04 November 2005
Post Database: Full Bore DSL Speeds?In the November 2, 2005 issue of Post Database, James Hein wondered why he wasn't getting the full bandwidth that he was "being charged for" (or more precisely, the 2.5 Mbps that his employer signed up for). The explanation lies in a practice called "sharing". Unbeknownst to many, you the customer are rarely alone in using the DSL circuit that you signed up for. Quite the contrary, more often than not you will be sharing your (sic) circuit with a dozen or more people.
Occasionally, but not always, ISP's will issue a disclaimer to this effect. For example, Ji-Net states this (grammar and comprehensibility notwithstanding): "Package DSL has been shared bandwidth. The speed may reduce by 10%-20% as a result of the OVERHEAD from the IP or traffic in the Internet".
Sharing is usually cited as a ratio. Most corporate packages implement sharing at about 1:5 to 1:10; i.e. 1 link shared by 5-10 customers. For home packages, this ratio is much greater and can range anywhere between 1:20 and more than 1:100. I won't name names here, but I suggest you ask the sales/customer service department of your ISP since there's no easy way to determine this ratio otherwise. Obviously, it's not something that ISP's wish to advertise because of its negative implications. Besides, keeping mum probably gives them flexibility in adjusting sharing ratios at a later date as well.
This sharing may explain why DSL prices began plummeting and why throughput could be suffering as a result. To be fair though, you get what you pay for. It's hardly likely that any ISP could make money providing you with a dedicated link while charging less than Baht 1,000 a month. Also, before you get too indignant, be advised that this sharing practice is hardly unique to Thailand.
Finally, in light of the above, I would suggest that you take into consideration more than just speed and cost while evaluating/comparing DSL packages. In addition, I would also look at: a) link speed/# shared users (a measure of performance), and b) link speed/# shared users/monthly fees (a measure of performance per unit cost).
19 September 2005
Post Database: Identifying Unknown Music CD'sThere was a letter to Post Database's Helpdesk last week (September 14, 2005) asking how one can identify songs on unlabelled music CD's. Wanda gave a logical answer for CD's containing MP3's, but left up in the air what to do with regular music CD's. Actually, there's a very simple solution: just install any music CD ripper that's capable of looking up Gracenote's CDDB music database or something similar like freedb. The ripper will attempt to identify the CD by looking up these databases and provide information on the artist, album title, date released, song titles, etc. This works best with popular commercial music CD's; local music CD's may not be sufficiently documented in these databases.
I know for a fact that this works with CDex, although you will need to enter your email address first before CDDB will work. Another more adventurous alternative is to download and run Apple's iTunes. You can run this even if you don't have an iPod, although naturally you can't use of all of its features. iTunes, aside from everything else is does, is still a multi-purpose media player, ripper, etc. It defaults to ripping to AAC, which while still a lossy format, should still be an improvement over plain vanilla MP3, especially if you rip at 160Kbps or higher. Unfortunately, not many hardware music players support this format besides the iPod. The same holds true for software players, although the full (but still free) version of WinAmp CAN play these AAC/MP4 files without a problem.
Note: Music files downloaded from the iTunes Music Stores, while still in AAC format, are encrypted and won't be playable on WinAmp (unless you know what to do them). But that's another story.
17 July 2005
Post Database: The Ins and Outs of Software LicensesIn the July 6, 2005 installment of "Computer Currents", James Hein wrote about his friend's trials and tribulations in obtaining Microsoft software here in Thailand. While I'm hardly a Microsoft insider, I have had the opportunity to purchase Microsoft software over the years, both on a personal and a professional basis. As such, I thought I'd offer a few observations.
(Caveat Emptor/Mea Culpa Department: Although most of my observations are based on what I know/believe to be fact or are from personal experience, I do occasionally make educated guesses. I've tried to make it clear when I do though.)
1) As a general rule, Microsoft doesn't sell directly to end-users. (I believe exceptions are with gargantuan accounts who require direct Microsoft involvement.) Instead, it relies on layers of resellers and dealers (I may have the terminology wrong here, but hopefully you get the point). The top layer resellers each focus on some segment(s) of industry. If you're a corporate or institutional buyer, chances are you'll deal with this level of reseller. On the other hand, if you buy boxed software from a store, you're buying from a dealer who's further down the "food chain".
