As a consultant, I love the fact that they've managed to take something simple and reliable that almost never has problems (wires connecting keyboards and mice), and replaced it with something prone to all sorts of failures... dead batteries, interference, poor range, and generally poor manufacturing quality.
Every wire they eliminate equals more billable hours for me! I'm all for wireless peripherals!
-R
ps- In all seriousness, I make sure my clients understand this before purchasing one...
I would have agreed to you up until V10. As a consultant I administer MANY windows domains, and I've had a significant issues with V10 that I never had with earlier versions at a number of customers, especially on the server (the worst place). I'm using Trend Micro SMB now... and couldn't be happier. They've even built in automatic removal of other A/V software (including Symantec) into their own client installer. My only gripes is that their spam filtering stinks (so does Symantec's), and it doesn't support IMAP at the client.
Nowhere have I said that a new windows install is never necessary, and I am certainly aware of the virtues of a clean install. I'm not saying I never do them, but rather that in most circumstances that I find myself in as a consultant, I rarely find them necessary or in the customer's best interests. BTW- I'm speaking exclusively of Win2k and XP here... the story is completely different for previous versions, which I rarely encounter these days. If the registry is corrupted and there's not a recent system restore backup available, or if they hard drive has failed badly enough to prevent grabbing a good image, or of the customer truly doesn't care for their configuration or data and there's a quick OS restore function... then I'll happily do a re-install.
But what gets my goat are consultants who do them because they think they know best, without taking into account the customer's actual situation and needs. I know this happens frequently because of the horror stories I hear from my clients. I feel confident that I'm not leaving behind the same unwitting "trail of destruction" because my business is driven entirely at this point by repeat business and referrals- I'm actually turning away business these days. And my rates are higher than average. If I were not meeting my customer's needs, I doubt they'd be coming back as they do.
I'll offer two possible explanations for what clearly is a massive cognitive gap between us: 1. I guess I really just don't find cleanups that difficult or time consuming. If they took 3-4 hours then I'd be agreeing with you. But what I commonly find is that, especially with recent fast machines, I can do a cleanup in 1-1.5 hours, including install of decent protection software. I'm really not kidding. And I'm not getting called back for re-dos (which I don't charge for) because the machine is still infected. When I compare that with the cost, risk, and time involved in a rebuild, it rarely makes sense to do the latter.
2. There are aspects to my business which may differ significantly from most. First, I work in a rather affluent area, and typically on-site... I'm not a "carry-in" shop. Most of my clients are not "grandma" types who just do e-mail and web... many are work-at-home professionals, and thus have comparatively elaborate systems. Another generalization... they prefer "full service" rather than "fast and cheap". Thus, when I do a rebuild, I'm not leaving folks with a bare windows install- rather I'm putting the machine back as close to the functionality they previously had, which typically includes office suite, e-mail client (configured for their host), A/V, anti-spyware, backup, printer drivers, digital camera / photo software, and personal finance software, just to name the most common things. So a rebuild is typically a more expensive, elaborate affair than the typical carry-in shop does.
Anyway, I would love to take you up on your offer of a "rebuild" vs "cleanup" face-off... sounds like fun! Name the time and place (as long as it's in Metro D.C....). You give me a badly infected machine. I'll give you a typical "real world" configuration scenario from one of my customers. Of course, to be fair, you get to wait while they track down the install media in a badly-organized file drawer with software dating back to 1992, only to discover their daughter probably took it with her when she went off to college. And call their accounting software company for the product key they've lost. And find their ISP dialup information they use when they're on the road but forgot to mention before you started the re-install. And... and...
Well, I didn't think I was doing anything that unusual, but given how many people are shocked by my claims of removing infections quickly, maybe I should start a training course! See one of my earlier posts in this thread for a summary of my technique... the reason why it doesn't take so long is that I don't use long-running (and often inneffective) file scanners.
I am constantly aware of how important it is to make a computer repair cost-effective... that's one of the reasons I charge a flat two-hour (not four) rate for cleanups, so that if I run into something that's truly novel, I can take the time to (a) do the right thing by the customer and (b) educate myself so that I'll know how to do it in the future.
But for more serious / time consuming problems, I do have a threshold for which it's "not worth" fixing a computer... what you've said is absolutely right in some cases. I don't do what I call "band-aid" work... I want to be sure that the customer is going to be fully satisfied with the result, and it just doesn't make sense to invest a lot of time into older machines.
I offer a money-back guarantee... if the infection predates my previous cleanup, the second cleanup is free.
If someone is stupid enough to get re-infected after sitting through the well-honed lecture I give while cleaning up their machines, I'm happy to take their money again. It's happened twice in memory, and both of them were teenagers of wealthy parents that really didn't care. Most "normal" folks are scared to death of getting infected, and get religion after I tell them about keystroke loggers and such. Although getting the right software on the machine is a big part of staying safe, modifying user behavior is the really important change.
There's no way Ballmer (if he's doing his "day job") has any real experience cleaning up malware. In fact, most regular corporate IT people don't really either, because (as so many have pointed out) a re-image is a much better route, if you've got the infrastructure and organization in place in advance.
There certainly are engineers at MS that could run circles around me, and I'm sure can easily clean up any of the widely-distributed infections my clients are likely to get (as opposed to the scary 0-day stuff used in corporate espionage).
But can you imagine what a mess that machine must have been in after Ballmer trying to fix it for two days? I doubt the infection was the real problem by that point. But who's got the cajones to tell that to him?
I've actually been to Redmond a few times- and not for conferences. The MS campus is really nice, but I wouldn't want to work there.:-)
This sounds like a justification... "if they haven't made the appropriate precautions... they get what they deserve".
This really misses the point. The scenario you describe is so far removed from the way non-geeks (ie, most people) understand and interact with their computers to be laugable.
This is squarely Microsoft's and other vendor's fault:
1. For not including recovery CDs with every computer (don't make the users burn their own recovery media) *inside the case*, where it can't get lost.
2. For writing crappy software that doesn't store files in the profile, exclusively.
3. For putting critical data (e-mail, browser configurations, etc...) in literally hidden directories that an average user doesn't even know is there. For putting user preferences in a registry file that can't simply be copied by the user to back them up.
