may or may not be planned and scheduled).
UNIX, and Linux doubly-so, aims for source compatibility and improved architectural simplicity at the cost of some administrative complexity, aka 'Worse is Better'
Which "UNIX" would that be ? Of all the remotely mainstream "UNIXes", Linux is the only one that has serious problems with binary compatibility.
The amount of computing power, let us not get into the Aero fiasco, needed for someone to 'get back to even' with Vista is much more than what it took from 9x -> XP.
Not really. A fast Windows 95 experience could be had with a 100Mhz Pentium and 16M RAM. For Windows 98 you would probably want to triple, maybe quadruple, that (~3-400Mhz P2 and 48-64M RAM). For similar performance in Windows XP you would need to multiple it by a factor of 8-10 (~700Mhz-1Ghz P3, 512M RAM).
Now, from that XP "baseline", you only need a machine maybe 5-6x as powerful (~1.6Ghz dual core, 2-3G RAM) for Vista to be "as fast". A $30 video card gets you all the "Aero capable" GPU you need.
(Obviously if you're comparing a Windows 98 machine you bought in, say, 2001 then the numbers are going to be a lot different, but that's hardly a fair comparison.)
By neither a relative, nor absolute, measure, does Vista have high hardware requirements on release. If anything, the opposite was true.
Vista has a few problems but from where I sit it's major problem was it's hardware requirements. From the 9x series to XP users not only got a huge boost in stability and features but they did not have to take out another mortgage to finance a new computer to run it.
The first sub-US$1000 mainstream PCs started appearing around the middle of 1997. In 1995, when Windows 95 was released, a PC capable of running it well would have probably been around US$1500 (or about US$2000 adjusted to 2007 dollars).
When Vista was released at the beginning of 2007, a machine capable of running it well (dual cores, 2G RAM) cost about US$800.
Or, to put it another way, in real terms the necessary hardware to run Vista cost less than half as much as the hardware to run Windows 95.
The biggest issue with vista compatibility is that with User Account Control, you can't write into the "Program Files" directory, even as administrator.
A program being "Run as Administrator" can most certainly write to %ProgramFiles%.
What's wrong with Vista is that it isn't really backwards compatible; old programs keep on crashing regularly. Given this, why not go with Linux and Wine instead ?
Because the number of programs that don't work in Vista are vastly outnumbered by the number of programs that don't work on WINE.
It is under active development, so its compatibility is likely to keep on improving, and the platform itself is far more stable than Vista.
And you think Microsoft won't keep working to improve Vista's backwards compatibility ?
What is in question is: does the DRM subsystem contribute to the observed problems of Vista (such as bad network performance) ? Do the DRM drivers get loaded, and do they monitor the state of the machine, even when no DRM'd content is being played ? And if yes, how much resources does this consume ?
And the questions have been answered: No, no and none.
This of course assumes that the 64-bit version of Windows 7 is better than the 64-bit version of Windows XP. It also assumes that either is better than the 64-bit support of Wine.
All of which are more than reasonable assumptions.
I fail to see the logic in your statement. If every big business runs on a (let's simplify) 5 year upgrade cycle, one would expect n / 5 businesses to upgrade any given year when there are n businesses. Or did you mean to say that every big business is on the same upgrade schedule?
The cycle includes both hardware replacement AND software certification. Not only does it take years for large businesses to evaluate, test and certify something as large as an across the board OS upgrade, but they won't even _start_ considering it until said OS has been "in the wild" for an appreciable amount of time.
years after release.
Which is why no intelligent person would expect Vista to have a significant presence in business for a good 3-4 years. Just like the previous versions did, and the ones before that. Just like major OS upgrades take years to filter through on other platforms, as well. Because big business is very risk averse, and major OS upgrades are risky propositions.
Following that same logic, they would have never upgraded from 2000 to XP.
