cluberti

True, but since 2003 the only real self-propagating Windows exploit was Conficker, and that attacks netbios shares, mostly the ADMIN$ share. Folks running a firewall (assuming they aren't letting in random anonymous SMB requests) are safe from this, as would they have been against Blaster, NIMDA, Code Red, Sobig, and any of the other self-propagating worms out there. The fact that the Windows firewall in XPSP2 (August 2004) and up, in Vista, and in Windows 7 will prevent these sorts of attacks means the self-propagating network exploit is just not that much a security risk anymore, if at all. I know many people who run without antivirus software actively running (myself included), and it is true that a firewall (at least host-based) and safe computer use habits are enough to be safe. Also note that Symantec recently released information on vulnerability statistics (aka infected machines stats), and found that the bulk of infected machines were XPSP1 and XP RTM, meaning folks who have old software vulnerable to the exploit and haven't upgraded to patch levels are the ones being attacked. I know that there are other ways to get this (autoplay, infected files on network shares) but as an automated network / internet propagating virus, I don't need antivirus software to protect me here - I just need to use the built-in firewall that came with the product I paid for (or I can use a 3rd party one if I don't like this).Again, it's true there could be a vulnerability out there that is unpatched that will leave us open until Microsoft patches it, but without direct network access to my machine it'll be hard to get a virus on it automatically. Which leaves us with social engineering, to which I'm not likely to fall for (but know plenty of family members who might, hence why they get MSE installed in all new builds as part of the first round of post-Windows-install software). If you feel safer running antivirus, by all means do so - this is not a post to try and sway anyone for or against. However, it is worth knowing that an ounce of prevention can be enough to not have to run it at all for a good number of folks (like most people here, who are security-conscious and knowledgeable enough to not be fooled by social exploits).

You'll find most boasts about being able to "boost" or "optimize" windows are little more than smoke and mirrors, if they're even going that far. Nothing a properly configured system can't be configured to do - defrag the disk (which 7 does automatically already), keep from installing/removing/reinstalling applications like some crazy people, and you will be fine .

If that's the case, it's just a bunch of vbs scripts that disable services and components - nothing you couldn't do on your own with the GUI or even powershell or write your own vbs to disable services and components) and you wouldn't need a warez'd version of Windows to do it with either .

I would start with a network trace (simultaneous) from client and server to see what happened on the wire when the problem occurred before doing anything else. If a client is running an app that requires SQL (just guessing from the problem description) and cannot connect (and needs a reboot to fix), assuming you didn't have to restart SQL the likelihood it's a client-side problem goes up. I'd also suggest the usual - remove apps that install network LSPs like 3rd party firewall, antivirus, and/or antimalware products to make sure none of them are destroying the connection prematurely.

Interesting - good to know.

More than likely no, because so far no one has posted they've heard of such a thing. I'd love to help, but given the little information, no examples, or anything of real assistance, it's hard to help. There's no .net way to "optimize" Windows via script or anything like that. There are apps (a lot of them here) that can help, but if those aren't what you're talking about I'm afraid we'll need more information.