4. For creating a convoluted software installation system, which prevents one from simply copying a folder to move an application from one computer to another.
5. For not making easy-to-use backup a truly integrated part of the OS, that you have to work hard to avoid and gives *useful* error messages when it stops working.
I could go on, but users are already being punished every day simply by having to use these incredibly confusing, unreliable machines for things that are really critical to them. The last thing we should do, as the technological elite, is blame them.
Actually, I really do know what I'm talking about. The theory is simple: all malware has to get launched. There are a limited number of places in windows something can stick itself to get launched at boot time. This is what tools like HijackThis and autoruns allow you to look at. If you use them frequently enough, it's pretty easy to pick out what doesn't below.
Kernel rootkits made this a bit more challenging, since you can't trust what Windows is telling you about what you see through these tools. The answer to that is to boot to trusted media (like a BartPE disk) and check out the potential autostart locations from that. Since the malware isn't loaded, it can't make the system lie to you.
Thanks to the newer kernel rootkits that hide files in NTFS alternate data streams, you have to scan for those as well from trusted media. But that's doable- there are tools which work from BartPE which will enumate files with ADS streams, and you can also check for ADS paths in the list of drivers which load a boot-time. Then when you think you've got it clean, throw a sniffer on the machine's LAN connection and see if anything unexpected happens.
Thus, it's really quite possible to manually discover and "kill" most infections with a reasonably high degree of confidence, if you have enough practice and experience. The level of assurance isn't high enough for a corporate server, but it's usually sufficient for home users and in small businesses, considering the cost of a from-scratch rebuild.
I do admit I'm a bit more qualified than most techs who do this... I've got a computer science degree, and have been doing IT stuff professionally for 14 years. You've got to have a pretty solid grasp of windows internals to do this well.
No kidding. I happen to be of the opinion that most modern UNIX variants do this wrong as well... back when I was doing UNIX sysadmin stuff at Rice University, we had a beautiful software installation convention called/usr/site... the upshot was that by forcing all applications (at compile time) to use version-specific install paths to refer to their own support files.
For example, say/machinename2 is your software install partition.
Each app is installed in a directory/machinename2/appname-version.
There's a directory called/user/site which has symlinks like this:
Same thing for/usr/local/lib,/usr/local/etc... etc... ALL of the references to the app or app files in the system are symlinks to/usr/site/appname/...
The app is compiled to refer to use the install prefix as/usr/site/appname-version.
The beauty of this sytem is that you can easily have multiple versions of an app installed simultaneously. Need to put a new version into production? In most cases, all you need to is change/usr/site/appname to point to a different/usr/site/appname-version. Same thing with a "downgrade".
This is actually a slightly simplified version of what we used, since it can be extended to "do the right thing" in environments where you're supporting multiple processor architectures off a single/machinename2 software installlation NFS share. If you google/usr/site, you can find complete documentation here.
For slightly larger companies, you're absolutely right. Imaging is great where it's practical. But for very small companies (15 computers) and home users, it rarely is. They've typically got a scattershot collection of machines with OEM licenses, different HALs, etc... Most small businesses this size are only buying 1-2 machines per year. Moreover, because of the diversity of roles within the company, few computers have the same software installs... the bookkeeper needs Quickbooks.
In addition, people need to have privileges to do their own maintenance and software installs, since they don't have on-site IT help to go to when they need their new PDA hooked up, or want to install an update to their accounting program. So no matter what I put on the image, it's going to be badly out of touch with what the user actually needs.
The good news is that a backup system that can back up every bit on every machine is affordable, since the total data storage is usually well under.5TB. I push those *hard*. But even then a full system restore is a pain because most backup systems don't cope with the proprietary boot sectors and hidden system partitions found on most major OEM installs these days. And a lot of small business folks are cheapskates and have to suffer a hard drive failure before they recognize the need for a real backup system.
I'm a consultant who helps small business and home users. I can't tell you how many times I have talked to customers who (in the past) have had another tech come along and do a re-install without understanding all of the implications.
There is value in a machine's configuration! The customizations, tweaks, and even icon arrangements people create to make their systems work and lives easier are time-consuming to recreate, and there can be a major loss of productivity if they have to re-do it all from scratch. I'm a professional, and it's not uncommon for it to take me 3-5 hours to do a good job of getting all of the software, utilities, and configuration changes done for a typical business machine. Just because you can rebuild your own gaming rig from scratch in two hours (because you do it once a month) doesn't meant that this is a course of action that makes sense for everyone.
This is why I always recommend *full* backups of the entire system... not just "important" documents. And it's why I do a full re-install as an absolute last resort. I can count the number of re-installs I've been forced into in the last *year* on one hand.
The good news is that if you know what you're doing (unfortunately many techs don't) VERY few problems require a rebuild. It's very possible to clean off even the "worst" infections fairly quickly, with high confidence that everything is gone. I charge a two-hour flat rate for *any* infection cleanup (including kernel rootkits), and that usually works out to my advantage. Hard drives often have only failed in a few sectors... I commonly am able to image the failed drive to a new one, and repair the windows install using a combination of sfc, system restore, misc subsystem fixes, and (in the worst cases) a repair re-install.
The benefit to the user is that they get their machine back *exactly the way it was*, the same day, without a large repair bill. The benefit to me is that the customer is happy and calls me back the next time they have a problem... instead of cursing me the whole time they are trying to rebuild their system the way they had it.
If you are a tech and haven't learned this stuff, you are doing your customers and yourself a disservice.
There's a link in the article to the company's website. They've developed a motorless feedpump system, and there's a rather elaborate flash animation that describes specifically how it works, and several possible sources of energy... solar water heaters, sub-boiling geothermal sources, or even wood stove waste heat. The point is that they think it can work efficiently with a 50 degree temperature differential above ambient temperatures, which is pretty easily achievable without a lot of elaborate heat/cold storage.
The point of their system is not to be more efficient than solar panels, but to be MUCH CHEAPER. We don't have a shortage of energy from the sun... we have a shortage of cost-effective ways to harnass it.
