I see alot of hand-waving in your post, but very little of specifics.
In what specific ways are you claiming that 'Microsoft is responsible for making their platform insecure'?
Who is the purveyor of the most popular development tools for use on Windows? Microsoft. And these tools are full of lots of guidance to make sure your apps run as non-admin, and how to do so. And have been for years.
Who is the purveyor of the most popular development training materials for use by budding Windows developers? Microsoft. And these training materials are full of encouragement and guidance to make sure your apps run as non-admin, and how to do so. And have been for years.
Who certifies Microsoft Certified Developers? Duh. Microsoft. And if you are a certified MCSD, then you have been trained to make your apps run as non-admin, and how to do so. And have been for years.
But I'm not really sure what that has to do with anything, as the vast, vast majority of software written for windows was not written by folks with MCSD certs.
Who is supposed to be leading their ISVs by example? Microsoft. And they have been, for years. By the time office 2003 came out (you know, 5 years ago), the vast majority of MS apps ran fine under non-admin. Even back to Office 2000, Office ran fine as non-admin, but there were some minor quirks.
What's the common denominator here? Microsoft. I'm not sure what that has to do with anything, other than Microsoft is the most involved with the Microsoft ecosystem. Big surprise there.
They are responsible for training developers to use unnecessary security elevations. Can you be specific about this? The vast majority of developers for windows apps arent trained by Microsoft at all. Most have absolutely no interaction with MS whatsoever.
And they do it themselves. Can you provide specifics? Hopefully better ones than the links you provided down below, which werent relevant to the discussion.
Did you even read either of those two links you included? I dont think they say what you expect them to say.
The first is an article originally published with the first release of WinXP, over 5 years ago. The second is a random search on MS.com, most of the results of which dont have anything to do with your argument.
instead of MS directly pressuring software developers themselves or with their dev tools, Microsoft has minimised it's value to computer users. How exactly would MS 'directly pressure software developers'?
The vast majority of 3rd party dev companies have absolutely no relationship with MS whatsoever. At most, they use Visual Studio, but even a very large percentage use VS6, which really predates the 'everyone should run as non-admin' in the mainstream IT mindset.
The problem is that there is really NO way for MS to pressure developers. There just isnt any.
So MS is forced into this terrible choice. They have to draw a line in the sand somewhere, but unfortunately, the bulk of the cost up front is going to be borne by users in the form of irritation and confusion over UAC.
It's a crappy situation, but I havent seen any better alternatives proposed. 'Pressuring software developers' sounds great on paper, but isnt really practical in the real world.
If Microsoft was really interested in security they would have done more and better sandboxing of applications. Nearly everything you suggest would end up in exactly the same situation we're in now.
Because most app developers wouldnt go to the trouble to use this sandbox method. Most wouldnt even know it exists.
The facilities available on windows are already sufficient to sandbox at whatever level you want. But it requires the app developer to participate. The problem that results in where we're at now is that most app devs dont. They dont have a clue about how windows works, or how to write an installer, or how to make an app run as non-admin.
And they dont care.
Your suggestion would require them to learn even more about windows new sandboxing mode. The problem is, the good devs/isv's who are willing to learn how to do it right have already done so, or are already working on it. The ones who arent, wouldnt even participate or know about your new sandbox mode.
There are already specific and thorough guidelines and resources published about how to correctly write software for windows, and not trip over any of this stuff. If you've been doing this all along, then your software wouldnt even have to change to run on windows with no UAC prompts.
UAC nags you for every little piece of rubbish. 99.999% of those requests are ok. By definition, if UAC is nagging you, then its not OK. Either you're purposefully doing something that prompts the system (ie, everything is OK), or some software you're using is doing something bad. Writing user preferences in C:\Program Files\DumbAssApp\prefs.ini is not okay.
The problem is that the bulk of the 3rd party software developers in the ecosystem use practices that violate the published guidelines and best-practices for the platform, and often use techniques that are indistinguishable from malware.
Instead of finding a way to give the user a secure system, MS just shifted the blame. You kind of argued yourself in a circle there.
Alot of hand waving about how bad UAC is, it maligns the users, etc etc. And then 'something should be done about it', but no substantive suggestions along those lines.
Propose a valid alternative that doesnt involve time travel, and your argument might have some weight.
And whats this stuff about 'blame'? There's no blame, just costs. How would you suggest Microsoft makes incompetent 3rd party developers pay the cost for their sloppy code writing without involving the user in any way?