That one is indeed a pain, although I'm hoping it's fixed with 8. Given the servicing engine, I don't see how they can make this happen for Vista and 7 though, which I guess was a tradeoff of getting a more modern servicing engine out the door with Vista. I think if MS made the SP integrated ISOs available to everyone as they do with trial server OS ISOs, this would be less of a problem. While that would be nice, there are legal hurdles to this. Not to mention getting other vendors to actually buy in and lose control of their crapware distribution systems. Again, I hope they fix this for Win8, although it would probably eat into installs of App-V and MED-V, so until the risk outweighs the reward I'm guessing no. Given the MSI engine's requirements for file scanning, I'm not sure this is possible yet. I know it's something we'd all want, but again, due to the design of the MSI engine it really (currently) has to be online. Maybe if more packages conformed to CBS and the MSU standard rather than MSIs? That'd work. Given how many times this is filed on connect and rejected, I doubt we'll see this anytime soon. Silly, but if you're waiting for Win<version> for this to change you'll be waiting a long time ... True, although this is less an issue than the rest you posted. I wish they'd spend more time on the UI as well, although it's gotten fairly close to really good in most places. While I would counter-argue that it can be made the same with one click of a slider, and the default settings are pretty good, I think the neutering of UAC to be less invasive will be a wait and see if it was a good idea or not. It might be OK, but if not we can thank the whiners for reducing security for convenience (which is what they usually want anyway). Given the architecture of IE, this isn't likely to change ever either. Accelerators are pretty interesting, and running a WPF app in silverlight is also really interesting, but neither are true plug-ins the way an activex control is. And, given the path of previous and current versions, I don't see a plug-in architecture above and beyond what we've currently got for v.next either, for better or worse.

Nice find, but it seems their site isn't available to everyone atm (I'm getting a host not found error, and a domain placeholder page).

Aaaah, I remember those.

Never heard of such a thing, honestly.

? Modifying Windows via vbs? I think you might be thinking of people who turn off services or remove components, but none of those are specific to vbs files. Perhaps if you were more specific in what you wanted to do (and you might want to hunt around the vLite section of the forum as well) we might be able to help.

Or it's loading it from somewhere else, which is always possible. I do have a question though - is this the Portuguese version that has the English text, or are we editing an English version? I apologize I've not looked at a Portuguese MUI'd ISO yet, but the post doesn't indicate either way. I'd assume it's a Portuguese install, but it left me wondering if we're trying to edit an English install, and then of course "why".

There were still folks that wanted 3.x's program manager, even in W2K and XP, so this isn't odd, it's par for the course. Change causes a lot of people to resist, whether it's good change or not.

If you open the registration reminder (you're getting asked to re-register because technically, a repair install is still a reinstall of the OS), you'll see there's a phone option.

See if anything in this thread helps you - I know at one time there was quite a bit of buzz over getting it to work, although I've not been tracking it for awhile.

Hate to add additional steps, but does installing your DVD copy into a virtual machine give you the same results, or does it work there?

Correct on both counts - as it stands, the only really current standards-compliant browser left for Win9x users is Opera (it scores 100 on Acid3 and renders Acid2 properly, for instance). I believe v10.1 is the latest version you can use on Win9x, although I may be wrong on the minor version number that's compatible.

Yes, in fact, if you seek you shall find. Please search before posting next time, as this has indeed been discussed here many times (that link just has the most useful info, IMO).

Honestly, this sounds like a good candidate to try Win7 on (the CPU is a little long in the tooth, but it would only really be apparent in heavy CPU-related tasks, and that would be the same in XP as well anyway). 2GB of RAM is more than enough to run the x86 version of Win7, and even the x64 version would run fine (it would have larger memory footprint as all binaries are larger, but not drastically so). XP is still a good OS however, so you should definitely check the Win7 upgrade advisor to see if you have any drivers or applications that would have problems under Win7 - if not, it's probably worth the plunge. If so, sticking with XP won't necessarily be a bad thing either. As to the domains question, no, Win7 Home Premium cannot join a traditional Windows domain, you would need Pro or Enterprise/Ultimate for that. However, if it's just a home situation, and the other machines are also Win7, you can set up a HomeGroup that would make it easier to do most of the things you'd use a domain for amongst your Win7 machines.

Understood. Honestly, most sites out there don't really use anything more than the Acid2 test would test for, but that's partially because there are enough browser users out there that would have trouble with a site if it did , hence the catch-22. Once sites do start moving en-masse to more HTML5, DOM2 and DOM3, and CSS3 features will you run into a problem (and I don't see that being a reality really until IE9 becomes the dominant IE version, and/or one of the other browsers gets more than 60% of the browser marketshare, neither of which is likely to happen in the next year or two at the very least).