2) Most of the big software companies (including Microsoft) sell software in either of two ways: a) as "boxed" software that usually comes with physical media and documentation, or b) as "software licenses" - i.e. pieces of paper that acknowledge your right to use whatever software in whatever quantities that was paid for. For more information, check out: https://www.microsoft.com/licensing/programs/open/default.mspx.
3) Microsoft software sold in Thailand has been sourced from Singapore for as far back as I can remember. Apparently, Singapore is a major manufacturing and distribution point for Microsoft software in the Southeast Asian region (and maybe even farther afield). Thus, it's not surprising that Microsoft software in Thailand should come from or at least be routed through there.
While in-country resellers and dealers do keep boxed software in stock, I doubt that they do so in any great quantity owing to the stocking costs involved, the limited volume of demand, and the sheer range of products in Microsoft's stable. More often that not, they'll just stock the most popular items and place orders for everything else - hardly a surprising business strategy.
Software licenses are dealt with in a different way. For one thing, there is no physical "stock" of licenses. My guess is that the process of generating a software license involves the customer placing an order with a reseller, the reseller forwarding the order details to the in-country Microsoft office, who then forwards the order to a Microsoft regional office (perhaps Singapore or perhaps even directly to the U.S.). Once the customer's information is duly recorded in a licensing database and the transaction approved, the license flows back down to the customer. This multi-step process takes time - three weeks is mentioned in the article. (Incidentally, I have a Microsoft re-order in the works and have been waiting for 3 weeks already as well.) In this day and age, this may seem agonizingly slow, but then again, consider how much longer it takes for a new magazine subscription to start!
Purchasing software in this manner is different from walking into a store and buying a box of software off the shelves or ordering software from a reseller or vendor who has it in stock and can deliver it in a matter of days. In the latter case, you can expect instant or near-instant gratification, but you can also expect to pay A LOT more for it as well.
4) Is there a comprehensive price list for Microsoft software? Of course there has to be one, but I suspect few local buyers ever get to see or access it. For me, this isn't a major problem since I've always been able to get an approximate price over the phone with my reseller, and it never takes more than a few hours to get an "official quotation" faxed to me (probably the time to get a higher up's approval for the pricing and/or the time to generate/send the fax). If I were to wager a guess, my reseller accesses a secure Microsoft website where they obtain the most current prices.
Why doesn't Microsoft publish this or at least make it publicly accessible? I suspect: a) the sheer number of Microsoft products makes this list cumbersome to distribute and update, especially in paper form, b) I'm sure it can be argued that digesting such a massive/complicated price list would be daunting for us mere mortals (sort of like airline pricing), c) Microsoft products are always priced in US Dollars, but always quoted in local currency - thus, prices can change at a moment's notice, d) there MIGHT be regional differences to the prices of some products, something Microsoft may not wish to highlight, e) the same products are often priced differently depending on how they're sold. I have a sneaky suspicion that f) limiting access to the prices also provides a modicum of "flexibility" when it comes to adjusting prices.
Actually, if you need a ballpark figure, U.S. retail prices are easy to come by. Just go to Amazon.com, Buy.com, any office supply website, any computer mail order website, and do a search. Sometimes these prices are discounted, sometimes they aren't. Sometimes they clearly state what the list prices are and what THEIR prices are. Of course, these aren't necessarily Thailand prices (which tend to be higher due to import taxes, shipping costs, etc.), but with some experience you should be able to guesstimate what the Thailand prices would be by multiplying it by some factor.
Even U.S. volume/licensing prices can be determined, albeit they're a lot harder to find. One website that conveniently lists prices for the various types of "Open Licenses" (e.g. business, academic, government, charity) is https://www.wasatchsoftware.com. These prices are practically identical to the prices quoted to me locally (at least the academic prices), which seems to confirm my supposition that licensing prices are based on a global/common US$ price list.
5) Are there differences in prices offered by resellers/dealers? My guess is "No" and "Yes". "No" in that I assume resellers are required to use the official Microsoft price as a guideline and sell only within a set range. "Yes" in that they probably have some leeway to reduce prices - taken from THEIR profit margin, I would think. If this is true, whatever differential exists probably isn't too significant, although I've never verified this because I've only dealt with one reseller at a time over the years.
However, price differences between boxed software and software licenses CAN be substantial. (This is especially true: a) if you're not a business, and b) compared with prices of local boxed software.) That's why buyers who are prepared to buy at least a minimum quantity (5 units at the outset, but just 1 unit for additional orders) opt for volume pricing. Given that this is a licensing scheme, it doesn't involve physical media per se, although you are permitted to buy or not buy these at your discretion.
