Microsoft rescues XP users with emergency browser fix | Hacker News Reader

Microsoft rescues XP users with emergency browser fix

news.msn.com

39 pointsquackerhacker63 comments

XP is turning out to be quite the headache for Microsoft. Undoubtedly it is one of the best versions of Windows ever put out, but it's 13 years old now and not even corporate conglomerates have an excuse for not updating. Yes, the alternative to XP is only a meagre few years old and I understand it costs money to upgrade, especially a large-scale corporate network spread out over multiple locations, regions and continents.

But then we need to ask ourselves this: at what point does supporting an older operating system like XP cost you more money then it would to spend the money upgrading to a later version of Windows? Surely there is a point where you're spending more money on maintaining the older version on-top of added security expenses and unexpected expenses due to loss because of an undiscovered vulnerability.

My CNC machines aren't so easy to update.

If you need Internet Explorer updates for those you are using them wrong.

Do not underestimate the incompetence of someone who designed a big metal-cutting robot that runs Windows.

Why would you need to update your CNC machine anyway? Surely such a device is locked down and wouldn't fall prey to the kinds of vulnerabilities a corporate PC would. Presumably you're referring to the fact your CNC machine is setup to be operated remotely? The risk of someone accessing your CNC machine is probably lesser than that of someone infiltrating a corporate network and wreaking havoc on files, stealing information and credentials.

I'm in the same boat where I work with XP running on our CNC machines. You can only lock them down so much and still have support from the manufacturer. It's just a standard xp box like any office pc would be running. The machines are all on the office network and are connected to the internet. It is just too valuable to keep them off the network. The foreman who is operating the machine can look at shop drawings on the manager's computer. Someone in the office draws a piece of furniture in sketchup and can export a cut list directly the the CNC machine ready to run. Run a sheet of material on the machine and it deducts it from the materials inventory. etc etc etc

Sounds like a good use-case for a VLAN.

By all means let the XP machines communicate with the inventory server and the blueprints NAS, but don't let them route to the Internet.

For closed systems I think there is no need to upgrade.

What is the harm of not upgrading, what is the worst thing that can happen?

Of course, this is for a system that is not connected to the outside world.

Exactly, if it isn't networked then it is basically just an appliance. Just as you don't need to update the software in your microwave oven, you probably don't need to update your CNC machine. The exception to this is probably new features, such as supporting new file formats perhaps.

There might be a slight concern that the CNC machine could be infected by USB-borne malware. The worse case there is probably just infecting other computers through additional USB sticks. That, or industrial sabotage, but that sort of threat was relatively unheard of until recently.

> what is the worst thing that can happen?

An operator that plugs in an infected USB drive that injects a virus that will destroy your Uranium enrichment facility? Air gaps are not what they used to be.

The USB drive was presumably infected on a computer that is or could be up to date though, indicating the strong possibility that even if the CNC machine were or could have been up to date, it would still be vulnerable.

What if you get a head crash on a boot drive? Can you reinstall from the original DVD? Maybe yes, but will the authentication code process, which needs a Microsoft server over the Internet, still work? What about getting the XP updates from the data of the DVD you have to the last XP updates? Are those still available? Can you get actual copies of the updates you can apply later without Internet access?

One solution is to backup XP with NTBACKUP. Ever done that? Did you ever try to restore from such a backup? Did it work? My guess is no on all of these. I do XP backups with NTBACKUP occasionally, and have successfully restored from such a backup, but that was not easy -- there was some severe mud wrestling involved. Indeed, the backup I had was useless. How to make a backup using NTBACKUP that will restore? There are deep, dark secrets there!

Live backup tools like NTBackup is only needed if the machine needs to be availble 24/7.

If the machine can be turned off for a couple of hours (not unreasonable for a CNC machine that would require routine maintenance anyway), the easiest way to back it up is to take an image of the entire hard drive. No need to reinstall, no need to mess with fickle filesystem-level backup tools. Just dump the raw binary contents of the hard drive. When needed, you just dump it back to the new drive and reboot. Everything will work as if it was a routine reboot.

Boot up a GPartEd Live CD, plug in an external HDD to store the backup, and do "dd if=/dev/sda | gzip -c > /external/drive/backup.gz". (If you don't need the whole drive, just backing up the MBR and C: partition works fine, too.)

