When I went hopping across India two years ago, and connecting from erratic and unsecure cybercafes, I’d spend half of my connected time clearing private details¹ from those public machines. Some computers allowed me to install Firefox, while some didn’t. Some allowed me to plug my USB stick in—for using my portable browser off it, while some didn’t. In most instances, I’d have access only via an outdated default version of Internet Explorer—5 or 6 typically. I used cybercafes not just for quick connectivity, but also to check my email, and buy flight tickets for my next unscheduled destination.²

Private information left by candid users was scary. It was too easy, and too stupid to be true. And yet, there were these computers in cybercafes full of saved usernames, passwords, chat histories, telephone numbers, credit card numbers in browser history!

I lived in a paranoia for a month while on the move—hoping and praying that none of those machines I used were keylogged, or that I hadn’t forgotten to erase any of my private interaction data. And I kept checking my bank statements for any fraudulent transactions.

If you get my drift, then this incognito mode feature, especially in Google Chrome—together with its installation location³ is a step in the right direction.

1. Web addresses, usernames, passwords, and autocomplete texts, among others.
2. I was in a situation in which, I could not plan my travel, because it depended on responses from certain authorities I was dealing with.
3. It allows any user to install Google Chrome without requiring Admin intervention—even on Admin locked machines.

It used to be like this:

http://www.google.com/reader/view/

Now, it is like the following (line wraps marked » –Ed.):

http://www.google.com/reader/view/#stream/feed%2F »
http%3A%2F%2Fwww.slate.com%2Frss%2Ffeed.aspx%3Fid%3D2126999

If it is a tag group, then it would look like this:

http://www.google.com/reader/view/#stream/user »
%2F123456%2Flabel%2Fscientific¹

From the above URLs, it is now obvious, they paint a detailed picture of my reading sources and habits. Not sure this is a wonderful thing—particularly if you’re concerned about privacy, and if you’re accessing from behind the wall. The only way to avoid the log is to click on “All items.” That would look like this:

http://www.google.com/reader/view/#stream/user »
%2F123456%2Fstate%2Fcom.google%2Freading-list²

1. The funny characters in the URLs are a no-brainer: %3A is a colon; %2F is a forward slash, and so on; See more in Table 1, ISO 8957 character table.
2. Before you jump up and point out, don’t worry, I’ve dumbed the Reader assigned user ID in all the URLs.

Much like auto-discovering feeds², an Opensearch-standard compliant browser now auto-discovers the availability of search on this site.

Try loading this site in a supported³ browser. You’ll either notice a subtle search icon highlighted in the default browser theme—suggesting that there’s something you can add to your list of search plugins or more visibly via a “+” sign, if you use a theme like Winstripe. Pull the search menu down and you’ll notice this: Add “ckunte.com”.

Enabling this on any site is easy. You’ll need to add the following⁴ in your site’s header part of XHTML code. I’m using my (this) site’s code for example:

<link rel="search" type="application/opensearchdescription+xml" href="/ckuntecom.xml" title="ckunte.com" />

(I had previously explained the construction of Opensearch XML file—ckuntecom.xml, which enables search auto-discovery.)

As a user, for sites that do not offer this feature, you can always use OpensearchFox. But if you’re a site owner, it is probably a good idea to provide this option to your regular readers and referrers.

1. I am beginning to test Sneha’s patience now.
2. Via the orange feed icon on the right end of the address or location bar.
3. Available in Opensearch supported browsers—Firefox and Internet Explorer 7.
4. Change file location, file name and title as applicable.