A great number of the capababilities in their primary product - ERD Commander - have now been duplicated in BART PE-based bootdisks like Ultimate Boot CD for Windows (UBCD4Win). I recently evaluated ERD Commander specifically against UBCD4Win, and in the hands of a well-trained user the latter did most of what we needed to do. It wasn't as slick, and there were some ways in which ERD Commander was much better, but the very high price of ERD Commander (at least for a small consulting shop like mine) made it very difficult to buy with the marginal additional functionality it adds. On the other hand, without UBCD4Win, it would have been quite attractive.
I'm guessing we're not the only ones noticing this... so the sale makes sense. Better sell out to MS rather that try to compete with 'free'.
-R
ps- Note my caveat "did most of what we needed to do". I'm not claiming that UBCD4Win is an ERD Commander replacement for everyone...
If you're (like me) one of the, umm, fortunate souls who get to clean up rootkit-infested machines regularly, there's a tool you should know about: LADS, for "list alternative data streams"
I haven't tried it yet, but it looks like it should work from a win32 bootdisk (like BARTPE). So you should be able to boot from a clean win32 environment and scan the computer's hard disk to find any files with ADSs. Fortunately, use of this feature within NTFS is not widespread, so malware should stand out pretty obviously.
Netstumbler only shows you other *wireless networks*. Wi-Spy shows you *all* RF interference in the 2.4 GHz spectrum, such as that caused by cordless phones, microwave ovens, etc...
I'm a small office / home office IT consultant, and I bought one of these a few weeks ago after stumbling across it on ThinkGeek. It's fabulous for my needs, which are simple: figure out if interference is the reason someone's wireless network is flaky.
Wi-Spy does a great job of doing this. I fired it up at a downtown client and saw there was a strip of intense interference down in channel 1. Moved them up to 11- problem solved. I've also done some tests at home... it's very easy to tell the difference between a microwave, spread spectrum phone system, video sender, and other wifi networks... they have rather distinctive appearences in the graphs Wi-Spy produces. Now that I know what they look like, I can take an educated guess, where before, I was grasping at straws.
For those of you getting your panties in a wad about it not matching a $5000 spectrum analyzer: Duh? Of course it doesn't. But that doesn't mean it's useless... there are a lot of folks (like me) for whom the cost of a "real" spectrum analyzer is completely unjustifiable. But I can spend $100 easily, and *for what I do*, which is occasionally troubleshoot SOHO wireless networks, it provides most of the functionality I need.
The really interesting fact is that this thing defines an entirely new product category: inexpensive spectrum analyzers. I would really like to see what could be done for $500... that's still an order of magnitude cheaper than the existing solutions, but I bet you could add a bunch of features.
You can't just add access points to increase capacity... the limitation is the radio frequency space available. Remember there is only room for 3 wifi channels (1, 6, 11) in the 2.4GHz spectrum. Add a forth into the same space, and you're just stepping on the others and causing interference. Of course I'm assuming 802.11b/g here, as 802.11a has 20 distinct channels.
The other issue that people have mentioned is outside interference. Microwave ovens can be a real bummer. So can the little cordless 2.4GHz headsets executives seem to like. And you better hope nobody sets up a 2.4Ghz video sender for their security system in the vicinity. Or a nearby cell tower, or radio station. You could be working perfectly for a year, and then suddenly have your network permanently broken by something completely outside your control or ability to change.
There's a reason you don't hear of many people doing this.
I've got FIOS and my traditional phone line now runs over the fiber They completely removed the existing phone box on the house and put the ONT in it's place... it has a similar block for wiring the house phone wiring to it. This is why the FIOS install comes with a UPS- so that your phone line will keep working if the power goes out. They didn't actually tear out the copper wire from the ground, but hooking it back up would be a project.
However, he's gone a bit too far with the regulatory fear-mongering. Yes, the fiber line is excempt from the regulations passed in 96 that forced the phone companies to allow competitive access to the copper that enabled Covad, Northpoint, and others to start building out DSL networks of their own. However, the FIOS phone line is still a tariffed / regulated service, with the same Public Utility Commission oversight as before.
I know of at least one ISP which does this- Cox cable in the N. VA suburbs of DC. Several times I've cleaned out home users computers that were infested with malware, only to discover that their Internet access was still being blocked. It's rather difficult to get them to remove a block quickly once it's in place... the L1 support folks disclaim all knowledge and assume it's anything else but a block, but if they talk with higher-level staff they eventually realize that's what's going on. Anyway, that's the way it was the last time it happened, which was probably a year ago.
Well, I think you may have missed my larger point due to some of the details. Sorry I mistook you for a Windows noob. Yes, I've done my time with old Sun machines too... nothing like a little VME backplane jumper reconfiguration to brighten your day!
Just a few points in response:
1. I'm responsible for supporting several hundred WIndows machines in several dozen very small companies and scores of homes. Although one can certainly screw up an XP or 2000 machine such that you have to reboot daily, this simply is no longer the norm for properly managed machines, even if you have a relatively unskilled user running as a local admin. For most users, stability is not a major issue any more.
2. Setup *is* a big part of usability. Most computers are not some fixed configuration that never changes- part of using a computer is connecting new things to it and adding software, and even unsophisticated users want to do new things regularly. There's a misconception that some huge percentage of the population out there only does e-mail and web browsing and that's it. If that were the case, Internet appliances would have been a lot more popular. Again, speaking from direct experience (this literally is my job) the "setup" problems are a deal killer for Linux for most "normal users", even if someone comes in and gets the system up and running for them to begin with. Believe me, I've tried.
3. I never said that supporting windows didn't suck. It is no paragon of UI consistency. Outlook in particular is a nightmare of rapidly shifting configuration interfaces. But it's lunacy to suggest that Linux even approaches the level of consistency in windows.
Perhaps you didn't get to the end of my post, but I ended up concluding that this isn't a bad thing, and in now way am I questioning your platform choice- I prefer Linux myself. But it simply is "not there" for the average user. I don't think it needs to be... the restrictions and strict standards necessary to get it "there" would kill off a great deal creativity and sponteneity that makes the "bazaar" work.