What MS has done here is to force the costs of sloppy coding by 3rd party developers to become visible, whereas prior to UAC, if you didnt run as non-admin, you never saw those costs. They were invisibile. MS just made them visible. So now users are bearing the costs of sloppy coding by 3rd party developers, in the hope that the pressure will then be passed on to these devs.
Unfortunately, MS doesnt have any direct relationship with these vendors, there's no place to have leverage, to make the 3rd party devs do 'the right thing'.
Overall, it sounds to me like you're just posting here to join in the 'look how much Micro$oft is teh suck' bandwagon, but without actually contributing anything to the conversation. Suggest an alternative thats more substantive than 'something should be done'.
What I think you'll find right now, is that in the realm of tablet/convertible machines, Vista is much better (as long as you dont have any crappy old software that was designed poorly and only works right when run as a local admin). The pen/tablet functionality of Vista is noticeably better than XP Tablet Edition.
For laptops and portables, its an even trade. If you get a corporate class machine with good drivers, and enough power, then its a better experience overall (though with some annoyances, mostly around the explorer shell). But stay away from the consumer level garbage laptops with Vista. They're mostly crap now.
Companies don't like upgrading hardware before their time.
Agreed, but what does that have to do anything? I dont care if every Vista session included a happy ending, if it required hardware changes, then businesses wont roll it out until they're ready. That is the correct response, and was exactly how they all handled the move to XP, and 2000 before that, etc.
I have a client who has an employee who got a new laptop with Vista Business on it. They didn't like some crap that came with it, so they dumped that and put Vista Ultimate on it, despite the fact that the machine wasn't certified for Ultimate, only Business. So the employee ended up with mouse freezes that drove him crazy. I reloaded some of the drivers and the problem appears to have gone away. But this demonstrates that installing Vista on existing machines not certified for even that particular version is a crapshoot.
You may or may not have actually experienced this, but the conclusion you come to is not accurate. There is no difference in drivers between the versions (excepting x86 vs x64). There is no such thing as a 'Vista Ultimate' driver. There are just Vista x86 drivers and x64 drivers.
In fact, by your own story, if re-loading the same drivers fixed it, then the problem wasnt with the drivers, it was something else.
As numerous analysts have said, there's really no great reason to do so, since much of the Vista changes can either be obtained with third party XP software (cheaper than buying a new copy of Vista) or aren't that important in any event.
Thats such nonsense. What 3rd party software gives you UAC? How about registry and program file virtualization? What about transactional NTFS? What about moving most of the drivers to userspace? Or a composited window manager? What about tablet functionality? What about much improved stability and better lifespan in a laptop scenario with 3-5 suspends/hibernates per day? What about a new kernel scheduler and a new I/O scheduler? What about a new tcp/ip stack that dynamically optimizes itself when moving from a fast, low-latency gigabit lan to a fast, high-latency vpn across the internet to a file server?
Mind you, just like it was when we moved to XP, there is some driver stability issues. More so in fact, since the driver situation changed so dramatically between XP and Vista. Add to that the fiasco with the 'Vista Capable' nonsense, and its taking a while to stabilize.
Large corporates not moving to Vista has very little to do with Vista, and more to do with the dynamcis of change in large corporates. For most of these folks, the bulk of their desktops are used for a small number of task-specifc roles. Once the company has their apps stabilized on a platform, there's very little reason to change those kinds of computers.
But the marketing department still uses Macs. Developers still use whatever the hell they want. And the early adopters in the IT staff were on Vista within a week of it being available through their Technet subscription download.
Business have enough trouble with XP right now. They don't need more trouble by trying to install Vista on XP machines. As you say, Vista will creep in - but only because M
I cant speak to any country version other than US. But here, someone like you should be buying from the Small & Medium Business site, where you can still get XP.
And that way, you stay away from the crappy consumer level stuff.
That's why:
1. always partition everything
2. never use Maxtor drives
3. never buy Dell I would strongly argue that you're learning the wrong lesson here.
The lesson to be learned is to have your data backed up.
With a secondary lesson in that if downtime is expensive for you, then have redundancy built in (ie, mirroring or better).
My guess is they're taking bits and pieces from XP and Vista, reworking them as modular sections. Basically, "Windows 7" is going to be Windows XP Service Pack 4 with backports of some of the Vista stuff that wasn't a total disaster. Not really. MS has been quite open in that they have wanted to move to a 'modular' windows design for years.