James seems perplexed by this, especially by Microsoft's supposed suggestion that his friend to get a temporary copy from Panthip (believe me, I've heard this as well). This may seem odd in light of Microsoft's oft heard exhortations to avoid software of this ilk (due to legality issues, due to fear of viruses, due to lack of support, etc.). But if I may play devil's advocate for the moment, I think the bottom line is this: as long as you pay the Microsoft piper (or your pound of flesh, depending on how you look at it), no one cares any more where you get your software from. Can't update to Windows XP SP2 because you're using a Panthip CD key - that's your problem.
Re: James' incredulity that software could be purchased without media, again I emphasize that this is a licensing scheme. If you were to license 20 copies of a given software, you don't really need to keep 20 copies in your possession to prove you're a legal user. Rather, your licenses can be verified simply by accessing Microsoft's eOpen licensing website (https://www.eopen.microsoft.com). No doubt, Microsoft prefers that end-users track/maintain licenses this way - when it comes to audits, it's far less laborious than counting physical disks, determining their authenticity, matching COA's (some of which are stuck to computers), and maintaining lists of CD keys.
In practice, when licensing software, one usually buys just one copy of the physical media and then one legally makes as many copies as needed, up to the quantity licensed. In fact, we order a single media set, copy a working set and then stick the originals in a safe. There's almost no paperwork to deal with, only one CD key per software product, and no product activation!
6) James' article mentions that his friend was required to pay up front for the software he ordered. If he placed his order with a dealer, I wouldn't be surprised at all. If he ordered from a reseller in the name of his company, that would be a bit more unusual, but still understandable if he hadn't had a previous relationship with the reseller. Remember, what the reseller is ordering/buying on the customer's behalf is a piece of paper imprinted with the prospective buyer's name. If the order gets cancelled, the reseller may end with something he can't get resell elsewhere and may not be able to return to Microsoft either. Presumably, once you're already "in the system" and have a history of purchases with your reseller, this requirement might be relaxed - but it all depends on the reseller.
* * * * * * * * * *
All in all, James' friend did the right thing (although he may not have known it at the time). It wouldn't have made sense to buy a boxed version of his Windows 2003 Server. Besides being nearly impossible to find in stores (it would have to be ordered), it would have been expensive to purchase at local retail prices. Besides, he would have had to go through a reseller anyway since Client Access Licenses (CAL's) aren't sold in boxes - you have to go the licensing route for this.
With his proposed purchase of server software + CAL's, he should have had no problem meeting the minimum buy-in point for Open Licensing. And with his software licenses recorded on eOpen, he doesn't need to deal with original program CD's and doesn't need to fear losing or damaging them either. He has a single CD key for each software product (not for each PIECE of software), and doesn't need to go through the detested activation process during installation.
As for his up front payment and lengthy waiting time for fulfillment, granted those may have been surprising, but they were just part of familiarizing oneself with the modus operandi of obtaining software licenses.
15 July 2005
Post Database: NOD32 Anti-VirusFor quite a while now, I've noticed that many Helpdesk readers tend to use free anti-virus solutions such as Grisoft's AVG and Alwil's Avast! I love using free software myself, but after trying/using these programs, not to mention the biggies such as McAfee's VirusScan (VS) and Symantec's Norton Anti-Virus (NAV), I've settled on what I consider to be a good compromise: Eset's NOD32 ("NOD" for short).
Why did I abandon these AV freebies and AV biggies? The freebies were knocked out of contention due to a question of trust. After all, what are the free versions of AVG and Avast! except feature-reduced versions of commercial programs offered by their respective companies. While I can justify the use of feature-reduced programs for many applications, I believe it penny-wise/pound-foolish to use a less-safe AV program.
Shouldn't this be a prescription to stick with the biggies only then? Not necessarily. In fact, I had used VS and NAV for years before I was tempted by the freebies. One major annoyance of the biggies is that their performance hit on your computer can be fairly significant. Furthermore, after a period of free updates, you'll be forced to pay a pretty penny for an annual update subscription (at least $25/year). This is above and beyond the original cost of the software itself (about $50).
My choice of AV programs was also influenced by my role as an administrator of several dozen computers at a company I help out at. From a licensing standpoint, the freebies couldn't be used in a corporate environment anyway - you HAD to buy the commercial versions. This put the freebies into the same boat as the biggies - too big and too expensive.
