ctrl.altered.mind

Having switched from a desktop to a notebook as my primary computing device (not considering the PS3), being conscious about the added vulnerability of personal data is now imperative. It would be terrible if the laptop was lost or stolen, of course, but there is now the more probable risk of accidentally damaging the portable machine (not that I am sloppy, but it is reassuring to know that I am protected from hard drive damage due to dropping the laptop, or from accidentally yanking it off the coffee table).

Some habits are really worth adopting in order to minimize the pain of going through data loss and/or recovery - even more so in the case of theft (fact: about 97% of stolen computers are never recovered). There are scores of data recovery services available, but they generally seem to take advantage of the customers’ cornered situations and charge a heavy premium for fetching data from a damaged laptop (often more than the cost of a new computer). Of course, there are some pre-emptive solutions available such as PCPhoneHome or LaptopLocate.net, which can help track down machines by monitoring their location. With the iAlertU freeware application, you can even add “car alarm” or security camera protection for an iSight-equipped Apple notebook.

The first and most vital thing to do for any computer user, is to use a backup storage device. The simplest form of this would be to connect an external hard drive. Hard drives with encryption features allow locking down of personal information as all data stored on the drive is inaccessible without a master password. While it is possible to back up data on optical media like DVDs, this is a cumbersome repetitive chore and requires a lot of sorting and organization for the backed up information to be easily accessible.

External drives can play well with automated backup software, thus making what would otherwise be a daily chore an invisible background task for the computer. My external hard drive is network-attached, which eliminates the need to constantly plug it into a laptop, and additionally allows moving all media files to the external drive so they are accessible from all shared computers on the home network. On the flip side, this does have the disadvantage of not being able to access your media outside the home (unless you open up the drive for internet access).

I often find myself minimizing the kind of personal data that is stored on the notebook’s hard drive. I typically use my Gmail account as a network archive as this makes my files retrievable from any computer with an internet connection (more on how to make this process more convenient in a future tutorial). This benefits both security and accessibility, and prevents losing years of data and pictures along with the laptop. There are new web services cropping up now, that recognize this activity and endeavor to help organize data tucked away in your email attachments. Xoopit is one such service that ties in with a Gmail account.

What is your favorite strategy?

Feel like you could use a good joke? How about one custom-crafted to suit your exact taste of humor?

A team at the University of California at Berkeley has been running a collaborative filtering system for recommending jokes. It’s called Jester, and is currently at version 4.0 in its evolution. The idea behind Jester is to recommend jokes to the user based on their ratings of previous jokes. The system can begin recommending anecdotes once the user rates a training set of 8 jokes. Apparently my tastes make for textbook cases of classification, for I found the jokes served up by Jester to be spot on hilarious, right after the initial set. A significant portion of the content in the database has been in wide circulation, so chances are you have read them before - the system might eventually be expanded to allow user submission of content. Content-based filtering, which is a form of collaborative filtering, has become one of the central mechanisms behind pioneering current-generation applications on the Web today.

Jester uses an algorithm called Eigentaste 5.0, which dynamically adapts the order in which items are recommended. The basic filtering algorithm uses universal queries to apply real-value user ratings on a common set of items. Those familiar with Principal Component Analysis (PCA) will recognize the techniques applied to the dense subset of the ratings matrix. PCA helps reduce multidimensional data sets to lower dimensions for analysis.

The dataset used by Jester has been made freely available for research use, while the framework is also being used to develop new predictive applications.

Finally, here’s one of the anecdotes served by Jester that had me smiling:

A group of managers were given the assignment to measure the height of a flagpole. So they go out to the flagpole with ladders and tape measures, and they’re falling off the ladders, dropping the tape measures–the whole thing is just a mess. An engineer comes along and sees what they’re trying to do, walks over, pulls the flagpole out of the ground, lays it flat, measures it from end to end, gives the measurement to one of the managers and walks away.