Besides, it's already been done... OS X is out there, if you really want a UNIX that is consistent and easy to use. Let Linux be what it is, which is wonderful.
It really is impossible for someone that's unfamiliar with a particular system to judge how "easy" or "difficult" it is in the absolute sense, compared with the system that they are already most comfortable using (and likely prefer). The article is not seeking to judge the platonic usability of Linux- rather he's honest about evaluating it strictly from the perspective of whether it is a usable system for windows-familiar users to switch to. So to answer your rhetorical question, "articles like this" don't evaluate the difficulties of windows because they're evaluating the claim that Linux is something practical for *windows users* to switch to- people who are already able to overcome windows deficiencies (at least to some extent).
Your assumption that the "prison cell" feeling when you use windows is largely due to the unfamiliarity of the system is absolutely right. However, it disqualifies you from an unbiased judgement- you would feel like windows was a prison cell no matter what.
I'm one of those annoying people who is truly and thoroughly proficient in both. I worked as a UNIX system administrator for 4 years, and know UNIX-based systems inside and out. I've got a credit in the sendmail source code. I've built a "Linux from scratch" system. However, I currently work as a small business computer consultant, spending 100% of my professional time in windows, and have an entire practice built around helping people navigate the incredible pain that is keeping windows systems running reliably in undstandardized environments. So believe me when I say that I know the pain, and I'm not a defender of the windows way of doing things.
But challenging as windows is, my opinion as a fairly unbiased observer and user is that Linux really is more difficult. To pick one very recent example, I've got a computer science degree, and it still took me hours to get my canon printer working with Linux. I'm not laying blame here... Linux has a much tougher road to hoe when it comes to usability because of several inherent factors:
1. Market share disadvantage- few manufacturers package drivers for their hardware 2. Fundamental conflicts between the GPL and software patents- multimedia codecs and the like 3. Total and complete lack of UI standardization- there are few if any UI conventions between different projects- even with simple text configuration files, the basic syntax is hugely different from system to system
I can see someone might argue that the third is a fundamental, structural issue like the first two. But I think it is an inherent result of the great strength of open source software, which is the constantly evolving, creative process of innovation. The fact that there are dozens or shells or window managers is a byproduct of hundreds of thousands of volunteer programmers saying "I know a better way to do this", and the best parts of what they come up with eventually spread widely.
This is why it confuses me when Linux folks get "up in arms" about usability complaints from Windows users. Linux is harder to use than windows. So what? Why do you care? You don't use Linux because it's easy. You use Linux because it's better, more creative, and gives you more control. In a lot of ways, control and usability are conflicting goals. Automatic transmissions sure are easy to use, but a lot of people prefer the control and efficiency of a stick shift. Manual trasmissions aren't going away anytime soon, and Linux doesn't have to defeat windows and recruit all of the "normal users" who value 'easy' above everything else to be successful.
If you want a user-friendly UNIX, get a Mac. Enjoy Linux as it is, and be secure in your superiority...
UltraVNC is part of my standard install set, and I've never seen substantial CPU utilization on machines that have the display driver correctly installed. Have you verified that it's loaded and actually working? You can tell by right-clicking the VNC helper icon in the system tray and choosing "Properties" (not Admin properties).
-R
Obviously designed by programmers for programmers
on
Microsoft PowerShell RC1
·
· Score: 5, Insightful
Although I haven't played with it, I've read a bit about this shell, and there was something that bothered me about it, and I finally just put my finger on it: this thing was designed by programmers.
I know that the line between "programmer" and "system administrator" is often blurry. And the line between "shell" and "interactive script interpreter" is as well. But when you start requiring people to understand concepts like objects (which may seem like old hat to a programmer), you're already presuming a relatively sophisticated understanding that an "average user" has no grasp of. And the.Net libraries are vast and complex... looking at some of the sample msh scripts, I understand how a windows programmer would think they were an amazingly powerful simplification, but damn there's a lot I have to know to get basic things done.
Ye olde csh and sh are great because they provide a simple way to put programming logic around the set of operations users spend their entire day in and are already familiar with. The learning curve is very incremental: you can master the basic UNIX commands, and then start to add in variable subtitutions (!$ anyone?) and loops (foreach) and such as needed.
In other words, the jump from basic UNIX user knowledge to simple scripting is very small, because the scripting is presented in *exactly* the same context and using the syntax the user does day-to-day work in. But as a competant windows admin who doesn't know VB and hasn't written a line of.Net code in my life, I see almost nothing familar when I read.msh scripts. It appears to require an entirely new body of knowledge to do simple things, and bears little or no relationship to the interfaces and paradigms I use day to day. Yes, I know those interfaces are graphical. Seems to me there's bound to be some way to do it (or would be if there were any logic or consistency to the organization of the everyday administative interfaces in Microsoft's products).
Don't get me wrong... I understand that the goal of an intuitive scripting tool is in many ways at odds with providing a rich and powerful development environment that can complete with something like perl, but I had hoped there was something a little closer to "ground level" coming.
I realize I left the key words "for backup" out of my first comment. It's a ridiculously slow, small storage method if you're just short on space.
But it has some really nice qualities for backup- it's geographically separated from your systems, so a tornado or fire won't cause you to lose everything. You can automate it... a flash drive isn't a very good backup if the lightning strike that takes out your computer also takes out the flash drive plugged into it. The encryption I mentioned handles confidentiality issues. And if the service is free? Use two. So even if one flakes out on you, you've got another backup.
It's only for that small subset of really critical, irreplaceable documents, and should be complemented by larger capacity, more frequent local backup, but if these services really are free, they'll be a boon for cheap backups.
I'm all for MS entering this field... if they (and Google) can drive down the price of online storage through increased competition, the better for everyone. And there's no reason to have to trust their security... rather than syncing your files directly, encrypt the hell out of them and just upload / sync the resulting files.
Slashdot Mirror
As a consultant, I love the fact that they've managed to take something simple and reliable that almost never has problems (wires connecting keyboards and mice), and replaced it with something prone to all sorts of failures... dead batteries, interference, poor range, and generally poor manufacturing quality.
Every wire they eliminate equals more billable hours for me! I'm all for wireless peripherals!