So far, I haven't addressed the question of virus fighting effectiveness. Using the results of Virus Bulletin (https://www.virusbtn.com) as a guide, the freebies - in fact, their commercial versions! - tend to have less than stellar performance, at least historically. Among the biggies, NAV performs the best, but even it isn't the top performer.
Which finally brings me back to NOD. I've known about NOD for years before I started using it in earnest. Previously, the need to re-license it every year was a bitter pill to swallow, but my mind was changed when NOD officially came to Thailand a few years ago. The "Home" version is now available for a mere 250 Baht/year. This was a significant discount over the regular $40 buy-in and $27 renewal price for buying NOD through regular channels.
Why do I like NOD so much:
- It's very affordable. Granted, it's not free, but at 250 Baht/year, the home version costs as little as a 1/4 tank of gasoline. The corporate pricing is more expensive at approximately 1,000 Baht/year/computer, but it also includes the ability to do "local" updating (i.e. only one central computer downloads updates from the Internet - all other networked client computers update from this "mirror" at LAN speeds). With NAV, this feature is only available on the considerably more expensive Corporate Edition. But even NAV's standalone version is more expensive than NOD's corporate version.
- NOD is VERY light on your computer. The performance hit is very small, even though it defaults to updating its definitions HOURLY (you can change this if your Internet connection is slow).
- There is only one version of the program - no "lite" versus "full" versions. Granted, there are home and corporate versions - not to mention cross-platform and specialty server versions - but the virus fighting features are essentially the same. The corporate version simply adds hooks for local updating, remote administration, etc.
- Lastly, and most importantly, its virus fighting capabilities are top notch. (Note: When browsing Virus Bulletin's test results, I suggest you look at both current AND past results.) And I can vouch for this: in the past year, I've never had to lift a finger whenever a new virus scare arose. This is also why I install it on every computer I'm asked to fix/reinstall/build.
- If there is a downside to NOD, it's the fact that it's rather techie and filled with tons of configurable options. Its interface is hardly aimed at the casual user, but fortunately, it works pretty well "out of the box", in an "install and ignore" mode. Also, it's not really an anti-spyware tool, even though the latest version 2.5 has added some adware/spyware detection capabilities. Unfortunately, it's too early to tell how well this works.
10 July 2005
Post Database: Slow Modem DisconnectsIn the July 6, 2005 installment of HelpDesk, reader Monton wrote about delays in disconnecting from a modem-initiated Internet connection. No mention was made as to what version of Windows was being used, but if he's using Windows 2000 or XP, I have a few additional ideas.
IF he's also: a) experiencing slow logins (e.g. it takes a minute or more for the login to be authenticated after the modems handshake), and/or b) he's using a large "hosts" file, there are two things he can try.
First, replace the large hosts file with a much smaller one (e.g. the default hosts file that comes with Windows, that's less than 1KB in size). Second - and this is probably the better solution, especially if he intentionally chose to use a large hosts in the first place - disable the DNS Client service. To do this, run "services.msc" from Start > Run, look for an entry called "DNS Client", double-click it, click the "Stop" button, and then in the "Startup Type" field, choose "Disabled".
If this doesn't work or if other problems arise, revert the DNS Client service back to its original settings - and alas, wait for another solution from somewhere/someone else.
09 July 2005
Post Database: Internet Browsing from Public ComputersIn the June 22, 2005 installment of HelpDesk, George Masaoka asked whether there was any way to safely use computers in Internet cafés for such things as online banking or online buying. Clearly, this wouldn't be an issue IF one could be certain that Internet café computers were clean of viruses and/or spyware.
The best way to ensure this is to ignore the operating system installed on the Internet café computer and to boot/run/operate from a known/clean environment instead. There are a few ways to accomplish this, but by far the easiest involves the use of a "Live" (i.e. bootable) Linux CD. Even if the computer were infested with Windows viruses or spyware, the chances that the "Live" Linux operating environment would (or even could) become infected is next to zero.