> the easiest way to back it up is to take an image of the entire hard drive.

You mean, say, just via XCOPY? So far for my boot partitions (on one hard drive, C: is an old Windows 2000 installation; D: and E: are Windows XP installations; and my usual boot partition is E:) I have NTBACKUP of E: on I: on a separate hard disk. I have successfully restored an NTBACKUP of a boot partition back to the same drive letter and boot partition and had a good running system. Yes, NTBACKUP can backup a live system, but what's really involved in backup and recovery of a boot partition, especially master boot records, I don't know.

No, not xcopy. By the time you run xcopy, Windows has already booted. It's too late.

My suggestion was to use the 'dd' Unix command (outside of Windows, of course) to back up the raw 0s and 1s on the disk, so that it doesn't matter what kind of OS you have or what kind of paritions you have. It's a very powerful method that lets you make a verbatim copy of every single bit on the hard drive, but of course that also makes it easier to fuck up something.

Thanks for the clarification.

On my system, especially considering the EOL of XP, the main issue is just backing up a 'boot partition', that is, a hard disk partition that is bootable, that has the usual XP directory tree, and that otherwise has as little as possible. For everything other than a boot partition, I have a workable system based on WinZip and DVDs. A 3TB external hard disk with some suitable software, maybe just XCOPY, would be better.

Since I have two XP bootable partitions, I could boot one of the two and get a backup of the other, bit for bit, if I could find or write a utility such as the one you mentioned for Unix.

But, then, much of this backup interest is recovering after a hard drive failure, and if I replace a hard drive then the new one will no doubt have to be larger than the old one in which case I wonder if the bit for bit, or record for record, copy I have would boot if copied to the new drive.

That NTBACKUP will backup a live system is convenient but not essential. What is important about NTBACKUP for me is that, finally, I got it to work, all the way, from backup to restore.

I also have a copy of Acronis which likely I should unwrap, study, work with, and use.

XP is 13 years old, but a viable alternative is only 5 years old; Windows 7 only came out in 2009. For many industries, replacements are measured in decades.

And I think Microsoft actually stopped selling Windows XP one year later, in October 2010.

They were going to stop selling it much earlier, but they became terrified about Linux getting a foothold in netbooks, which at the time were starting to become quite popular.

OSX v10.1 came out around the same time and nobody seems to say the same thing about Apple which dropped support years ago. Ubuntu is used by tons of huge corporations and its LTS versions are only supported for 5 years yet nobody blames Canonical. Yes XP runs enterprise but 13 years is more than reasonable to stop support. Its not like MS suddenly dropped this on these companies either, they gave years of warning to them but the companies didn't listen and now blame MS. And if these huge companies are planning their IT infrastructure decades at a time, in an OS industry that is only around 30 or so years old, they need to rethink their strategy and hire a new CTO.

I work at a telco. We just upgraded to Win 7 last week, and even then, more that 40% of machines are still not done, and likely won't be until late in the 3rd quarter.

"Do it as late as possible" is the motto when upgrading your machines doesn't earn more income.

Large bank here. our Win7 upgrade is mostly done, but still ongoing.

upgrade to Win7 which is already 5 years old...

You don't let your friend upgrade to Win8.

You'd be shocked how not-cutting-edge telcos are. We are an ISP for tens of thousands of customers, it's all running on XP/very old tech/hardware.

> Undoubtedly it is one of the best versions of Windows ever put out

It might be better to say "most successful," not "best."

OK, name the versions of Microsoft Windows which are better than XP.

Windows 2000. Faster in every way I could come up with, particularly local file and SMB network access, and not crippled to not support raw sockets like XPSP2 is.

But the XP end-of-life event has turned out to be pretty lucrative for them, with companies and government agencies opening their wallets in exchange for extended support. I wonder if they'll be enticed to shake that tree again more frequently in future.

I think this is one of those "Some money is too expensive" kind of situations, really. The drain on Microsoft's resources, and the knock-on effects of having increasingly vulnerable windows installs around, are probably worth far more, in the long run, than any amount anyone would ever pay them.