-R
ps- In all seriousness, I make sure my clients understand this before purchasing one...
I would have agreed to you up until V10. As a consultant I administer MANY windows domains, and I've had a significant issues with V10 that I never had with earlier versions at a number of customers, especially on the server (the worst place). I'm using Trend Micro SMB now... and couldn't be happier. They've even built in automatic removal of other A/V software (including Symantec) into their own client installer. My only gripes is that their spam filtering stinks (so does Symantec's), and it doesn't support IMAP at the client.
-R
Nowhere have I said that a new windows install is never necessary, and I am certainly aware of the virtues of a clean install. I'm not saying I never do them, but rather that in most circumstances that I find myself in as a consultant, I rarely find them necessary or in the customer's best interests. BTW- I'm speaking exclusively of Win2k and XP here... the story is completely different for previous versions, which I rarely encounter these days. If the registry is corrupted and there's not a recent system restore backup available, or if they hard drive has failed badly enough to prevent grabbing a good image, or of the customer truly doesn't care for their configuration or data and there's a quick OS restore function... then I'll happily do a re-install.
But what gets my goat are consultants who do them because they think they know best, without taking into account the customer's actual situation and needs. I know this happens frequently because of the horror stories I hear from my clients. I feel confident that I'm not leaving behind the same unwitting "trail of destruction" because my business is driven entirely at this point by repeat business and referrals- I'm actually turning away business these days. And my rates are higher than average. If I were not meeting my customer's needs, I doubt they'd be coming back as they do.
I'll offer two possible explanations for what clearly is a massive cognitive gap between us:
1. I guess I really just don't find cleanups that difficult or time consuming. If they took 3-4 hours then I'd be agreeing with you. But what I commonly find is that, especially with recent fast machines, I can do a cleanup in 1-1.5 hours, including install of decent protection software. I'm really not kidding. And I'm not getting called back for re-dos (which I don't charge for) because the machine is still infected. When I compare that with the cost, risk, and time involved in a rebuild, it rarely makes sense to do the latter.
2. There are aspects to my business which may differ significantly from most. First, I work in a rather affluent area, and typically on-site... I'm not a "carry-in" shop. Most of my clients are not "grandma" types who just do e-mail and web... many are work-at-home professionals, and thus have comparatively elaborate systems. Another generalization... they prefer "full service" rather than "fast and cheap". Thus, when I do a rebuild, I'm not leaving folks with a bare windows install- rather I'm putting the machine back as close to the functionality they previously had, which typically includes office suite, e-mail client (configured for their host), A/V, anti-spyware, backup, printer drivers, digital camera / photo software, and personal finance software, just to name the most common things. So a rebuild is typically a more expensive, elaborate affair than the typical carry-in shop does.
Anyway, I would love to take you up on your offer of a "rebuild" vs "cleanup" face-off... sounds like fun! Name the time and place (as long as it's in Metro D.C....). You give me a badly infected machine. I'll give you a typical "real world" configuration scenario from one of my customers. Of course, to be fair, you get to wait while they track down the install media in a badly-organized file drawer with software dating back to 1992, only to discover their daughter probably took it with her when she went off to college. And call their accounting software company for the product key they've lost. And find their ISP dialup information they use when they're on the road but forgot to mention before you started the re-install. And... and...
-R
Well, I didn't think I was doing anything that unusual, but given how many people are shocked by my claims of removing infections quickly, maybe I should start a training course! See one of my earlier posts in this thread for a summary of my technique... the reason why it doesn't take so long is that I don't use long-running (and often inneffective) file scanners.
I am constantly aware of how important it is to make a computer repair cost-effective... that's one of the reasons I charge a flat two-hour (not four) rate for cleanups, so that if I run into something that's truly novel, I can take the time to (a) do the right thing by the customer and (b) educate myself so that I'll know how to do it in the future.
But for more serious / time consuming problems, I do have a threshold for which it's "not worth" fixing a computer... what you've said is absolutely right in some cases. I don't do what I call "band-aid" work... I want to be sure that the customer is going to be fully satisfied with the result, and it just doesn't make sense to invest a lot of time into older machines.
-R
I offer a money-back guarantee... if the infection predates my previous cleanup, the second cleanup is free.
If someone is stupid enough to get re-infected after sitting through the well-honed lecture I give while cleaning up their machines, I'm happy to take their money again. It's happened twice in memory, and both of them were teenagers of wealthy parents that really didn't care. Most "normal" folks are scared to death of getting infected, and get religion after I tell them about keystroke loggers and such. Although getting the right software on the machine is a big part of staying safe, modifying user behavior is the really important change.
-R
There's no way Ballmer (if he's doing his "day job") has any real experience cleaning up malware. In fact, most regular corporate IT people don't really either, because (as so many have pointed out) a re-image is a much better route, if you've got the infrastructure and organization in place in advance.
:-)
There certainly are engineers at MS that could run circles around me, and I'm sure can easily clean up any of the widely-distributed infections my clients are likely to get (as opposed to the scary 0-day stuff used in corporate espionage).
But can you imagine what a mess that machine must have been in after Ballmer trying to fix it for two days? I doubt the infection was the real problem by that point. But who's got the cajones to tell that to him?
I've actually been to Redmond a few times- and not for conferences. The MS campus is really nice, but I wouldn't want to work there.
-R
This sounds like a justification... "if they haven't made the appropriate precautions... they get what they deserve".
This really misses the point. The scenario you describe is so far removed from the way non-geeks (ie, most people) understand and interact with their computers to be laugable.
This is squarely Microsoft's and other vendor's fault:
1. For not including recovery CDs with every computer (don't make the users burn their own recovery media) *inside the case*, where it can't get lost.
2. For writing crappy software that doesn't store files in the profile, exclusively.
3. For putting critical data (e-mail, browser configurations, etc...) in literally hidden directories that an average user doesn't even know is there. For putting user preferences in a registry file that can't simply be copied by the user to back them up.
4. For creating a convoluted software installation system, which prevents one from simply copying a folder to move an application from one computer to another.