A few caveats though. First, you may need to learn your way around Linux a bit. For this particular need though, it shouldn't be too difficult - just find the web browser (these days, probably Firefox) and run it. Second, if your target websites require Internet Explorer you're probably out of luck. In this day and age though, this should be less of a problem (all of the Internet banking/credit card websites I use work pretty well with Firefox). Third, the Internet café computer MUST have a CD drive, and it must be configured to boot from the CD device BEFORE the hard disk (granted, this isn't too unusual a setup). Lastly, and more problematical is the fact that booting from a Live CD will override any usage/time metering that may be running under Windows. As long as you can come to an amicable agreement with the Internet café proprietor on how to calculate your time used, this shouldn't be a problem either.
Two related matters. First, is it possible to create a Windows "Live" CD? The answer is: "perhaps". If the name "Bart Lagerweij" means anything to you, you'll probably suspect that it IS indeed possible - even though it would hardly be a simple undertaking and may not even be legal as per Microsoft's EULA. I suggest you ignore this possibility.
Second, is there any way to create a pristine "bubble" all the while running the (presumably Windows) operating system installed on the Internet café computer? In my opinion: probably not. Granted, you could reduce your risks by running a lesser-known web browser on a thumb drive, or you could use the likes of the "Anonymizer" service (https://www.anonymizer.com) to minimize the risks on the Internet side. But you would still be at the mercy of any malware that exists on the local computer. Short of installing anti-virus/anti-spyware software and then going through the cleansing process on every computer you intend to use (assuming you were even allowed to do so), this too is an untenable option.
Having said all that, if you frequent Internet café's, I would still recommend that you to look into the possibility of running your own web browser from a thumb drive. Chances are most Internet café computers aren't likely to have any web browser other than Internet Explorer. Worse still, said computers aren't likely to be fully patched against Windows or IE vulnerabilities. A better alternative would be to run a safer web browser on your own portable device. Introducing: "Portable Firefox". Portable Firefox is designed to run from writeable, removable storage device such as thumb or Zip drive. It's not an officially sanctioned/supported release of Firefox, but it IS based on official Firefox builds.
Portable Firefox was developed by John Haller and is available from his website (which unfortunately is currently inaccessible). Alternatively, it can be obtained from the Major Geeks website at: https://www.majorgeeks.com/download4424.html.
Final note: Portable Firefox is also excellent for use on school/university computers that may not have Firefox installed. It's also an awfully useful way of having your bookmarks handy wherever you go.
06 January 2005
Post Database: Firefox Questions ReduxI re-wrote and expanded upon yesterday's blog entry for submission as a letter to the Post Database. Here's how it reads:
"In the December 22 issue of HelpDesk, there were a few questions regarding Firefox. One, by Robert Legrand, questioned why Firefox's default starting page shows up in Thai and asked how it could be changed to English. The cause of this behaviour is that Firefox's starting page uses Google. Google checks your IP address to determine which country you're accessing it from and then "helpfully" provides web pages in the language of that country. Since reader Legrand is no doubt "calling" from Thailand, he gets a page with Thai text. This isn't a problem unique to Firefox though - Internet Explorer (IE) can manifest this problem as well when accessing Google.
I wrote up the fix to this problem in a blog entry (24-Nov-2004) on my website (/index.html). But to recap: go to https://www.google.come (which will redirect you to https://www.google.co.th) and then click the first "English" link you see (in the middle of the page). This will create a cookie that sets a language preference and will cause Firefox's default starting page to display in English only (plus displays of any subsequent Google pages), as long as the cookie remains.
A second query by "A Reader" wondered why Firefox seems to generate network activity every 15 seconds or so, while IE doesn't. I don't have the perfect answer to this question, except to say that based on packet analyses, Internet Explorer does in fact exhibit this behaviour as well, albeit to a lesser degree. As a matter of fact, with NO browser loaded, network activity still occurs. My complete response to this issue is rather technical and is best read in my 05-Jan-2004 blog entry. Suffice to say though, it is my GUESS that this is NOT anomalous behaviour as long as you're sure that adware or spyware isn't running on your system.
One final Firefox-related comment. Recently, Wanda Sloan wrote that IE tends to be more "forgiving" than Firefox. In some instances, this is true - IE will sometimes gloss over badly coded HTML and display it in a palatable way, while Firefox is much more of a stickler for standards and won't bend over backwards for you. Another dimension to this is the fact that IE - like Netscape before it - often runs roughshod over standards and unilaterally creates conventions that it expects the rest of the industry to slavishly follow (or not - they don't care). Unfortunately, the designers of Firefox often refuse to play along and you get websites specifically coded to Microsoft standards that display badly under other web browsers (not just Firefox).