Firstly, 2000 to XP was a minor change. Secondly, following my logic they would not upgrade from 2000 to XP (or, more likely, NT4 to XP) for several years while the cycle played out (ie: as the "mostly known quantity" was turned into a "completely known" quantity). Which is, lo and behold, exactly what happened.
A funny remark, but you do not reply to GP's point.
I do, by giving it the ridicule it deserves. Every year, there has been some analyst making the same general statement that because of Microsoft Product X which has been poorly received (usually only to the biases of the writer) and due to improving compatibility on alternate platform B (typically still offering less than 100% drop-in compatibility), then this will be the year that alternate platform B takes over.
Businesses aren't going to move from Windows to Linux (or anything else) because it can be "just as good, maybe". It needs to be "better, a lot better" to justify the massive additional expense of a platform migration over an upgrade.
Now, when someone comes out about this year being the year of Linux on the desktop along with some reasons why migrating to Linux will deliver 30% up-front cost savings, 50% lower annual costs and enabling employees to be 20% more productive, *then* I might take them seriously. As will the business world.
With ads in a national newspaper?
Was it an ad for "A DELL MINI 9 running Linux", or an ad for "A Dell Mini 9 RUNNING LINUX". Was Linux noted as a standard feature like the Atom CPU and a gig of RAM, or was it highlighted as a special feature that justified buying the product ?
An advertisement for a machine that has features A, B, C, D, Linux, X, Y, Z is a very different thing to an advertisement for a machine running Linux that has other features A, B, C, D.
Incidentally, the move away from Windows to Linux, if it ever happens, will be driven from the business world, not the consumer world where toys like the Mini 9 are sold. Businesses aren't going to start using Linux because their employees at home are using it. People using Linux at work, however, might be inclined to start using it at home.
Slashdot Mirror
It also encourages people to be able to speak freely without fear of persecution. Without anonymity it would be impossible for whistleblowers to out evil empire corps without losing their jobs and probably never being hired again.
I agree completely. Anonymity is a cornerstone of free society.
Without anonymity we wouldn't have vitriolic bloggers; we wouldn't have this fantastic forum of discourse where we can speak our minds and not worry about being smacked with a lawsuit (well, not including the video professor).
Of course, that doesn't mean we should encourage people writing inflammatory bullshit just for the sake of it, because they're trolls and know they can hide their bias behind the veil of anonymity.
Or, in other words, anonymity definitely has its uses, but that doesn't imply it should be encouraged.
And anyway, non-anonymity is vapid and trite.
No, it's honest and expected.
Why not encourage anonymity?
Because it also encourages the lack of accountability that goes along with it.
Or, put more crudely.
Filesystems are fundamentally engineered to cope with the high latency of hard drives [...]
In what ways ?
not going to happen dude. perhaps if you had mentioned more trains, more mass transport, ok. but you are asking people to exert a lot of effort. you realize that, right? you want some office worker to bike 10 miles a day? really? are you serious?
While it's certainly not going to happen, it is difficult to conceive of any genuinely negative results should the average office worker start riding 10 miles to work.
Umm, Linux does quite a bit better for graphic design. Especially bigtime movie producers (pixar, etc) don't run Mac. They run linux.
I doubt you'll find many people in the industry think the a bunch of machines in a render farm are doing "graphic design".
You have to show that people truly won't be offended by the extremes towards which you are heading.
The burden of proof is (or should be) on the person asserting that the "extremes" necessary to offend people will ever be reached.
What you are describing is the reason the "slippery slope" is a fallacy to begin with. It's begging the question.
This way they can continually have it in testing until it fades from the publics mind then mothball it.
I think you meant until the next election, at which point it will be rolled out again to garner the religious vote.
I never smile anyway, but what's with this "you can't wear glasses" rule?
Same rationale as passport photos.
It's pretty hard to take the glasses off a photo. OTOH, it's pretty easy to take them off your face when requested.
He was only fined A$3000 and required to behave himself, and no costs were awarded. So there isn't really a heavy hand here.