> Undoubtedly it is one of the best versions of Windows ever put out

I agree with this. XP was really the first version of Windows that came from the NT/2000 line (rather than the Win95/98 line) that was consumer-friendly. It's hardware requirements were (and are) pretty meager. I know I only upgraded once Windows 7 came out and it became apparent that (unlike Vista) it didn't completely suck.

> at what point does supporting an older operating system like XP cost you more money then it would to spend the money upgrading to a later version of Windows?

Going back to the 90s, it used to be that a new computer was essentially useless within about 3 years, such was the thirst for more power, memory and storage.

But something happened about 5-10 years ago. Ever since then, pretty much any computer you can buy is "enough" for most users, particularly corporate users (which largely comes down to running a browser and MS Office). Microsoft became successful in an era where they could rely on increasing market size and churn to get rid of their old versions. Not so any more.

Some will say "just spend the money and upgrade" but think about it this way: upgrading to Win7 will mean not only buying licenses but in many cases new computers. What if those computers don't actually offer any significant cost savings or improved capabilities for most workers? Think of a hospital that could spend $X million upgrading all their PCs or they could spend that money on an MRI machine or something else which saves lives. In comparison, a computer upgrade can be hard to justify.

One of the reasons I switched to Chrome was for the automatic updates. It's also why I switched all my family members. No one wants to manually run updates. No one even wants the Firefox version of "Install update now?" when you start it (by virtue of you starting Firefox means you want to do something other than wait for it do stuff).

Microsoft needs to integrate the philosophy of continual transparent updates. Unfortunately they're addicted to the revenue of new license sales.

Just recently I looked into buying Photoshop. It's kind of a must-have for digital photography but it's really expensive (for personal use). I discovered that last year Adobe had switched from a license model to a subscription model. Some complained about this. Adobe was famous for doing things like not updating the RAW converter for previous versions of Photoshop to force you to upgrade.

Thing is, their subscription is as little as $10/month. Normally these companies want to recoup the purchase cost in <12 months or something similarly unreasonable. I'm perfectly happy to pay $10/month for Photoshop.

(Some) consumers are willing to pay a service fee of $15/month for TiVo. That's after purchasing the hardware and before paying for any actual content. Why isn't the market just as willing to pay a similar price for something that's meant to protect your private information, credentials photos, identity and financial data?

To me, this is the real power of, say, ChromeOS. Many people only need a Web browser (particularly at work) and don't want to have to spend the time, effort and money to manage their corporate deployment.

And of course MS aren't doing themselves any favours with the clusterfk that is Windows 8...

> But something happened about 5-10 years ago. Ever since then, pretty much any computer you can buy is "enough" for most users, particularly corporate users (which largely comes down to running a browser and MS Office).

I think it was the internet.

Upgrading from XP is worse than that, much worse than just the money.

I'm typing this on XP. For Windows 7, I have a full, retail purchased, on DVD copy of Windows 7 Professional. Okay, but so what? To install Windows 7:

(1) Find a boot partition -- actually I do have one available.

(2) Wonder if Windows 7 will run on my hardware with a 32 bit processor and a 1.8 GHz clock. E.g., I've not seen a statement from Microsoft on what hardware Windows 7 needs.

(3) Especially wonder what of my crucial old software will run on Windows 7. E.g., I don't want to get into the 'security mud wrestling' of Vista.

(4) What's new in Windows 7 I will have to waste time learning?

(5) What is in XP that I am using that will not be in Windows 7?

(6) What of my hardware, e.g., a FAX modem, Windows 7 won't work with?

(7) What if anything I could want will I get from Windows 7 I don't have with XP? I haven't seen any such list from Microsoft.

(8) As far as I can tell, the main reason to install Windows 7 is just so that bug fixes will continue to be available. This matter of bugs is getting too convenient for Microsoft's bottom line.

(9) What about major software such as SQL Server I am running now on XP? Installing SQL Server was massively wasteful mud wrestling on XP, and I fear loss of some days of time getting SQL Server running and, then, loss of some weeks of time, literally, getting the system running again after the installation of SQL Server ruins my system requiring me to format a boot partition back to empty, reinstall XP from the DVD, install XP updates, no longer available?, that is, maybe the 'updates' I have still just pull the actual updates from Microsoft. Yes, SQL Server installation was bad enough it wiped out my favorite boot partition for XP and all the software I'd installed on it. My experience is that installing SQL Server is a big threat to all other software on the system.