That was one of the big changes they were trying to make with Vista before the reset. They tried to do too much at once, and the result was very poor, so they did the reset back to the win2003/xp-64 codebase for vista.
However, with Server 2008, they've set alot of the foundations for a more modular OS. You can see that in the 'Server Core' business. It's not completely modular, but they've taken some steps down that road, and will be able to continue down that path in the next version.
A complete rewrite would take them another five years like it did Vista. Vista wasnt a complete re-write and was never meant to be. They did try to make some big changes, a modular system, etc, but then they did a reset. The result was a pretty vanilla 3-year development cycle resulting in Vista, after the reset.
Most businesses, especially small business, really have utterly no use for Vista - XP is fine for most of them, despite the pathetic quality of the OS. What most business have no need for is unnecessary change. Vista will creep in. Once you have machines that have 3-4GB of memory and a 2.0 or better C2D processor in the $700-1000 price range, Vista is quite nice.
In other words, given stable quality drivers and enough hardware power, and Vista is noticeably better than XP. Particularly on laptops and tablets/convertibles.
I should have known better, given your twitter-esque tendency to quote yourself as if you were a reputable source.
But then I went to one at random, and found an article comparing two completely unrelated products, and full of factually inaccurate claims about the Microsoft product.
Thats incorrect. Companies in that situation dont pay for XP separately. They just pay for the OS with the hardware, then their VL contract allows them to just rip the Vista install off and put whatever they want on it.
With WinXP Prof EOL this year June, what's the alternative to Vista? Depends on which customer base you're talking about.
For consumers, there is no alternative to Vista.
For businesses, the alternative to Vista is XP Pro. MS just requires that the corps license Vista w/ SA, and then they get downgrade rights to XP. Without SA, MS will not give out downgrade rights to XP. This has been a big level for companies to buy SA on top of their existing programs.
XP Pro will likely be part of the VL media package for 5 years.
There are certainly conceptual problems with it. Microsoft isn't the only company that has tried to come up with a file system built on a database; Oracle has taken stabs at it, too. If the two of them have trouble coming up with something, and the FOSS world hasn't had any significant success, either, then there must be something difficult to it. All filesystems are built on a database. They're just not typically relational databases in the typical sense of the word. But they're very similar to an ISAM database.
Databases have latencies that have to be overcome. Microsoft knows this well from trying to move Exchange to a SQL-based storage mechanism instead of the Jet Engine base they've been using for more than a decade. They have had test builds that work, but performance takes a huge hit, so they haven't done it. See above wrt databases and file systems.
Exchange does not, and has never, used Jet for storage. Exchange uses ESEDB, which for a short period of time in its history was code named 'Jet Blue'.
ESE DB (aka Jet Blue) has nothing to do with Jet (aka Jet Red) or what you think of as the db underlying MS Access.
The fact that they for a brief period had similar names had to do with the marketing/product-planning of the time, not about any underlying similarity or shared technology.
I wish more people had critical thinking skills like you do.
Do business have their sales people do oil changes, tire rotations, etc on their company cars? Of course not, you have it done professionally.
Do businesses have clerical stuff repair the Air Conditioners or Plumbing when they break? Of course not, you have it done professionally.
But computers, operating systems, networks, and software. These are orders of magnitude more complicated than HVAC, cars, or plumbing. Yet people want to self-manage?
Rigorous lockdown doesn't work for us at the hospital where I work. Too many apps that we're locked-in to won't run properly without admin rights. You probably need better IT people.
For the vast, vast majority of software that supposedly needs admin rights, some simple investigation using Process Monitor (regmon, filemon) and some group policy enforced ntfs and registry acl tweaks solves it.
It's a one time investment of research and configuration time, and then it works fine as non-admins.
There are some pieces of business software that load their own drivers in such an ultra-crappy that the above technique doesnt always work. Even then, you can usually give selective user rights to users who need to do this.
Mind you, there are some pieces of software, even in a business, that you cant easily work around. But its very rare, and almost always is a runtime loaded driver issue. Crap software, in other words.
Okay, I hate to feed the cowards, but that was not a reasonable response.
No sane company does backups of the individual machine's OS directories or files.
Most sane companies dont do backups of local machines at all (rarely connected laptops excepted). Business files go on the server.
And running A/V wont stop people from getting hit by malware. It's just one line in the defense. Far more effective is to have the machines autopatch, and dont run as admin. When you do those two things, A/V is a very rarely used fallback defense.