It's set a pretty terrifying precendent, however.
Think about it, by the "logic" used, everyone who watches a film where someone does is an accessory to murder.
One of the most onerous problems facing users of Windows are the 2GB per process and 4 GB total limits.(often only 3.5GB or so actually usable) This seems to be a corporate decision by Microsoft rater than an actual limitation of the hardware, as all of the modern processors will allow 38 bit addressing.
It's an engineering decision. Lots of drivers for cheap and nasty hardware break when PAE is enabled.
PAE can be enabled in server versions of Windows and all versions before XP. However, if you've got 4GB+ of RAM, you're better off using x64.
I believe the answer is due to poor OS design. As far as I know, linux will much more effectively use nearly all of your RAM before it tries to swap anything out to disk. Not so with windows.
Windows will "pre-emptively" swap pages to disk so that memory can be freed up more quickly (ie: without having to swap), if necessary.
Whether or not that is "bad design" depends on your objectives...
whilst many people would be capable of making the same deductions, many people are not.
I think the word you're after here is 'paranoia', not "deductions".
I don't see GitTorrent offering any meaningful advantage. In any managed software project, the "choke point" is the person or person who decides what goes into a given relaese, not anything technical.
How you came to that conclusion is beyond me?
Because the question asked was what systemic improvements regarding systems management were introduced in Vista, not what good (or bad) experiences have people had using it.
By all means, cherry pick points and ignore those which don't reflect favourably on your point of view.
I'm ignoring the points which are irrelevant, or grossly exaggerated by your obvious bias.
I am yet to see a "Microsoft" deployment plan that gives me half as many options as Arconis.
When was the last time you looked ?
I use it as a straight up imaging tool for identical laptops. 10 minutes and there is a new XP image ready to be put on the domain and into service (2.5 GB from a DVD). Vista's massive space requirements make this difficult..
8G for a default install. Heady stuff, indeed, when you can barely even find new machines with drives smaller than 160G.
The Vista installed DVD has about 2.5G of data on it. That should give you some idea as to why you should be looking at recommended deployment methods, rather than ones you have come up with in the past.
Yes I could move images to a network or USB drive but still, this tends to make it slower or less convenient then DVD plus increases my need for storage and reliable backups where as a DVD can be copied cheaply and easily as many times as needed.
If re-imaging from a network source is slower for you than a DVD, you have bigger problems to deal with.
The machines and AD are configured correctly. UAC is at best security theatre.
These two statements are in conflict. UAC provides exactly the functionality, for the same reasons, at the same times and in basically the same way, as other temporary privilege escalation functions on other OSes (eg: sudo). (Although it may well be that you think sudo is "security theatre" as well, but I doubt it.)
As I said (and I do hate repeating myself) users have already learned to instinctively click "allow" and not even bother reading what UAC has to say.
If your machines are in a Domain, users can't just "click allow" to dismiss a UAC prompt. They need to provide Administrator (or some other) credentials. Unless, of course, you've deliberately configured their systems otherwise.
Then they come and bother me as to why they need to do this 10 times an hour.
If it's happening that often (which is seriously doubt) then you should fix the root cause (which would almost certainly be some program trying to write data to system areas).
Its not that the irreplaceable data isn't backed up, its that they will need to waste valuable work time re-copying it, also the data users consider replaceable will not be moved to servers. You also fail to account for the "road warriors" that need to keep data on local storage because servers are not always available. Beyond this getting users to keep all data on the servers is like herding cats especially when you start dealing with multi-gigabyte data sets.
Actually, it's pretty easy. Firstly, you have the appropriate managerial body institute a policy that all data must be located on managed, centralised servers (with appropriate exceptions for laptop users, along with synchronisation facilities). Then you tell them that any data that isn't on a server is 100% their responsibility, and you won't even attempt to recover it if something goes wrong.
A professional what?
A professional systems administrator. You know, what this whole discussion is about.