The issues are not just buying Windows 7, guys; I've already done that, and the copy is sitting unopened on a bookshelf.

Microsoft: Document your products. Tell me what I need to know. Be clear. I've already bought your product and, still, due to lack of enough documentation am scared stiff to use it; I fear a disaster to my work. Did I mention you need to document your products, tell me what I need to know, and, by the way, document your products and tell me what I need to know? Did I mention those things?

> (2) Wonder if Windows 7 will run on my hardware with a 32 bit processor and a 1.8 GHz clock. E.g., I've not seen a statement from Microsoft on what hardware Windows 7 needs.

http://windows.microsoft.com/en-us/windows7/products/system-...

I read it quickly; it was somewhat helpful. At one point they implied "if you like your XP applications then you can keep your XP applications" or some such promise. Good.

XP is over a decade old and you seem tech savvy enough that I can't tell if you're joking or not. There are plenty of legitimate reasons for not upgrading due to hardware or certain software but your post reads as either being stubborn or a joke.

Most of the questions you asked are readily available via a simple Google search.

I'm not joking, but Google searching all that stuff might be a lot of work and, still, might not really tell me what is going on.

I just tried a little, and it looks like Google is working better than I expected and more like you suggested:

(1) At

http://indianexpress.com/article/technology/technology-other...

"You’ll still be able to run XP computers and install past updates. If you need to reinstall XP from scratch, you can do so if you still have the discs that came with your computer."

Well I do have the original install disk and the codes. Hmm ..., but my install disk is an 'upgrade' from Windows 2000, so if I lost my boot drive, with three partitions, one for 2000 and two for XP, maybe I would need to install 2000 before installing XP. I have the 2000 media and codes but would wonder if the Microsoft authorization server would still be up.

(2) For my most important software tool, my favorite editor KEDIT, the first search result was from Wikipedia and clearly stated that KEDIT will run on Windows 7 64 bit (presumably both 7 and KEDIT will run on a 32 bit processor).

So, good news so far.

I intend to buy another computer fairly soon. Then computer A will do its backup to disk space on computer B, and B to A. Main point: If computer A fails, then I still have computer B still working and with all the data on computer A. So, fix (or replace) computer A, copy the data back from B, and continue.

The reason I got a copy of Windows 7 is that it will be the OS I will run on my next computer, computer B. Then with computer B just sitting there, I will be able to do experiments in installation, configuration, system management and administration, etc. while computer A still has access to the Internet for Google, MSDN, various fora, etc.

Then via the Microsoft BizSpark program, I will install on computer B, on its own disk partition, the most recent, appropriately sized, copies of Windows Server, IIS, SQL Server, etc. That will be my Web server for my startup.

For now, my computer A with XP I hope will keep running until I get computer B running. Why? I'm concentrating on the rest of the software development for my startup and don't want to take a detour into more 'system management'.

Generally, though, the main way I have in mind to solve problems of software installation, configuration, compatibility, etc. is just to have a second computer (so that I will still have a working computer) to run experiments on until I discover what the situation is. E.g., I suspect that the way a high end user of, say, SQL Server gets installations done is just to have a computer with an appropriate version of Windows Server and then just install over and over until everything is smooth, document is, and hopefully automate it.

Net, I need a second computer. Soon.

I have a copy of Acronis 'True Image' or some such; maybe I should dig into it and get a 'true image' of my Windows 2000 partition in case it needs to be there in case the boot drive fails and I need to reinstall my Windows XP 'upgrade'. Then I will be able to reinstall my NTBACKUP image of my favorite XP boot partition. I.e., my computer has three hard drives, one for boot partitions, one for disk to disk backup, and one for the real work. I'm well enough backed up that if any one disk fails I will still be able to recover (at least if I grab a copy of the Windows 2000 partition).

In your case, it probably won't make sense to upgrade your current computer. The reality is though, eventually you'll be forced to upgrade. There are numerous components inside your computer that could fail at any time, and your priorities will change as software versions are progressively iterated.