I'd hope that, say, software engineers actually understood a thing or two about computers. You'd think so, but I find that its usually not the case.
So very many developers are so ridiculous overspecialized that they dont really understand the underlying operating systems, or little things like relational databases, or HTTP protocols, etc.
This is part of why, IMO, you see so many developers moving to OSX. Because they dont have to understand or think about the OS, they can just write code.
It's really sad though because these are the people writing software, software installers, database apps, etc. But ask them how any of those technologies they are supposedly working with work, and you get a blank stare.
It's quite sad, actually.
I've worked with supposedly high end engineering programmers who will rant and rave at you that Linux is 'The Only Way', yet they cant figure out how to flip their machine from DHCP to static IP, or similar low level silliness.
Well, two options that jump right off the page at me:
1. Samba project tools: smbpasswd
2. Web based password change functionality. Commonly known as IISADMPWD, its a set of asp pages that are included with IIS (can optionally be installed on iis6/win2003). Part of every windows server install.
I've got several sharepoint installs at client sites who use Firefox, and it works fine for them.
The only difference I'm aware of is the wiki editor. On IE you get a nicer WYSIWYG rich text editor and you can optionally write wiki markup (much like html markup), on Firefox you have to use the wiki markup editor.
We've got plenty of PDFs in ours, and they work just fine.
Search engine works good for us, but we dont use it real heavily.
Also note that both Alfresco and Sharepoint are free. So they're both the same, zero cash, assuming you've got a windows server to run it on.
I think you are confusing OOXML and Microsoft.doc file format? I am slightly conflating.doc and.docx, sure. Both are published specs now though, and there are tons of apps that can read/write, import/export to both.
You can't "leave" unless you can take your documents with you. If Microsoft refuses to write ODF or any other format that you can take, you are locked in. Well thats just not true at all.
Despite it not being a standard, and office not writing to ODF, there are a ton of available software packages that can read and write, import and export to both.doc and.docx.
OpenOffice and iWork come to mind right off the top of my head.
None of them work perfectly, but thats just par for the course. Even the MS apps dont work perfectly from version to version.
And there are third party apps whose sole purpose is to do batch conversions of one doctype to another. They're for pay, but they exist. They cover the precise 'I'm leaving' scenario.
Not to mention word docs you can export to rtf or pdf or html and take them with you. And PowerPoint you can export to PDF or HTML and take it with you.
Of course, with things like Excel and Access, businesses will have tons of code written in VBA inside these apps, and that'll never be portable. But then again, the scripting language inside these apps are never portable. Even if its in a broadly available language, the object model to interact with the document itself will be different from suite to suite.
And overall... you CAN leave. I could make a decision to move my business off of MS Office and I guarantee you that I could make it work. And the presence of OpenOffice and ODF does in fact provide me with leverage from MS. Not that I've used it, but its there should I need it.
But the bottom line is that in the real world, it IS possible to leave MS Office and take your docs with you. It does come with a fair amount of pain though (conversion, retraining, compatibility), and you have to measure that against the benefits you think you'll get (licensing fees, etc).
How about you lay off the angry juice for a bit. My comment was about the ISO processes themselves, not about whether OOXML is a good standard proposal or not.
And please... just because there are google results for the words OOXML and bribe doesnt mean that bribery was used. Thats a completely nonsensical argument.
This is FAR from the first report of "creative rule bending" mid-game rule rewrites and outright rule violations and mis-informing people about votes to prevent them from showing up to cast their votes and bribery and national "No" votes being falsely recorded as "Yes" votes and assorted other shenanigans from countries all across the globe. And ALL of the "voting irregularities" all stack one way. They all involve either a country being shoved up into the "yes" column or blocked from the "no" column and diverted up into the "abstention" column. Part of my point is that these orgs are not very public or transparent. I cant look at the rules for the Norway ISO body and see what their rules are, and then compare their minutes to the rules, to see if it went according to the rules.
This means that unless you were there, physically in the room, then you (and I mean the generic person you, not you specifically) do not have first hand information.
And there's a lot of FUD on both sides on this. All the brouhaha on the Norway vote... how the admins locked out the attendees and made a vote. That has been all over the techy news. But its starting to appear as if that original story was completely wrong. That the vote in question was an administrative matter about a prior vote.
Oh, and I happen to be a programmer. Likewise. I'm also a software business owner.