Any good Sysadmin would know that you must always consider the end user [...]
From what you've said so far, you don't put much consideration into helping your end users.
So I say again, where is the improvement if its forced me to go back to the tool I already had.
Because, apparently, it makes better attempts to fix itself and allows the end user to also do some basic
Technically correct, but any malware that runs purely with user privileges is going to be pretty tame [...]
Just what is it you think the average piece of malware might want to do, that it can't do with regular user-level privileges ?
[...] - especially as it should be DEAD easy to find and kill (admittedly therefore some kind of "antivirus" software would actually be needed, but nowhere near the complex scale of current things)
The rather large assumption here is that the average user would even notice that something was wrong, let alone start looking for a cause.
Also off the top of my head:
Anecdotes are interesting, but ultimately irrelevant. The OP wanted to know what systemic improvements were available with Vista, not what good or bad experiences people have had.
Still not equal to Arconis. Only good if you do things the "Microsoft" way and the "Microsoft" way only.
As opposed to "the Arconis way" ?
Great way to make my users angry and frustrated. It's security theatre at best, all our current Vista users already igrore it and click allow by default.
Then maybe you should configure their machines properly. An Active Directory domain would probably be a good starting point.
This is not needed in an enterprise/SMB.
Having seen it successfully and productively used in same, I have to disagree.
I beg to differ, in fact I'd say it was a massive leap backwards (see my comments above). Microsoft assumes you are running a single large drive for Vista (stupid, because if Vista goes belly up it takes all your data with it).
Not as stupid as keeping irreplaceable data on client machines.
Many organisations including mine use a small system drive and large data drive. Yes you can redirect the CSC folder but the procedure is beyond the average user [...]
We're not talking about 'the average user', we're talking about professionals.
[...] and who's dumb idea was it to make the folder completely inaccessible?
Probably the same guy who was sick of getting support calls from people who tried to be clever by indiscriminately deleting files they didn't think were important. Of course, as evidenced by your Linux Live CD comments, you can never underestimate the ingenuity of fools.
Like taking out telnet so I cant even check for the ability to connect to a web server?
So change your standard system build to install it.
You mean like the "network and sharing centre" not being able to tell me that it cant get an address from DHCP?
I'm pretty sure it does, although you might need to try and 'repair' it first. Far easier to just run ipconfig from a commandline.
I've yet to see any improvement in the Event Viewer. No improvements here I'm afraid
Then you couldn't have looked very hard, since even a trivial glance would show the much larger range of event logs. It can now also send events back to a centralised logging host and use events as triggers for other tasks.
It sounds like you are trying to manage your Vista machines with knowledge, processes and tools that date from the days of Windows 95 workgroup networks.
I wish people would stop parroting this fallacy all the time. Market share has nothing to do with how easy it is to break into a system.
No-one ever suggests that it does.
If you have something like windows where security is bolted on after the fact, and OS that was never meant to be a multi-user OS connected to the internet (all these were added as features later on and done poorly) [...]
This is simply false.
UNIX on the other hand was designed from day one to be networked multi-user OS, and security and separation of concerns was there from beginning.
The very first versions of UNIX were not multiuser.
You really had put it on a system it did not belong on to see it do poorly. Case in point I just finally retired a K6 system with 256M of RAM that ran XP SP2 well enough to do 'the basics'. Web, mail, basic office stuff. Hell I even had Starcraft, UT, and at one point ran Dungeon Siege on it. (DS really pushed it to it's max limit but it did run.)
If all you want to do is "the basics", then a fast PC from 1999-2000 (1Ghz+ CPU) will suffice for Vista, although it will probably require some minor upgrades from its original configuration, like more RAM (to 1Gb) and a better video card (maybe $30).
And, yes, I have tried Vista on a PC like this, just for the hell of it - an 800Mhz P3 with 768Mb RAM and a 64M GeForce 5600 (my old DOS games box). It was marginally slower than my 1Ghz G4 iBook with 768Mb.