5. For not making easy-to-use backup a truly integrated part of the OS, that you have to work hard to avoid and gives *useful* error messages when it stops working.
I could go on, but users are already being punished every day simply by having to use these incredibly confusing, unreliable machines for things that are really critical to them. The last thing we should do, as the technological elite, is blame them.
-R
Actually, I really do know what I'm talking about. The theory is simple: all malware has to get launched. There are a limited number of places in windows something can stick itself to get launched at boot time. This is what tools like HijackThis and autoruns allow you to look at. If you use them frequently enough, it's pretty easy to pick out what doesn't below.
Kernel rootkits made this a bit more challenging, since you can't trust what Windows is telling you about what you see through these tools. The answer to that is to boot to trusted media (like a BartPE disk) and check out the potential autostart locations from that. Since the malware isn't loaded, it can't make the system lie to you.
Thanks to the newer kernel rootkits that hide files in NTFS alternate data streams, you have to scan for those as well from trusted media. But that's doable- there are tools which work from BartPE which will enumate files with ADS streams, and you can also check for ADS paths in the list of drivers which load a boot-time. Then when you think you've got it clean, throw a sniffer on the machine's LAN connection and see if anything unexpected happens.
Thus, it's really quite possible to manually discover and "kill" most infections with a reasonably high degree of confidence, if you have enough practice and experience. The level of assurance isn't high enough for a corporate server, but it's usually sufficient for home users and in small businesses, considering the cost of a from-scratch rebuild.
I do admit I'm a bit more qualified than most techs who do this... I've got a computer science degree, and have been doing IT stuff professionally for 14 years. You've got to have a pretty solid grasp of windows internals to do this well.
-R
No kidding. I happen to be of the opinion that most modern UNIX variants do this wrong as well... back when I was doing UNIX sysadmin stuff at Rice University, we had a beautiful software installation convention called /usr/site... the upshot was that by forcing all applications (at compile time) to use version-specific install paths to refer to their own support files.
/machinename2 is your software install partition.
/machinename2/appname-version.
/user/site which has symlinks like this:
/usr/site/appname-version -> /machinename2/appname-version
/usr/site/appname -> /user/site/appname-version
/usr/local/bin, you instead put symlinks like this:
/usr/local/bin/executablename -> /usr/site/appname/bin/executablename
/usr/local/lib, /usr/local/etc... etc... ALL of the references to the app or app files in the system are symlinks to /usr/site/appname/...
/usr/site/appname-version.
/usr/site/appname to point to a different /usr/site/appname-version. Same thing with a "downgrade".
/machinename2 software installlation NFS share. If you google /usr/site, you can find complete documentation here.
For example, say
Each app is installed in a directory
There's a directory called
Rather than putting executabels in
Same thing for
The app is compiled to refer to use the install prefix as
The beauty of this sytem is that you can easily have multiple versions of an app installed simultaneously. Need to put a new version into production? In most cases, all you need to is change
This is actually a slightly simplified version of what we used, since it can be extended to "do the right thing" in environments where you're supporting multiple processor architectures off a single
-R
For slightly larger companies, you're absolutely right. Imaging is great where it's practical. But for very small companies (15 computers) and home users, it rarely is. They've typically got a scattershot collection of machines with OEM licenses, different HALs, etc... Most small businesses this size are only buying 1-2 machines per year. Moreover, because of the diversity of roles within the company, few computers have the same software installs... the bookkeeper needs Quickbooks.
.5TB. I push those *hard*. But even then a full system restore is a pain because most backup systems don't cope with the proprietary boot sectors and hidden system partitions found on most major OEM installs these days. And a lot of small business folks are cheapskates and have to suffer a hard drive failure before they recognize the need for a real backup system.
In addition, people need to have privileges to do their own maintenance and software installs, since they don't have on-site IT help to go to when they need their new PDA hooked up, or want to install an update to their accounting program. So no matter what I put on the image, it's going to be badly out of touch with what the user actually needs.
The good news is that a backup system that can back up every bit on every machine is affordable, since the total data storage is usually well under
So I've gotten pretty good at repairs...
-R
Sorry to go on an off-topic rant here, but...
I'm a consultant who helps small business and home users. I can't tell you how many times I have talked to customers who (in the past) have had another tech come along and do a re-install without understanding all of the implications.
There is value in a machine's configuration! The customizations, tweaks, and even icon arrangements people create to make their systems work and lives easier are time-consuming to recreate, and there can be a major loss of productivity if they have to re-do it all from scratch. I'm a professional, and it's not uncommon for it to take me 3-5 hours to do a good job of getting all of the software, utilities, and configuration changes done for a typical business machine. Just because you can rebuild your own gaming rig from scratch in two hours (because you do it once a month) doesn't meant that this is a course of action that makes sense for everyone.
This is why I always recommend *full* backups of the entire system... not just "important" documents. And it's why I do a full re-install as an absolute last resort. I can count the number of re-installs I've been forced into in the last *year* on one hand.
The good news is that if you know what you're doing (unfortunately many techs don't) VERY few problems require a rebuild. It's very possible to clean off even the "worst" infections fairly quickly, with high confidence that everything is gone. I charge a two-hour flat rate for *any* infection cleanup (including kernel rootkits), and that usually works out to my advantage. Hard drives often have only failed in a few sectors... I commonly am able to image the failed drive to a new one, and repair the windows install using a combination of sfc, system restore, misc subsystem fixes, and (in the worst cases) a repair re-install.
The benefit to the user is that they get their machine back *exactly the way it was*, the same day, without a large repair bill. The benefit to me is that the customer is happy and calls me back the next time they have a problem... instead of cursing me the whole time they are trying to rebuild their system the way they had it.
If you are a tech and haven't learned this stuff, you are doing your customers and yourself a disservice.
-R
There's a link in the article to the company's website. They've developed a motorless feedpump system, and there's a rather elaborate flash animation that describes specifically how it works, and several possible sources of energy... solar water heaters, sub-boiling geothermal sources, or even wood stove waste heat. The point is that they think it can work efficiently with a 50 degree temperature differential above ambient temperatures, which is pretty easily achievable without a lot of elaborate heat/cold storage.