Given issues (3) and (9), my recommendation is to eventually buy a new computer and perform a wholesale virtualization of your current computer onto it. This also gives you the choice of switching paths to an entirely different operating system (e.g. Mac OS X or Linux) rather than to Windows 7 or newer.

The cool part here is that you can start isolating components -- for example, building a virtual machine specifically to run SQL Server.

> In your case, it probably won't make sense to upgrade your current computer. The reality is though, eventually you'll be forced to upgrade.

In a few months, I intend to get a second computer as the production computer for my startup. First, before going live, I will install Windows 7 and see what it does/doesn't do. Then I will install from the Microsoft BizSpark program appropriate copies of Windows Server, IIS, and SQL Server.

I hope to keep running XP on my present computer at least until my startup is live. Then, with some Windows 7 experience, I may install Windows 7 on my present computer -- I have three boot partitions and can put Windows 7 on one of them. If I really like Windows 7, then I will just junk XP.

Have you considered a hardware upgrade and switching to a virtual OS workflow? It sounds like it might be worth your time to consider.

I have a 4 year old AMD cpu (not sure which one), 16 GB RAM (cost about $80) and SSD (best upgrade in years)

my bare metal OS is win8, on which I have minimal software installed. I use multiple VMware images for software I don't use often or which installs too much cruft which I don't like slowing down my machine.

For example, I only need VS occasionally. For some reason even a minimal install of VS comes along with 10s of MS service packs SQL server IIS etc. Basically junk that I don't need. So now it lives in a VM that I might fire up once in a while.

Benefits of a VM:

  -- Easy backups
  -- Snapshots

On a mid range system VMware works flawlessly and gives seamless access to host resources such as networking and USB etc.

Running a XP VM would offer multiple benefits such as access to your old software and VM sandboxing for security issues that go along with running an EOLd OS

I understand virtual machines from my old IBM days, but so far I have yet to do anything with them on PCs. Likely for my startup, as the work moves along (today made good progress on some de/serialization issues -- wrote some small test programs that showed that de/serialization was working like it should; now to find out why it was not working in the code of a Web page talking to one of my 'servers', now just software on the same computer but intended to be on a separate computer, TCP/IP communications in both cases).

The sand boxing security sounds good. Maybe in practice for years VMs are the only easy path we have to security from 'malware', etc.

> Have you considered a hardware upgrade and switching to a virtual OS workflow?

I intend to buy a second computer fairly soon. I will install the copy of Windows 7 Professional I already have and then get used to the computer, arrange that each of my computers can copy backup files to the other one. And/or get a hard drive connected vis USB and some appropriate software, or maybe just use XCOPY or WinZip.

With two computers, if one of them is down for some reason, then I can still get to the Internet with the other one.

The second computer will also be the first server for my startup, to go live fairly soon -- the software is all ready except I just need to debug the code for the last two Web pages.

So this second computer will also have appropriate copies from the Microsoft BizSpark program of Windows Server, IIS, and SQL Server for production for my startup.

> at what point does supporting an older operating system like XP cost you more money then it would to spend the money upgrading to a later version of Windows?

Well... when expensive capital equipment runs XP, that point is essentially "never."

For example: Old versions of LabVIEW run on XP, but not 7. A license for one desktop runs $10-50k, depending on options. Lots of sciencey equipment requires software that only runs on XP. Want to upgrade to 7? Ok, here, buy a new $100k laser / $20k UV-vis-IR spectrometer / $10k multi-channel scaler / $10k oscilloscope / whatever. Another user mentioned his CNC machine (those aren't cheap).

I mean, you can do that, but $10k buys you a lot of extended support (which is curiously unavailable to everyone who isn't a big business). I'm sure most university labs would jump at the chance to pay a few hundred dollars per machine to support instruments incompatible with 7.

Seriously. A few years ago (2010ish) I was roped into helping out with a microprobe array that would have cost about $150k to replace, whose controller software would only run on Windows 95... not OSR2, meaning that USB wouldn't even work (we used ZIP disks to move sequence files on and off of it). The fellow in charge of the thing would grab systems being surplussed from elsewhere in the department that might run Win95 and clone to their drives... when I left he had 4-5 spares in a closet for the day the current controller died.