After the engineer has gone, one manager turns to another and laughs. “Isn’t that just like an engineer? We’re looking for the height and he gives us the length.”

American Express is one of the major global financial service providers and is best known for its credit card and traveler’s cheque businesses. Naturally, I was shocked to learn that the 74th largest company (Fortune 500, 2007) has such poor focus on enhancing user security for its financial account portal.

The AMEX website permits users to only set passwords with a maximum of 8 characters, with no non-alphanumeric characters supported. This is very strange for a corporation the size of American Express, and especially so for a financial services company. I figure that with these constraints, any kind of password set up on the AMEX site will only qualify as having “weak” strength.

However, I’m sure most people would not even encounter these constraints while selecting their passwords. In a study on password security conducted with undergraduate and graduate students, it was found that over half of reporting users (52.70%) never change their passwords if not required by the system. These numbers drop with increasing frequency, with about 12% users changing their password every three months.

Furthermore, some common practices have been observed to be followed by users when choosing a password. Most users only use a combination of lower case letters and numbers, where the number is usually a personally meaningful one (such as birth dates or phone numbers). Over half of all users also reported using the same password for multiple accounts (around 33% use some variation of the password).

With most websites (AMEX, for example) not enforcing stricter rules for generating passwords, user tendencies are not going to see much change. Passwords are generally considered a pretty basic form of security themselves, and therefore one should be very careful in selecting a password that effectively appears as random as possible to a potential intruder.

Password strength is the amount of security that a password can provide against password-guessing attacks, and is measured in bits of entropy.

Common guidelines for choosing good passwords, as listed on Wikipedia are:

* Include numbers, punctuation, and upper and lower case letters
* Use passwords with at least 8 characters
* Avoid passwords based on repetition, dictionary words, letter or number sequences, usernames, or biographical information like names or dates.

Wikipedia also lists some examples of weak and strong passwords (which are probably already included in password-cracking databases, so please don’t choose one of them for your own).

General alertness is also of extreme importance while using online portals, as even the strongest passwords are of no use when it comes to protecting users against certain forms of attacks like phishing or keystroke logging.

I just signed up for Jaman, a service that brings people access to movies from around the world, including Bollywood. The website attempts to build an online community around its large library of independent and international feature films and documentaries. Jaman was founded and is currently led by Gaurav Dhillon, the former CEO of Informatica Corporation - a company that provides data integration and delivery services. The name Jaman has been derived from the Hindi word Jamun, a tree and the purple plum fruit of the same name that it bears.

My first impression of Jaman’s website has been very favorable - it sports a polished, yet lean interface that is careful not to over-indulge in cliched Web 2.0 design elements. The movie catalog is easy to browse and is neatly categorized by genre, geographical regions, and ratings. Jaman also has a neat movie discovery feature that allows you to tune your preferences in order to find a movie to watch - complete with an ‘I’m Feeling Lucky’ button, as made famous by Google.

Jaman is however, primarily a movie rental service that employs its own DRM and P2P network platform that enables users to rent and watch movies in high definition on their Mac OS X or Windows computers. While the site is free to browse, the Jaman player needs to be installed in order to watch licensed movie titles (which come in at $1.99 for a 7-day rental, or to purchase for $4.99). New users get three free ‘movie tickets’ when they sign up, which can be applied towards rentals. You also get two free tickets for every friend that you introduce to Jaman (so please do let me know in the comments if you intend to sign up after reading this, and I’ll email you an invite :P)

The Jaman team has also developed an unofficial AppleTV plugin that lets you stream content from your computer to the TV (if you own an AppleTV, of course). While the AppleTV has received a tepid response thus far, that scene has been predicted to change come January, once Apple announces their much-anticipated movie rental service via iTunes.

I previously mentioned that Jaman has built a social network around its media collection - this network enables users to get in touch with other movie buffs who may share common interests, fans, as well as filmmakers. People can also submit and share their original professional-quality content for distribution via this channel.