Unless you explicitly change the security in a zone to allow script to access local resources, no browser-based script can read from the registry or local disk (an activex control can, because it's got an out-of-process COM server and can be written to allow such access). Most checks look at the UA, jscript to check the navigator.userAgent, or try specifically to set a property that specific browser versions can't use (similar to the previous CSS property check). If some page or site does the first two, the registry change to modify the UA you send is sufficient to spoof yourself. If some page or site does the last, there's no way you're going to get around that without replacing the rendering engine (and at that point, might as well use a different browser anyway). The only time hacking .dlls would change anything would be for locally-running apps or activex controls that actually check binary versions for browser version determination - it's not going to help you any with fooling a web site. I guess the question is which site or app you're trying to fool, and why, to make a determination on how best to attack that.

Well, Microsoft has never released an update for XP, Vista, or Win7 for that matter that modified the shell reg key/value pair. Either it was caused by malware, 3rd party software, or perhaps the user themselves inadvertently (the last bit is probably unlikely, but you can never rule anything out). Most people complain about the registry, but having the registry means memory-mapping sections of a VERY large settings database which is very performant (more so that reading/writing files), provides a standard interface for developing and storing/retrieving said information without having to know anything else about a program or the OS, and also allows for centralized control via security ACLs, reporting/auditing, and Group Policy control. It's not perfect, but it's no better or worse than having hundreds or thousands of config files scattered about the disk (or even in one directory like OS X does). The registry is pretty robust and secure as far as settings databases go, although I do agree it could probably be protected better. The problem is with hundreds of millions of users, you might have the same percentage of idiots that'll hork their boxes, but .5% of 500 million is a lot more than, say, .5% of 10 million.

Why use one of these instead of your ISP? Well, what happens when your ISP uses DNS to (inadvertently or on purpose) throttle access to certain sites based on the lookup like Time Warner does here in the US? I've had issues specifically with youtube, netflix, and XBox Live performance and routing that magically go away when I switch from TW's DNS to OpenDNS. Consider this - I'm in the southern US on the east coast, and the Netflix streaming servers are just outside NYC for the east coast. If I use TWs DNS, I'm routed through Texas to Netflix servers in California based on TWs DNS structure, whereas when I use OpenDNS I get DNS resolved to hosts that are on the east coast, and performance is VASTLY improved. I have similar wierd routing issues specifically with Youtube servers and games over XBL, because TWs DNS servers end up routing me to the west coast A LOT for sites that are geolocated, like most large media and gaming services are. The extra features of OpenDNS that can block ads, spam, and certain other types of sites are what keep me using OpenDNS over Google's servers, but having an additional option (or options) is always a good idea. I think most cable ISPs in the US have poor DNS service in general (I've been on TW, Comcast, and even Adelphia when they were still a separate company), and always used non-ISP DNS servers after awhile because of their poor performance. I don't know if it's network size that causes the issues, or if they do it on purpose (and having worked for one of these big 3, I know at least that particular company *does* do it on purpose), but ultimately it doesn't matter. Performance is almost always better with a DNS service, in my experience.

Most people with camstudio problems are usually trying to capture full screen video with a fairly underpowered CPU or GPU, or have problems with codecs. I've never personally had an issue with the software in the year and a half or so I've used it, but the PC it's on is fairly new (Core2 Quad 9550, 8GB RAM). I didn't have trouble on an old P4 3GHz machine w/2GB of RAM I ran it on before this without issue, for what it's worth.I would say use it, and if you have problems, look around for an alternative. However, if you've got a decently muscled PC and/or don't plan on capturing video at HD resolutions, you should be fine.

The update is caused because at some point in the past, the registry entry for launching explorer.exe as the desktop has been corrupted in such a way that the data is there, but cannot be read once the updates change security slightly. Prevx has a utility to fix this, but this black destkop issue would not go away at a lower resolution like it does for the OP, hence this is not the same issue.