EDIT: And we did attempt to virtualize it, but at least at the time were unable to do so successfully (timing issues, I think).

I don't think custom support is that cheap. There is typically a floor of at least $100k for Custom Support Standard.

It's not. But from what I've heard, it's much less than that per seat.

I'm saying that the PIs of most university labs would gladly pay several times the retail price of XP just to be able to continue having support for multi-thousand-dollar instruments.

But I think this nitpicking over a relatively minor point--the important part is that upgrading from XP to 7 costs tens or hundreds of thousands of dollars per computer for some people.

Edit: In case you were referring to the sentence "$10k buys a lot of...," I can see how my intent was unclear (and I was sort of talking about both small and large labs at the same time). My understanding was that there is a minimum buy-in for extended support (something like $100k per year), but that such a contract would cover hundreds or thousands of computers. That's on the order of $100 to $1k per PC.

A medium-sized university lab might have dozens of instruments costing on the order of $10k to $100k. Now, paying $100k to support $100k worth of instruments for a year doesn't make too much sense, but paying a few hundred dollars per $10k instrument does. Even at $1k per PC per year, avoiding an instrument purchase buys a least a decade of extended support. It's just a question of getting a critical mass (or, by convincing Microsoft to open up the program to smaller organizations).

It's breathtaking to see such a blunder: a Forune 100 company telling nearly 20% [1] of its installed user base, it's second largest desktop OS customer base, to stop and make a financial decision about what to do, instead of giving them a low-cost, path of least resistance option to continue in some way similar to the status quo. For many people still on XP, the users are not technical enough to understand how to do an OS upgrade or how to migrate user data to a new machine, let alone understand why they would want to when the current system appears to work fine.

I'm surprised that they're not trying to monetize their current XP user base with some sort of "XP extended support" fee-based subscription so they don't force users to look elsewhere for a desktop OS -- between Windows 8's blunders and a Mac, I suspect many of those XP users will consider a Mac. Or simply by offering a ~$40 upgrader app to get to Windows 7 "lite" for XP users that works on the same hardware and drivers...

[1] http://techcrunch.com/2014/04/02/discontinued-windowsxp-stil...

Microsoft is going to have to take a stand and make this the last XP patch (for anyone not paying for extended support like certain governments) forRealThisTime if they ever want to move on.

In this case, I can understand the reprieve, if only because it came like 20 days after the EOL. In fact, you almost wonder if this wasn't publicized until after the EOL in an attempt for malicious parties to do maximum damage.

But this won't be the last hole. Others will exist and if Microsoft doesn't stop from now on, it won't disappear until web standards change to such an extent that all versions of IE that run on XP won't render modern sites at all.

Kind of a good reason, in retrospect, to decouple the web browser from the OS, huh?

> Microsoft is going to have to take a stand

And risk having all those corporate drones move to Linux, Android, ChromeOS or anything non-Microsoft or, worse, Microsoft-hostile? Risk spawning a non-Microsoft ecosystem around vast IT-deficient corporations (any reasonably managed IT infrastructure has moved on long ago) that would undermine their other side businesses?

They'll never do that.

No other desktop is suitable for the enterprise desktop environment out-of-the-box. Much as Windows is maligned, OSX doesn't have the range of tools available for enterprise use, and free desktops still aren't quite there for regular users en masse.

There may be many Windows-specific functionalities missing, but it's certainly possible to work around those. Why do you think OSs like Ubuntu or Red Hat are not suitable to be used as enterprise desktops?

I worked with in the real world at a non tech focused company regularly struggle with WordPress, Twitter, and Email. Just because HN is tech focused doesn't mean everyone's coworkers are.

My housemate is a regular user (not power user) and a known potterer. He loves figuring things out and patching things together, and wanted to try a linux desktop, so I set him up with one. He didn't want me to help him out, as he wanted to find things out for himself, to try it without me coming in and 'just fixing it'. I watched him struggle for months, then he admitted defeat and went back to Windows. He still (theatrically) shivers if we discuss the topic.