The point of their system is not to be more efficient than solar panels, but to be MUCH CHEAPER. We don't have a shortage of energy from the sun... we have a shortage of cost-effective ways to harnass it.
-R
A great number of the capababilities in their primary product - ERD Commander - have now been duplicated in BART PE-based bootdisks like Ultimate Boot CD for Windows (UBCD4Win). I recently evaluated ERD Commander specifically against UBCD4Win, and in the hands of a well-trained user the latter did most of what we needed to do. It wasn't as slick, and there were some ways in which ERD Commander was much better, but the very high price of ERD Commander (at least for a small consulting shop like mine) made it very difficult to buy with the marginal additional functionality it adds. On the other hand, without UBCD4Win, it would have been quite attractive.
I'm guessing we're not the only ones noticing this... so the sale makes sense. Better sell out to MS rather that try to compete with 'free'.
-R
ps- Note my caveat "did most of what we needed to do". I'm not claiming that UBCD4Win is an ERD Commander replacement for everyone...
If you're (like me) one of the, umm, fortunate souls who get to clean up rootkit-infested machines regularly, there's a tool you should know about: LADS, for "list alternative data streams"
It can be found buried in this FAQ about the NTFS ADS feature: http://www.heysoft.de/nt/ntfs-ads.htm
I haven't tried it yet, but it looks like it should work from a win32 bootdisk (like BARTPE). So you should be able to boot from a clean win32 environment and scan the computer's hard disk to find any files with ADSs. Fortunately, use of this feature within NTFS is not widespread, so malware should stand out pretty obviously.
Have fun!
-R
Netstumbler only shows you other *wireless networks*. Wi-Spy shows you *all* RF interference in the 2.4 GHz spectrum, such as that caused by cordless phones, microwave ovens, etc...
-R
I'm a small office / home office IT consultant, and I bought one of these a few weeks ago after stumbling across it on ThinkGeek. It's fabulous for my needs, which are simple: figure out if interference is the reason someone's wireless network is flaky.
Wi-Spy does a great job of doing this. I fired it up at a downtown client and saw there was a strip of intense interference down in channel 1. Moved them up to 11- problem solved. I've also done some tests at home... it's very easy to tell the difference between a microwave, spread spectrum phone system, video sender, and other wifi networks... they have rather distinctive appearences in the graphs Wi-Spy produces. Now that I know what they look like, I can take an educated guess, where before, I was grasping at straws.
For those of you getting your panties in a wad about it not matching a $5000 spectrum analyzer: Duh? Of course it doesn't. But that doesn't mean it's useless... there are a lot of folks (like me) for whom the cost of a "real" spectrum analyzer is completely unjustifiable. But I can spend $100 easily, and *for what I do*, which is occasionally troubleshoot SOHO wireless networks, it provides most of the functionality I need.
The really interesting fact is that this thing defines an entirely new product category: inexpensive spectrum analyzers. I would really like to see what could be done for $500... that's still an order of magnitude cheaper than the existing solutions, but I bet you could add a bunch of features.
-R
You can't just add access points to increase capacity... the limitation is the radio frequency space available. Remember there is only room for 3 wifi channels (1, 6, 11) in the 2.4GHz spectrum. Add a forth into the same space, and you're just stepping on the others and causing interference. Of course I'm assuming 802.11b/g here, as 802.11a has 20 distinct channels.
The other issue that people have mentioned is outside interference. Microwave ovens can be a real bummer. So can the little cordless 2.4GHz headsets executives seem to like. And you better hope nobody sets up a 2.4Ghz video sender for their security system in the vicinity. Or a nearby cell tower, or radio station. You could be working perfectly for a year, and then suddenly have your network permanently broken by something completely outside your control or ability to change.
There's a reason you don't hear of many people doing this.
-R
I've got FIOS and my traditional phone line now runs over the fiber They completely removed the existing phone box on the house and put the ONT in it's place... it has a similar block for wiring the house phone wiring to it. This is why the FIOS install comes with a UPS- so that your phone line will keep working if the power goes out. They didn't actually tear out the copper wire from the ground, but hooking it back up would be a project.
However, he's gone a bit too far with the regulatory fear-mongering. Yes, the fiber line is excempt from the regulations passed in 96 that forced the phone companies to allow competitive access to the copper that enabled Covad, Northpoint, and others to start building out DSL networks of their own. However, the FIOS phone line is still a tariffed / regulated service, with the same Public Utility Commission oversight as before.
-R
I know of at least one ISP which does this- Cox cable in the N. VA suburbs of DC. Several times I've cleaned out home users computers that were infested with malware, only to discover that their Internet access was still being blocked. It's rather difficult to get them to remove a block quickly once it's in place... the L1 support folks disclaim all knowledge and assume it's anything else but a block, but if they talk with higher-level staff they eventually realize that's what's going on. Anyway, that's the way it was the last time it happened, which was probably a year ago.
-R
Well, I think you may have missed my larger point due to some of the details. Sorry I mistook you for a Windows noob. Yes, I've done my time with old Sun machines too... nothing like a little VME backplane jumper reconfiguration to brighten your day!
Just a few points in response:
1. I'm responsible for supporting several hundred WIndows machines in several dozen very small companies and scores of homes. Although one can certainly screw up an XP or 2000 machine such that you have to reboot daily, this simply is no longer the norm for properly managed machines, even if you have a relatively unskilled user running as a local admin. For most users, stability is not a major issue any more.
2. Setup *is* a big part of usability. Most computers are not some fixed configuration that never changes- part of using a computer is connecting new things to it and adding software, and even unsophisticated users want to do new things regularly. There's a misconception that some huge percentage of the population out there only does e-mail and web browsing and that's it. If that were the case, Internet appliances would have been a lot more popular. Again, speaking from direct experience (this literally is my job) the "setup" problems are a deal killer for Linux for most "normal users", even if someone comes in and gets the system up and running for them to begin with. Believe me, I've tried.
3. I never said that supporting windows didn't suck. It is no paragon of UI consistency. Outlook in particular is a nightmare of rapidly shifting configuration interfaces. But it's lunacy to suggest that Linux even approaches the level of consistency in windows.