I have to mention Bollywood too - while Indian filmmakers and movie studios are waking up to the command that Internet audiences have in terms of bringing in revenue, Jaman is perhaps the first to offer audiences access to (a few) Bollywood movies on the Web. Zeroing in on the Bollywood section on Jaman, I see about 48 movies currently available - these range from the 70s (like Sharmilee) to contemporary releases (like Black and Paheli). Hopefully, this section will grow over time as more people start using the service.

Unlike other free and commercial offerings on the Internet for video consumption (such as Babelgum, Hulu, Joost, and a slew of others), Jaman is focused on providing a Web-based alternative to the Hollywood-dominated mainstream, and simultaneously improve the meager percentage of world cinema that is available to movie enthusiasts in the US.

I think I shall use my first free Jaman movie ticket to watch Inside Iraq: The Untold Stories, a film by citizen journalist Mike Shiley.

Joost is an exciting new platform that has the potential to revolutionize the way TV shows and other video content is distributed over networks. If a few more popular and informative TV networks join the system, I may never have to subscribe to cable television ever again.

If you haven’t the first clue about Joost, it’s a P2PTV-based program (by the founders of Skype and Kazaa) that lets you watch near-TV resolution video content over the internet, converting a computer into an instant on-demand TV set (without a set top box and its tight storage constraints).

Oh, and the program is still in beta testing, under the title ‘Joost for Friends’. As a beta tester, I do have a few invites, so if you’d like one, feel free to leave your name and email address in the comments. Don’t worry, I won’t send you spam. You’re not that important. Just kidding!

If there is space, it needs to be mapped. Everyone needs direction at some point and in some form, and needless to say, a good variety of tools exists as far as geographical mapping needs need to be satisfied. As an orientationally challenged individual, I find myself resorting to online mapping services much more than the average person, although I’m sure no one would complain as these services (online and otherwise) are constantly improved upon and expanded. While Mapquest has existed for a long time, the online maps sector was just evolving linearly until Google stepped in with a revolutionary interface and provided the thrust that has boosted efforts tremendously in this area. Everyone, be it Microsoft, Yahoo, and a slew of others, are investing heavily in the field.

While Google added most desirable features like fluid drag and zoom abilities to their map interface, they also went a step further and enhanced maps with ideas like satellite and hybrid views. The innovation does not stop here, of course. Microsoft’s Live Search has added 3D capability to their online map feature, which gets lets you view regions at a fly-by angle, eerily close enough to be able to accurately identify the make and model of each car on any road in the US (if not the license plate, not yet). Live Search has definitely seen a great deal of innovation in its interface since its launch around September this year, and the current interface is slick and functional without compromising on ease of use.

The most important benefit that has yet to be tapped, however, is the ability to map just about any spatial expanse and region, not just sections of the Earth’s surface. For example, real-estate businesses might be interested in online solutions to create highly interactive maps of their houses and locales for sale. Neurologists would like an easy way to create a map of a section of the brain. K-12 School teachers and educationists most definitely would like to create maps too, such as one for a tree or the human body. I was pleased to hear about a free, community-based service called MapLib.net (thanks, rs6) that lets anyone create an online, interactive custom map based on Google’s mapping API.

According to MapLib.net’s introduction:

” [the service] turns any image you uploaded as large as 6000*6000 into a custom Google Map and makes it zoomable & draggable with fantastic viewing experience, in really simple two steps. You can make annotations for it by adding markers, as well as embed it in your own web pages or blog.”

MapLib.net is a bilingual website currently operating in English and Chinese. It is currently in beta, but I definitely would like to see it being polished and upgraded soon. This would be a very good candidate for Google to incorporate under their flag.