Maybe that is what MS likes to say but when I've seen real world systems, such as your example system above, have XP vs Vista the difference is huge. A ~1.6Ghz dual core with 2-3G of RAM will run noticeably faster with XP than with Vista.
Benchmarks say they're basically identical. It's true Vista requires (relatively) a lot more RAM, but IMHO it's worth it, given how cheap RAM is. My current workstation is a ~3-year-old 1.8Ghz dual-core Opteron with 4G RAM and it's fine. Plus, since it also has three 1920x1200 monitors attached and I typically have dozens of windows and tabs open, it's being driven much harder than the average Vista box would be.
Finally you should keep in mind that I'm referring to the cost of computers upon Vistas release. Looking back I am sure I did not make that clear enough but that is my point. Sure now it's easy to say that cost of what it takes to make a decent Vista machine is minimal but when the OS was released dual core PCs with more than 2G of RAM were not cheap.
When Vista was released, dual-core machines with 2G RAM started at about US$800, but within a matter of months they'd dropped to US$5-600. A fast (3Ghz+ or equivalent) single-core would also suffice, and they were around the same price, if not a bit cheaper.
That is in comparison to XP which, as I pointed out, could be installed on machines that were running 9x and do just fine.
"Running 9x" covers a massive range of hardware (5-6 years worth, during one of the fastest periods for performance improvement and price reduction in computing ever seen). A machine that was "running 9x" could be anything from a 100Mhz 486 with 32Mb RAM to a 1Ghz P3 with 1Gb RAM. XP will be fine on the latter, but unusable on the former.
And that doesn't happen in civilised countries, basically, ever (short of storms bringing down lines).
You might want to expand on "civilised countries" here, so we can tell how much of an ass you're being.
What kind of government would survive if it allowed 'brownouts'?
Electricity supply is privatised (typically with negative results) in many countries. Hardly something the Government has a huge amount of control over.
That's half-true. The power company MUST meet peak demand, as well the rest of the electrical system. You can't say to people "sorry, you won electricity tonight because your neighbor is consuming too much power". That's absurd and ridiculous.
Say what ? That's EXACTLY what happens - when power companies can't meet demand you get brownouts and blackouts (which may or may not be planned and scheduled).
UNIX, and Linux doubly-so, aims for source compatibility and improved architectural simplicity at the cost of some administrative complexity, aka 'Worse is Better'
Which "UNIX" would that be ? Of all the remotely mainstream "UNIXes", Linux is the only one that has serious problems with binary compatibility.
The amount of computing power, let us not get into the Aero fiasco, needed for someone to 'get back to even' with Vista is much more than what it took from 9x -> XP.
Not really. A fast Windows 95 experience could be had with a 100Mhz Pentium and 16M RAM. For Windows 98 you would probably want to triple, maybe quadruple, that (~3-400Mhz P2 and 48-64M RAM). For similar performance in Windows XP you would need to multiple it by a factor of 8-10 (~700Mhz-1Ghz P3, 512M RAM).
Now, from that XP "baseline", you only need a machine maybe 5-6x as powerful (~1.6Ghz dual core, 2-3G RAM) for Vista to be "as fast". A $30 video card gets you all the "Aero capable" GPU you need.
(Obviously if you're comparing a Windows 98 machine you bought in, say, 2001 then the numbers are going to be a lot different, but that's hardly a fair comparison.)
By neither a relative, nor absolute, measure, does Vista have high hardware requirements on release. If anything, the opposite was true.
Vista has a few problems but from where I sit it's major problem was it's hardware requirements. From the 9x series to XP users not only got a huge boost in stability and features but they did not have to take out another mortgage to finance a new computer to run it.
The first sub-US$1000 mainstream PCs started appearing around the middle of 1997. In 1995, when Windows 95 was released, a PC capable of running it well would have probably been around US$1500 (or about US$2000 adjusted to 2007 dollars).