I have looked at portions of the proposed OOXML standard. It is riddled with literally THOUSANDS of minor errors and substantial flaws. It is a total mess. There is no way in hell it belongs on the ISO fast track process. It would take YEARS of review and edits on the standard mainline ISO process to get it UP to the level of merely being a bad standard. Even the SUPPORTERS essentially admit this, saying yeah it's riddled with problems but pass it anyway and let the maintenance process worry about fixing everything. All that is true, but also doesnt mean much.
OOXML is a documentation of a current de facto standard. It's the documentation of a representation thats had more than 15 years of evolution, never once until the last few years with any thought towards public documentation.
And even with its flaws, its arguably more useful than ODF.
Look at the differences:
ODF is a nice, clean standard. However, its not really used in the real world outside of a very very tiny subset of the technical population. And despite it being a standard, if you use the format, then the vast majority of the world wont be able to read your documents.
OOXML would make a fairly crappy standard. However, its used by the vast majority of the people in the world. And despite it not being a standard, if you send it to arbitrary other pople, the likelihood that they will be able to read it with their current software is very high.
All that being said, I do think ODF is a nice standard (in theory), and I dont think OOXML should have been fast-tracked.
So, the likelihood of ODF being a meaningful document storage format over the next 10 years is very low, at least in my opinion. It's a great idea, but probably just wont play out in the market. Microsoft is just too damn good at marketing, sales, and deal-making.
However, even with that, its 'A Good Thing' that ODF exists. Because it puts competitive pressure on companies like Microsoft.
ODF being present probably wont sway most companies, they'll still buy MS Office. But it gives them a stick to use against MS and other companies. With it, you can always leave. This helps keep companies like that more honest and competitive than they would be otherwise.
Slashdot Mirror
In what specific ways are you claiming that 'Microsoft is responsible for making their platform insecure'? Who is the purveyor of the most popular development tools for use on Windows? Microsoft. And these tools are full of lots of guidance to make sure your apps run as non-admin, and how to do so. And have been for years. Who is the purveyor of the most popular development training materials for use by budding Windows developers? Microsoft. And these training materials are full of encouragement and guidance to make sure your apps run as non-admin, and how to do so. And have been for years. Who certifies Microsoft Certified Developers? Duh. Microsoft. And if you are a certified MCSD, then you have been trained to make your apps run as non-admin, and how to do so. And have been for years.
But I'm not really sure what that has to do with anything, as the vast, vast majority of software written for windows was not written by folks with MCSD certs. Who is supposed to be leading their ISVs by example? Microsoft. And they have been, for years. By the time office 2003 came out (you know, 5 years ago), the vast majority of MS apps ran fine under non-admin. Even back to Office 2000, Office ran fine as non-admin, but there were some minor quirks. What's the common denominator here? Microsoft. I'm not sure what that has to do with anything, other than Microsoft is the most involved with the Microsoft ecosystem. Big surprise there. They are responsible for training developers to use unnecessary security elevations. Can you be specific about this? The vast majority of developers for windows apps arent trained by Microsoft at all. Most have absolutely no interaction with MS whatsoever. And they do it themselves. Can you provide specifics? Hopefully better ones than the links you provided down below, which werent relevant to the discussion.
Did you even read either of those two links you included? I dont think they say what you expect them to say.
The first is an article originally published with the first release of WinXP, over 5 years ago. The second is a random search on MS.com, most of the results of which dont have anything to do with your argument.
The vast majority of 3rd party dev companies have absolutely no relationship with MS whatsoever. At most, they use Visual Studio, but even a very large percentage use VS6, which really predates the 'everyone should run as non-admin' in the mainstream IT mindset.
The problem is that there is really NO way for MS to pressure developers. There just isnt any.
So MS is forced into this terrible choice. They have to draw a line in the sand somewhere, but unfortunately, the bulk of the cost up front is going to be borne by users in the form of irritation and confusion over UAC.
It's a crappy situation, but I havent seen any better alternatives proposed. 'Pressuring software developers' sounds great on paper, but isnt really practical in the real world.
Because most app developers wouldnt go to the trouble to use this sandbox method. Most wouldnt even know it exists.
The facilities available on windows are already sufficient to sandbox at whatever level you want. But it requires the app developer to participate. The problem that results in where we're at now is that most app devs dont. They dont have a clue about how windows works, or how to write an installer, or how to make an app run as non-admin.
And they dont care.