Ordinarily, the thought of asking my mother to take a few pictures using the digital camera, zipping them into an archive, firing up a mail client and emailing the file over, would have been followed by the visualization of a long-drawn tutorial phone conversation that perhaps would never see the said task list brought to fruition. Don’t get me wrong, my mom is as enthusiastic as can be when asked to send me some pictures over the internet, but when the process involves anything more than switching three applications and a dozen clicks of the mouse, I know my mom’s no longer part of the audience.

To a geek like me, the process of say, setting up a simple web server is no big deal. I’ve done it many times before, on Windows of course. You begin with installing the latest apache httpd web server… oh wait, did I lose you as well? So you see the point. Every person who owns a computer would like to see it being used for so many applications, that are simply beyond reach - a majority of people end up purchasing computing power that vastly exceeds their needs. If only it were a bit more simple.

I got my mom to set up a web server yesterday. Yes, she is a web administrator now - it just took a couple of clicks, thanks to the Mac. I’ve always admired the power in the simplicity of Mac OS X, but today I’m a much bigger fan - for they’ve bundled an apache web server distribution with the OS.

All you need to do on a Mac, is this:

1. Pull up System Preferences (via the dock or the Apple menu (top left))
2. Click on Sharing, under Internet and Network
3. Check the Personal Web Sharing item under Services
4. Importantly, check the Firewall tab and make sure the built-in firewall is turned on - and that Personal Web Sharing option is checked
5. Close the System Preferences window

Thats it. You’re done. Your Mac is now running a personal web server. To check if its working correctly, fire up a web browser window and point it to the URL mentioned when you select the Personal Web Sharing option under System Preferences, which looks like this (see screenshot below):

View this computer’s website at http://192.168.1.2/ or your personal website at http://192.168.1.2/~aidoor/

You can even type in http://localhost/ at the same machine to check. To access your web server from any other machine on the internet, you will need the ISP-assigned IP address similar to the one in the note above. The easiest way to find your IP address is to visit http://www.whatismyip.com.

If things are set up correctly, your browser should bring up the default Apache index page. You can now build a website by designing and placing html files in a specific folder - under your main drive, select Library > WebServer > Documents

Each user account on your Mac also gets a personal folder for web sharing as well - for this you need to put your html files in the Sites folder located in the user’s Home folder. The URL for each user’s web account looks like http://your.IP.addr.ess/~username

The next step would be to register for a domain name using one of the many registry services on the internet (I use GoDaddy), so you can have a nice name to give out to people, instead of a 12 digit number.

Unbelievably simple. My mom loves it.

(Screenshot of System Preferences window)

Feedmap.net is a new blog-related service that brings a geographical perspective to blogs, and is an addition to the vast array of web mashups that seek to add a mapping overlay to just about every online service conceivable - the most useful of them (in my opinion) probably being HousingMaps (which marries Google Maps to craigslist.org real-estate and housing ads). Strictly speaking, Feedmap is not a mashup as it does not really produce a confluence of services, but it gets pretty darn close, given that most mashups do use Google Maps as their main ingredient. ProgrammableWeb.com, which tracks mashups, has an interesting grid view of the hybrid services available on the Web today.

Coming back to Feedmap (also called Blogmap), the service lets bloggers create buttons for their sites to indicate neighboring blogs and feeds available on the internet (although I’m not quite sold on how much the popularity of blogging in a neighborhood would matter). Also, the service would find popularity with bloggers who would not mind revealing their geographical location in their blogs, which I would guess makes up a smaller part of the web-trotting community. Most bloggers use their online journals as an outlet for the escapist in their minds - adding the geographical overlay only pulls them back to physical reality. Nonetheless, the service would be useful enough for the rest of us (myself included) that do list our identities on the web. It would be interesting, after all, to learn of the blogs that get published around us and from specific parts of the world.

From the Blogmap About page:

Here is a list of things you can do with BlogMap:

* Geo-code your blog feed using the submit page (and get your own BlogMap badge).
* Browse blogs by location using the browse page.
* Search for local blogs using the search page.
* Find bloggers in your neighborhood!
* Get local BlogRoll in OPML format.