When Vista was released at the beginning of 2007, a machine capable of running it well (dual cores, 2G RAM) cost about US$800.
Or, to put it another way, in real terms the necessary hardware to run Vista cost less than half as much as the hardware to run Windows 95.
The biggest issue with vista compatibility is that with User Account Control, you can't write into the "Program Files" directory, even as administrator.
A program being "Run as Administrator" can most certainly write to %ProgramFiles%.
What's wrong with Vista is that it isn't really backwards compatible; old programs keep on crashing regularly. Given this, why not go with Linux and Wine instead ?
Because the number of programs that don't work in Vista are vastly outnumbered by the number of programs that don't work on WINE.
It is under active development, so its compatibility is likely to keep on improving, and the platform itself is far more stable than Vista.
And you think Microsoft won't keep working to improve Vista's backwards compatibility ?
What is in question is: does the DRM subsystem contribute to the observed problems of Vista (such as bad network performance) ? Do the DRM drivers get loaded, and do they monitor the state of the machine, even when no DRM'd content is being played ? And if yes, how much resources does this consume ?
And the questions have been answered: No, no and none.
This of course assumes that the 64-bit version of Windows 7 is better than the 64-bit version of Windows XP. It also assumes that either is better than the 64-bit support of Wine.
All of which are more than reasonable assumptions.
I fail to see the logic in your statement. If every big business runs on a (let's simplify) 5 year upgrade cycle, one would expect n / 5 businesses to upgrade any given year when there are n businesses. Or did you mean to say that every big business is on the same upgrade schedule?
The cycle includes both hardware replacement AND software certification. Not only does it take years for large businesses to evaluate, test and certify something as large as an across the board OS upgrade, but they won't even _start_ considering it until said OS has been "in the wild" for an appreciable amount of time. years after release.
Which is why no intelligent person would expect Vista to have a significant presence in business for a good 3-4 years. Just like the previous versions did, and the ones before that. Just like major OS upgrades take years to filter through on other platforms, as well. Because big business is very risk averse, and major OS upgrades are risky propositions.
Following that same logic, they would have never upgraded from 2000 to XP.
Firstly, 2000 to XP was a minor change. Secondly, following my logic they would not upgrade from 2000 to XP (or, more likely, NT4 to XP) for several years while the cycle played out (ie: as the "mostly known quantity" was turned into a "completely known" quantity). Which is, lo and behold, exactly what happened.
A funny remark, but you do not reply to GP's point.
I do, by giving it the ridicule it deserves. Every year, there has been some analyst making the same general statement that because of Microsoft Product X which has been poorly received (usually only to the biases of the writer) and due to improving compatibility on alternate platform B (typically still offering less than 100% drop-in compatibility), then this will be the year that alternate platform B takes over.
Businesses aren't going to move from Windows to Linux (or anything else) because it can be "just as good, maybe". It needs to be "better, a lot better" to justify the massive additional expense of a platform migration over an upgrade.
Now, when someone comes out about this year being the year of Linux on the desktop along with some reasons why migrating to Linux will deliver 30% up-front cost savings, 50% lower annual costs and enabling employees to be 20% more productive, *then* I might take them seriously. As will the business world.
With ads in a national newspaper?
Was it an ad for "A DELL MINI 9 running Linux", or an ad for "A Dell Mini 9 RUNNING LINUX". Was Linux noted as a standard feature like the Atom CPU and a gig of RAM, or was it highlighted as a special feature that justified buying the product ?
An advertisement for a machine that has features A, B, C, D, Linux, X, Y, Z is a very different thing to an advertisement for a machine running Linux that has other features A, B, C, D.
Incidentally, the move away from Windows to Linux, if it ever happens, will be driven from the business world, not the consumer world where toys like the Mini 9 are sold. Businesses aren't going to start using Linux because their employees at home are using it. People using Linux at work, however, might be inclined to start using it at home.