Where is that site?
There's a lot of talk right now about international trade. Everyone wants to know where the things they buy come from. But do you know where the Web pages you visit are really located? This tool can help you find out where you're actually connecting to when you visit a Web site.
Enter a domain name and press "Submit."
Important notes
The biggest lesson you can learn from this tool is that the Web is truly international. Few sites have just one physical location, and many depend on servers in more than one country, even to serve a single Web page. Without freedom of speech across international borders, the Web as we know it could not exist.
This tool uses best efforts to determine the locations associated with domains and IP addresses, based on information from public databases, but we cannot guarantee that the results are accurate. Do not rely on the results for any major decisions.
If a site blocks our scanner, or returns a Javascript redirect page that the scanner cannot decode to find the real front page of the site, then the results will only include partial information.
The "flag counts" numbers are, literally, the number of times each flag appears in the "detailed analysis" section. Not all of those appearances are equally significant. The location of the main Web server for a site may be much more relevant to where the site "really is," than for instance the vanity domain registration of a minor support server from which the site happens to load an image file. As a result, the flag counts may not really be very useful. They are provided just because they are interesting.
Many sites use Content Distribution Networks (CDNs) which cache pages in multiple servers around the world and send them to readers from whichever server is nearest. Data in CDNs will appear in this tool as coming from whichever CDN server answers the request from our testing server (which is located in Toronto); as a result, CDN-hosted data will often appear as coming from Canada or the United States even if that is not where the "owner" of the data may be located, and even if you will receive the same page from some other CDN server when you connect to the site yourself.
Some small or low-population countries allow foreign entities to register within their national TLD space in order to get a desirable domain name, despite the registrants not having any other connection to the countries in question. Common examples include Tuvalu (.tv), the British Indian Ocean Territories (.io), and Niue (.nu). These kinds of places tend to be disproportionately represented in the scan results.
Some notes on things found in the results:
- "national TLD": This means that the domain name ends in a standardized two-letter abbreviation like .ca (Canada) or .se (Sweden). These are recognized solely by examining the names (not by connecting to the sites) and so there can be results even for domains that do not actually exist.
- IP geolocation: any time an IP address comes up in the scan, if possible the tool will show its location, looked up in a local database
- "MX record": these records name the servers that will handle email addressed to the domain; often they are separate from the Web servers and located in different places
- "header links reference": domains mentioned in the HREF attribute of LINK elements in the head of the Web site code are listed this way. These are typically the servers that store style sheets and similar resources used by the Web site.
- "media sourced from": domains mentioned in the SRC attributes of various HTML tags; these include the servers for graphics that appear on the page, and those providing Javascript support (such as most advertising networks)
- "CDATA references": the scanner does not attempt to actually execute Javascript from the site being scanned, which would be dangerous and stupid, but many sites load important parts of themselves only through Javascript and would be missed if we trashed scripts entirely. So as a compromise, the scanner examines scripts (and anything else that parses as CDATA) without running the code, and looks for anything that resembles a domain name. This will sometimes give false positives on strings that resemble domain names but actually are not.
Results for each site are cached to prevent spamming the sites under test, but details of which incoming connections asked to scan which sites are not recorded for privacy reasons.
Please support North Coast Synthesis Ltd. by buying our synthesizer products. Without sales we cannot afford to provide public services like this one.