I've seen this experience is seen in the wider desktop space - the only time a free desktop is suitable for low support usage is when it's so locked down it's not really a proper desktop environment anymore; that locked down, it's more like an appliance. When something goes wrong with a free desktop, it's not just a matter of figuring out what went wrong, but which are the right steps to fix it. This error on Gnome, will this fix I found for Fedora work on Debian? Is the problem in Gnome desktop or in the distro?

My colleague once had a problem with something in Ubuntu, and found that there were four different solutions depending on which of the last four releases were in use. When your changing releases are years apart (Windows) this is less of a problem than when they're 6 months apart. I was pleasantly surprised when I loaded up ubuntu 14.04 and saw that it looked the same as before.

Note that I'm not in the enterprise environment, only that I've had to help a lot of people with changes from OS version to version, and understand the need for stability... which is a reason why the path Win8 took particularly puzzles me, given Microsoft's strong position in the business world. I think that the free desktops are getting there, but haven't quite arrived yet.

Your anecdote may tell us something about the "typical grandma end-user" situation, but it is totally irrelevant for situations that arise in large "enterprise" IT facilities, where the admins set everything up for you (in some central, network-booted/installed image), and the users don't really have to figure out anything. Configurations are centrally managed with cfengine etc.

Where I work we have around 10k RHEL workstations, and they are all centrally managed. Users neither install nor uninstall software, they don't set up printers, they don't set up their own hardware configurations, ...

Another example is Disney animations/pixar, where all animators use RHEL workstations.

Your colleague may have given up on ubuntu because he couldn't get "desktop dungeons" or "plants vs. zombies" to run without windows, but in a coorporate setting, management does not generally consider that a priority.

You make valid points, but please stop saying that naive users are grandmas. I even specified he was my housemate - how many elderly people do you know share a house? It's even the wrong gender. The world's non-power computer users are not limited to grandmothers (who seem to be all about desktop use) and toddlers (who seem to be all about iOS).

Also, my colleague didn't give up on Ubuntu. He just noted four different ways to fix the given problem in four consecutive six-month releases.

I have another anecdote for you. My mom, a 79 year-old retired tourist agency owner and occasional tour guide had a Windows machine. When I gave up on fighting viruses on her Windows XP machine, I moved her to a Mac running OS9. Then I set her up with an Ubuntu box.

She does her upgrades through the GUI (after the machine prompts her), runs mostly Firefox, imports her photos into Shotwell, shares them with her friends (on e-mail - she doesn't like Facebook), writes letters (and mail the printed copies) and seems generally comfortable with that.

She moved from Gnome 2 to Unity without thinking twice. She prefers Unity.

My housemate didn't have problems with browsing or photos either. Things like a pair of bluetooth headphones, which had linux support, were a nightmare (for him) to get sorted. The desire to use a pair of bluetooth headphones really shouldn't require a power user.

It's probably better for Microsoft if those companies did move to an operating system like RHEL. They can still sell Office 365 subscriptions, and don't have to be responsible for security patches any more.

Except at a certain point, none of their products will work with XP and as they shift from perpetual to subscription licenses and service contracts, the value of a shop that hasn't paid in X years is disappearing all the time.

Losing those customers might be a good thing, at least if urge can still retain some business.

They gave the OS 13 years. Ask RedHat to do that. Or Google. (not even going to mention Apple for mass enterprise installations, that's not their target anyway).

Moreover, Microsoft can still argue that it's cheaper to migrate to. New Windows than a brand new system. Or you know, try to offer a VMWare or Citrix style solution to cut down on machine/licensing costs.

I'm guessing that those still using XP after EOL are those who don't care about security.

It's pretty obvious that they did this because people were being advised not to use IE - without specifying a version.

If they didn't update IE on XP then even IE on Win 7/8 would get hit by advice not to use it.

Right. Of course, the fact that that advice came from DHS, which still has a ton of XP/IE6-based internal apps probably made it that much worse. The government is a big client. Having them move to another browser, even if they haven't agreed to upgrade all the clients to the new Windows yet, is just a bad idea.

What I'm wondering is how many times is this going to happen? Where they're more-or-less forced to continue emergency patch mode for major vulns?

Out of support should mean out of support -- no one off patches.

ETA: Organizations will use these 'emergency patches' as yet another reason not to upgrade.

Reply on news.ycombinator.com