Your suggestion would require them to learn even more about windows new sandboxing mode. The problem is, the good devs/isv's who are willing to learn how to do it right have already done so, or are already working on it. The ones who arent, wouldnt even participate or know about your new sandbox mode.
There are already specific and thorough guidelines and resources published about how to correctly write software for windows, and not trip over any of this stuff. If you've been doing this all along, then your software wouldnt even have to change to run on windows with no UAC prompts.
The problem is that the bulk of the 3rd party software developers in the ecosystem use practices that violate the published guidelines and best-practices for the platform, and often use techniques that are indistinguishable from malware. Instead of finding a way to give the user a secure system, MS just shifted the blame. You kind of argued yourself in a circle there.
Alot of hand waving about how bad UAC is, it maligns the users, etc etc. And then 'something should be done about it', but no substantive suggestions along those lines.
Propose a valid alternative that doesnt involve time travel, and your argument might have some weight.
And whats this stuff about 'blame'? There's no blame, just costs. How would you suggest Microsoft makes incompetent 3rd party developers pay the cost for their sloppy code writing without involving the user in any way?
What MS has done here is to force the costs of sloppy coding by 3rd party developers to become visible, whereas prior to UAC, if you didnt run as non-admin, you never saw those costs. They were invisibile. MS just made them visible. So now users are bearing the costs of sloppy coding by 3rd party developers, in the hope that the pressure will then be passed on to these devs.
Unfortunately, MS doesnt have any direct relationship with these vendors, there's no place to have leverage, to make the 3rd party devs do 'the right thing'.
Overall, it sounds to me like you're just posting here to join in the 'look how much Micro$oft is teh suck' bandwagon, but without actually contributing anything to the conversation. Suggest an alternative thats more substantive than 'something should be done'.
In other words, right now XP is better.
Better for whom? In what circumstances?
What I think you'll find right now, is that in the realm of tablet/convertible machines, Vista is much better (as long as you dont have any crappy old software that was designed poorly and only works right when run as a local admin). The pen/tablet functionality of Vista is noticeably better than XP Tablet Edition.
For laptops and portables, its an even trade. If you get a corporate class machine with good drivers, and enough power, then its a better experience overall (though with some annoyances, mostly around the explorer shell). But stay away from the consumer level garbage laptops with Vista. They're mostly crap now.
Companies don't like upgrading hardware before their time.
Agreed, but what does that have to do anything? I dont care if every Vista session included a happy ending, if it required hardware changes, then businesses wont roll it out until they're ready. That is the correct response, and was exactly how they all handled the move to XP, and 2000 before that, etc.
I have a client who has an employee who got a new laptop with Vista Business on it. They didn't like some crap that came with it, so they dumped that and put Vista Ultimate on it, despite the fact that the machine wasn't certified for Ultimate, only Business. So the employee ended up with mouse freezes that drove him crazy. I reloaded some of the drivers and the problem appears to have gone away. But this demonstrates that installing Vista on existing machines not certified for even that particular version is a crapshoot.
You may or may not have actually experienced this, but the conclusion you come to is not accurate. There is no difference in drivers between the versions (excepting x86 vs x64). There is no such thing as a 'Vista Ultimate' driver. There are just Vista x86 drivers and x64 drivers.
In fact, by your own story, if re-loading the same drivers fixed it, then the problem wasnt with the drivers, it was something else.
As numerous analysts have said, there's really no great reason to do so, since much of the Vista changes can either be obtained with third party XP software (cheaper than buying a new copy of Vista) or aren't that important in any event.
Thats such nonsense. What 3rd party software gives you UAC? How about registry and program file virtualization? What about transactional NTFS? What about moving most of the drivers to userspace? Or a composited window manager? What about tablet functionality? What about much improved stability and better lifespan in a laptop scenario with 3-5 suspends/hibernates per day? What about a new kernel scheduler and a new I/O scheduler? What about a new tcp/ip stack that dynamically optimizes itself when moving from a fast, low-latency gigabit lan to a fast, high-latency vpn across the internet to a file server?
Mind you, just like it was when we moved to XP, there is some driver stability issues. More so in fact, since the driver situation changed so dramatically between XP and Vista. Add to that the fiasco with the 'Vista Capable' nonsense, and its taking a while to stabilize.
Large corporates not moving to Vista has very little to do with Vista, and more to do with the dynamcis of change in large corporates. For most of these folks, the bulk of their desktops are used for a small number of task-specifc roles. Once the company has their apps stabilized on a platform, there's very little reason to change those kinds of computers.