That's why I continue to keep IE handy as a backup browser - and recommend that other people do so as well. Indeed, I need it to access EGV's new website, which now complains that my Firefox doesn't have Flash installed (excuse me, but it does). Of course, it doesn't complain when I use my Flash enabled IE. Go figure."
05 January 2005
Post Database: Firefox QuestionsIn the 22 December 2004 issue of Post Database's HelpDesk, there were a couple of questions regarding Firefox. One, by Robert Legrand, questioned why Firefox's default starting page shows up in Thai and asked how it could be changed to English. A "no brainer" way to fix this is to point Firefox's home page to another (presumably English) page. But of course, this just skirts the issue.
Looking a bit closer, I noticed that Firefox's starting page actually uses Google. The solution to this problem is actually identical to the Thai/Google problem that I described in 24-Nov-2004 blog entry. That is, go to https://www.google.com (which actually takes you to https://www.google.co.th) and then click the first "English" link you see. This will create a cookie that sets a language preference and will cause Firefox's default starting page to display in English only - as well as displays of other Google pages (as long as the cookie still exists).
A second query by "A Reader" wondered why Firefox seems to generate network activity every 15 seconds or so. Turning off automatic upgrade checks for both Firefox and Firefox extensions alleviates this problem to an extent. You do this by going to: Tools > Options > Advanced > Software Update and then unchecking the boxes in front of "Firefox" and "My Extensions and Themes".
But it doesn't eliminate all network activity. And contrary to what "A Reader" found, I discovered that both Firefox AND Internet Explorer (IE) tend to tickle the network at intervals. To take a closer look at what sort of packets were flying around, I installed a freeware packet sniffer called "Ethereal" to determine how many packets were being detected by my network card and what kind of packets they were.
For the record, I ran Ethereal with Firefox v1.0 and then with Internet Explorer v6 SP1 (i.e. separately), with automatic updating turned off and with both browsers pointing to a static HTML file on my hard disk. Ethereal reported that over a 15 minute period, the number of packets detected with Firefox running were indeed greater than for IE by a factor of about 50% (82 vs 52 packets). While 82 packets over a 15 minute period averages out to about one packet every 11 seconds, it should be noted that some packets would occur ANYWAY even if NO browser were loaded (61 packets were detected when no browsers were running at all).
The protocols of the packets found in both the Firefox and IE runs were similar (i.e. ARP, BROWSE, DNS, HTTP, LANMAN, NBNS, NBSS, SNMP, TCP) and nothing seemed amiss to my not very trained eyes. My GUESS therefore is that the network activity detected by "A Reader" isn't unusual, especially since IE manifests the same activity - although I still don't have an explanation why Firefox is more "chatty" than IE. Still, it should be emphasized that on a typical computer, numerous programs generate network activity for perfectly valid reasons. Of course, it is also possible (especially in these times), that adware or spyware are sending or retrieving information for less valid reasons as well.
15 November 2004
Article: Spyware Testing (Part 1)I've just written a full length article, that details results of some spyware testing. By the way, if the article sounds like I'm writing for someone else, you're right. I actually wrote this with an eye for a reprint in Post Database.
17 August 2001
Random Ramblings: Letter to Post DatabaseAlthough I haven't written anything in a while, I did pen this facetious little letter to the Post Database today.
11 July 2001
Post Database: Getting Rid of Pop-Up WindowsThere was a letter in Post Database from Phillip Ahrens today asking how to dispense with pop-up advertising windows. Lately on Yahoo!, there's this recurring ad for an X10 mini camera which is incredibly annoying. Under Windows at least, you can get rid of this with with the help of a program called "Pow!" from Analogx (https://www.analogx.com/contents/download/network/pow.htm). Pow! works by scanning the window title and automatically closes any window that matches your "kill" list. Another useful freebie from the good folks at AnalogX.
Actually, there are several other techniques to accomplish this as well. First, you can disable Active and/or Java scripting. Unfortunately, there are probably times when you DO want this to work, so odds are, you'll be toggling this on and off ad infinitum. With Internet Explorer, this is adjusted under "Internet Options", "Security", "Custom Level".
Second, if you know the address of the offending ad, you can always add a line to your HOSTS file (most Windows installations don't have this file, but there IS a HOST.SAM sample file you can use as a template) to point this site to 127.0.0.1, which happens to be the dummy or "loopback" IP address of your own computer. For more information on this technique, read this.