Perhaps you didn't get to the end of my post, but I ended up concluding that this isn't a bad thing, and in now way am I questioning your platform choice- I prefer Linux myself. But it simply is "not there" for the average user. I don't think it needs to be... the restrictions and strict standards necessary to get it "there" would kill off a great deal creativity and sponteneity that makes the "bazaar" work.
Besides, it's already been done... OS X is out there, if you really want a UNIX that is consistent and easy to use. Let Linux be what it is, which is wonderful.
-R
It really is impossible for someone that's unfamiliar with a particular system to judge how "easy" or "difficult" it is in the absolute sense, compared with the system that they are already most comfortable using (and likely prefer). The article is not seeking to judge the platonic usability of Linux- rather he's honest about evaluating it strictly from the perspective of whether it is a usable system for windows-familiar users to switch to. So to answer your rhetorical question, "articles like this" don't evaluate the difficulties of windows because they're evaluating the claim that Linux is something practical for *windows users* to switch to- people who are already able to overcome windows deficiencies (at least to some extent).
Your assumption that the "prison cell" feeling when you use windows is largely due to the unfamiliarity of the system is absolutely right. However, it disqualifies you from an unbiased judgement- you would feel like windows was a prison cell no matter what.
I'm one of those annoying people who is truly and thoroughly proficient in both. I worked as a UNIX system administrator for 4 years, and know UNIX-based systems inside and out. I've got a credit in the sendmail source code. I've built a "Linux from scratch" system. However, I currently work as a small business computer consultant, spending 100% of my professional time in windows, and have an entire practice built around helping people navigate the incredible pain that is keeping windows systems running reliably in undstandardized environments. So believe me when I say that I know the pain, and I'm not a defender of the windows way of doing things.
But challenging as windows is, my opinion as a fairly unbiased observer and user is that Linux really is more difficult. To pick one very recent example, I've got a computer science degree, and it still took me hours to get my canon printer working with Linux. I'm not laying blame here... Linux has a much tougher road to hoe when it comes to usability because of several inherent factors:
1. Market share disadvantage- few manufacturers package drivers for their hardware
2. Fundamental conflicts between the GPL and software patents- multimedia codecs and the like
3. Total and complete lack of UI standardization- there are few if any UI conventions between different projects- even with simple text configuration files, the basic syntax is hugely different from system to system
I can see someone might argue that the third is a fundamental, structural issue like the first two. But I think it is an inherent result of the great strength of open source software, which is the constantly evolving, creative process of innovation. The fact that there are dozens or shells or window managers is a byproduct of hundreds of thousands of volunteer programmers saying "I know a better way to do this", and the best parts of what they come up with eventually spread widely.
This is why it confuses me when Linux folks get "up in arms" about usability complaints from Windows users. Linux is harder to use than windows. So what? Why do you care? You don't use Linux because it's easy. You use Linux because it's better, more creative, and gives you more control. In a lot of ways, control and usability are conflicting goals. Automatic transmissions sure are easy to use, but a lot of people prefer the control and efficiency of a stick shift. Manual trasmissions aren't going away anytime soon, and Linux doesn't have to defeat windows and recruit all of the "normal users" who value 'easy' above everything else to be successful.
If you want a user-friendly UNIX, get a Mac. Enjoy Linux as it is, and be secure in your superiority...
-R
UltraVNC is part of my standard install set, and I've never seen substantial CPU utilization on machines that have the display driver correctly installed. Have you verified that it's loaded and actually working? You can tell by right-clicking the VNC helper icon in the system tray and choosing "Properties" (not Admin properties).
-R
Although I haven't played with it, I've read a bit about this shell, and there was something that bothered me about it, and I finally just put my finger on it: this thing was designed by programmers.
.Net libraries are vast and complex... looking at some of the sample msh scripts, I understand how a windows programmer would think they were an amazingly powerful simplification, but damn there's a lot I have to know to get basic things done.
.Net code in my life, I see almost nothing familar when I read .msh scripts. It appears to require an entirely new body of knowledge to do simple things, and bears little or no relationship to the interfaces and paradigms I use day to day. Yes, I know those interfaces are graphical. Seems to me there's bound to be some way to do it (or would be if there were any logic or consistency to the organization of the everyday administative interfaces in Microsoft's products).
I know that the line between "programmer" and "system administrator" is often blurry. And the line between "shell" and "interactive script interpreter" is as well. But when you start requiring people to understand concepts like objects (which may seem like old hat to a programmer), you're already presuming a relatively sophisticated understanding that an "average user" has no grasp of. And the
Ye olde csh and sh are great because they provide a simple way to put programming logic around the set of operations users spend their entire day in and are already familiar with. The learning curve is very incremental: you can master the basic UNIX commands, and then start to add in variable subtitutions (!$ anyone?) and loops (foreach) and such as needed.
In other words, the jump from basic UNIX user knowledge to simple scripting is very small, because the scripting is presented in *exactly* the same context and using the syntax the user does day-to-day work in. But as a competant windows admin who doesn't know VB and hasn't written a line of
Don't get me wrong... I understand that the goal of an intuitive scripting tool is in many ways at odds with providing a rich and powerful development environment that can complete with something like perl, but I had hoped there was something a little closer to "ground level" coming.
-R
I realize I left the key words "for backup" out of my first comment. It's a ridiculously slow, small storage method if you're just short on space.
But it has some really nice qualities for backup- it's geographically separated from your systems, so a tornado or fire won't cause you to lose everything. You can automate it... a flash drive isn't a very good backup if the lightning strike that takes out your computer also takes out the flash drive plugged into it. The encryption I mentioned handles confidentiality issues. And if the service is free? Use two. So even if one flakes out on you, you've got another backup.
It's only for that small subset of really critical, irreplaceable documents, and should be complemented by larger capacity, more frequent local backup, but if these services really are free, they'll be a boon for cheap backups.
-R
I'm all for MS entering this field... if they (and Google) can drive down the price of online storage through increased competition, the better for everyone. And there's no reason to have to trust their security... rather than syncing your files directly, encrypt the hell out of them and just upload / sync the resulting files.
-R