But the marketing department still uses Macs. Developers still use whatever the hell they want. And the early adopters in the IT staff were on Vista within a week of it being available through their Technet subscription download.
Business have enough trouble with XP right now. They don't need more trouble by trying to install Vista on XP machines. As you say, Vista will creep in - but only because M
I cant speak to any country version other than US. But here, someone like you should be buying from the Small & Medium Business site, where you can still get XP.
And that way, you stay away from the crappy consumer level stuff.
1. always partition everything
2. never use Maxtor drives
3. never buy Dell I would strongly argue that you're learning the wrong lesson here.
The lesson to be learned is to have your data backed up.
With a secondary lesson in that if downtime is expensive for you, then have redundancy built in (ie, mirroring or better).
That was one of the big changes they were trying to make with Vista before the reset. They tried to do too much at once, and the result was very poor, so they did the reset back to the win2003/xp-64 codebase for vista.
However, with Server 2008, they've set alot of the foundations for a more modular OS. You can see that in the 'Server Core' business. It's not completely modular, but they've taken some steps down that road, and will be able to continue down that path in the next version. A complete rewrite would take them another five years like it did Vista. Vista wasnt a complete re-write and was never meant to be. They did try to make some big changes, a modular system, etc, but then they did a reset. The result was a pretty vanilla 3-year development cycle resulting in Vista, after the reset. Most businesses, especially small business, really have utterly no use for Vista - XP is fine for most of them, despite the pathetic quality of the OS. What most business have no need for is unnecessary change. Vista will creep in. Once you have machines that have 3-4GB of memory and a 2.0 or better C2D processor in the $700-1000 price range, Vista is quite nice.
In other words, given stable quality drivers and enough hardware power, and Vista is noticeably better than XP. Particularly on laptops and tablets/convertibles.
Wow, pimp your own blog much?
I should have known better, given your twitter-esque tendency to quote yourself as if you were a reputable source.
But then I went to one at random, and found an article comparing two completely unrelated products, and full of factually inaccurate claims about the Microsoft product.
Hallelujah. A breath of logic and reason on slashdot. It's such a nice thing to see now and then in this pit of reactionary ignorance.
I'd mod you up if I hadnt posted several times already.
Thats incorrect. Companies in that situation dont pay for XP separately. They just pay for the OS with the hardware, then their VL contract allows them to just rip the Vista install off and put whatever they want on it.
There is no requirement to pay twice.
For consumers, there is no alternative to Vista.
For businesses, the alternative to Vista is XP Pro. MS just requires that the corps license Vista w/ SA, and then they get downgrade rights to XP. Without SA, MS will not give out downgrade rights to XP. This has been a big level for companies to buy SA on top of their existing programs.
XP Pro will likely be part of the VL media package for 5 years.
Exchange does not, and has never, used Jet for storage. Exchange uses ESEDB, which for a short period of time in its history was code named 'Jet Blue'.
ESE DB (aka Jet Blue) has nothing to do with Jet (aka Jet Red) or what you think of as the db underlying MS Access.
The fact that they for a brief period had similar names had to do with the marketing/product-planning of the time, not about any underlying similarity or shared technology.
THANK YOU.
I wish more people had critical thinking skills like you do.
Do business have their sales people do oil changes, tire rotations, etc on their company cars? Of course not, you have it done professionally.
Do businesses have clerical stuff repair the Air Conditioners or Plumbing when they break? Of course not, you have it done professionally.
But computers, operating systems, networks, and software. These are orders of magnitude more complicated than HVAC, cars, or plumbing. Yet people want to self-manage?
Give me a break.
For printing? Does the printer have an ethernet port built in?
If so, I'd bet that the HP Universal Printer Driver would work on it.
If you want scanning, faxing from the network, thats a whole different ball game.
For the vast, vast majority of software that supposedly needs admin rights, some simple investigation using Process Monitor (regmon, filemon) and some group policy enforced ntfs and registry acl tweaks solves it.
It's a one time investment of research and configuration time, and then it works fine as non-admins.
There are some pieces of business software that load their own drivers in such an ultra-crappy that the above technique doesnt always work. Even then, you can usually give selective user rights to users who need to do this.
Mind you, there are some pieces of software, even in a business, that you cant easily work around. But its very rare, and almost always is a runtime loaded driver issue. Crap software, in other words.
Okay, I hate to feed the cowards, but that was not a reasonable response.
No sane company does backups of the individual machine's OS directories or files.
Most sane companies dont do backups of local machines at all (rarely connected laptops excepted). Business files go on the server.
And running A/V wont stop people from getting hit by malware. It's just one line in the defense. Far more effective is to have the machines autopatch, and dont run as admin. When you do those two things, A/V is a very rarely used fallback defense.
So very many developers are so ridiculous overspecialized that they dont really understand the underlying operating systems, or little things like relational databases, or HTTP protocols, etc.
This is part of why, IMO, you see so many developers moving to OSX. Because they dont have to understand or think about the OS, they can just write code.
It's really sad though because these are the people writing software, software installers, database apps, etc. But ask them how any of those technologies they are supposedly working with work, and you get a blank stare.
It's quite sad, actually.
I've worked with supposedly high end engineering programmers who will rant and rave at you that Linux is 'The Only Way', yet they cant figure out how to flip their machine from DHCP to static IP, or similar low level silliness.
Well, two options that jump right off the page at me:
1. Samba project tools: smbpasswd
2. Web based password change functionality. Commonly known as IISADMPWD, its a set of asp pages that are included with IIS (can optionally be installed on iis6/win2003). Part of every windows server install.
If the sharepoint is critical to your business, then you just bring up more than one, and run them as a farm.
Thats a bit of an exaggeration.
I've got several sharepoint installs at client sites who use Firefox, and it works fine for them.
The only difference I'm aware of is the wiki editor. On IE you get a nicer WYSIWYG rich text editor and you can optionally write wiki markup (much like html markup), on Firefox you have to use the wiki markup editor.
We've got plenty of PDFs in ours, and they work just fine.
Search engine works good for us, but we dont use it real heavily.
Also note that both Alfresco and Sharepoint are free. So they're both the same, zero cash, assuming you've got a windows server to run it on.
Sharepoint doesnt have its own set of user/passwords, they're domain accounts.
So how did you change your password on the domain before sharepoint? It'd be the same now.
Despite it not being a standard, and office not writing to ODF, there are a ton of available software packages that can read and write, import and export to both
OpenOffice and iWork come to mind right off the top of my head.
None of them work perfectly, but thats just par for the course. Even the MS apps dont work perfectly from version to version.
And there are third party apps whose sole purpose is to do batch conversions of one doctype to another. They're for pay, but they exist. They cover the precise 'I'm leaving' scenario.
Not to mention word docs you can export to rtf or pdf or html and take them with you. And PowerPoint you can export to PDF or HTML and take it with you.
Of course, with things like Excel and Access, businesses will have tons of code written in VBA inside these apps, and that'll never be portable. But then again, the scripting language inside these apps are never portable. Even if its in a broadly available language, the object model to interact with the document itself will be different from suite to suite.
And overall
But the bottom line is that in the real world, it IS possible to leave MS Office and take your docs with you. It does come with a fair amount of pain though (conversion, retraining, compatibility), and you have to measure that against the benefits you think you'll get (licensing fees, etc).
And please
This means that unless you were there, physically in the room, then you (and I mean the generic person you, not you specifically) do not have first hand information.
And there's a lot of FUD on both sides on this. All the brouhaha on the Norway vote
OOXML is a documentation of a current de facto standard. It's the documentation of a representation thats had more than 15 years of evolution, never once until the last few years with any thought towards public documentation.
And even with its flaws, its arguably more useful than ODF.
Look at the differences:
ODF is a nice, clean standard. However, its not really used in the real world outside of a very very tiny subset of the technical population. And despite it being a standard, if you use the format, then the vast majority of the world wont be able to read your documents.
OOXML would make a fairly crappy standard. However, its used by the vast majority of the people in the world. And despite it not being a standard, if you send it to arbitrary other pople, the likelihood that they will be able to read it with their current software is very high.
All that being said, I do think ODF is a nice standard (in theory), and I dont think OOXML should have been fast-tracked.
So, the likelihood of ODF being a meaningful document storage format over the next 10 years is very low, at least in my opinion. It's a great idea, but probably just wont play out in the market. Microsoft is just too damn good at marketing, sales, and deal-making.
However, even with that, its 'A Good Thing' that ODF exists. Because it puts competitive pressure on companies like Microsoft.
ODF being present probably wont sway most companies, they'll still buy MS Office. But it gives them a stick to use against MS and other companies. With it, you can always leave. This helps keep companies like that more honest and competitive than they would be otherwise.